Advances in Child Development and Behavior Volume 31

ADVANCES IN CHILD DEVELOPMENT AND BEHAVIOR Volume 31 Contributors to This Volume David A. Beaulieu Megan M. McClell...

Author: Robert V. Kail

41 downloads 1424 Views 4MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

ADVANCES IN CHILD DEVELOPMENT AND BEHAVIOR

Volume 31

Contributors to This Volume David A. Beaulieu

Megan M. McClelland

Daphne Blunt Bugental

Frederick J. Morrison

Gedeon O. Dea´k

Lenna L. Ontai

William J. Friedman

Kevin A. Pelphrey

Maureen Kessenich

J. Steven Reznick

Brett Kessler

Ross A. Thompson

Gary W. Ladd

Rebecca Treiman

Deborah J. Laible

Theodore D. Wachs

Lynn S. Liben

ADVANCES IN CHILD DEVELOPMENT AND BEHAVIOR

edited by Robert V. Kail Department of Psychological Sciences Purdue University West Lafayette IN 47907 USA

Volume 31

Amsterdam Boston Heidelberg London New York Oxford Paris San Diego San Francisco Singapore Sydney Tokyo

This book is printed on acid-free paper.

Copyright ß 2003, Elsevier Science (USA). All Rights Reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the Publisher. The appearance of the code at the bottom of the first page of a chapter in this book indicates the Publisher’s consent that copies of the chapter may be made for personal or internal use of specific clients. This consent is given on the condition, however, that the copier pay the stated per copy fee through the Copyright Clearance Center, Inc. (222 Rosewood Drive, Danvers, Massachusetts 01923), for copying beyond that permitted by Sections 107 or 108 of the U.S. Copyright Law. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Copy fees for pre-2003 chapters are as shown on the title pages. If no fee code appears on the title page, the copy fee is the same as for current chapters. 0065-2407/2003 $35.00 Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone: (+44) 1865 843830, fax: (+44) 1865 853333, e-mail: [email protected]. You may also complete your request on-line via the Elsevier Science homepage (http://elsevier.com), by selecting ‘‘Customer Support’’ and then ‘‘Obtaining Permissions.’’

Academic Press An Elsevier Science Imprint. 525 B Street, Suite 1900, San Diego, California 92101-4495, USA http://www.academicpress.com

Academic Press 84 Theobald’s Road, London WC1X 8RR, UK http://www.academicpress.com International Standard Book Number: 0-12-009731-1 PRINTED IN THE UNITED STATES OF AMERICA 03 04 05 06 07 08 9 8 7 6 5 4 3 2 1

Contents Contributors .............................................................................................

ix

Preface .....................................................................................................

xi

Beyond Point And Shoot: Children’s Developing Understanding of Photographs as Spatial and Expressive Representations LYNN S. LIBEN I. A Photographic Vision ............................................................................ II. Beneath the Photographic Surface ............................................................. III. Empirical Investigations of Photographic Development ................................. IV. The Larger Developmental Picture ............................................................ References ............................................................................................

1 3 12 33 39

Probing the Adaptive Significance of Children’s Behavior and Relationships in the School Context: A Child by Environment Perspective GARY W. LADD I. Introduction .......................................................................................... II. Primary Premises and Findings from Research Guided by ‘‘Main Effects’’ Models ........................................................................... III. Primary Premises and Findings from Research Guided by ‘‘Child by Environment’’ Models .............................................................. IV. Summary and Conclusions....................................................................... References ............................................................................................

44 47 73 84 95

The Role of Letter Names in the Acquisition of Literacy REBECCA TREIMAN AND BRETT KESSLER I. Introduction .......................................................................................... II. Letter Names in English and Other Alphabetic Writing Systems ..................................................................... III. Children’s Learning of Letter Names ......................................................... IV. Children’s Learning of Letter Sounds ........................................................

v

105 108 113 119

vi V. VI. VII.

Contents Role of Letter Names in Learning to Read Words ....................................... Role of Letter Names in Learning to Spell Words........................................ Conclusions .......................................................................................... References ............................................................................................

123 129 132 134

Early Understandings of Emotion, Morality, and Self: Developing a Working Model ROSS A. THOMPSON, DEBORAH J. LAIBLE AND LENNA L. ONTAI I. Internal Working Models and Relationships ............................................... II. Internal Working Models and Cognitive Growth ......................................... III. Developing a Working Model .................................................................. IV. Conclusion: The Impact of Relationships on Emotion, Morality—and the Self ............................................................................ References ............................................................................................

139 141 150 165 168

Working Memory in Infancy KEVIN A. PELPHREY AND J. STEVEN REZNICK I. Introduction .......................................................................................... II. Toward a Definition of WM in Infants ...................................................... III. Research on WM in Infants ..................................................................... IV. Frontiers of Research on WM in Infants .................................................... References ............................................................................................

173 175 182 202 217

The Development of a Differentiated Sense of the Past and the Future WILLIAM J. FRIEDMAN I. The Past and Future in Children’s and Parents’ Language............................. II. Differentiation of the Past and of the Future .............................................. III. Differentiation Between the Past and the Future .......................................... IV. Conclusions and Directions for Future Research ......................................... References ............................................................................................

232 235 253 261 266

The Development of Cognitive Flexibility and Language Abilities GEDEON O. DEA´K I. What is Flexible Cognition? ..................................................................... II. Developing Toward . . . ? Adults’ Flexible Cognitive Processing of Meanings and Messages ....................................................... III. Toward a Model of Flexible Representation in Language Processing ...............

273 279 283

vii

Contents IV. V.

Children’s Flexible Thinking About Meanings and Messages ......................... Questions and Conclusions ...................................................................... References ............................................................................................

285 316 320

A Bio-Social-Cognitive Approach to Understanding and Promoting The Outcomes of Children with Medical and Physical Disorders DAPHNE BLUNT BUGENTAL AND DAVID A. BEAULIEU I. Introduction .......................................................................................... II. Theoretical Models Employed to Understand the Experiences of Children with Medical and Physical Disorders ......................................... III. Optimizing the Outcomes of Children with Medical and Physical Disorders ..... IV. Integration ............................................................................................ References ............................................................................................

329 330 349 355 356

Expanding Our View of Context: The Bio-ecological Environment and Development THEODORE D. WACHS I. Starting with the Psychosocial Environment ................................................ II. Going Beyond the Psychosocial: The Bio-ecological Environment ................... III. The Bio-ecological Environment and Children’s Development ........................ IV. Bringing it All Together: A Converging Structural and Conceptual Framework ..................................................................... V. Conclusions .......................................................................................... References ............................................................................................

365 368 371 388 394 395

Pathways to Early Literacy: The Complex Interplay of Child, Family, and Sociocultural Factors MEGAN M. MCCLELLAND, MAUREEN KESSENICH, AND FREDERICK J. MORRISON I. Introduction .......................................................................................... II. Child Factors and Early Literacy Development ........................................... III. Parenting and Early Literacy Development ................................................. IV. Sociocultural Factors and Early Literacy Development ................................. V. Schooling Influences and Children’s Early Literacy Development.................... VI. Dynamic Relations Between Child, Family, & Sociocultural Factors, Schooling Influences and Early Literacy Skills ............................................. VII. Conclusion ............................................................................................ References ............................................................................................

431 439 440

Author Index ........................................................................................

449

Subject Index ........................................................................................

473

Contents of Previous Volumes ..................................................................

489

412 416 420 426 430

This Page Intentionally Left Blank

Contributors Numbers in parentheses indicate the pages on which the authors’ contributions begin.

DAVID A. BEAULIEU

Department of Psychology, University of California, Santa Barbara, Santa Barbara, California 93106 (329) DAPHNE BLUNT BUGENTAL

Department of Psychology, University of California, Santa Barbara, Santa Barbara, California 93106 (329) GEDEON O. DEA´K

Department of Cognitive Science, University of California, San Diego, 9500 Gilman Dr., La Jolla, CA 92093-0515 (271) WILLIAM J. FRIEDMAN

Department of Psychology, Oberlin College, Oberlin, Ohio 44074 (229) MAUREEN KESSENICH

Department of Pediatrics, Perinatal Center/Neonatal Developmental Follow-Up Clinic, Loyola University Medical Center, Maywood, Illinois 60153 (411) BRETT KESSLER

Department of Psychology, Washington University, St. Louis, Missouri 63130 (105) GARY W. LADD

Department of Psychology and Department of Family and Human Development, Arizona State University, Tempe, Arizona 87287 (43) DEBORAH J. LAIBLE

Department of Psychology, Southern Methodist University, Dallas, Texas 75275 (137) LYNN S. LIBEN

Department of Psychology, The Pennsylvania State University, University Park, PA 16802 (1) MEGAN M. MCCLELLAND

Department of Human Development and Family Sciences, Oregon State University, Corvallis, Oregon 97330 (411) FREDERICK J. MORRISON

Department of Psychology, University of Michigan, Ann Arbor, Michigan 48109 (411)

ix

x

Contributors

LENNA L. ONTAI

Department of Human and Community Development, University of California, Davis, California 95616 (137) KEVIN A. PELPHREY

Department of Psychiatry, University of North Carolina School of Medicine, Chapel Hill, North Carolina 27599 (173) J. STEVEN REZNICK

Department of Psychology, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599 (173) ROSS A. THOMPSON

Department of Psychology, University of Nebraska, Lincoln, Nebraska 68588 (137) REBECCA TREIMAN

Department of Psychology, Washington University, St. Louis, Missouri 63130 (105) THEODORE D. WACHS

Department of Psychological Sciences, Purdue University, West Lafayette, Indiana 47907 (363)

Preface The amount of research and theoretical discussion in the field of child development and behavior is so vast that researchers, instructors, and students are confronted with a formidable task in keeping abreast of new developments within their areas of specialization through the use of primary sources, as well as being knowledgeable in areas peripheral to their primary focus of interest. Moreover, journal space is often simply too limited to permit publication of more speculative kinds of analyses that might spark expanded interest in a problem area or stimulate new modes of attack on a problem. The serial publication Advances in Child Development and Behavior is intended to ease the burden by providing scholarly technical articles serving as reference material and by providing a place for publication of scholarly speculation. In these critical reviews, recent advances in the field are summarized and integrated, complexities are exposed, and fresh viewpoints are offered. These reviews should be useful not only to the expert in the area but also to the general reader. No attempt is made to organize each volume around a particular theme or topic. Manuscripts are solicited from investigators conducting programmatic work on problems of current and significant interest. The editor often encourages the preparation of critical syntheses dealing intensively with topics of relatively narrow scope but of considerable potential interest to the scientific community. Contributors are encouraged to criticize, integrate, and stimulate, but always within a framework of high scholarship. Although appearance in the volumes is ordinarily by invitation, unsolicited manuscripts will be accepted for review. All papers—whether invited or submitted—receive careful editorial scrutiny. Invited papers are automatically accepted for publication in principle, but usually require revision before final acceptance. Submitted papers receive the same treatment except that they are not automatically accepted for publication even in principle, and may be rejected. I acknowledge with gratitude the aid of my home institution, Purdue University, which generously provided time and facilities for the preparation of this volume. I also thank Thomas J. Berndt and Linda Smith for their editorial assistance. Robert V. Kail

xi


BEYOND POINT AND SHOOT: CHILDREN’S DEVELOPING UNDERSTANDING OF PHOTOGRAPHS AS SPATIAL AND EXPRESSIVE REPRESENTATIONS

Lynn S. Liben DEPARTMENT OF PSYCHOLOGY THE PENNSYLVANIA STATE UNIVERSITY UNIVERSITY PARK, PA 16802

I. A PHOTOGRAPHIC VISION II. BENEATH THE PHOTOGRAPHIC SURFACE A. PHOTOGRAPHS AS REPRESENTATIONS B. PHOTOGRAPHS AS SPATIAL REPRESENTATIONS C. PHOTOGRAPHS AS MEDIA-SPECIFIC REPRESENTATIONS D. PHOTOGRAPHS AS COMMUNICATIVE REPRESENTATIONS E. FROM ANALYTIC ARGUMENT TO EMPIRICAL EVIDENCE III. EMPIRICAL INVESTIGATIONS OF PHOTOGRAPHIC DEVELOPMENT A. CONTRASTING PHOTOGRAPHIC PAIRS B. REPRODUCING PHOTOGRAPHIC MODELS C. TRANSLATING VERBAL DESCRIPTIONS TO PHOTOGRAPHIC IMAGES D. PROVIDING AND EXPLAINING PHOTOGRAPHIC PREFERENCES E. EXTENDING EMPIRICAL WORK IV. THE LARGER DEVELOPMENTAL PICTURE A. COMING TO UNDERSTAND REFERENTIAL MEANING AND REPRESENTATIONAL QUALITIES B. EXPERIENTIAL FACTORS IN PHOTOGRAPHIC UNDERSTANDING C. THE ROLE OF CONSTRUCTIVE PROCESSES D. A DEVELOPING PICTURE REFERENCES

I. A Photographic Vision Take a moment to look up from the words on this page and glance around the room. Given the time and distance that divide my experience

ADVANCES IN CHILD DEVELOPMENT AND BEHAVIOR, VOL. 31

1

Copyright 2003, Elsevier Science (USA). All rights reserved. 0065-2407/03 $35.00

2

Lynn S. Liben

of writing these words from your experience of reading them, I cannot, of course, know with certainty what you will see. But there is a high probability that you can see an open book on the surface of a desk; that you can look through a window out toward some urban or country vista; that you can see a collection of office equipment such as chairs and bookcases, pencils, piles of paper, computer keyboard and monitor. What you see depends in part on what is truly ‘‘there,’’ on the perceptual endowment of our species in general (e.g., the human eye, optic nerve), on your perceptual endowment in particular (e.g., your own visual acuity), and on the collection of life experiences that enables you to interpret virtually instantaneously the various objects that surround you. Importantly, though, what you see also depends on the direction in which you look, and on the particular object(s) on which you focus your attention. Thus, for example, you have very different visual experiences when you look at the window (e.g., focusing on the window sill or window panes) versus when you look through the window (e.g., focusing on a tree or a car in the street below). Now take a moment to look at Figure 1. Assuming that this photograph was made using a film (rather than a digital) camera and was not later manipulated with computer technology, it, too, provides information about what was ‘‘there’’ at the time the film was exposed. Again, what you see when you look at this image depends in part on your perceptual endowment and life experiences. But the vantage point and focus are no longer under your own control. The photographer, Kenneth Josephson, has fixed these permanently for you and for all other viewers. You may chose not to look at the photograph at all, and of course you may view the paper from different angles, but nothing you can do will change the fact that the image itself

Fig. 1. Chicago, 1988. Photograph by Kenneth Josephson, in Wolf, 1999, p. 156. Reproduced by permission of the photographer. ß Kenneth Josephson.

Beyond Point and Shoot

3

allows you to see a particular book from a particular distance and a particular direction and from a particular viewing angle, and that the movement that occurred at the time the exposure was made appears as it does because of the shutter speed that was used. You cannot—by changing your head position or by accommodating the lens of your eye—decide to look at the book’s pages face-on, or to examine the writing on the binding, or erase the evidence of the pages’ movement. Furthermore, it is unlikely that you will be able to ignore entirely the aesthetic impact of the book’s contours and textures, an impact that you probably would have overlooked had you simply glanced up while the pages of the book were in motion. Photographs provide a fascinating and surprisingly understudied form of graphic representation in psychology in general, and in developmental psychology in particular. My goal in this chapter is to draw attention to several components of photographs that raise—and may be exploited to address—fundamental questions about both cognitive and social development. To do so, in the next section of the chapter, Beneath the Photographic Surface, I identify several ways in which photographs are more than simple depictions of their referents, offering illustrative developmental questions relevant to each. In the following section, Empirical Investigations of Photographic Development, I illustrate empirical approaches to these questions by describing a selection of research on photography that my colleagues and I have been conducting. This research allows interesting conclusions about age-related and experience-related components of photographic mastery, raises many additional questions about the roles of ontogenetic development and social experience, and offers methodologies that may be useful for answering some of these questions in future research. In the final section, The Larger Developmental Picture, I place the conceptual and empirical work on photography within a broader model of children’s developing understanding of spatial-graphic representations.

II. Beneath the Photographic Surface At the level of anecdotal observation, most people seem to assume that photographs are very simple kinds of representations. One reason for this assumption may be the ease with which photographs can be made. The availability of Instamatic, Polaroid, point-and-shoot, and even disposable cameras makes the process of creating photographic images a simple one. However, just as an ability to speak single words is not the pinnacle of language development, so too, the ability to produce an image with a camera is not the zenith of photographic mastery. In this section of the chapter, I consider developmental questions relevant to four aspects of

4

Lynn S. Liben

photographs that go beyond their denotative meaning: photographs as (a) representations, (b) spatial representations, (c) media-specific representations, and (d) communicative representations. A. PHOTOGRAPHS AS REPRESENTATIONS

An illustration of one of the most interesting features of photographs is provided, with apologies to Magritte, in Figure 2. What is, of course, amusing about this image is that it makes salient the fact that a photograph of a pipe (like Magritte’s painting of a pipe) is not identical to its referent. That is, it forces the viewer to confront the representational surface of the image, rather than processing it simply as a pipe, so that the viewer processes Figure 2 as a photograph of a pipe. The degree to which viewers are cognizant of the representational nature of graphics probably varies considerably across different media, in part driven by the diversity of experiences people have with different kinds of graphics. In the realm of maps, for example, people in the United States typically encounter only a very limited sample. They are especially likely to see small scale political world maps, centered on North America, oriented with north at the top, and rendered in a Mercator projection. In this light, it is not surprising that many people develop the view that maps are ‘‘accurate’’ cartographic renditions of Earth, and that maps that do not have these prototypical qualities are ‘‘distorted’’ (Downs, 1981; Liben, 2001; Liben & Downs, 1989). In contrast, in the realm of painting and drawing, viewers typically encounter a far greater assortment of representations. In some cases, the

Fig. 2. Photograph after Magritte’s Ceci n’est pas une pipe.


5

subject matter itself may be banal. For example, the vases and grapes that are depicted in still-life paintings are unlikely to strike viewers as inherently interesting, novel, or informative. As a result, viewers are led to look elsewhere for paintings’ raison d’eˆtre. Even when the subject matter itself is inherently interesting, viewers may be led to attend to the representational surface because of the dramatically different surfaces they encounter. For example, seeing both Goya’s Disasters of War and Picasso’s Guernica is likely to lead viewers to infer that paintings and drawings are ‘‘about’’ something more than the referents they depict. In the realm of photography, most peoples’ experiences probably center on snapshots of social gatherings or vacations, or on photographs in newspapers. In these contexts, photographs are aimed at archiving or communicating information about people, places, or events. For example, having photographed her former roommates at a college reunion, Martha hands the camera to Karen for the next shot so that the final collection of reunion photographs will include them all. Or, having taken a photograph of Pam in front of the Eiffel Tower, Nick is ready to move on to the Arc de Triomphe. When multiple photographs are taken in these situations, it is typically to ensure that all critical people or objects have been included (completeness), or to guard against a malfunction of the camera or photographer (redundancy), rather than to achieve diversity in representational surfaces. In addition to differences among media with respect to the experiences people have with them, there are also differences among media with respect to the perceptual similarity between the referent and the representation. In an earlier paper (Liben, 1999), I suggested that the representational nature of a graphic becomes increasingly salient to viewers as the graphic becomes perceptually less like its referent. For example, a highly abstract rendition of a human body such as Duchamp’s cubist Nude Descending a Staircase is likely to draw more attention to the representational surface than is a canonical photograph of a human nude. Thus, although becoming aware of the surface of any representation may be challenging, photographs may be particularly challenging because of the high level of perceptual similarity between the referent and photographic image. Consistent with this idea is Beilin’s (1991) report that some young preschool children impute qualities of referents to their photographic images, as when children assert that a photograph of an ice cream cone is itself cold. As adults in contemporary Western society, we come to recognize that despite the many similarities between photographs and their referents, the former do not in actuality reproduce the latter. (As implied by the cultural and temporal modifier, this depiction of mature understanding should not be taken as universal. Thus, the current chapter addresses the development of photographic

6

Lynn S. Liben

understanding within a particular, rather than a universal, context. Crosscultural and historical contrasts also offer fascinating questions for study; for examples, see Deregowski, 1968; Igou, 2001; Krippner, 1973; Price & Wells, 1997; Turner, 1987). In short, one important aspect of photographs is that they are representations of referents, not the referents themselves. From a developmental perspective, coming to understand photographs—like coming to understand other kinds of representations—is therefore a complex developmental process that includes not only the ability to identify the referential content of the image (e.g., that ‘‘it is a pipe’’), but also the ability to appreciate the symbolic nature of the representation and that its surface entails more than denotative information about the referent. B. PHOTOGRAPHS AS SPATIAL REPRESENTATIONS

The photograph shown in Figure 2 also illustrates a second important quality of photographs: Any given photograph depicts its three-dimensional referent from a particular distance, viewing angle, and viewing azimuth. These three concepts and their impact on a given graphic representation have been discussed most explicitly with respect to the ‘‘cartographic eye’’ (Downs, 1981). First, any map selects a particular viewing distance, affecting the scale of the map. For example, compare a large-scale map of a garden landscape versus a small-scale map showing the world’s major rivers. Second, it uses a particular viewing angle. For example, compare the overhead or vertical view used on most road maps versus the oblique view often used on city tourist maps that show local attractions with perspective line drawings. Third, it fixes a particular viewing azimuth or direction. For example, compare a contemporary map of Africa with north at the top of the page to one from the 1500s, with south at the top, a direction more consistent with the perspective of explorers traveling toward Africa from Europe (see Turnbull, 1989/1993). The spatial elements of the cartographic eye translate directly to the spatial elements of what might be called, analogously, the photographic eye. The particular combination of these spatial elements in any given photograph creates a viewpoint-specific image. The photograph in Figure 2, for example, may be described as using a short viewing distance (it is a ‘‘close up’’ shot), an elevation or eye-level viewing angle (the pipe is seen at an eye-level angle rather than, say, from directly overhead or from an oblique angle), and an azimuth that faces toward a side of the pipe (rather than, say, facing the pipe bulb or stem). From a developmental perspective, a number of questions concerning the view-specific nature of photographs are of interest. At the broadest level,


7

one may ask whether children even discriminate among photographs that vary in vantage point when the depicted referent is the same ‘‘thing.’’ Other questions concern children’s abilities to interpret particular spatial qualities of the photograph. If, for example, viewing distance is great enough to render the depicted referent as small, does the viewer presume that the referent itself is small? Can children use depicted sizes of known objects to help them interpret sizes or even the identity of other depicted objects? Can they infer spatial relations among the depicted referents? Do individuals come to understand the relation between the photographer’s actions in the referential space and the resulting photograph, and if so, when and how? From the perspective of comprehension, can children infer what actions were taken by the photographer to create a particular view-specific image? From the perspective of production, can children control their own actions to create intended view-specific images? C. PHOTOGRAPHS AS MEDIA-SPECIFIC REPRESENTATIONS

As implied by the earlier discussion of photographs in the context of other media, photographs share some qualities with other kinds of twodimensional graphic representations. For example, through framing and focus, all static two-dimensional graphic representations fix what is available in the representation for the viewer to see. In addition, some qualities are media-specific. In the case of painting, for example, the qualities of the image depend on the particular type of paint (e.g., water color, tempera, acrylics, oil paint), the tool with which it is applied (e.g., a pallette knife, fine or broad brush, spray can), the surface upon which it is placed (e.g., wood, paper, canvas), and so on. In the case of photography, the qualities of the image depend not only on the qualities of the referent and on where the camera is positioned, but also on movement of the camera (panning or still), the kind of lens, shutter speed, aperture, the type of film, the presence or absence of polarizing or other filters, the type of paper, the details of the printing process, and so on. Students of photography are, of course, taught about these media-specific qualities directly. They read about and see illustrations of these factors. If they are involved in studio work, they experience the consequences of manipulating these qualities in their own creations. Novices, however, are likely to be unaware of many media-specific variables. One reason may again be that they have experienced only a small slice of the possibilities. For example, most people probably use a particular brand of film (e.g., Kodak, Fuji) throughout their lives. Their choice is more likely to be driven by family traditions or drugstore marketing practices than by considering the effects of the differential color balances used by different film manufacturers.

8

Lynn S. Liben

A second reason for failing to appreciate the consequences of mediaspecific manipulations is that most viewers will have seen their effects over too great a spatial or temporal divide to become cognizant of their power. For example, someone familiar with the photography of Ansel Adams may presume that the different images they have seen are the product of different negatives. It may require seeing different images juxtaposed and explained to appreciate that different printing processes with the identical negative may yield dramatically different images (as achieved in the exhibition, Ansel Adams at 100, see Szarkowski, 2001). Juxtaposition of multiple images of the same scene may be essential in coming to appreciate the effects of other media-specific techniques such as lighting, size of aperture, or use of filters. (Excellent discussions and photographic examples of these and a wide range of other techniques may be found in London & Upton, 1998.) In summary, the appearance of any given photograph—like the appearance of any particular image created in any particular medium—is affected by a range of factors. From a developmental perspective, interesting questions concern if, when, and how children recognize that the ultimate appearance of a given photograph is the product not only of the referents and vantage points, but also of the various media-specific techniques that are involved in producing it. Can children disentangle which features of a photograph are rooted in the world of the referents, and which are rooted in the world of the medium? Are some medium-specific qualities easier to appreciate than others? What mechanisms are responsible for progressive understanding of these variables? D. PHOTOGRAPHS AS COMMUNICATIVE REPRESENTATIONS

The arguments I have offered thus far make clear that fully understanding any given photograph involves not only determining its referential content and recognizing its representational status, but also understanding the myriad of viewpoint-specific and media-specific decisions that are entailed in each. Once having recognized that photographs do not simply ‘‘capture’’ some singular reality, it may become apparent that photographs serve a communicative purpose. The inverse holds as well: once having developed an appreciation for alternative communicative messages, it may become apparent that different messages may be achieved by exploiting the viewpoint-specific and media-specific qualities of photographs. 1. Defining the Focus Before turning to a discussion of photographs as communicative representations, I offer several preliminary comments. First, it is important to note again that I am focusing on photographs as communicative


9

representations in contemporary, mainstream, Western society. I do not, for example, discuss religious or spiritual uses of photographs. Second, my approach is unabashedly psychological. This means that the work that my colleagues and I have done focuses primarily on the ways that photographs are understood by, and affect individuals, as well as on the ways that qualities of individuals may affect their understanding of photographs (see Liben & Szechter, 1999, 2001, 2002). Although cultural and social variables are important in this work, they are considered primarily with respect to how they affect individuals directly (as in influencing the kinds of photographs that individuals encounter), or indirectly (as in affecting the ways that parents mediate their children’s photographic experiences). Our psychological approach thus differs from much of the work on photography in many other disciplines which places cultural and society variables at the center. In sociology, for example, photographs may be used to index societal values (e.g., Bordieu, 1965/1990; Halle, 1991), or studied for what they reveal about social networks (e.g., Giuffre, 2001); in communication they may be examined as devices for reinforcing the societal status quo (e.g., Kress & van Leeuwen, 1996) or as tools to reveal and hence transform social injustice (e.g., see Dewdney & Isherwood, 1994). Thus, the communicative roles discussed here are only a portion of the many that could be studied. 2. Archival Functions As already implied by the earlier discussion of snap shots of social gatherings and trips, one of the most common communicative functions served by photographs is derived from their ability to archive people, objects, or events. Photographs may thus provide the means of recalling personal experiences, and of sharing or informing viewers about events they had not personally experienced because of separation in time (e.g., earlier historical periods) and/or distance (e.g., for most Americans, the Taj Mahal). Photographs may allow communication about referents that are otherwise entirely inaccessible because of distance in space or time or because of the constraints of the human perceptual system. The former is illustrated by photographs of Mars; the latter by photographs of actions too fast to be seen by the human eye (e.g., Edgerton’s well known photograph of a drop of milk as it hits a surface). 3. Emotional Communication Photographs may also be used to communicate emotions, for example, a sense of serenity conveyed by a sunset, of majesty conveyed by a mountain landscape, or of sorrow conveyed by a mourning mother. A related (or perhaps subsumed) function is communicating an aesthetically pleasing or

10

Lynn S. Liben

interesting surface as in photographs of patterns of color or of light and dark or of interesting shapes (e.g., see Figure 1). Third, photographs may be used as tools for acquiring and communicating psychological insights about self or others, as when therapists encourage clients to take and discuss photographs of themselves and family members to discover and explain personal relationships (e.g., Cosden & Reynolds, 1982), or as when teachers ask children to create portraits of themselves to discover and reflect on their own sense of identity (e.g., Dewdney & Lister, 1988; Ewald, 2001; Newbury, 1996). 4. Communicative Roles Identifying the communicative power of any particular photograph cannot, however, be divorced from considering the people who fill roles in the communicative process. Most obvious is the role of the creator of the photograph. The photographer may have a specific goal in creating an image (e.g., to record a keepsake of a friend, to create an image that conveys a particular emotion) or may be focused on the referent with no thought about manipulating view-specific or media-specific qualities. In some cases, the photographer may be uninterested in the product altogether, instead using the act of photography for some other purpose. For example, there may be circumstances in which the act of ‘‘being a photographer’’ is used to improve access to a person or event (e.g., obtaining front row seats at concerts, games, or political events), or, as suggested by Sontag (1977), to protect against social interaction or to provide an excuse for voyeurism. A second role, one often overlooked, is that of curator, a term I use broadly to refer to anyone who selects, displays, or otherwise shows photographs. The importance of this role is most obvious in museums in which a ‘‘real’’ curator structures an exhibit. The very act of placing photographs on museum walls itself has a significant impact on how those photographs are viewed. This point was made convincingly by an exhibit in which snapshots from family albums were framed and mounted at the New York Metropolitan Museum of Art, thereby imbuing them with far more importance than they would otherwise have had (Walther, 2000). In addition to selecting which photographs to display, curators control the way in which photographs are grouped and labeled. Although not focused on photography per se, the general point that curators play a powerful role in creating meaning for viewers is made dramatically by installation artist Fred Wilson. Using objects that are already owned by museums, Wilson groups and labels them in ways that project new meanings and that force visitors to recognize the messages about culture, art, history, and politics that are implicit in traditional museum or gallery exhibits


11

(see Corrin, 2001). Museum curators are not, however, the only mediators between photographs and viewers. Someone who selects and arranges photographs for the family album or for the living room wall also has an impact on what photographs communicate to the viewer. In short, the salience and communicative impact of any single photograph is profoundly affected by the context in which it is displayed. The third essential role is that of the viewer. Is the communication directed to others (as in showing a photograph to someone else, or the reverse), or to self (as in using a photograph to retrieve a memory or to develop a personal insight)? What characterizes the viewer? With what expectations does the individual viewer approach the photograph? Is the viewer prepared to extract referential meaning, emotions, and aesthetic value? What knowledge base does the viewer have about the depicted referent and the medium? Does the viewer consider the photographer’s intent, and does this consideration affect the viewer’s own experience? 5. Developmental Questions The prior analysis of photographs as communicative representations again raises a host of interesting developmental questions. The broadest question is if, when, and how individuals come to appreciate the range of functions that photographs serve, and come to appreciate and enact roles as creators, curators, and consumers. More specific questions derive from changes in children’s developing cognitive and social achievements. For example, does a developing theory of mind contribute to children’s ability to understand the photographer’s communicative intent? Do advances in perspective-taking skills enable children to recognize that there may be alternative interpretations of the same photograph? Is a general progression from concrete to abstract thought reflected in mastering some kinds of communicative functions (e.g., archival) sooner than others (e.g., emotional or aesthetic)? E. FROM ANALYTIC ARGUMENT TO EMPIRICAL EVIDENCE

The preceding discussions of the representational, spatial, media, and communicative qualities of photographs provide analytic support for my earlier assertion that photography is more than pushing the exposure button on a camera and identifying the referent of an image. My analysis also implies that development and experience should affect the production and comprehension of photographs. In the next section of the chapter, I describe empirical work designed to test these expectations about photographic development.

12

Lynn S. Liben

III. Empirical Investigations of Photographic Development No single program of research nor single chapter can address the wideranging set of questions concerning photographic production and comprehension raised above. Thus, the current section provides only illustrative work that my colleagues and I have conducted in this area (Eisenberg & Liben, 1995; Liben & Szechter, 1999, 2001, 2002; Szechter, 2003; Szechter & Liben, 2003; Szechter, Liben, & Rogers, 1998). Our work puts photographs per se at the center of our research, thus placing it under the umbrella of the study of children’s developing understanding of external graphic representations more generally (e.g., Ittleson, 1996; Liben, 1981, 1999; Sigel, 1978). This approach to photography is not unique to our work (see also, for example, Beilin, 1991; O’Connor, Beilin, & Kose, 1981; Seidman & Beilin, 1984; Sroka, 1995; and Thomas et al., 2001), but it does differ from much developmental research in which external graphic representations are used as stimuli to study some other construct of interest (e.g., environmental knowledge, emotional sensitivity, or face recognition). The description of our empirical work is divided into four sections, each devoted to a different type of task. Included are tasks in which participants are asked (a) to identify what (if anything) differs between two photographs of the same referent, and to account for the identified differences; (b) to produce photographs that match photographic models; (c) to create photographs of particular objects or scenes that satisfy some verbal description; and (d) to make and explain judgements about photographs that they like or dislike.

A. CONTRASTING PHOTOGRAPHIC PAIRS

Earlier in this chapter I explained several photographic qualities by contrasting two or more photographs of the same referent. Underlying this graphic-rhetorical device is my belief that juxtaposing alternative depictions of the same referent draws attention away from photographic referents, and to photographic surfaces. The same rationale undergirds one of the tasks we developed for our empirical work—contrasting photographic pairs. In these tasks, participants are shown two photographs, asked whether the two are different, and if so, in what way or ways. Follow-up questions are used to explore participants’ understanding of how these differences were produced. As discussed next, we have used pair tasks to assess participants’ sensitivity to the spatial qualities of photographs and understanding of the photographer’s vantage point, and to study participants’ sensitivity to photographic technique.


13

1. Understanding Vantage Point As I described earlier in discussing the photographic eye, the vantage point for any given photograph may be described with respect to viewing distance, viewing angle, and viewing azimuth. To examine sensitivity to each, we prepared pairs of photographs that differed along one of these dimensions. The first three photographic pairs reproduced in Figure 3 provide illustrations of each. Participants were asked (a) whether the photographs were the same or different, (b) if different, whether the change resulted from something that occurred in the referent (e.g., ‘‘something that changed in this place’’) or because of something that the photographer did, and (c) to explain either what had happened or what the photographer had done. Some pairs were composed of two identical photographs so that in some cases, the correct answer was that the two photographs were the same. Some pairs were different because of something that happened in the referent rather than something that the photographer had done. The final example shown in Figure 3 illustrates a referent-change pair. In this pair, the size difference between the two depicted balloons results from different amounts of inflation. That the larger-looking balloon is actually larger is evident from the fact that the sizes of other components of the image (e.g., the duck, the floor tiles) remain unchanged. Remaining pairs differed because of a change in viewing distance, viewing angle, or viewing azimuth. Five pairs differing by viewing distance were shown to children aged 3-, 5-, and 7-years, and to a sample of college students. Among the 3-year-olds, about 75% could not explain the role of distance on any of the items, and the remainder were correct on only a single item. By the age of 5, there was far more range in performance, with about half the children responding correctly on all or almost all items. By 7 years, most were making only one (if any) error; and by adulthood, virtually all participants explained correctly that the photographer moved closer or further away as appropriate. As in any procedure that requires participants to verbalize their understanding to be scored as correct, these data may underestimate young children’s understanding. However, the kinds of erroneous explanations provide reasonable assurance that failures are at least not entirely due to children’s general inability to understand the question or to a general discomfort about explaining their thoughts to the investigator. When shown photographs of a statue that had been taken from two different viewing distances, for example, one 3-year-old said: ‘‘This man is smaller and this man is bigger.’’ Although this particular response might have been a statement about the man’s image rather than the imaged man, other responses seem to demonstrate even more convincingly that young children

14

Lynn S. Liben

(a)

(b)

(c)

(d) Fig. 3. Sample stimuli for Contrasting Photographic Pairs task. Illustrated are pairs that change: (a) distance, (b) angle, (c) azimuth, and that show (d) a change in referent. Actual photographs were in color.

did not fully understood the effects of changing viewing distance. For example, when shown the two photographs of dandelions reproduced in Figure 3, a 5-year-old responded that the photographs differed, explaining: ‘‘Well, this one only has dandelions and this one has dandelions and part of a tree.’’ The investigator responded: ‘‘That’s exactly right. And how did the photographer get the tree in this one?’’ The child answered: ‘‘Um, see, he


15

had a very big camera and he took a picture of the tree too, and the dandelions.’’ The explanations given by young children have some inherent ambiguities, but they contrast with older children’s and adults’ ability to provide clear statements that the photographer ‘‘got closer’’ or ‘‘moved further back.’’ (Interestingly, and consistent with the suggestion that even most adults give little thought to media-specific techniques, none of the adults suggested that the photographer had changed the camera’s lens or that the print had been enlarged and cropped.) In the cases of viewing angle and viewing azimuth, the data suggest similar, but more protracted developmental progressions. For pairs distinguished by viewing angle, none of the 3-year-olds provided even a single correct explanation. By 5 years, some children showed understanding, although nearly half the children were unable to explain a change in viewing angle. Children’s comments reveal their focus on the referential content rather than on the surface appearance of the photographs. For example, even though they were asked about differences between the photographs and had the contrasting images in view, 3-year-olds typically simply commented on the content of the picture in some way. Shown the angle pair in Figure 3, for example, one 3-year-old said: ‘‘I like tulips. My mom likes tulips. These ones are yellow [pointing to the straight-ahead view] and these ones are red [pointing to the overhead view].’’ Another 3-year-old said: ‘‘They’re both the same.’’ When the investigator responded ‘‘Well, they’re both of tulips, but is there anything different about the pictures?’’ the child responded: ‘‘Nope this one [points] has the same stuff.’’ When 5-year-olds failed to give evidence that they understood the difference in the photographer’s viewing angle, they were at least more likely than the 3-year-olds to demonstrate that they had noticed a difference between the photographs. However, they erroneously attributed the difference to something in the referent rather than in the way the referent was depicted. For example, one 5-year-old child said that the difference was because the photographer ‘‘took this one when they were all curled up and those ones when they were all blooming.’’ Another commented: ‘‘The tulips are up and the tulips are down. . . First, in the spring they were closed and then in the summer they came out again.’’ Still another commented: ‘‘This one is closed and this one is open.’’ In response to the investigator’s probe, ‘‘How did the photographer do that?’’ the child answered: ‘‘He took one sometime when they were closed and one sometime when they were open.’’ Some of the 5-year-olds did, however, demonstrate understanding, as illustrated by the child who explained: ‘‘Um, that one you’re looking that way [points straight ahead] and that one you’re looking down, this way [bends over the picture]’’ or another who said ‘‘Oh I like these. . . . He [the photographer] went sort of on the side of them and then like up above them

16

Lynn S. Liben

to get the middle [points hand down on top of the tulips and rises in chair]. . . because um, like this one is like straight across and this one’s like you’re looking down [flexes hand to point all fingers down onto the tulips].’’ These kinds of responses became more common by 7 years, and by adulthood, performance was virtually perfect. The data from the azimuth pairs are—with one exception—highly similar to those from the angle pairs. That is, at the youngest ages, many children even failed to see that there was any difference between the two photographs, and when they did, often attributed it to a change in the referent. For example, one 3-year-old said: ‘‘Hey, that’s not the same rooster! There he’s putting his head up and he’s not the same!’’ Even more commonly, children said that there had been a change in the position of the referent. For example, for the sample item shown in Figure 3, a common response was that the tile itself had been turned. This answer fails to appreciate the counter-evidence about orientation provided by the direction of the woodgrain of the background. The major difference between data from angle and azimuth pairs alluded to previously is that on the azimuth pairs even adults evidenced considerable difficulty. Like the children, they often erred by attributing the difference in the photographs to movement of the referent (i.e., using the same example, the turning of the rooster tile) rather than a change in the position of the photographer. That adults continue to have difficulty in interpreting changes in representations caused by a change in viewing azimuth is completely consistent with extensive literature from other paradigms showing that even adults struggle in using other azimuth-sensitive images such as maps (e.g., Levine, Marchon, & Hanley, 1984). 2. Understanding Media-specific and Communicative Photographic Qualities The photograph pairs methodology may also be used to explore nonspatial variations in photographs. The structure of the method remains similar: participants are asked to compare two photographs, and, for any differences they observe, to explain what they believe produced them, and/or to explain what consequences these different images may have on those who view them. Pair tasks of this kind were given to adults and their 7- to 13-year-old children (Szechter & Liben, 2003) as part of research on the role of parent–child communication in the development of aesthetic awareness (Szechter, 2003). Pairs differed with respect to a variety of media-specific techniques including length of exposure, type of lens, camera movement, and use of filters. Similar to the findings from the vantage-point photograph pairs discussed earlier, children were more likely than adults to offer referent-based


17

explanations for the differences between the two photographs. Illustrative were findings from a pair in which a woman had been photographed using different filters. Among the children, 47% offered a referent-based explanation of the difference by suggesting that the woman in the photograph had changed her clothes (e.g., ‘‘In this one she was wearing a darker scarf’’), whereas among the adults, only 12% offered a referent-based explanation of this kind. Only adults explicitly identified some kind of photographic technique as the reason that the photographs appeared different: 7% correctly specified the use of filters and another 20% suggested some other technique such as a change in exposure. Although these percentages are relatively low, they contrast markedly with the complete absence of such responses by children. A common response among both children and adults was to describe the contrasting appearance of the two photographs without unambiguously identifying the reason for the contrast. Even within these responses, however, there was a difference in the quality of explanations: children tended to focus more on simple description (e.g., ‘‘And the hat is darker in this one’’) whereas adults tended to focus more on the effect that different surface features of the image have on the viewer (e.g., ‘‘The color of the cape and the background in this picture [points] are harmonious and they’re also harmonious in this one, so it makes you feel like it’s a different time, like it somehow changed’’). The patterns of explanations from this and other technique-based contrasting pairs reveal age-linked increases in attention to the photographic medium. Further research is needed to evaluate which differences are due to accumulating experience with the photographic process per se, which are due to a growing appreciation for representational media in general, and which may reflect adults’ greater willingness to attend to and reason about photographic qualities (e.g., shadow directions) during the interview itself. B. REPRODUCING PHOTOGRAPHIC MODELS

The pairs method used in the studies I have described thus far is aimed at exploring the participant’s ability to interpret the causes and consequences of spatial and media variables in photographs that were produced by others. As noted earlier, this methodology is highly dependent on verbalization, and provides no feedback to participants about the viability of their hypotheses. The second task—reproducing photographic models—avoids the strong verbal component and provides an opportunity for participants to discover that changing the vantage point has an effect on the image. In this task, participants were given model photographs, and were asked to try to

18

Lynn S. Liben

Fig. 4. Photographs used for the Reproducing Photographic Models task. Actual photographs were in color.

produce identical images. A digital camera was used in this work so that participants would be able to see their photographs immediately. We have used this procedure with 8-year-old children and college students to investigate age differences in the ability to exploit camera position to control the spatial qualities of photographs. The four model photographs were scenes taken within a campus building, and are reproduced in Figure 4. The model photographs were intentionally designed to avoid what we assumed would be likely canonical photographs in this setting. We did so to reduce the possibility that participants’ photographs would match the models simply because both were canonical images of the referents. For example, we presumed that visitors to this building would be unlikely to photograph the mail box at all, but even if so, almost certainly not


19

do so from this particular angle. Or, if visitors were to take photographs of Otto’s cafe´, they would be far more likely to photograph it with the sign centered and straight ahead than they would to photograph it from the direction shown in the model. (As implied by the word ‘‘presumed,’’ we did not, however, empirically identify canonical views for this task.) By the time they reached this building, participants had already received lessons and practice with the digital camera that had been used to create the model photographs. Participants were given the model photographs one at a time and asked to create as close a match as possible. The models were available to participants throughout the process if they wished to consult them. After taking one picture (using the viewfinder), participants were asked to look at their image on the camera’s display, compare it to the model, and then to take a second picture to try to produce an even closer match. To provide a sense of the range of performance on this task, Figure 5 displays four sample responses for one model photograph. Consistent with the notion that there is early sensitivity to the referential contents of photographs, all participants succeeded in reproducing the focal content of the model photographs in their own images with a single exception of a single photograph. That is, all but one photograph correctly included the correct flag, mailbox, Otto’s sign, and table as shown in the model images. There was, however, far more variability among participants with respect to their success in reproducing the precise way that the content was depicted. For example, even something as simple as camera orientation showed variability. Collapsing over the first photographs for all four models, correct camera orientation (that is, a horizontal orientation for the photographs of Otto’s cafe´ and the flag; a vertical orientation for the mailbox and table) was achieved in 93% of the adults’ photographs, but in only 63% of the children’s photographs. To summarize performance, points were awarded to photographic qualities that depended on correct viewing distance, viewing angle, viewing azimuth, and camera orientation. Scores for the four photographs were then summed to create a total score that could range between 0 and 46. A distribution of total scores is given by age and by trial in Figure 6. Apparent from this distribution is the generally lower performance by children than adults (mean scores of 18.7 vs. 29.0, respectively) and the overall improvement from trial 1 to trial 2 (22.0 vs. 25.8, respectively, with the improvement in trial 2 significant among children only). The distributions also reveal considerable variability within age groups as well as considerable overlap between them. Although additional research is needed to study performance on this photo model task across a greater age range, our research already allows the broad conclusion that there are age-linked progressions in understanding

20

Lynn S. Liben

Fig. 5. Sample responses given by children (top row) and adults (bottom row). Scores for these responses, based on a maximum of 12 were, top row: left 3, right 7; bottom row: left 7, right 11. Actual photographs were in color.

and controlling the viewing distance, angle, and azimuth of the photographic eye. Interestingly, our work also revealed dramatic individual differences within any given age. These data thus add to research on other graphic representational domains showing that challenges remain even into adulthood (e.g., see discussion of adult performance on mapping tasks in Liben, Kastens, & Stevenson, 2002), and raise interesting questions about the role of individual differences in spatial skills and experience. C. TRANSLATING VERBAL DESCRIPTIONS TO PHOTOGRAPHIC IMAGES

The third task we have used is one in which participants are asked to produce photographs of a particular referent in a way that fulfills some


21

Fig. 6. Distributions of participants’ total scores (maximum ¼ 46) on Reproducing Photographic Models task. (Categories on x-axis indicate highest score so that ‘‘6’’ includes scores ranging from 0 to 6; ‘‘12’’ includes 7 to 12, etc.)

verbally described characteristic. The items for this Verbal Description task were designed to encourage participants to think about both spatial and communicative qualities of their photographs. Given that participants were again using a digital camera that allowed control of vantage point only (rather than, say, lighting or lens focus), of interest was how participants would use viewing distance, angle, and azimuth to implement our requests. Unlike the prior tasks, however, there was no a priori, uniquely correct solution to these requests. Thus, our goal was to see the range of strategies that children and adults might use in responding to the task demands. These data are interesting not only with respect to children’s understanding of the referential information contained in different views (i.e., a predominately cognitive issue), but also with respect to children’s understanding of the different kinds of emotions that may be communicated by different views (a predominately affective issue). Specifically, the method involved taking individual 8-year-olds and college students on a predetermined route through the Penn State campus. They were told before beginning that in addition to taking photographs of their own choosing, at some locations they would be asked to take particular photographs. The first request occurred at a fountain sculpture. To begin, participants were simply asked to photograph it. After the participant and investigator viewed the resulting photograph on the screen of the digital camera, the participant was asked to take a second photograph, with the specific instruction designed to require a change in vantage point. If the initial photograph showed all or most of the fountain,

22

Lynn S. Liben

the participant was asked ‘‘What if you only wanted to see that spiral shaped part on top [gesturing a sphere up in the air] in your picture? What could you do to make that part take up most of the picture?’’ Irrespective of how participants answered this question, they were asked to ‘‘Try it.’’ If the initial vantage point had produced a photograph showing only part of the fountain, the instruction was designed to ask participants what they could do to get the entire fountain in the picture, including the large circular pool in which the entire sculpture rested (with a downward gesture outlining the perimeter of the pool). In both cases, the request for the second photograph was meant to motivate the participant to change viewing distance and/or viewing angle. Interestingly, with few exceptions, both children and adults chose to take their initial photographs from a relatively long distance so that their resulting photographs contained the entire sphere and the triangular support on which it stands, with all but two adults including either all or some of the bottom round base as well. To provide an index of viewing distance, each photograph was printed as a 1.500 200 print, and the diameter of the representation of the sphere was measured (extrapolated beyond the picture frame when necessary). What is most striking about the data is that—again, with the single exception of the two adults mentioned above— the sphere representations in the first photograph were relatively small, and the patterns of data were highly similar for children and adults (mean diameters for children and adults, respectively, were .27 and .33 inches). Because all but two participants (both adults) had begun with a long viewing distance, all but these two were asked to take the second photograph so that the sphere would take up most of the picture. Children and adults were differentially successful in complying with the request, with adults showing far more dramatic changes in the size of the depicted sphere. Figure 7 shows prototypical examples of children’s and adults’ first and second photographs. To quantify participants’ success in implementing the verbal request, we examined the diameter of the second sphere in relation to the diameter of the first sphere separately for each participant. Among children, only 65% succeeded in increasing the size of the sphere, a proportion that includes the 8% of the sample who increased the size of the sphere by an almost imperceptible amount (less than 10% larger than the first). The children who failed included 11% whose second sphere was identical to or trivially (less than 10%) smaller as well as 24% whose second sphere was noticeably (more than 10%) smaller than the first. On average, the diameter of the second depicted sphere was only 1.02 times larger than the first. Among the adults who were asked to have the sphere take up more of the image, all implemented the request successfully. On average, the diameter of the second depicted sphere was 3.20 times larger


23

Fig. 7. Prototypical examples of the first and second photographs of fountain by children (top row) and adults (bottom row). Actual photographs were in color.

than the first. These data suggest that as a group, adults are strikingly better than children in controlling viewing distance to create a specified effect. A second verbal request was made as participants approached the campus art museum at which large sculptures of lions’ paws flank the main entrance. Here participants were asked to ‘‘get just one paw in the picture.’’ Of interest was not only whether participants would be successful in carrying out the request, but also what strategies they would use to do so. Many children had difficulty in implementing the request: 32% produced photographs that had two rather than one paw in the photograph. Although 10% of the adults also produced photographs with two paws, these were cases in which the second paw was so close to the edge that it would not have been visible through the viewfinder. (There is an imperfect match in the digital camera between what is visible in the viewfinder and what is recorded and hence visible in the print.) Successful participants were very creative in finding strategies for carrying out the request. The samples shown in Figure 8, all taken by children, illustrate the use of noncanonical viewing distances, viewing azimuths, viewing angles, or a combination of two or more of these strategies. Most

24

Lynn S. Liben

Fig. 8. Illustrative responses by children to verbal request to have ‘‘just one paw’’ in the photograph. Actual photographs were in color.

changes in viewing angle involved tilting the camera toward the sky, although one child did the opposite by climbing up on one of the paws and taking a photograph angled down. One additional child provided a unique solution saying that he ‘‘waited for people walking by to block one of the paws from view.’’ The third and fourth verbal requests were both made at a statue of a mountain lion that serves as the university mascot (the Nittany Lion). Participants were first asked to take a photograph so that ‘‘it looks like we’re seeing the lion peeking out from behind a tree.’’ With the exception of a single child, all participants were able to take a photograph so that the lion’s head seemed to be emerging from behind a tree. Adults, however, were more likely than children to use a viewing azimuth in which the head was the only part of the lion that showed in the photograph (58% vs. 34%). Although this quality was not explicitly requested, these images suggested ‘‘peeking’’ more effectively than did photographs in which the back of the lion could be seen as well. The final verbal request was intentionally very open ended, and was directed at exploring the kinds of strategies children and adults might use when asked to photograph a referent to convey some affective tone. Specifically, participants were asked to take a photograph in a way that would make the lion look ‘‘kind of scary.’’ A priori, we had expected that one way to make the lion look scary would be to take a close-up photograph. To examine children’s and adults’ use of this strategy, we measured the head of the lion following the same procedure that we had


25

used to measure the sphere of the fountain: each photograph was printed in a fixed size (1.500 200 ), a circle was drawn to encompass the lion’s head (extrapolating beyond the picture boundaries when necessary), and the diameter recorded. The resulting data revealed that, as a group, adults tended to move in closer to the lion statue than did children. The average diameter head sizes for children’s and adults’ photographs were, respectively, .3700 and .8200 . To provide a sense of how these numbers translate into graphics, the top row of Figure 9 presents photographs that approximate, respectively, these two mean sizes. The data showed that children, but not adults, took photographs from a very great distance (so that the lion was very small), but both children and adults took close-up shots. Even within the same viewing distance, however, photographs varied along other dimensions. The two photographs shown in the bottom row of Figure 9 illustrate this point insofar as they are taken from roughly the same distance, and yet seem to convey a different expressive tone. We thus scored photographs in three additional ways.

Fig. 9. Illustrative responses to verbal request to make the lion look ‘‘kind of scary.’’ Top row shows sample photographs of approximately average head size by children (left) and adults (right). Bottom row shows a child’s (left) and adults’ (right) photograph at a similar viewing distance.

26

Lynn S. Liben

First, we scored all photographs with respect to viewing azimuth. The data showed that adults’ and children’s photographs differed significantly with respect to the proportion of photographs in which the camera was, versus was not, facing directly toward the lion’s face (for adults, proportions were 67% vs. 33% respectively; for children, they were 35% vs. 65%). Second, we scored viewing angle. A similar age-linked contrast held. Children were more likely to take their photographs from an eye-level angle than with the camera tilting up (65% vs. 35%) whereas adults were less likely to use an eye-level angle than to tilt the camera up (22% vs. 61%). (The remaining adults angled the camera slightly down, an angle more readily achieved by adults given their greater height.) Finally, we asked another group of college students to rate the photographs (blind to age of photographer) for ‘‘scariness’’ using a 7-point scale ranging from ‘‘not scary at all’’ to ‘‘very scary.’’ Data showed that, on average, the photographs taken by children were rated as significantly less scary than those taken by adults (with mean ratings of 3.1 vs. 3.7, respectively). Taken together, these data provide clear evidence for both individual and age-linked differences in the degree to which photographs convey a desired affective tone such as scariness. Additional research is needed to describe the relation between spatial factors and emotional impact more precisely, and to identify the many nonspatial factors that influence the emotional tone of a given photograph (e.g., patterns of shadows, inclusion of differing background content such as buildings vs. flowers). Before closing the discussion of this task, it is also interesting to add that apart from providing the photographic data themselves, the procedure we used spontaneously elicited a few revealing verbal comments from children. One child commented that ‘‘If you want to make it look scary you can’t take a picture of the face because the face isn’t scary at all. The eyes are just circles and the mouth isn’t growling. You have to take other angles to make it look scary.’’ This comment might be taken as a sophisticated one insofar as it seems to reflect awareness of vantage point. However, the comment might be better interpreted as a relatively unsophisticated one insofar as it addresses only the qualities of the referent (that is, the face itself ‘‘isn’t scary’’). Consistent with the belief that the face needs to be avoided to create a scary photograph, this child photographed the lion from its side, apparently missing the possibility that the appearance of the face might be manipulated by a change in vantage point or by some other technique. A second child also offered the opinion that it would be difficult to make the lion—as is—look scary. After taking her own pictures, this child climbed up on the lion to ‘‘give him horns to make him scary,’’ and asked the investigator to take a photograph.


27

In contrast to these two children, a third child proposed using a photographic ‘‘technique’’ for making the lion look frightening. This child said he was going to make the picture look blurry so ‘‘it looks like the lion’s going to jump out at you.’’ Implementation was unsuccessful, however: the child shook the camera immediately before and after, but not while pressing the exposure button. Taken together, these spontaneous verbal comments suggest that it would be valuable in future research to ask all participants to comment on, and evaluate the efficacy of their strategies. D. PROVIDING AND EXPLAINING PHOTOGRAPHIC PREFERENCES

The three methods discussed so far all yield responses that may be scored with respect to some externally defined criterion of success. This is particularly true for the photographic pairs task (in which participants either correctly or incorrectly identify and explain similarities and differences between the photographs), the photographic model task (in which the participants’ photographs may be judged for similarity to the model), and the first two of the verbal requests (in which the photographs can be judged with respect to whether the photograph includes appropriate referents, respectively, the sphere or base of the fountain and only one paw). The second two verbal requests (the peeking lion and the scary lion) are more open ended, allowing far more room for alternative strategies and personal opinion about whether the goal has been achieved (e.g., whether close-up photographs do, indeed, make the lion look more frightening). The final method discussed here moves entirely away from the idea of ‘‘correct’’ or ‘‘incorrect’’ solutions, and instead explores participants’ personal reactions to photographs. By characterizing the patterns of images that participants like relatively more or less, and by categorizing participants’ explanations of their own preferences, it is possible to explore which photographic qualities and functions are attended to and valued, and whether these shift with age or experience. To provide one means of learning about which characteristics of photographs are important to participants, children and adults were asked to comment on photographs that they, themselves, had created. These data were collected at the end of the campus walk in the study with 8-year-olds and college students described previously. Participants returned to a room in the building at which they had begun their tour, and downloaded the images they had just taken. After looking at all their own images, participants were asked to select their three most favorite and three least favorite, and to explain the reasons for their choices. Sample photographs selected as most-liked and least-liked by children and adults are shown, respectively, in Figures 10 and 11. Among children,

28

Lynn S. Liben

Fig. 10. Illustrative photographs selected as most-liked (top row) and least-liked (bottom row) by children. Actual photographs were in color.

Fig. 11. Illustrative photographs selected as most-liked (top row) and least-liked (bottom row) by adults. Actual photographs were in color.


29

explanations generally focused on the referential content of the photograph, rather than on the qualities of the image. Reasons given for the sample most-liked images shown in Figure 10 included: ‘‘Because it was my favorite team’s mascot, because there is my favorite summer sport in it . . . planting flowers’’ (participant #4); ‘‘Cause it shows a pretty sign and pretty flowers’’ (#6); and ‘‘I love beetles, and plus it is a yellow beetle and yellow is one of my favorite colors’’ (#11). All of these are referent-based reasons. Children’s reasons given for the illustrative least-liked images include: ‘‘It doesn’t have that many artistic things’’ (#7); ‘‘Mom and dad don’t like flags . . . they like the pointy flags more than square ones’’ (#15); and ‘‘Because it has the word ‘love’ on it’’ (#23). Although the first of these is ambiguous about whether the child is expressing discontent with the referent or with the image, the second two clearly focus on the referents themselves. The explanations given by adults provide a contrast to those of the children insofar as they are addressed more directly to the surface of the photograph itself. Explanations given for the best-liked selections in Figure 11 include: ‘‘Cause it just makes him look really cool. The close up of it. Just the effect of it’’ (#44); ‘‘The way the sun beats down, it kind of distorts the image. You can see like spectrum colors’’ (#48); and ‘‘It just reminds me of something I would do; lie on the ground and stare up at a tree and a really blue sky. Also I like the colors. Also usually you look at pictures straight on [gestures straight ahead], but you’re looking up’’ (#57). Reasons given for the least-liked selections in Figure 11 include: ‘‘No depth to it. Kind of took it at a bad angle. Should’ve gotten a side angle to see more depth to it’’ (#42); ‘‘It’s kind of hard to look at. You don’t know where your eye’s supposed to be focused, and the water fountain is just not that exciting from that point of view’’ (#52); and ‘‘Because with the colors you can’t see what it is very easily . . . the metal blends in with the concrete of the building’’ (#58). The observation concerning differences in children’s and adults’ foci is consistent with what others have noted about children’s tendency to focus on referential meaning in other visual arts (e.g., Freeman & Parsons, 2001; Gardner, 1970; Gardner & Gardner, 1970; Parsons, 1987; Winner, 1982). Adults’ explanations also sometimes allude to referential content (e.g., see #57), but they far more commonly allude to the visual image itself, including comments about depth, angle, and the focal point of attention. To characterize these explanations more systematically, we developed a 25-category coding system, further organized into three major categories (plus an ‘‘uncodable’’ one). Comments expressing like or dislike related to the object or event depicted in the image were categorized as ‘‘referent’’ explanations. Illustrative is the explanation of participant #15, whose comment concerns the shape of flags per se, not the shape of their

30

Lynn S. Liben

appearance in the photograph. Included within this category were comments about the identity of the referent, memories of the referent, and various qualities of the referent (e.g., about its beauty or behavior). Explanations that concerned the appearance of the image (irrespective of the process by which the image was created) were categorized as ‘‘surface’’ explanations. The explanation given by participant #52 illustrates this category insofar as the comment concerns an evaluation of the central focal point of the image itself. Detailed codes for this category included comments about how the photograph itself looked, such as color contrast, sharpness of the image, depth of field, and balance. Finally, comments that focused on the actual process of creating the image fell into the category of ‘‘technique’’ explanations. The second part of the explanation given by participant #42 illustrates this category because the explanation addresses what the participant failed to do in the act of taking the photograph (‘‘should have gotten a side angle’’). Detailed codes in this category included comments about the photographer’s position (e.g., distance from the object, angle from which it was photographed, orientation in which the camera was held), movement (e.g., camera shake), or about strategies used to reproduce one of the photographic models. The explanation given by participant #42 provides an illustration of the possibility of individual explanations including more than one type of reason. In this example, the first part of the explanation—‘‘No depth to it’’—is a comment about the image rather than the process, but the second part of the explanation—‘‘Kind of took it at a bad angle’’—is a comment about the process of taking the photograph. Each explanation was coded according to its major thrust, with coders blind to whether it had been given by a child or adult. Results are shown in Figure 12. One striking pattern seen in these data, and consistent with the illustrative justifications quoted earlier, is that for most-liked images, children gave greater attention to referents than did adults. A second is that children tended to pay more attention to technique when discussing their least-liked rather than their most-liked images. The relatively large percentage of ‘‘uncodable’’ explanations offered by children in discussing their least-liked images reflects the problem that it was often impossible to decide reliably whether a particular explanation addressed the referent itself or the image. For example, the word ‘‘it’’ in an explanation ‘‘I just like it’’ might refer either to the referent or to the photograph. Thus, unless the participant expanded on the explanation in response to follow-up probes, responses like these were categorized as uncodable. Taken together, these data suggest that there are interesting age-related differences in the salient aspects of photographs, similar to the age-related differences that have been reported in responses to other visual arts.


31

Fig. 12. Distributions of types of explanations given by children (C) and adults (A) for mostand least-liked photographs.

In the course of collecting the data just described, we were struck by how often the ‘‘best-liked’’ choices seemed pedestrian (or worse), and by how images we found to be aesthetically pleasing were often ignored (or even selected as ‘‘least liked’’) by the participant. A potentially important difference between our own responses and those of the participants is that the participants had personal access to their purpose in taking the photograph. Thus, participants may have been evaluating the image with respect to how well it implemented their intent, rather than with respect to how much they liked the finished product. Future research is needed to examine whether there are developmental changes in sensitivity to photographers’ (self or other) intentions, and in evaluating the degree to which various intentions are or are not conveyed in a given image. In addition, future research is needed to address age-linked responses to others’ photographs, even apart from intent. We have already studied experience-linked differences among adults in the valence of responses to others’ photographs (Liben & Szechter, 2002). College students were asked to use a 7-point scale to characterize their personal responses (from negative to positive) to 25 photographs, and to explain the reasons for their ratings. Students were asked to rate photographs at both the beginning and the end of the semester. Participants were recruited from introductory psychology classes, a photography appreciation class, and from studio photography classes. These classes were expected to differ in two ways: first, they were expected to attract students with differing levels of interest and experience in photography, and second, they were expected to provide

32

Lynn S. Liben

students with different amounts of photography-related experience during the semester. Of interest was whether students’ personal responses to photographs would differ initially, and whether responses would change differentially during the semester as a function of differing levels of photography-related experience. As expected, responses to almost all photographs differed either in relation to class, to time (beginning or end of the semester), or to the interaction of these factors. For example, as predicted, students from photography classes rated a traditional tourist picture of the Eiffel Tower less positively than did students from the psychology classes. Photographs of this kind are typically viewed as hackneyed by expert photographers. Also as predicted, photographs that had more disturbing content (e.g., a photograph by Sean Kernan entitled Man and Kitten, West Virginia Penitentiary) were responded to more positively by the photography students than by the psychology students. In both these cases, the responses from students in the studio course were most differentiated from responses of students in the psychology course. Although these data fit predictions well, not all response patterns were as neatly explained. For example, for some photographs, changes in ratings over the semester were greater among students in the photography appreciation course than among students in the studio photography course, whereas for other photographs, this pattern was reversed. Although much additional research is needed to explain these patterns (e.g., the photographs may differ with respect to how saliently they illustrate some technical photographic process that would be particularly obvious to students in a studio photography class), the completed research has already demonstrated the viability of the methodology. Similar methodologies might be adapted to study age-linked effects in conjunction with experience-linked effects, for example by collecting rating data, longitudinally, from children who do and do not participate in photography classes. E. EXTENDING EMPIRICAL WORK

In summary, we have used a variety of methods to explore age-linked and experience-linked differences in the ways that people understand and create photographs. Much additional work is needed even at the descriptive level to characterize how sensitivity to various qualities of photographs develops, and how individuals become increasingly able to manipulate photographic qualities in achieving some specified goal, be it a goal aimed at controlling referential content and appearance (as in the photographic models task) or at effecting an intended communicative impact (as in producing a photograph that conveys some emotional tone as in the ‘‘scary lion’’ request). Although the data reviewed above cannot yet tell a complete story,


33

they provide a foundation on which to base programs of further empirical work.

IV. The Larger Developmental Picture In the prior sections of this chapter, I argued that photographs are an important substantive focus of study in their own right. Although the corpus of developmental empirical work on photography is limited, the work described here and in related work (see Beilin, 1991, 1999; Kose, 1985; Sroka, 1995; Thomas et al., 2001) already documents important age-linked and experience-linked change. In this final section of the chapter, I consider how these developmental changes fit within the developing understanding of external spatial representations more generally, using a conceptualization offered earlier (Liben, 1999). The definitions and developmental matrix of that conceptualization are summarized in Tables I and II. A. COMING TO UNDERSTAND REFERENTIAL MEANING AND REPRESENTATIONAL QUALITIES

The original presentation of the developmental trajectory described in Tables I and II was focused largely on children’s developing ability to use representations to gain information about the referents of the representations. In taking an informational stance to representations, the major goal is to see through the representational surface to the referential meaning that lies beneath it. The viewer may accomplish this in a number of ways. At a relatively primitive level (see Table I, point I), the viewer may essentially remain unaware of the surface features of the representation altogether, interpreting the perceptual display as if it were the referent itself. This is presumably what happens when babies behave toward a representation similarly to the way they behave toward the referential object itself. Illustrative are cases in which babies attempt to pick up patterns on flat surfaces such as fabric, paper, or the floor (e.g., Church, 1961; DeLoache et al., 1998; Liben & Downs, 1992; Ninio & Bruner, 1978). At a more advanced level (see Table I, point IV), the viewer is aware of surface features, but in addition to interpreting the referential meaning of some features correctly, infers referential meaning of other features incorrectly (Liben & Downs, 1989; Liben, 2001). For example, a preschooler may correctly interpret a line on a road map as carrying meaning about the existence and general location of a road, but may incorrectly infer that the line’s red color means that the referential road is itself red (see Liben & Downs, 1989). At a still more advanced level (Table I, point VI), the viewer

34

Lynn S. Liben TABLE I Progressive Competencies in Understanding External Spatial Representations

I. Referential content. The viewer begins to identify the referential meaning of the representation, with varying ease depending upon the physical similarity of representation and referent. Thus, the viewer ‘‘understands’’ the representation in the sense of identifying the denoted referent, but appears to confuse them (as in trying to pick up a depicted object). II. Global differentiation. The viewer identifies the denotative meaning of the representation, distinguishes the representation and referent, and responds to each differentially. The viewer does not, however, reflect upon the correspondence between the two. The ‘‘stand for’’ relation is implicit in identification, but not generally subject to intentional manipulation. III. Representational insight. The viewer distinguishes between representation and referent, and intentionally interprets or assigns ‘‘stand for’’ meaning to the representation. Representational insight occurs first for objects that are inherently representational (as a photograph) and only later for objects that do not normally function as representations, but rather are most salient as objects in their own right (as a scale model). IV. Attribute differentiation. The viewer comes to appreciate that some, but not all attributes of the representation are motivated by attributes of the referent, and that some, but not all attributes of the referent motivate graphic attributes of the representation. Until doing so, the viewer inappropriately expects that attributes of the representation necessarily mimic attributes of the referent (as in inferring that a red line means a red road) and that attributes of the referent will necessarily be mimicked by attributes of the representation (as in expecting that a large building will appear large in the representation). V. Correspondence mastery. The viewer extends the prior understanding of attribute differentiation to develop understanding of the formal representational and geometric correspondences between representation and referent. The former allows the viewer to understand the referential content of symbols; the latter allows the viewer to understand the referential meaning of graphic space. VI. Meta-representation. The viewer is able to reflect upon the mechanisms by which, and the purposes for which, graphic representations are created, including understanding that different correspondence rules and conventions are used in different media (as in maps versus graphs), different traditions (as in Western versus Asian art), and different renditions (as in a world map in a Mercator versus a Peters projection). As a result, the viewer is able to understand representations not simply as convenient substitutions for referents, but rather as cognitive tools that enrich understanding of the referent, and to select among them appropriately for particular purposes. Note: Reproduced from Liben, 1999, with permission.

is able to understand why representational (surface) features may be manipulated to emphasize or reveal different qualities of the referent. Photographs are likely to be among the easiest kinds of graphic spatial representations for children to understand with respect to informational meaning. As noted earlier, the perceptual similarity between referents and photographs (albeit not equivalence, see Gibson, 1980) enhances the relative ease with which one can ‘‘see through’’ a photograph (i.e., enhances its ‘‘transparency’’). Importantly, however, the very features that make it easy

TABLE II Developmental Progression in Understanding External Spatial Representations Competency

Infants Toddlers Preschoolers Young Children Older Children Adolescents þ

Referential content

Global differentiation

Representational insight

Attribute differentiation

Correspondence mastery

Metarepresentation

e f f f f f

e e f f f

e e f f

e e f

e e


Age group

Notes: Shaded cells indicate focal competency under development. Cells marked with an open circle indicate that considerable development in the competency is continuing or beginning. Cells marked with a closed circle indicate that the basic competence has been achieved, although further minor development may still be occurring. Blank cells indicate that little development is yet under way. Definitions of competencies are given in Table I. Reproduced from Liben, 1999, with permission.

35

36

Lynn S. Liben

for viewers to see through photographs to the referents that lie ‘‘behind’’ them, may make it difficult for viewers to notice the surfaces of photographs themselves. Thus, when one is primarily focused on children’s appreciation of the qualities of the representation itself, photographs may prove a particularly challenging graphic medium relative to other media such as drawing or painting. B. EXPERIENTIAL FACTORS IN PHOTOGRAPHIC UNDERSTANDING

In the original presentation of the developmental sequence reflected in Tables I and II (Liben, 1999), I suggested five factors that could be expected to facilitate children’s attention to and understanding of representations in general. By extension, these should also be relevant to children’s developing understanding of photographs in particular. I briefly consider each in turn. First, I suggested that representational understanding would be enhanced by exposure to many different kinds of external representations, particularly when they all refer to the identical referent. Illustrative would be experience in seeing one’s own house represented on a map, photograph, drawing, and blueprint. The opportunity to see that the same referent can be symbolized by many different kinds of representational surfaces would be expected to draw attention to the representational surfaces themselves, and lead children to consider the idiosyncracies of what each symbolic mode may offer. Developing this kind of general understanding of and appreciation for different graphic media would thus provide a strong foundation on which to build an understanding of and appreciation for photography in particular. Second, I suggested that representational development would be facilitated by exposure to alternative representations of the identical referent within a single medium. Although the original examples were drawn from cartography (e.g., arguing for the salutary effect of seeing many different kinds of maps), they can as easily be drawn from photography instead. Here the notion is that exposure to many different photographs of the same referent would help viewers to come to differentiate the qualities that adhere in the referent from those that adhere in the representation. For example, seeing one’s own well-known dog photographed at long and short viewing distances, with color and black-and-white film, with high and low speed film, at slow and fast shutter speeds, in direct and indirect lighting, and so on should draw attention to photographic surfaces and techniques. Third, I suggested that representational understanding would be developed by socially mediated experiences that draw the child’s attention to surface features of representations. As noted earlier in this chapter (see Section II.D), most people seem to focus on the content of photographs


37

rather than on their identity as graphic and aesthetic representations. Informal observation (in need of empirical test) likewise suggests that adults tend to direct their children’s attention similarly. For example, when looking at the family photo album, parents seem more likely to ask questions about people or events (e.g., ‘‘Can you find Aunt Jane?’’ or ‘‘Who is playing baseball?’’) than about the medium (e.g., ‘‘Why does Aunt Jane’s foot look so big in this photograph?’’ or ‘‘Why does the baseball bat look so blurry?’’). Past research has documented dramatic individual differences among parents in the degree to which they scaffold their children’s attention to an understanding of the drawings in picture books, and of the relation between these parental behaviors and children’s cognitive outcomes (Szechter & Liben, 2002). Additional research is needed to evaluate the ways in which parents may similarly scaffold their children’s understanding and appreciation of photographs both as spatial and aesthetic representations (see Szechter, 2003). Fourth, I suggested that understanding representations would be enhanced by experience in creating representations or in learning about graphic techniques. Consistent with this general prediction in the realm of drawing are Callaghan’s findings (1999; Callaghan & Rankin, 2002) that children who were given drawing lessons in producing graphic symbols show better performance in comprehending others’ symbols. Data consistent with this point in the realm of video photography are found in Troseth’s (2003) research showing that when 2-year-old children were given experience with live photography in their homes, their performance was facilitated on a symbolic representational task given in the laboratory two weeks later. Positive effects of experience with the techniques of still photography are evident from the data reported earlier on the impact that photography curriculum had on college students’ responses to photographs (see Section III.D) and from research described elsewhere (Liben & Szechter, 1999) showing that 8-year-old children who were given photography lessons performed better on spatial representational tasks than did children who had not received photography lessons. Finally, I suggested that children’s understanding of representations might be enhanced by exposure to ego-deictic representations, that is, representations that in some way point to themselves as representations. From the realm of drawing and painting, illustrations of ego-deictic materials include trompe l’oeil art and etchings by Escher (1992) that depict spatially impossible relations. Within the realm of photography it is more difficult to identify ego-deictic representations because of the changing nature of the technology. For example, when photographs were first introduced, their identity as representations was probably very salient. In the ensuing decades, as photographs became commonplace, their representational nature was probably relatively invisible. The introduction of

38

Lynn S. Liben

digital photography has probably again served to make the representational nature of photographs salient. As the power to manipulate images spreads, assumptions that photographs are records of ‘‘truth’’ (e.g., see Beilin, 1991; O’Connor, Beilin, & Kose, 1981; Price & Wells, 1997; Turner, 1987) may be expected to give way to assumptions that photographs are manipulable, symbolic representations. Earlier developmental research had demonstrated that young children ‘‘mistakenly’’ believe that photographs are updated to represent changing realities (Thomas, Nye, & Robinson, 1994; Zaitchik, 1990). Perhaps new technologies will lead us to re-conceptualize what we take to be a mature understanding of the photographic medium. C. THE ROLE OF CONSTRUCTIVE PROCESSES

As illustrated by the five experiential factors suggested above, the perspective I am advancing here is consistent with the general position that socially mediated experiences are critical for developmental outcomes (e.g., Gauvain, 2001; Vygotsky, 1978). An endorsement of the importance of socially driven experience is not, however, meant to deny the importance of constructive, self-directive processes (e.g., Piaget, 1970). Indeed, I have argued that child-driven processes assume a central position, as evidenced in the Embedded Model proposed to characterize the child’s developing understanding of external spatial representations (Liben, 1999). Particularly important in this model is the explicit point that children’s developing perceptual and cognitive processes affect not only the way that the child interacts with actual objects, but also the way that the child interacts with representations of those objects. Implicit in the model is the notion that these perceptual and cognitive processes are themselves undergoing constructive restructuring. Thus, for example, as children’s facility with projective and metric spatial concepts develops (e.g., see Liben, 2002), children would be expected to show improved understanding of vantage points in photographs as well. Finally, the Embedded Model proposes that effects are reciprocal: qualities of the child both affect—and are affected by—interactions with referents and their representations. D. A DEVELOPING PICTURE

In conclusion, my hope is that the concepts, methods, and data described in this chapter will provide a useful basis for future empirical work. But even stronger is my hope that they will support the more general point that the developmental study of photography is not some highly specialized and isolated substantive domain, but is instead a domain that speaks (and listens) to many fundamental concerns of developmental psychology,


39

including issues of representation, space, communication, perception, parent–child interaction, and individual and group identity. Just as the study of photographic development has much to offer basic research in developmental psychology, it also has much to offer to those interested in applying developmental psychology to education. It is all too common to dismiss education in graphic media like photography as being part of an expendable luxury of art education. Such a view may be challenged on two grounds. First, in the emotional realm, art is an essential piece of our humanity. Indicative are the dramatic descriptions of concentration camp inmates maintaining their lives through art and the many reports of the respite that art provided after 9/11. Second, in the cognitive realm, the graphic domain provides an important inroad to fostering representational and spatial thinking. In an era in which we rely increasingly on graphic representations, skill in understanding graphic spatial representations is arguably as important as skill in understanding verbal and numerical representations. I thus end this chapter where I began, by urging readers to look up from the words on this page. Photographs, and spatial graphic representations more generally, are worth bringing into scholarly focus.

ACKNOWLEDGMENTS My association with the Social Science Research Council Program on the Arts, funded through the Rockefeller Foundation, has provided an important context for much of the work described in this chapter. I am grateful to these institutions, and particularly to Frank Kessel (of SSRC) and Joan Shigekawa (of the Rockefeller Foundation) for their support. Likewise, I thank fellow members of the SSRC Arts Committee and of the SSRC working group on photography, and my Penn State colleagues Lisa Szechter (Department of Psychology) and David Ebitz (School of Visual Art) for their far-reaching contributions to my thinking about photography. Finally, I express my gratitude to my father, Jay Liben, for modeling the pleasures of a photographic eye, and for fostering my own love of photography from my Starflex Brownie Camera to my Nikon. It is only one of the many joys he has given me in life.

REFERENCES Beilin, H. (1991). Developmental aesthetics and the psychology of photography. In R. M. Downs, L. S. Liben, & D. S. Palermo (Eds.), Visions of aesthetics, the environment, and development: The legacy of Joachim F. Wohlwill (pp. 45–86). Hillsdale, NJ: Lawrence Erlbaum Associates. Beilin, H. (1999). Understanding the photographic image. Journal of Applied Developmental Psychology, 20, 1–20. Bordieu, P. (1965/1990). Photography: A middle-brow art. Stanford, CA: Stanford University Press. Originally published as Un art moyen: Essai sur les usages sociaux de la photographie. Les Editions de Minuit.

40

Lynn S. Liben

Callaghan, T. C. (1999). Early understanding and production of graphic symbols. Child Development, 70, 1314–1324. Callaghan, T. C., & Rankin, M. P. (2002). Emergence of graphic symbol functioning and the question of domain specificity: A longitudinal training study. Child Development, 73, 359–376. Church, J. (1961). Language and the discovery of reality. New York: Random House. Corrin, L. G. (Ed.) (2001). Mining the museum: An installation by Fred Wilson. New York: The New Press. Cosden, C., & Reynolds, D. (1982). Photography as therapy. Arts in Psychotherapy, 9, 19–23. DeLoache, J. S., Pierroutsakos, S. L., Uttal, D. H., Rosengren, K. S., & Gottlieb, A. (1998). Grasping the nature of pictures. Psychological Science, 9, 205–210. Deregowski, J. (1968). Pictorial recognition in subjects from a relatively pictureless environment. Africa Social Research, 5, 356–364. Dewdney, A., & Isherwood, S. (1994). Photography in education. In S. Isherwood & N. Stanley (Eds.), Creating vision: Photography & the national curriculum (pp. 21–45). Manchester, UK: The Arts Council of Great Britain. Dewdney, A., & Lister, M. (1988). Youth, culture, and photography. Houndsmills, UK: Macmillan Education. Downs, R. M. (1981). Maps and mappings as metaphors for spatial representation. In L. S. Liben, A. H. Patterson, & N. Newcombe (Eds.), Spatial representation and behavior across the life span: Theory and application (pp. 143–166). New York: Academic Press. Eisenberg, J., & Liben, L. S. (1995, June). Children’s understanding of photography: Effects of spatial concepts and technical instruction. Poster presented at the annual symposium of the Jean Piaget Society, Berkeley. Escher, M. C. (1992). The graphic work. Germany: Benedikt Taschen. Ewald, W. (2001). I wanna take me a picture. Boston: Beacon Press. Freeman, N. H. (1995). The emergence of a framework theory of pictorial reasoning. In C. Lange-Ku¨ttner & G. V. Thomas (Eds.), Drawing and looking (pp. 135–146). New York: Harvester Wheatsheaf. Freeman, N. H., & Parsons, M. J. (2001). Children’s intuitive understandings of pictures. In B. Torff & R. S. Sternberg (Eds.), Understanding and teaching the intuitive mind: Student and teacher learning (pp. 73–91). Mahwah, NH: Lawrence Erlbaum Associates. Gardner, H. (1970). Children’s sensitivity to painting styles. Child Development, 41, 813–821. Gardner, H., & Gardner, J. (1970). Developmental trends in sensitivity to painting style and subject matter. Studies in Art Education, 12, 11–16. Gauvain, M. (2001). The social context of cognitive development. New York: Guilford. Gibson, J. J. (1980). Foreword: A prefatory essay on the perception of surfaces versus the perception of markings on a surface. In M. A. Hagen (Ed.), The perception of pictures (Vol. 1, pp. xi–xviii). New York: Academic Press. Giuffre, K. (2001). Mental maps: Social networks and the language of critical reviews. Sociological Inquiry, 71, 381–393. Igou, B. (2001). The Amish & photographs. Amish Country News, Retrieved January 1, 2003, from http://www.amishnews.com/amisharticles/amishand%20photos.htm. Halle, D. (1991). Displaying the dream: The visual presentation of family and self in the modern American household. Journal of Comparative Family Studies, 22, 217–229. Ittleson, W. H. (1996). Visual perception of markings. Psychonomic Bulletin and Review, 3, 171–187. Kose, G. (1985). Children’s knowledge of photography: A study of the developing awareness of a representational medium. British Journal of Developmental Psychology, 3, 373–384.


41

Kose, G., Beilin, H., & O’Connor, J. M. (1983). Children’s comprehension of actions depicted in photographs. Developmental Psychology, 19, 636–643. Kress, G. & van Leeuwen, T. (1996). Reading images. London: Routledge. Krippner, S. (1973). Sentics, Kirlian photography, and psychosomatic illness. Gifted Child Quarterly, 17, 293–296. Levine, M., Marchon, I., & Hanley, G. (1984). The placement and misplacement of You-AreHere maps. Environment and Behavior, 16, 139–158. Liben, L. S. (1981). Spatial representation and behavior: Multiple perspectives. In L. S. Liben, A. H. Patterson, & N. Newcombe (Eds.), Spatial representation and behavior across the life span: Theory and application (pp. 3–36). New York: Academic Press. Liben, L. S. (1999). Developing an understanding of external spatial representations. In I. E. Sigel (Ed.), Development of mental representation: Theories and applications (pp. 297–321). Mahwah, NJ: Lawrence Erlbaum Associates. Liben, L. S. (2001). Thinking through maps. In M. Gattis (Ed.), Spatial schemas and abstract thought (pp. 44–77). Cambridge, MA: MIT Press. Liben, L. S. (2002). Spatial development in children: Where are we now? In U. Goswami (Ed.), Blackwell handbook of childhood cognitive development (pp. 326–348). Oxford, UK: Blackwell Publishers. Liben, L. S., & Downs, R. M. (1989). Understanding maps as symbols: The development of map concepts in children. In H. W. Reese (Ed.), Advances in child development and behavior (Vol. 22, pp. 145–201). New York: Academic Press. Liben, L. S., & Downs, R. M. (1992). Developing an understanding of graphic representations in children and adults: The case of GEO-graphics. Cognitive Development, 7, 331–349. Liben, L.S., Kastens, K. A., & Stevenson, L. M. (2002). Real-world knowledge through realworld maps: A developmental guide for navigating the educational terrain. Developmental Review, 22, 267–322. Liben, L. S., & Szechter, L. E. (1999, October). Teaching children photography. Poster presented at the biennial meeting of the Cognitive Development Society, Chapel Hill. Liben, L. S., & Szechter, L. E. (2001, October). Understanding the spatial qualities of photographs. In L. S. Liben (Chair), Cognitive development: A photographic view. Symposium conducted at the biennial meeting of the Cognitive Development Society, Virginia Beach, VA. Liben, L. S., & Szechter, L. E. (2002). A social science of the arts: An emerging organizational initiative and an illustrative investigation of photography. Qualitative Sociology, 25, 385–408. London, B., & Upton, J. (1998). Photography (Sixth Edition). New York: Longman. Newbury, D. (1996). Reconstructing the self: Photography, education and disability. Disability & Society, 11, 349–360. Ninio, A., & Bruner, J. (1978). The achievements and antecedents of labeling. Journal of Child Language, 5, 1–15. O’Connor, J., Beilin, H., & Kose, G. (1981). Children’s belief in photographic fidelity. Developmental Psychology, 17, 859–865. Parsons, M. J. (1987). How we understand art: A cognitive developmental account of aesthetic experience. Cambridge: Cambridge University Press. Piaget, J. (1970). Piaget’s theory. In P. Mussen (Ed.), Carmichael’s manual of child psychology (pp. 703–732). New York: Wiley. Price, D., & Wells, L. (1997). Thinking about photography. In L. Wells (Ed.), Photography: A critical introduction (pp. 9–64). London: Routledge. Seidman, S., & Beilin, H. (1984). Effects of media on picturing by children and adults. Developmental Psychology, 20, 667–672.

42

Lynn S. Liben

Sigel, I. E. (1978). The development of pictorial comprehension. In B. S. Randhawa & W. E. Coffman (Eds.), Visual learning, thinking, and communication (pp. 93–111). New York: Academic Press. Sontag, S. (1977). On photography. New York: Anchor Books. Sroka, I. (1995). Young children’s photographic practice: Representation and world knowledge. Doctoral dissertation, Graduate School, City University of New York, NY. Szarkowski, J. (2001). Ansel Adams at 100. Boston: Bullfinch Press. Szechter, L. E. (2003). Artistic expertise, parent–child interaction, and the development of aesthetic sensitivity. Doctoral dissertation, Pennsylvania State University. Szechter, L. E., & Liben, L. S. (2002). Children’s understanding of spatial-graphic representations: The role of parental guidance. Unpublished manuscript, Pennsylvania State University. Szechter, L. E., & Liben, L. S. (2003, May). Parent–child interaction and the development of aesthetic awareness. Poster presented at the convention of the American Psychological Society, Atlanta. Szechter, L. E., Liben, L. S., & Rogers, J. E. (1998, August). Children’s understanding of photographs: Interventions and implications. Poster presented at the annual convention of the American Psychological Association, San Francisco. Thomas, G. V., Nye, R., & Robinson, E. (1994). How children view pictures: Children’s responses to pictures as things in themselves and as representations of something else. Cognitive Development, 9, 141–164. Thomas, G. V., Sharples, M., Davison, L., & Rudman, P. (2001, June). Children as photographers 2000. Birmingham, UK: University of Birmingham. Troseth, G. L. (2003). TV guide: Two-year-old children learn to use video as a source of information. Developmental Psychology, 39, 140–150. Turnbull, D. (1989/1993). Maps are territories: Science is an atlas. Chicago: University of Chicago Press (originally published in Victoria, Australia by Deakin University Press). Turner, P. (1987). History of photography. Greenwich, CT: Brompton. Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Cambridge, MA: Harvard University Press. Walther, T. (2000). Other pictures: Anonymous photographs from the collection of Thomas Walther. Sante Fe, NM: Twin Palms Press. Winner, E. (1982) Invented worlds: The psychology of the arts. Cambridge, MA: Harvard University Press. Wolf, S. (1999). Kenneth Josephson: A retrospective. Chicago: The Art Institute of Chicago. Zaitchik, D. (1990). When representations conflict with reality: The preschooler’s problem with false beliefs and ‘‘false’’ photographs. Cognition, 35, 41–68.

PROBING THE ADAPTIVE SIGNIFICANCE OF CHILDREN’S BEHAVIOR AND RELATIONSHIPS IN THE SCHOOL CONTEXT: A CHILD BY ENVIRONMENT PERSPECTIVE

Gary W. Ladd DEPARTMENT OF PSYCHOLOGY AND DEPARTMENT OF FAMILY AND HUMAN DEVELOPMENT ARIZONA STATE UNIVERSITY, TEMPE, ARIZONA 87287

I. INTRODUCTION A. DEFINING ADJUSTMENT IN THE SCHOOL CONTEXT B. INVESTIGATING THE INTERPERSONAL ANTECEDENTS OF SCHOOL ADJUSTMENT II. PRIMARY PREMISES AND FINDINGS FROM RESEARCH GUIDED BY ‘‘MAIN EFFECTS’’ MODELS A. MAIN EFFECTS PREMISE 1: CHILDREN WITH AGGRESSIVE AND WITHDRAWN BEHAVIORAL STYLES ARE PRONE TOWARD MALADJUSTMENT B. MAIN EFFECTS PREMISE 2: CHILDREN WITH POOR PEER RELATIONSHIPS ARE PRONE TOWARD MALADJUSTMENT III. PRIMARY PREMISES AND FINDINGS FROM RESEARCH GUIDED BY ‘‘CHILD BY ENVIRONMENT’’ MODELS A. CHILD BY ENVIRONMENT PREMISE 1: CHILDREN’S BEHAVIORAL STYLES AND THEIR PARTICIPATION IN PEER RELATIONSHIPS CONTRIBUTE TO SCHOOL ADJUSTMENT B. CHILD BY ENVIRONMENT PREMISE 2: IN CONJUNCTION WITH BEHAVIORAL STYLES, ENDURING PEER RELATIONSHIPS HAVE GREATER EFFECTS ON CHILDREN’S ADJUSTMENT THAN BRIEF OR TRANSIENT PEER RELATIONSHIPS IV. SUMMARY AND CONCLUSIONS A. THEORY AND EVIDENCE: GOODNESS OF FIT? B. CONCEPTUAL AND EMPIRICAL AGENDAS C. IMPLICATIONS FOR FUTURE RESEARCH REFERENCES


43


44

Gary W. Ladd

I. Introduction A central aim articulated by advocates of a national program of prevention research (see Coie et al., 1993) is to supplant simple predictive models of child health and dysfunction with paradigms that offer a developmental-process explanation of human adaptation and its role in shaping children’s adjustment. This position is based on the criticism that ‘‘main effects’’ models, in which adjustment has been forecasted from either early child characteristics (i.e., ‘‘child effects models’’) or environmental inputs (e.g., ‘‘socialization effects models’’), only partially account for the mechanisms that are responsible for later health or dysfunction. An alternative perspective that researchers have begun to embrace is based on the view that human adaptation is a complex, developmental process, one that is triggered by multiple factors that are both internal and external to the child, and instrumental in shaping the child’s development over the life course. Paradigms that encompass this view have often been termed ‘‘child by environment models’’ or ‘‘child and environment models’’ because development is seen as progressing along complex pathways in which characteristics of the child and of the child’s milieu combine in interactive or transactional patterns to shape child outcomes. In the 1990s, researchers began to use child by environment perspectives to formulate hypotheses about how specific child characteristics and environmental factors combine in ways that influence children’s developmental trajectories, or orient them toward early- and later-emerging forms of adjustment. Many of these propositions are relatively novel and have received little or no empirical attention; moreover, few have been tested using prospective longitudinal data. Accordingly, my aims for this chapter are to: (1) delineate a series of premises that follow not only from the assumptions of main effects models but also from child and environment models, (2) review existing empirical support for these tenets, and (3) consider how information from this analysis can be used to create a new generation of frameworks that may guide researchers toward a better understanding of human adaptation and its role in shaping children’s adjustment. Each of these aims will be addressed in the context of research on children’s psychological and school adjustment. Understanding how children adapt psychologically and scholastically in the school context has been an important objective for researchers interested in the promotion of competence and the prevention of educational and psychological problems. Moreover, the school is an important context within which to evaluate the merits of child and environment premises because this setting provides access to a large

A Child by Environment Perspective

45

proportion of the population that is at risk for disorder at early stages in children’s development. Of children aged 5–18 years in the USA, 9 out of 10 attend school, and between 7.5 and 14 million (12%–22%) are clinically maladjusted (Achenbach & Edelbrock, 1981; Coie et al., 1993). Moreover, adjustment during the school years foreshadows many facets of child, adolescent, and adult health and dysfunction (Kohlberg, LaCrosse, & Ricks, 1972; Parker & Asher, 1987). Thus, schools are an ideal context for testing hypotheses about the origins and trajectories of children’s adjustment, and evidence gathered during the early school years is likely to have important implications for prevention research and programming.

A. DEFINING ADJUSTMENT IN THE SCHOOL CONTEXT

By design, schools are intended to be challenging contexts for children. Much of what we know about school as a developmental and socialization context, however, pertains to the instructional and curricular challenges that children face in classrooms, and their links with learning and achievement outcomes. Because the promotion of scholastic competence is one of the primary purposes of schooling, measures of achievement can be interpreted as evidence of how well the child has adjusted to this specific challenge. However, we can also conceive of schooling as a context that presents a complex array of challenges—demands that are both scholastic and interpersonal in nature—all of which require some degree of adaptation on the part of the child (see Ladd, 1989, 1996; Perry & Weinstein, 1998). These diverse challenges create demands that—depending on the child and the resources or constraints that are operating in the child’s environment—may affect children on many different levels, all of which may have important implications for their development and performance in the school context. Thus, we can define school adjustment as a multifaceted construct that encompasses more than just achievement as an indicator of how well children have adjusted to the school environment. For example, classroom challenges may be instrumental in shaping children’s: (a) perceptions and appraisals of school and the classroom environment, (b) psychological and emotional reactions in this context, including both internalizing and externalizing problems, (c) involvement and disengagement in classroom activities, and (d) achievement and academic progress. On the basis of this logic, I have argued that a more holistic assessment of school adjustment should include indicators of children’s adaptive success within each of these domains (see Ladd, 1989, 1996; Ladd & Burgess, 2001).

46

Gary W. Ladd

B. INVESTIGATING THE INTERPERSONAL ANTECEDENTS OF SCHOOL ADJUSTMENT

Only since the 1990s have researchers begun to examine empirically the possibility that multiple interpersonal factors—such as children’s social behavior in classrooms and relationships with peers in this context—may have an important bearing on their school adjustment. This is surprising when one considers that education is, in many respects, a social enterprise. Social interactions and relationships are an essential part of children’s learning and school experience. Increasingly, the application of curricula and other elements of educational programming have become dependent on peer-mediated activities, such as group exploration and learning activities, learning centers, peer collaboration and tutoring, and competitive and cooperative learning groups. Furthermore, especially among young children, relations with classmates may play an important role in establishing the psychological conditions (e.g., curiosity, interest, attention, motivation, support, and security) that enable children to identify or ‘‘bond’’ with school and learn in this context (Birch & Ladd, 1996). The remaining sections of this chapter are organized in a manner that is consistent with the aims that I have outlined above, and with the progression of scientific thinking that has guided the search for behavioral and environmental antecedents of children’s psychological and school adjustment. The purpose of Section II is to overview premises and findings that are central to main effects models. Of particular interest are empirical attempts to corroborate the following premises: (1) certain behavioral propensities in children constitute enduring styles of acting upon and reacting to environmental contingencies that, ultimately, lead to dysfunction, and (2) adverse peer relationships are enduring features of children’s social lives, and participation in such relationships antecedes maladjustment. Section III contains a survey of findings that have been generated by empirical applications of child by environment models, where the guiding premise is that the precursors of children’s adjustment originate not only within the child but also within the child’s relational environment. In this section, many of the behavioral dispositions and peer relationships that have been investigated from a main effects perspective are reexamined as potential conjoint rather than independent determinants of children’s psychological and school maladjustment. Finally, in Section IV, I attempt to summarize the extent to which main effects and child by environment models are corroborated by empirical evidence. I also consider the utility of these models, along with promising variants of the child by environment perspective, as purveyors of future empirical discoveries.


47

II. Primary Premises and Findings from Research Guided by ‘‘Main Effects’’ Models Initially, attempts to identify the interpersonal precursors of children’s psychological and school adjustment were guided by the logic of ‘‘main effects’’ models. Premises generated from this conceptual perspective tended to emphasize the causal priority of a single or closely related set of interpersonal variables as precursors of children’s adjustment. From the late 1960s through the 1980s, the interpersonal factors that received the most empirical attention as potential antecedents of children’s adjustment included two specific behavioral dispositions or styles (i.e., aggressive, withdrawn behavior), and one form of peer relationship (i.e., low peer group acceptance/peer rejection). Of these two factors, one can be viewed as an attribute of the child (i.e., a child’s behavioral dispositions) and the other as a contextual, relationship construct (i.e., a peer group’s rejecting attitudes/responses toward a child). Although both of these interpersonal dimensions were hypothesized to be precursors of later maladjustment, researchers typically investigated them separately, which encouraged the development of distinct research hypotheses, literatures, and traditions. A brief overview of these traditions, organized by exemplary premises and findings, follows. A. MAIN EFFECTS PREMISE 1: CHILDREN WITH AGGRESSIVE AND WITHDRAWN BEHAVIORAL STYLES ARE PRONE TOWARD MALADJUSTMENT

A sizeable proportion of the research designed to illuminate the role of behavioral styles in children’s adjustment was focused on aggressive and withdrawn behavioral dispositions. Of these two behavioral styles, aggression more than withdrawal has been an enduring focal point for theory and research on child development and adjustment. Consequently, the scope of the evidence that has accrued on aggressive behavior over the last several decades is greater than that which has been assembled on withdrawn behavior. 1. Evidence Linking Children’s Aggressive Behavior with Later Adjustment The premise that aggressive behavior is a manifestation of a specific disposition of children has been central to the logic of investigations based on a main effects model of adjustment. Whether an aggressive behavioral style is primarily attributable to inheritance, learning, or some transaction

48

Gary W. Ladd

between the two, the view proscribed by this model is that the propensity to engage in aggressive acts becomes, for many children, an enduring style of acting upon and reacting to environmental contingencies that, ultimately, leads to dysfunction (e.g., see Caspi, Elder, & Bem, 1987; Moss & Susman, 1980; Olweus, 1979). Substantiation of this position requires, in part, evidence of stable individual differences in children’s aggressive behavior. A sizeable number of longitudinal investigations have produced evidence consistent with the view that individual differences in aggressive behavior are stable over time. This is remarkable in that, for many psychological constructs, the magnitudes of stability coefficients typically decline dramatically as the lags between assessment intervals increase. One of the longest stability estimates ever obtained came from a 20-year follow-up of children who were prone to ‘‘explosive’’ behavior during childhood. In this study, Caspi, Elder, and Bem (1987) estimated that severe childhood tantrums correlated .27 with adult irritability and .45 with adult undercontrolled behavior. Similar or higher stability coefficients were obtained in more recent prospective longitudinal studies with measures that tap more differentiated forms of aggression. For example, Cairns et al. (1989) gathered yearly assessments of children’s aggressive behavior from teacher- and self-report measures across grades four through nine. In general, the obtained stability estimates declined only slightly over the 5-year span, were stronger for boys than for girls, and were highest when calculated from teacher- rather than self-report measures of aggression. To be specific, the reported stability coefficients for the teacher measure were .72 for boys and .33 for girls from grades 4–5, and .45 for boys and .51 for girls across the entire 5-year interval. Based on additional follow-up data, Cairns and Cairns (1994) further estimated that across the 8 years from grades 4 through 12 the median stability coefficients for boys and girls were .46 and .30, respectively. Coefficients of a similar magnitude have also been obtained with samples of much younger children. Ladd and Burgess (1999), for example, followed two moderately large samples of children from kindergarten to second grade and estimated the stability of aggressive behavior using a teacher-report measure. In one sample, the stability coefficients for aggressive behavior were .50 from kindergarten to grade 1, and .43 from kindergarten to grade 2. In the second sample, slightly higher coefficients were found across the same two intervals: .56 and .52, respectively. Subsequently, Ladd and colleagues estimated the stability of young boys’ and girls’ aggressive behavior over an even longer interval, from kindergarten to grade 6, using both teacher- and peer-report measures (see Table I). Across this 6-year period, the stability coefficients for the teacher-report measure were .45 for boys and .48 for girls. Using the peer-report measure, the coefficients were .51 for boys and .60 for girls.

49

A Child by Environment Perspective TABLE I Stability Coefficients for Aggressive Behavior from Kindergarten through Grade 6 by Gender and Type of Informant Boys Teacher reports Grade K 1 2 3 4 5

Peer reports

1

2

3

4

5

6

1

2

3

4

5

6

.59

.49 .53

.54 .50 .67

.53 .51 .54 .55

.59 .44 .49 .56 .64

.45 .45 .45 .47 .51 .61

.74

.64 .65

.63 .61 .78

.45 .52 .48 .52

.44 .56 .62 .62 .58

.51 .49 .57 .67 .50 .67

Girls Grade K 1 2 3 4 5

1

2

3

4

5

6

1

2

3

4

5

6

.53

.54 .57

.63 .62 .63

.59 .41 .46 .62

.33 .51 .33 .24 .33

.48 .33 .48 .44 .45 .30

.72

.76 .71

.75 .73 .77

.63 .61 .65 .65

.56 .55 .63 .55 .76

.56 .54 .57 .56 .58 .60

Note: The findings contained in this table were obtained from the Pathways Project (G. W. Ladd, Principal Investigator), and were first reported in an unpublished progress report submitted to the National Institute of Mental Health pursuant to grant 2-RO1MH-49223.

Thus, there is considerable support for the premise that individual differences in children’s aggressive behavior are stable over time. Moreover, findings that reflect on this proposition suggest that individual differences in children’s aggressive behavior may remain stable over many years, even into adulthood. A second tenet integral to the main effects perspective is that individual differences in aggressiveness during childhood predict later maladjustment. Historically, the association between aggressive behavior in childhood and later maladjustment has been examined both retrospectively in ‘‘followback’’ studies and predictively, either in ‘‘follow-forward,’’ or ‘‘prospective’’ longitudinal studies. Most investigators consider data gathered with followforward, and especially prospective longitudinal designs, to be more reliable than that obtained from follow-back designs (for a review of the merits and disadvantages of longitudinal designs for prediction studies, see Parker & Asher, 1987).

50

Gary W. Ladd

The principal aim in previous longitudinal studies has been to determine whether aggressiveness in childhood is associated with delinquency, psychopathology, or other forms of psychological maladjustment during adolescence or adulthood. In these studies, aggression toward agemates consistently has emerged as a predictor of later forms of maladjustment, particularly antisocial offenses in adolescence and adulthood (see reviews by Coie & Dodge, 1998; Coie, Dodge, & Kupersmidt, 1990; Kohlberg, LaCrosse, & Ricks, 1972; Parker & Asher, 1987). As illustrated in the sections that follow, the question of whether early aggressiveness is linked with later school maladjustment also has been investigated, but not as thoroughly. a. Antisocial Conduct and Psychopathology. Early studies of aggression as a predictor of dysfunction were primarily conducted with males and carried out with follow-back longitudinal designs. In general, these studies showed that boys who were known to be perpetrators of delinquent and criminal acts in adolescence often had histories of aggressive behavior that began in childhood (e.g., Conger & Miller, 1966; Mulligan et al., 1963). These findings were replicated and extended in a second wave of longitudinal studies that were conducted with follow-up designs (Ensminger, Kellam, & Rubin, 1983; Farrington, 1979; Magnussen, Stattin, & Duner, 1983). For example, Magnussen, Stattin, and Duner (1983) found that half or more of the boys they identified as aggressive in preadolescence (depending on the nature of the offense) were in trouble with the law during adolescence, as compared with only about 5% of boys who had previously been classified as nonaggressive. Other findings from early follow-back studies implicated childhood aggression (and, sometimes, withdrawn behavior patterns) as a precursor of adolescent and adult psychopathology, including schizophrenia (Flemming & Ricks, 1970). Linkages such as these, however, often were not corroborated by the results from later follow-up longitudinal studies (see Janes & Hesselbrock, 1978; Robbins, 1966), leading many researchers to conclude that aggressive behavior was more closely linked with conduct problems than with psychopathology (e.g., Kohlberg, LaCrosse, & Ricks, 1972; Kupersmidt, Dodge, & Coie, 1990; Parker & Asher, 1987). Although rare, a few of the studies conducted during this era were designed to distinguish the predictive efficacy of aggression as compared to other, related forms of ‘‘deviant’’ behavior. However, as a whole, findings from these studies were not uniform, possibly due to differences in the design and analyses, and the types of predictors and criteria that were utilized in these investigations. Moore, Chamberlain, and Mukai (1979), for example, found that children’s aggression in the family context was not as


51

effective as other antisocial behaviors (e.g., theft) for predicting future juvenile delinquency. In contrast, Roff and Wirt (1984) found that aggression that children exhibited in the peer context, as distinct from other forms of aggression, forecasted later conduct problems, even after controlling for other forms of earlier antisocial behavior. Evidence gathered in modern, prospective longitudinal studies has further elucidated what is known about the predictive links from early aggression to later maladjustment. It has become increasingly clear, for example, that aggressiveness not only predicts involvement in later externalizing problems, but also predisposes children toward developmental trajectories in which minor offenses often evolve into serious, frequent, and diversified forms of crime and violence (e.g., see Loeber & Farrington, 2000; Moffit, 1993; Tolan & Gorman-Smith, 1998). Although the manifestations of aggression that mark the onset of these trajectories may be traceable to early adolescence (i.e., late-onset trajectories), an increasing corpus of evidence indicates that similar markers can be found at very early ages (e.g., fighting with peers during preschool, kindergarten, and grade school; Dodge, 1991; Lahey et al., 1998; Tremblay et al., 1994; White et al., 1990). For example, Coie et al. (1989) found that aggression among third grade boys forecasted higher levels of self-reported conduct disorders by the end of sixth grade. Beginning with samples of fourth-graders, Cairns and Cairns (1994) compared subgroups of assaultive and violent children with matched controls and found that the aggressive groups were more likely to manifest multiple forms of maladjustment in adolescence and adulthood, including criminal arrests, teenage motherhood, and participation in residential psychiatric care. Somewhat more surprising are findings indicating that aggressiveness during early adolescence antecedes some forms of internalizing problems, such as suicide. Cairns and Cairns (1994), for example, found that aggressive adolescent males were prone to engage in life-threatening ‘‘games’’ and accidents, and were at greater risk than the norm to harm themselves physically. These findings were interpreted, in part, to a propensity for aggressive boys to act impulsively and rely on violence as a coping mechanism. Another trend was to parse the aggression construct into qualitative subtypes and investigate whether the different forms of aggression children exhibit are differentially linked with maladjustment. Coie et al. (1991), for example, proposed three subtypes, labeled instrumental, bullying, and reactive aggression. Others (see Cairns & Cairns, 1994; Crick & Grotpeter, 1995; Galen & Underwood, 1997; Lagerspetz, Bjorkqvist, & Peltonen, 1988) drew distinctions between aggression directly expressed toward others (e.g., direct, confrontational, or overt aggression) and aggression performed in less direct ways (e.g., covert, indirect, social, or relational aggression). At present, there is evidence to suggest that, over short periods of time, both

52

Gary W. Ladd

direct and indirect forms of aggression are moderately stable and predictive of maladjustment (Crick, 1996). Moreover, in an investigation with college students, Werner and Crick (1999) found that relational forms of aggression were associated with the symptoms of borderline personality disorders. These findings are, however, complicated by the fact that investigators have often found that these subtypes of aggressive behavior are substantially correlated (e.g., Dodge & Coie, 1987; Crick & Grotpeter, 1995). Whether measures of these subtypes of childhood aggression are of greater utility than aggregated measures of aggression for forecasting later maladjustment (cf. Rushton, Brainerd, & Pressley, 1983) is not yet well understood. Thus, the extent to which qualitatively different forms of aggression have differential predictive utility, or are useful for forecasting different types of adjustment outcomes, requires further empirical evaluation—especially in the context of long-term prospective longitudinal studies. b. School Adjustment. The predictive association between early aggression and later school adjustment has received less attention relative to other forms of health or dysfunction, and the evidence gathered on this criterion has typically been restricted to a narrow range of indicators such as high school drop-out rates, absences, truancy, and grade retentions. For example, in a follow-back longitudinal study, Lambert (1972) found that many high school children who exhibited adjustment problems (e.g., discipline problems, remedial instruction, dropping out of school) had been previously nominated by their third-grade classmates for negative (often aggressive) roles on a ‘‘class play’’ measure. Even stronger evidence of this linkage has been obtained in longitudinal studies that were conducted with a follow-up design (e.g., Feldhusen, Thurston, & Benning, 1973; Havighurst et al., 1962). Havighurst et al., for example, found that children who were identified as aggressive in grades six and seven, as compared with their nonaggressive schoolmates, were much less likely to complete high school. Some of these same linkages have been replicated in modern, prospective longitudinal studies (e.g., see Cairns & Cairns, 1994; Cairns, Cairns, & Neckerman, 1989; Kupersmidt, Dodge, & Coie, 1990). Findings reported by Cairns and Cairns (1994) showed that, across the period of seventh through eleventh grade, aggressive children were more likely to drop out of school than were nonaggressive children, or children who were members of families from lower SES backgrounds. However, other findings from the Cairns and Cairns (1994) study and from other studies (see Ensminger & Slusarcick, 1992) qualify these results by showing that children’s failure to complete high school was not predicted by aggression alone. Rather, these studies revealed that the combination of aggressive behavior and poor school performance was a better predictor of dropping out of high school. For example, in the


53

Cairns and Cairns (1994) study, children who could be identified as aggressive in grade 7 were even more likely to drop out of high school if they had had a history of academic difficulties (e.g., low achievement, frequent grade retentions). In the interest of establishing an empirical base for school-based prevention programs, many researchers have investigated the links between aggression and school adjustment during the preadolescent and earlyadolescent years. Coie and colleagues (Coie et al., 1987; Coie et al., 1992), for example, assessed both aggression and peer rejection at third- and fourth-grade, and found that both variables were predictive of adolescents’ adjustment difficulties after the transition into middle school. In another prospective longitudinal study, Kupersmidt and Coie (1990) followed children from grades five through ten and found that aggressiveness among fifth-graders forecasted higher levels of delinquency among tenthgraders, and that both aggression and absences were associated with later drop-out rates. There is also evidence to suggest that, prior to children’s entry into adolescence, aggression is a better predictor of dropping out of school for boys than for girls. Cairns and Cairns (1994) found that, when measured in grade 4, aggression and academic difficulties better predicted dropping out of high school for boys than for girls. For girls, poor academic performance rather than aggression was more closely linked with failure to complete high school. The question of whether aggressive behavior is correlated with adjustment problems at the earliest stages of schooling—such as the transition from preschool into kindergarten and from kindergarten into the primary grades—has received much less attention. This is surprising, given that many psychologists believe that children’s initial school experiences are an important determinant of later adjustment and progress (Alexander & Entwisle, 1988; Ladd, 1989). Early school adjustment problems can have lasting or cumulative effects (Ladd, Buhs, & Seid, 2000); problems that arise early in children’s school careers are often perpetuated by socialpsychological factors (e.g., low self-esteem, reputational bias, self-fulfilling prophesies), or are exacerbated when nascent academic or social difficulties undermine later progress (Connell, Spencer, & Aber, 1994; Ladd, 1989; Perry & Weinstein, 1998). Those researchers who addressed this agenda have found that, as early as preschool and kindergarten, aggression is common in classrooms and is a significant predictor of later school adjustment problems (Ladd & Mars, 1986; Ladd & Price, 1987). Ladd and Price (1987), for example, found that children who perpetrated more negative peer contacts (i.e., were aggressive toward many rather than a few classmates) in preschool classrooms were more likely to develop later social difficulties, after they entered kindergarten.

54

Gary W. Ladd

Similarly, Ladd and Burgess (1999) found that aggressive kindergartners were more likely than their counterparts in a matched risk-comparison group to develop and maintain social difficulties with classroom peers and teachers throughout the early primary grades. Even more compelling were findings reported by Ladd and Burgess (2001). These investigators identified aggressive children (i.e., those who exceeded a risk criterion) during the initial weeks of kindergarten, and followed these children and their classmates prospectively over a 2-year period. Children’s initial risk status on aggression was used to predict changes in school adjustment across this period, as reflected on measures of attentional problems, thought problems, behavioral misconduct, classroom participation, school attitudes or sentiments, and achievement. Children with elevated levels of aggression at school entrance were more likely to display increases in school maladjustment across the primary school years. To be specific, aggressive children exhibited gains in thought problems, misconduct, and classroom disengagement, and they were prone to underachievement and the formation of negative school attitudes. Furthermore, children who tended to be chronically aggressive (i.e., remained aggressive over multiple assessment occasions) were even more likely to exhibit these and other forms of school maladjustment. In sum, based on these findings and other, more comprehensive evaluations of past and recent evidence (see Coie & Dodge, 1998; Parker & Asher, 1987; Parker et al., 1995; Tolan & Gorman-Smith, 1998), many scholars have concluded that aggression in childhood is a fairly robust predictor of later psychological maladjustment. Furthermore, there is evidence to suggest that aggressive children are at risk for school maladjustment, even at very early stages in their school careers. However, in view of the larger patterns of findings reported in the aggression literature, support for the aggression–maladjustment hypothesis appears strongest when the type of dysfunction predicted belongs to an analogous class of behaviors such as antisocial and violent acts (e.g., conduct problems, delinquency, and crime). Another principal conclusion that has been advanced in this literature is that aggressive children often follow a developmental trajectory in which early, small aggressive acts often develop into larger and more serious patterns of antisocial behavior, including violence and criminality (Coie & Dodge, 1998; Loeber, 1990; Magnussen, Stattin, & Duner, 1983; Moffit, 1993; Tolan & Gorman-Smith, 1998). 2. Evidence Linking Children’s Withdrawn Behavior with Later Adjustment In contrast to children’s aggressive behavior, far less is known about the risks associated with early solitary or withdrawn behavioral styles, and almost nothing is known about the relation between early social withdrawal


55

and school adjustment. Near the inception of this research domain, investigators documented that some children tended to avoid peers or interact only minimally with agemates (Reznick et al., 1986; Robbins, 1966), but there was considerable controversy about whether withdrawn children were at risk for adjustment difficulties (cf. Robbins, 1966; Wanlass & Prinz, 1982). Since that time, it has become apparent that children who interact infrequently with agemates can be identified as early as the toddler and preschool years (Rubin, 1985; Rubin, Burgess, & Hastings, 2002), and that these children differ from their normative counterparts in numerous ways. Early studies revealed that, compared with more sociable peers, withdrawn children received lower mental age scores, had underdeveloped problemsolving skills, engaged in higher levels of egocentric speech, made fewer requests of peers, complied more during peer interactions, and were more often ignored by peers (Rubin, 1982; Rubin & Borwick, 1984; Rubin, Daniels-Bierness, & Bream, 1984). Based on these and other findings, some researchers (e.g., Parker et al., 1995; Rubin, LeMare, & Lollis, 1990) proposed that social withdrawal prevents children from participating in the mainstream of peer interaction where many important social abilities are learned and refined. This lack of engagement, in turn, was seen as inhibiting children’s interpersonal maturity and preventing them from mastering social skills and coping responses that would allow them to adapt to changing developmental tasks and challenges. From a main effects perspective, a key assumption behind this logic is that withdrawn behavior is reflective of an enduring disposition of the child. This tenet remains under investigation, and is closely intertwined with researchers’ efforts to define subtypes of withdrawn behavior, and determine the level of risk associated with different types of solitary behavior (e.g., see Gazelle & Ladd, 2003; Harrist et al., 1997). Unfortunately, a precise and consistent terminology for labeling different forms of withdrawn or solitary behavior in children has yet to be developed. Based on studies conducted with preschoolers, Rubin and colleagues proposed that withdrawn behavior can be classified into subtypes termed ‘‘isolated’’ (Rubin, 1985), and subsequently ‘‘solitary-passive,’’ ‘‘solitary-active,’’ and ‘‘reticent’’ (see Coplan et al., 1994; Rubin, Burgess, & Hastings, 2002). These solitary groups can be differentiated as follows: Isolated preschoolers tend to play alone. Preschoolers prone to solitary-passive behavior tend to explore objects and play alone in a constructive manner, whereas those disposed toward solitary-active behavior engage in repetitive or dramatic play that is often disruptive. In contrast, reticent children tend to be wary, or seek to maintain distance from peers. In studies conducted with kindergarten and primary-grade children, researchers have empirically classified participants into subtypes termed

56

Gary W. Ladd

passive-anxious, active-isolated, depressed, and unsociable (see Harrist et al., 1997; Ladd & Burgess, 1999), or asocial-withdrawn and aggressivewithdrawn (see Ladd & Burgess, 1999; Ledingham & Schwartzman, 1984). Of these groups, those that have been hypothesized to be at risk for adjustment difficulties include children who are withdrawn and anxious (see Harrist et al., 1997; Rubin, Burgess, & Hastings, 2002), asocial and withdrawn (Ladd & Burgess, 1999), withdrawn and aggressive (Ladd & Burgess, 1999; Ledingham & Schwartzman, 1984), and solitary and active (Rubin, Burgess, & Hastings, 2002). For anxious-withdrawn or solitaryanxious children, this hypothesis stems from the rationale that such children desire social contact, but often avoid peers because approach behaviors heighten their anxiety (see Coplan et al., 1994; Gazelle & Ladd, 2003; Ladd & Burgess, 1999). Children who are both withdrawn and aggressive are assumed to be at risk because they not only act aversively during interactions but also spurn peers’ overtures and are avoided by peers, or both (see Ladd & Burgess, 1999). The combination of withdrawal and depression has received less empirical attention, and has been conceptualized as either a comorbid risk profile, or as an association that has developed because depression caused children to become socially withdrawn. In the latter case, researchers have argued that depression antecedes or exacerbates withdrawn children’s solitary tendencies, and reduces their competence in social encounters (Cole et al., 1998; Harrist et al., 1997). Data that reflect on the stability of social withdrawal and, thereby the contention that such behavior is indicative of an enduring disposition or style, remain mixed. Extant findings are qualified by factors such as the form of withdrawn behavior, the data source, and the investigated developmental interval. Prior to research on withdrawn subtypes, Rubin (1985) followed 111 children from kindergarten to second grade and found that 10% of the sample could be classified as ‘‘isolated’’ in kindergarten or grade one, and eight children (7%) fit this classification across all three grades. Later, in a summary of the same investigation, Rubin (1993) estimated the stability of a construct termed ‘‘social withdrawal’’ using observational data, and found coefficients of .25 from kindergarten to grade 2 and .37 from grade 2 to grade 4 (estimates based on peer reports were somewhat higher; e.g., .53 from grade 2 to grade 4). In the only long-term follow-up investigation to date, Caspi, Elder, and Bem (1988) found that shyness ratings in late childhood correlated moderately and in predicted directions not only with convergent indicators in preadolescence, including sociability ( .26), reserve (.46), and somberness (.40), but also with a shyness composite that was obtained many years later in adulthood (.26). As researchers have begun to study subtypes of withdrawn behavior, a more differentiated picture of the stability of withdrawn behavior has


57

emerged. Coplan et al. (1994) correlated scores for 4-year-olds’ solitarypassive, solitary-active, and reticent behaviors across two closely spaced play sessions, and found greater cross-session consistency for reticent (.66) and solitary-active behaviors (.51) than for solitary-passive behaviors (.34). In the Harrist et al. (1997) investigation, children who were initially classified into four solitary subgroups (i.e., passive-anxious, active-isolated, depressed-withdrawn, and unsociable) were compared to nonsolitary controls on teacher-rated withdrawn behavior as they progressed from first through third grade. In grade 1, all but the unsociable solitary groups were more withdrawn than the control children. However, by grade 3, only the active-isolated and depressed-withdrawn groups were rated as significantly more withdrawn than the controls. Ladd and Burgess (1999) estimated the stability of asocial-withdrawn behavior with two samples of children across the period of kindergarten to grade 2, and found it to be quite modest (.22 and .14 from kindergarten to grade 1; .18 and .26 from kindergarten to grade 2). These investigators also calculated the mean-level trajectories for children who were classified as asocial- and aggressivewithdrawn from kindergarten through grade 2 and found that, although withdrawn behavior decreased from kindergarten to grade 1 for both groups, it subsequently increased significantly for children who were classified as aggressive and withdrawn. This increase was sufficiently large that, by grade 2, children in the aggressive-withdrawn group were found to be significantly more withdrawn than their counterparts in a normative control group. In one of the longest prospective studies of withdrawn subtypes, Gazelle and Ladd (2003) examined the stability of teachers’ ratings of anxious-withdrawn behavior over a 5-year period, from kindergarten to grade 4. Scores for this behavior became more stable as children grew older (see Table II). That is, anxious withdrawn behavior was less stable from kindergarten to grades 1 and 2 (.31, .36, respectively) than it was from grade 2 to grades 3 and 4 (.48, .38, respectively). In sum, the extant findings are apparently too sparse to draw reliable inferences about which of the various forms of withdrawn behavior constitute enduring dispositions of children. However, the stability estimates reported thus far suggest that there may be persistent individual differences in at least four subtypes of withdrawn children, including those classified as solitary-active, solitary-anxious or anxious-withdrawn, depressedwithdrawn, and aggressive-withdrawn. a. Psychopathology: Interpersonal and Internalizing Problems. As previously stated, investigators have proposed that children prone to withdrawn or solitary behavioral styles are subject to potentially debilitating cycles that eventually precipitate negative self-perceptions and internalizing problems

58

Gary W. Ladd TABLE II Stability Coefficients for Teacher-Reported Anxious-Withdrawn Behavior from Grades Kindergarten through 5 Grade K 1 2 3 4

1

2

3

4

5

.31

.36 .28

.51 .24 .48

.45 .32 .38 .47

.24 .16 .28 .31 .32

Note: The source for the data presented in this table is Gazelle, H., & Ladd, G. W. (2003). Anxious solitude and peer exclusion: A diathesis–stress model of internalizing trajectories in childhood. Child Development, 74, 257–278. Reprinted with permission from the Society for Research in Child Development.

(see also MacDougall et al., 2001). In particular, it has been hypothesized that children’s failure to develop specific skills, coupled with their growing awareness of these deficits, may foster negative self-perceptions which, in turn, fosters internalizing problems such as anxiety and depression (Gazelle & Ladd, 2003; Parker et al., 1995). Evidence that reflects on these propositions is limited, and comes largely from a small number of longitudinal studies in which investigators have examined the association between children’s withdrawn behavior styles and later psychological maladjustment. Some of the first investigators to address this question did so in the context of follow-back longitudinal studies, and found that schizophrenic symptoms in adulthood were often linked with shy or withdrawn behavior during childhood (e.g., Frazee, 1953; Ricks & Berry, 1970). However, as Parker and Asher (1987) note, most of these studies were conducted with clinic (i.e., non-normative) rather than community samples, and results were often based on measures of unknown reliability and validity. Moreover, findings from other, more rigorous investigations (e.g., those conducted with follow-up longitudinal designs) often failed to substantiate these results (see Janes et al., 1979; Michael, Morris, & Soroker, 1957; Morris, Soroker, & Burruss, 1954; Robbins, 1966), or did so only for children of one gender (i.e., girls; see Janes & Hesselbrock, 1978; John, Mednick, & Schulsinger, 1982). In subsequent studies, there was greater reliance on prospective longitudinal designs to study the psychological consequences of childhood social withdrawal. Another trend has been to determine whether childhood shyness and withdrawn behavior are predictive of a broader range of mental health criteria. In the first wave of these investigations, researchers examined the predictive links between global or unspecified measures of withdrawal and the development of negative self-evaluations, loneliness,


59

and internalizing problems. Rubin (1985) reported that kindergartners tended to develop low perceived competence and self esteem by grade 2, and Morrison and Masten (1991) obtained similar findings with children who were followed from childhood into adolescence. In a short-term longitudinal study, Renshaw and Brown (1993) found that social withdrawal during middle to late childhood was associated with concurrent and subsequent loneliness. In yet another study (Hymel et al., 1990) social isolation in grade 2 predicted children’s internalizing symptoms in grade 5. Subsequently, investigators have sought to determine whether children who belong to specific withdrawn subtypes are more at risk for later psychological maladjustment. One such subtype includes children who manifest a combination of affective difficulties and withdrawn behavior. In short-term longitudinal studies of preschoolers, withdrawn behavior coupled with anxiety (e.g., anxious withdrawal or reticence) has been linked with internalizing problems (Coplan et al., 1994; Coplan & Rubin, 1998; Coplan, 2000). Similarly, grade school children who are sad or anxious and withdrawn (e.g., depressed-withdrawn, anxious-withdrawn subtypes) appear to be at risk for maladjustment. Harrist et al. (1997) reported that kindergartners who fit a depressed-withdrawn profile were more likely to have social problems as they moved through the elementary grades. Using growth-curve analyses, Gazelle and Ladd (2003) found that elevated trajectories toward depression in middle childhood were more likely among children who manifested stable patterns of anxious-withdrawal during early grade school (from kindergarten to grade 2). Thus, there is some evidence to suggest that withdrawn behavior coupled with sad or anxious affect can be an enduring profile, especially as children grow older, and is associated with the development of interpersonal and internalizing problems. Comparatively, only a few investigators have undertaken longitudinal studies in which they have examined the adjustment of passive-withdrawn or unsociable children. Although Harrist et al. (1997) followed a subgroup of unsociable children from kindergarten to grade 3, no evidence of dysfunction was found during this period. b. Psychopathology: Externalizing Problems. Active isolation or aggressive-withdrawn behavior, as a subtype of social withdrawal, has been implicated as a risk factor for later externalizing problems (Coplan et al., 2001; Ledingham, 1981). Although not many long-term longitudinal studies have been undertaken, findings from a few studies conducted with preschool children are consistent with this tenet (Coplan, 2000; Coplan et al., 2001; Coplan & Rubin, 1998). The fact that this form of withdrawal has been linked with externalizing problems may be attributable, in part, to the

60

Gary W. Ladd

elevated levels of aggressive behavior that are typical of children who fit this subtype. Another possibility is that such problems are a response to the form of social withdrawal these children experience. Some investigators, for example, have argued that aggressive-withdrawn children tend to be alone because they are isolated, excluded, or rejected by peers (Asendorpf, 1993; Coplan et al., 2001). Consistent with the latter proposition, early studies conducted by Ledingham and colleagues (Ledingham, 1981; Ledingham & Schwartzman, 1984) showed that aggressive-withdrawn children tended to be disliked by peers. Furthermore, Ladd and Burgess (1999) found that by grade 2, children who had been aggressive-withdrawn in kindergarten had social difficulties with peers and teachers that were more serious than those exhibited by aggressive-only or withdrawn-only children. Harrist et al. (1997) followed kindergartners who were classified as active isolates over a four-year period and reported that these children not only developed social difficulties (e.g., peer rejection) but also they displayed social-cognitive deficits that were similar to those of aggressive children. c. School Adjustment. Historically, the question of whether shy or withdrawn children are at risk for school adjustment problems has been overlooked and, as a result, extant findings are limited in scope. In early follow-back longitudinal studies, some investigators (e.g., Kuhlen & Collister, 1952; Lambert, 1972) reported that children who dropped out of school tended to be withdrawn or aggressive earlier in high school (grade 9). However, these studies have been criticized because no effort was made to control for the possibility that some of the withdrawn children may also have been aggressive (e.g., withdrawn-aggressive) and, thus, aggression rather than withdrawal may have been responsible for this finding (see Parker & Asher, 1987). The few investigators who used follow-up longitudinal designs did so only with clinic samples (e.g., Janes et al., 1979; Morris et al., 1954) and failed to find significant links between withdrawn behavior (or aggressive behavior) and dropping out of school. It was more common in subsequent studies for researchers to consider whether children who are prone toward withdrawn behavior have difficulty adapting to school, or are more likely to develop school adjustment problems. Data from a long-term follow-up study of Swedish children (Kerr, Lambert, & Bem, 1996) showed that girls who were rated as shy by their mothers at ages 8 to 10, as compared to nonshy girls, developed lower levels of educational attainment. Ollendick et al. (1990) conducted a 5-year follow-up study and found that social withdrawal, assessed at age 10, was associated with dropping out of school. Researchers have also begun to investigate whether some forms of withdrawn behavior pose greater risk for school adjustment problems than


61

others. In some of the first studies of aggressive-withdrawn children, Ledingham and colleagues (Ledingham, 1981; Ledingham & Schwartzman, 1984) found that this subtype of withdrawn children were not only disliked by peers but also tended to suffer academic difficulties. Ladd and Burgess (1999) followed children who displayed two forms of withdrawal—asocial-withdrawn and aggressive-withdrawn behavior—from kindergarten to grade 2 and compared them to samples of aggressive children and normative, matched controls. Across all grades, children in the aggressive-withdrawn group were more likely to form and maintain teacher–child relationships that were high in conflict and dependency and low in closeness. The relationship difficulties of asocial-withdrawn children, in contrast, were found to be more transient. These children exhibited more dependent relationships with their teachers as they began kindergarten, but not thereafter. In sum, evidentiary support for the premise that withdrawn behavior is an antecedent of maladjustment is not as strong as that obtained for aggressive behavior. On the one hand, withdrawn behavior may be less of a risk factor than is aggressive behavior either because it is not a direct precursor of maladjustment, or because the forms of maladjustment that it does portend are less obvious, harmful to others, or measurable (e.g., internalizing rather than externalizing problems). On the other hand, certain forms of withdrawn behavior may inhibit adaptation and promote maladjustment, but such consequences remain undetected due to a paucity of longitudinal investigation. Firmer conclusions about this premise may be possible as a larger corpus of longitudinal evidence accrues. From what is known thus far, anxious- and sad-withdrawn children would appear to be at greater risk for developing internalizing problems and debilitating self-perceptions. Extant findings also suggest that children with aggressive-withdrawn behavioral styles, much like their aggressive counterparts, are likely to develop externalizing difficulties and school adjustment problems. It is less clear that shy-, passive-, asocial- or unsociable-withdrawn children are likely to manifest long-term adjustment problems. Again, these inferences are preliminary and subject to revision based on new evidence. B. MAIN EFFECTS PREMISE 2: CHILDREN WITH POOR PEER RELATIONSHIPS ARE PRONE TOWARD MALADJUSTMENT

The premise that children’s adjustment is influenced by their relationships with peers originated in socialization theories, particularly those that emphasized the importance of agemates in shaping the course of children’s development (e.g., see Asher & Gottman, 1981; Berndt & Ladd, 1989; Buhrmester & Furman, 1986; Hartup, 1970; Sullivan, 1953). Empirical attempts to elucidate the contributions of peer relationships to children’s

62

Gary W. Ladd

development, especially those implemented during the early years of this discipline (for historical summaries, see Ladd, 1999; Parker et al., 1995; Renshaw, 1981), were often guided by a main effects perspective. Essentially, investigators sought to determine whether poor or dysfunctional peer relationships during childhood were enduring features of children’s social lives, and whether participation in such relationships was associated with later-emerging symptoms of maladjustment. To address questions about the nature and functions of peer relationships in children’s development, researchers have attempted to differentiate among the types of relationships children form with peers, and among the processes that occur in these relationships that might interfere with adaptation and foster maladjustment. Included among the types of peer relationships that have been examined as antecedents of later dysfunction are peer group rejection and peer victimization. Friendship has been researched as well, but more often as a form of relationship that may protect children or prevent the development of adjustment problems. Conceptually, arguments have been made for distinguishing between friendship and peer acceptance/rejection as precursors of adjustment on both structural and psychological grounds (see Bukowski & Hoza, 1989; Ladd, 1988). In general, friendship has been defined as a voluntary, dyadic form of relationship that often embodies a positive affective tie (Buhrmester & Furman, 1986; Furman & Robbins, 1985; Howes, 1988), whereas peer acceptance/rejection has been defined as a child’s relational ‘‘status’’ in a peer group, as indicated by the degree to which they are liked or disliked by group members (Asher et al., 1979; Ladd, Kochenderfer, & Coleman, 1997). Peer victimization, in contrast, has been defined as a form of relationship in which a subset of the peer group (e.g., one or several peers) frequently aggresses against specific children (Perry, Kusel, & Perry, 1988). Implied in this definition is the assertion that ‘‘victims’’ are frequently the recipients of peers’ aggressive behaviors (see KochenderferLadd & Ladd, 2001; Olweus, 1993; Perry, Kusel, & Perry, 1988). Along with these distinctions, researchers have postulated that children may simultaneously participate in more than one form of peer relationship (e.g., be rejected but have a friend, be victimized and rejected, etc.; Perry et al., 1988; Ladd et al., 1997) and that different types of peer relationships may expose children to different processes or experiences (Furman & Robbins, 1985) and, so, make differential contributions to their adjustment. For example, it has been proposed that victimization may increase children’s fearfulness and thereby elevate their motivation to avoid school, whereas peer group rejection may isolate children and reduce their achievement (Buhs & Ladd, 2001; Ladd et al., 1997; Ladd, Birch, & Buhs, 1999). In this area of investigation, the logic of a main effects model implies continuity in children’s participation within specific types of peer relationships.


63

Support for this proposition has been reported in a number of longitudinal investigations in the form of stability coefficients for measures of children’s peer group acceptance/rejection, peer victimization, and participation in friendships. Most often, peer group acceptance or rejection has been assessed by asking group members (e.g., classmates) to rate or nominate individuals within their group based on a liking criterion (e.g., how much they like to associate with that person). From these data, researchers create measures of peer group acceptance and rejection by averaging and standardizing: (a) the nominations that children receive from peers to obtain a continuous measure of peer group acceptance (e.g., positive nominations) or peer rejection (e.g., negative nominations), or (b) the ratings received from peers to obtain a bipolar acceptance–rejection index. Another common approach is to assign children to acceptance–rejection groups using algorithms developed for specific classification schemes (e.g., popular, neglected, rejected, controversial, and average status categories; see Coie, Dodge, & Coppotelli, 1982; Newcomb & Bukowski, 1983, 1984). In general, the stability coefficients reported for nomination and rating measures have been moderate to high in magnitude. For example, in studies of preschool children, Asher et al. (1979) found 4-week stability coefficients of .81 for averaged acceptance ratings, and .56 and .42 for positive and negative nomination measures, respectively. Over a 7-week period, Poteat, Ironsmith, and Bullock (1986) reported estimates of .64 to .68 for nomination measures, and .55 to .66 for aggregated ratings. Similar or higher stability coefficients have been found with grade-school samples. In an ongoing prospective longitudinal study, Ladd and colleagues gathered data on several types of peer relationships over a 6-year period from kindergarten through grade 5 and found moderate stability for measures of children’s peer group acceptance and rejection (i.e., both rating and nomination measures; see Table III). Researchers have also estimated the stability of children’s membership in particular peer acceptance/status categories by calculating the percentages of children who retain their category designations over varying temporal intervals. Investigators such as Coie and Dodge (1983), DeRosier et al. (1994), and Ladd and Troop (in press) have found that, for some children, peer rejection can be a chronic problem that persists over many years. In a review of 12 longitudinal studies, Cillessen, Bukowski, and Haselager (2000) found that children of all ages (preschool through high school) who were classified as popular or rejected were among the most likely to retain their status over intervals ranging from 1 to 48 months. Research on the stability of peer victimization is at an early stage and most estimates have been derived from either self-report or peer-report victimization measures (see Ladd & Kochenderfer-Ladd, 2002). For estimates based

64

Gary W. Ladd TABLE III Stability Coefficients for Measures of Children’s Peer Relationships for 1- to 5-Year Intervals from Kindergarten through Grade 5 Peer relationship stabilities Grade

1

2

3

K 1 2 3 4

.50

.48 .52

K 1 2 3 4

.57

.61 .60

K 1 2 3 4

.67

.64 .57

K 1 2 3 4

.37

.37 .35

4

5

Peer acceptance nominations .42 .42 .43

.28 .36 .39 .41

.36 .37 .31 .44 .39

Peer rejection nominations .51 .46 .58

.36 .34 .42 .52

.37 .41 .44 .48 .46

.42 .44 .41 .52

.47 .48 .49 .57 .57

.35 .31 .21 .42

.30 .32 .27 .34 .38

Peer acceptance ratings .59 .55 .60

Mutual friendships .37 .28 .29

Note: The findings in this table were obtained from the Pathways Project (G. W. Ladd, Principal Investigator), and were first reported in an unpublished progress report submitted to the National Institute of Mental Health pursuant to grant 2-RO1MH-49223.

on self-report measures, stability estimates range from .24 for kindergarten children over a 5-month period (Kochenderfer & Ladd, 1996a) and .34–.36 for 9- to 12-year-olds over a 10-month period (Hawker, 1997). With samples of older children, Boulton and Smith (1994) gathered assessments of peer victimization on four occasions (i.e., October, March, and June of the same year, and October of the following year) and found stability coefficients ranging from .50 (October to March) to .80 (March to June). They also examined the stability of children’s peer victimization scores from one grade level to the next (June to October) and found a coefficient of .15 for girls and .66 for boys. Longer-term longitudinal findings reported by


65

Kochenderfer-Ladd and Wardop (2001) suggest that, during the early school years, chronic victimization occurs, but is not highly prevalent. These investigators followed children from kindergarten through grade 3 and found that less than 4% of their sample was persistently identified as victims over a four-year period. Stability coefficients calculated for consecutive assessments ranged from r ¼ .27 (from kindergarten to grade 1) to r ¼ .41 (from grades 2 to 3) suggesting that peer harassment may become more stable as children approach middle childhood. Peer-report measures tend to be constructed by obtaining victimization nominations from a number of peers, and there is some evidence to suggest that when investigators aggregate peer reports they typically obtain more reliable (consensual) estimates of children’s status as victims (Ladd & Kochenderfer-Ladd, 2002; Perry et al., 1988). Although rarely reported, stability coefficients for peer-report measures have been estimated at .93 for 9- to 12-year-olds over 3 months (Perry et al., 1988), and .30 to .71 for 9- to 12-year-olds over intervals of 10 months to 1 year (Boivin, Hymel, & Bukowski, 1995; Hawker, 1997). Ladd and Kochenderfer-Ladd (2002) computed stability coefficients for victimization measures obtained from different types of informants (e.g., self, peer, teachers, parents). Stability estimates obtained across a 5-year interval from kindergarten through grade 4 (see Table IVA) varied by source, with self-reports producing more stable estimates than peer reports with samples of young children (i.e., kindergarten and grade 1 samples). By grade 2, data from all four types of informants suggested that peer victimization is a moderately stable form of relationship during the middle-childhood years (see Table IVB). Children’s participation in friendships appears to become moderately stable, at least over brief intervals, during the preschool years. Early evidence indicated that many preschoolers develop ‘‘strong associates,’’ or preferred play partners that may be maintained over several weeks (Eckerman & Stein, 1982; Hinde et al., 1985; Howes, 1983), and that preschoolers’ and grade-schoolers’ reports of friendships are often corroborated by persons such as teachers and parents (Ladd & Emerson, 1984; Howes, 1988). The most reliable estimates of the stability of children’s participation in friendships have been based on measures of reciprocated or mutual friendship, in which both the child and his or her ‘‘friend’’ must agree that their relationship is a friendship (see Bukowski & Hoza, 1989; Parker & Asher, 1993). Data gathered with these measures has revealed that young children are capable of maintaining the same friendships over a period of years (Dunn, 1993; Howes, 1988; Howes & Phillipsen, 1992), and across major ecological transitions (e.g., preschool into kindergarten; Ladd, 1990). Estimates of the stability of children’s participation in mutual friendships from kindergarten through grade 5 are also shown

66

Gary W. Ladd TABLE IV Stability Coefficients for Self-, Peer-, Teacher-, and Parent-Report Victimization Measures by Grade and Gender

A. Stability coefficients for self- and peer-report measures Interval

K!G4 G1!G4 G2!G4 G3!G4

Self-Report measure

Peer-Report measure

Sample

Males

Females

Sample

Males

Females

.01 .23** .25** .31***

.06 .14 .21* .29**

.02 .33** .28** .39***

.15 .07 .39*** .49***

.17 .03 .32** .48***

.11 .12 .45*** .49***

B. Stability coefficients for self-, peer-, teacher-, and parent-report measures Grade

Self-Report victimization scale

Peer-Report victimization scale

Teacher-Report victimization scale

Parent-Report victimization scale

Sample G2!G3 G3!G4

.36*** .40***

.66*** .61***

.46*** .30***

.44*** .50***

Males G2!G3 G3!G4

.31*** .28***

.66*** .62***

.42*** .18*

.42*** .50***

Females G2!G3 G3!G4

.39*** .55***

.61*** .61***

.56*** .47***

.47*** .51***

Note: K ¼ Kindergarten; G1, G2, G3, G4 ¼ Grades 1, 2, 3, 4, respectively; *p < .05; **p < .01; ***p < .001. Adapted from Ladd, G. W. & Kochenderfer-Ladd, B. J. (2002), Identifying victims of peer aggression from early to middle childhood: Analysis of cross-informant data for concordance, estimation of relational adjustment, prevalence of victimization, and characteristics of identified victims. Psychological Assessment, 14, 74–96; see pages 79 and 86 (Tables 1 & 3), respectively, in this journal article. Reprinted with permission of American Psychological Association.

in Table III. As can be seen, these coefficients are not as high in magnitude as those obtained for measures of peer group acceptance and rejection, but are somewhat more substantial than those obtained for peer victimization. In sum, the relationships that children form with agemates appear to have a moderate degree of continuity. Accumulating evidence suggests that features such as children’s acceptance or rejection by peer group members and participation in mutual friendships are among the most stable forms of relationship that children experience during early and middle childhood. In contrast, the stability coefficients reported for peer victimization are more variable in magnitude, and not particularly strong over substantial intervals


67

(e.g., 3–4 years). These data suggest that, whereas some children appear to be persistently harassed from year to year, others may be harassed sporadically, and still others may encounter such treatment for a relatively brief or transitory period. 1. Psychopathology Early efforts to elucidate the contributions of children’s peer relationships to the development of psychopathology were focused principally on a single construct—poor peer group relations (e.g., low peer acceptance, peer rejection)—and conducted primarily with males in clinic samples (see Kupersmidt et al., 1990; Parker & Asher, 1987). Findings from early followback studies (see Parker & Asher, 1987 for a more extensive review), most of which were based on retrospective analyses of school and clinic records, suggested that many psychologically impaired men had poor peer group relations as children, including histories of peer rejection and neglect (e.g., Frazee, 1953; Roff, 1961, 1963). Similarly, Cowen et al. (1973) found that children with negative peer reputations in grade 3 were likely to receive mental health services as adults, and that this factor was linked more closely with later maladjustment than were many other indicators of childhood adjustment (e.g., teacher’s ratings of adjustment, children’s self-esteem, anxiety, etc.). However, follow-up longitudinal studies that were completed during this same era, or soon thereafter, did not always corroborate these findings. For example, evidence of an association between early peer group difficulties and later psychopathology was found in clinic samples by Janes et al. (1979), but not by Robbins (1966). Using a school-based sample, Roff and Wirt (1984) reported modest predictive associations between 8- to 10-year-olds’ sociometric status and later psychiatric hospitalization. Data from subsequent, prospective longitudinal studies (e.g., DeRosier et al., 1994; Hymel et al., 1990) tended to corroborate earlier evidence indicating that peer rejection is a relatively stable characteristic that predicts internalizing problems and externalizing problems during the grade school years (see also MacDougall et al., 2001). Ollendick et al. (1992), for example, followed children who belonged to specific peer acceptance groups (i.e., popular, rejected, average, neglected, and controversial) from ages 9 through 14 and found that rejected children were more likely than popular children to exhibit externalizing problems such as misconduct, delinquency, and substance abuse. Kupersmidt and Coie (1990) followed a small sample of peer-rejected 10-year-olds over a 7-year period and found that peer rejection forecasted later dysfunction, but that this link was stronger when maladjustment was defined broadly (i.e., aggregated over multiple indicators) rather than narrowly (i.e., used to predict specific forms of maladjustment). Findings reported by DeRosier et al. (1994), further

68

Gary W. Ladd

revealed that the severity of children’s internalizing and externalizing problems increased as a function of the chronicity and proximity of peer rejection. Links were also found between peer rejection and loneliness during both early (Cassidy & Asher, 1992) and middle childhood (Asher, Hymel, & Renshaw, 1984; Crick & Ladd, 1993). Participation in friendship, as well as the quality of children’s friendships, have been shown to be important predictors of children’s emotional wellbeing (Bukowski & Hoza, 1989; Bukowski, Newcomb, & Hartup, 1996; Parker & Asher, 1993) and adjustment trajectories during early and middle grade school (Ladd et al., 1996). Findings such as these are often attributed to the hypothesis that children benefit from having a close friend who affirms their positive attributes and downplays or discounts shortcomings (Buhrmester & Furman, 1986; Furman & Robbins, 1985). In contrast, children who remain friendless may be deprived of these resources and, thus, at greater risk for maladjustment. This view is corroborated by evidence indicating that children with close friendships see themselves more positively (Berndt & Burgy, 1996; Keefe & Berndt, 1996; Savin-Williams & Berndt, 1990), and children who have one or more close friendships tend to experience greater perceived social support and less loneliness (Ladd et al., 1996; Parker & Asher, 1993). In addition, children who have friendships that are high in positive features, such as intimacy and support, tend to have higher levels of self esteem (Berndt, 1996). Although the predictive significance of friendship has seldom been examined in the context of long-term follow-up studies, Bagwell and colleagues (Bagwell, Newcomb, & Bukowski, 1998) identified groups of children who either had friends or were friendless in grade 5, and assessed their adjustment 12 years later in early adulthood. Results showed that children with friends were better adjusted in grade 5 and in later life on a variety of indicators as compared with friendless children, including trouble with the law, family life, and overall adjustment. Prior friendship, however, was not found to be a unique predictor of later adjustment once children’s fifth-grade adjustment and peer group acceptance scores were taken into account. The investigation of peer victimization as a precursor of maladjustment began in Scandinavian countries during the late 1970s in response to public concern about peer abuse as a cause of child and adolescent suicide (see Olweus, 1978). Early findings suggested that the victims of peer abuse were at risk for a range of serious, potentially life-threatening mental health problems (e.g., depression, suicide; see Olweus, 1978, 1993). In the decades to follow, similar investigative agendas have been launched in other nations, including the United States (for a review, see Juvonen & Graham, 2001). Emerging findings show that peer victimization antecedes a number of adjustment difficulties during childhood and adolescence, including


69

loneliness, depression, anxiety, low self-esteem, and social problems (e.g., Alsaker, 1993; Bjo¨rkqvist, Ekman, & Lagerspetz, 1982; Boulton & Underwood, 1992; Egan & Perry, 1998; Graham & Juvonen, 1998b; Kochenderfer & Ladd, 1996a; Olweus, 1993). Furthermore, peer victimization occurs not only during middle childhood and adolescence, but also during early childhood. As early as school entry, peer victimization predicts both transient and enduring adjustment difficulties. Kochenderfer and Ladd (1996a), for example, found that the frequency of children’s peer victimization experiences as they began kindergarten forecasted significant gains in loneliness over their first year in school. Furthermore, these investigators found that increases in children’s adjustment difficulties cooccurred with the onset of victimization during the school year, and that children who had been victimized for longer periods of time had more severe adjustment difficulties than those who had been abused for brief periods. 2. School Adjustment Not until the 1990s did researchers investigate the possibility that children’s participation in different types of classroom peer relationships is associated with early as well as later forms of school adjustment (see Ladd, 1989, 1996, 1999; Perry & Weinstein, 1998). A guiding premise in this area of investigation has been that children’s relationships with classmates immerses them in processes that affect their ability to adapt to school challenges (e.g., participation vs. exclusion; receiving assistance vs. being ignored) which, ultimately, affects their development in this context (e.g., increases or decreases their sense of worth, competence, etc.). Because relationships bring different processes to bear upon children and create different resources or constraints, the adaptive significance of different types of peer relationships for school-related demands may vary (Furman & Robbins, 1985; Ladd et al., 1997). A growing corpus of findings link peer acceptance and rejection with indicators of later school adjustment. In a critical analysis of early research findings, Parker and Asher (1987) reported that low peer acceptance was a significant correlate of later school adjustment, particularly dropping out of high school. Subsequently, investigators began to test hypotheses about the role of peer acceptance and rejection in more proximal forms of school maladjustment. To illustrate, early peer rejection—at school entry—has been shown to predict problems such as negative school attitudes, school avoidance, and underachievement during the first year of schooling (Buhs & Ladd, 2001; Ladd, 1990; Ladd, Birch, & Buhs, 1999). Later, in the elementary years, peer acceptance has been linked with loneliness (Parker & Asher, 1993), peer interaction difficulties, lower emotional well being, and academic deficits (Ladd, Kochenderfer, & Coleman, 1997; Vandell &

70

Gary W. Ladd

Hembree, 1994). Evidence from other longitudinal studies suggests that peer rejection predicts absenteeism during the grade school years (e.g., DeRosier et al., 1994; Hymel et al., 1990), and grade retention and adjustment difficulties during the transition to middle school (Coie et al., 1992). Within ensuing studies, researchers attempted to distinguish the contributions of peer acceptance from those of other relationships. Ladd and colleagues (Ladd et al., 1997; Ladd, Birch, & Buhs, 1999) found that, even after controlling for other forms of peer and teacher–child relationships, peer rejection predicted children’s participation in the classroom that, in turn, forecasted later achievement. In a similar study, Buhs and Ladd (2001) found that children’s peer acceptance at school entry predicted changes in classroom participation that, in turn, predicted later academic and emotional adjustment. In general, these results provide support for the premise that peer acceptance promotes social inclusion in the classroom, which, in turn, yields resources (e.g., sense of belongingness; involvement in learning activities) that enhance interpersonal and scholastic adjustment. In research on classroom friendships, investigators have often measured features such as children’s participation in a close friendship, the number of mutual friends they have in their classrooms, the duration of these relationships, and positive and negative friendship features (see Berndt, 1996; Ladd, Kochenderfer, & Coleman, 1996; Parker & Asher, 1993). Growing evidence links one or more of these facets of friendship to children’s school adjustment. To illustrate, the presence of a pre-established friendship in a child’s classroom (e.g., starting school with a friend established during preschool) has been linked with better adjustment in new classroom environments (Ladd, 1990). Furthermore, as children enter school, those who form new friendships tend to form more favorable perceptions of school and do better academically than peers with fewer friends (Ladd, 1990). The processes that typify friends’ interactions have also been implicated in children’s school adjustment. Young children, especially boys who reported conflict within their friendships, have been shown to have adjustment difficulties, including lower levels of classroom engagement and participation (Ladd et al., 1996). Ladd et al. (1996) also found that when children saw their friendships as offering higher levels of validation (support) and aid (assistance) they tended to perceive classrooms as supportive interpersonal environments. Conversely, third- through fifthgraders whose friendships lacked supportive features were found to be lonelier in school (Parker & Asher, 1993). Although less well researched, evidence suggests that friendship may not always contribute positively to school adjustment. Berndt and Keefe (1995), for example, found that fighting and disruptiveness tended to increase if adolescents had stable friendships with peers who exhibited the same problems.


71

The probability that children will encounter peer abuse, or become the targets of peers’ aggressive behaviors, increases as they enter school and progress through the primary grades (see Kochenderfer & Ladd, 1996b; Kochenderfer-Ladd & Wardrop, 2001). Moreover, exposure to peer abuse during early and middle childhood appears to increase children’s risk for concurrent and later school adjustment difficulties (see Ladd, Kochenderfer, & Coleman, 1997; Kochenderfer-Ladd & Wardrop, 2001). Some have argued that frequent bully/victim interactions in the school context interfere with children’s ability to successfully adapt to school. For instance, it has been posited that frequent harassment leads children to become so preoccupied with fears, feelings of social alienation, and safety concerns that they have difficulty attending to school tasks and, eventually, develop negative school attitudes or higher levels of school avoidance (Hoover & Hazler, 1991; Kochenderfer & Ladd, 1996a, 1996b; Slee, 1994). Evidence of a link between peer victimization and school maladjustment has been found in a number of investigations. Boivin, Hymel, and Bukowski (1995) found that grade-schoolers who experienced gains in victimization over a year also reported higher levels of loneliness. Similarly, at school entrance, higher levels of peer victimization predicted increases in loneliness and school avoidance (Kochenderfer & Ladd, 1996a; Ladd et al., 1997). Moreover, when peer abuse is pronounced or prolonged (e.g., chronic), children appear to be at even greater risk for school maladjustment. Using growth-curve analyses, Kochenderfer-Ladd and Wardrop (2001) found that the adjustment trajectories of children who were exposed to victimization over longer intervals (assessments were conducted from kindergarten through grade 3) were more likely to feel lonely in school and less satisfied with their peer relationships in that context. Children who were victimized during the early grades but not thereafter did not always recover, or show improvements in their adjustment. These longitudinal findings are corroborated by cross-sectional evidence gathered with diverse age groups around the world. In Canada and England, for example, researchers have found that victims are more likely than nonvictims to report negative feelings and attitudes toward school and classroom tasks (Boivin & Hymel, 1997; Boulton & Underwood, 1992). Although further investigation is needed, especially across age groups and gender, the bulk of extant evidence conforms to the hypothesis that victimization is associated with a range of school-related difficulties. Thus, although far from being conclusive or exhaustive, there is evidence to suggest that children’s participation in several types of peer relationships are associated with school maladjustment. In addition to peer group acceptance, children’s participation in friendships and the features of children’s friendships have been linked with concurrent and subsequent

72

Gary W. Ladd

school adjustment indicators across a wide range of ages. Peer victimization, a form of relationship most closely resembling child abuse, appears to antecede several forms of maladjustment including school avoidance, loneliness, and negative attitudes toward classmates, school tasks, and the larger school environment. 3. Differential Contributions of Peer Relationships to Children’s Adjustment The tendency to study peer relationships individually has been supplemented by investigations in which researchers have examined multiple forms of peer relationship and the relative (differential) ‘‘contributions’’ of these relationships to specific adjustment outcomes. Initial efforts to investigate differential relationship contributions were focused on friendship and peer acceptance (e.g., see Parker & Asher, 1993; Vandell & Hembree, 1994). Although some investigators proposed that the provisions children derive from friendship and peer acceptance were both unique and overlapping (see Furman & Robbins, 1985), the weight of initial evidence suggested that these relationships contribute uniquely to children’s adjustment. Research with adolescents indicated that loneliness is more closely linked with friendship than peer acceptance, and feelings of isolation are more closely tied to peer acceptance than friendship (see Bukowski & Hoza, 1989). Among grade-school children, friendship and peer acceptance make separate contributions to the prediction of both socio-emotional adjustment and academic competence (Parker & Asher, 1993; Vandell & Hembree, 1994). With young children, Ladd (1990) found that friendship and peer acceptance uniquely predicted changes in kindergartner’s school perceptions, avoidance, and performance. Such findings are consistent with the hypothesis that the effects of friendship on children’s development are unique relative to those conferred by peer acceptance, and that these relationships differ in their adaptive value for specific adjustment outcomes. Even broader ranges of children’s classroom peer relationships have been researched in subsequent work. In one investigation (Ladd, Kochenderfer, & Coleman, 1997), four forms of peer relationships were examined simultaneously (i.e., two forms of friendship, peer group acceptance, and peer victimization) as predictors of changes in multiple indices of children’s school adjustment. These investigators found that several types of peer relationships were linked with most of the examined psychological and school adjustment criteria, but that the adaptive significance of particular forms of relationship (i.e., presence of unique vs. shared predictive linkages) varied across adjustment domains. To illustrate, when the four forms of peer relationship were examined individually rather than jointly


73

as predictors of children’s affect in school (e.g., loneliness), significant associations were found for number of friends, peer group acceptance, and peer victimization. However, when the contributions of these relationships were examined after adjusting for shared predictive linkages, some relationships were found to be better predictors than others (i.e., in the sense that they uniquely accounted for variance in criterion after controlling for the predictive power of other forms of relationship). Peer victimization, for example, predicted gains in children’s loneliness above and beyond associations that were attributable to the other three forms of peer relationship. Application of this same analytic strategy with indicators of children’s school performance revealed that peer group acceptance uniquely predicted improvements in children’s achievement. Overall, these findings were consistent with the view that peer relationships are both specialized in the types of resources or constraints they create for children, but also diverse in the sense that some resources may be found in more than one form of relationship.

III. Primary Premises and Findings from Research Guided by ‘‘Child by Environment’’ Models As I have illustrated in prior sections, two rather distinct literatures have grown up around main effects perspectives—one of which has been a response to the hypothesis that children’s ‘‘risky’’ behavioral dispositions contribute to later maladjustment, and the other of which has been guided by the premise that children’s participation in ‘‘risky relationships’’ is a precursor of later dysfunction. Previously, many investigators—depending on their theoretical orientations—have regarded the explanatory power of one of these two ‘‘main effects’’ perspectives as dominant over the other. For example, when considering how relationship difficulties contribute to adjustment, relative to children’s behavioral dispositions, the former have often been construed as ‘‘marker variables’’ or links ultimately attributable to children’s underlying behavioral propensities (see MacDougall et al., 2001; Parker & Asher, 1987). However, many researchers have subsequently begun to reexamine these ‘‘main effects’’ perspectives by conducting investigations guided by ‘‘child by relational environment’’ models. To reiterate, a central premise of a child by environment model is that the precursors of children’s adjustment originate not only within the child but also within the child’s relational environment. This perspective has enabled researchers to move beyond simple ‘‘main effects’’ explanations of maladjustment toward more complex, ‘‘processoriented’’ frameworks (see Coie et al., 1993).

74

Gary W. Ladd

A. CHILD BY ENVIRONMENT PREMISE 1: CHILDREN’S BEHAVIORAL STYLES AND THEIR PARTICIPATION IN PEER RELATIONSHIPS CONTRIBUTE TO SCHOOL ADJUSTMENT

With the advent of child by environment theories (see Caspi et al., 1987, 1988; Coie et al., 1993; Ladd, 1989; Parker et al., 1995), researchers began to investigate multiple antecedents of adaptation and, in particular, the interface between children’s behavioral styles and participation in peer relationships as predictors of psychological and school adjustment. A central assumption is that both the behavioral characteristics children display in interpersonal contexts and the nature of the relationships they form in such settings influence adjustment. One approach to further explicating the confluence of children’s behavior and relationships as antecedents of adjustment stems from the risk and resilience literature, where epidemiological models have been adapted for use in psychological research (see Garmezy, Masten, & Tellegen, 1984; Rutter, 1990). ‘‘Risk’’ factors are defined as aspects of children’s behaviors or relationships that increase the probability that they will develop adjustment problems, whereas ‘‘resources’’ (or protective factors) refer to features of children’s behavior or relationships that decrease the likelihood of dysfunction or improve functioning (Coie et al., 1993; Rutter, 1990). Within this perspective, behavioral and relational risks or resources can be conceptualized as factors that influence each other as well as children’s adjustment in different ways over the course of development. Researchers have attempted to elucidate the nature of these linkages by investigating whether child attributes and features of the child’s relational environment make additive, moderated (contingent), or mediated contributions to adjustment. An additive model implies that, separate from (i.e., partially overlapping or independent of) the contributions of children’s behavioral dispositions, the experiences they have in peer relationships may ‘‘add’’ to (i.e., increase or decrease) the probability that adjustment or maladjustment will develop. For example, for children who exhibit risky behavioral dispositions such as aggressiveness, exposure to abusive relational experiences (e.g., peer victimization) may increase or exacerbate the probability of maladjustment. In contrast, when aggressive dispositions are accompanied by a relational resource (e.g., receiving support from a friend), the former factor might be expected to independently reduce the probability of maladjustment, or ‘‘compensate’’ for the risk posed by aggressive behavior. In contrast, a moderator model is one in which the effects of one risk or resource on children’s adjustment is contingent upon others. Using the prior example as an illustration, we might hypothesize that the contribution of


75

relationship risks or resources to adjustment would vary depending on the degree to which children exhibit an aggressive behavioral disposition. For example, an investigator might find that peer abuse worsens the adjustment trajectories of aggressive children more than it does to those of nonaggressive children. Such a finding could be considered evidence of a moderated linkage among the behavioral and relationship antecedents of adjustment. In a mediator model, the principal tenet is that the effects of one risk/ resource on children’s adjustment are transmitted through a second risk or resource. We might hypothesize, for example, that the effect of children’s behavioral styles on their school adjustment are transmitted through classroom relational risks or resources. Support for such hypotheses might consist of findings indicating that children’s aggressive behaviors were strongly linked with classroom peer rejection, but that the latter variable emerged as the principal antecedent of negative school attitudes, after controlling for the association between aggression and school attitudes. Such a result would be consistent with the interpretation that the association between aggression and negative school attitudes was mediated by classroom peer rejection. 1. Psychopathology Thus far, most of the research conducted on the interface between behavioral and relational risk factors has been designed to explicate the relative contributions of confrontive aggression and peer rejection as antecedents of psychological maladjustment (see Ladd, 1999; MacDougall et al., 2001). The accumulated evidence tends to support an additive model, suggesting that in addition to aggression, peer rejection increases children’s risk for maladjustment. For example, in an analysis of longitudinal findings, McDougall et al. (2001) concluded that support for the additive contribution of peer rejection to future maladjustment was strongest for internalizing problems (e.g., see Coie et al., 1995; Lochman & Wayland, 1994; Renshaw & Brown, 1993), and that results consistent with an additive model were often more pronounced for boys than for girls. Even though this may be the case, other findings link aggression and peer rejection to later externalizing problems. Evidence from two longitudinal studies (e.g., Coie et al., 1992; Hymel et al., 1990) indicates that both aggression and peer rejection in grade school make unique contributions to maladjustment in early adolescence. Kupersmidt and Coie (1990), however, found that the strength of these linkages varied with the type of adjustment outcome examined: whereas aggression in middle childhood best predicted delinquency in adolescence, both aggression and peer rejection anteceded other types of externalizing problems.

76

Gary W. Ladd

Support for the hypothesis that the risk posed by aggressive or withdrawn behavior is moderated by relational adversity is not as prevalent. There is some evidence, however, to suggest that chronic peer victimization predicts greater gains in dysfunction (internalizing, externalizing problems) for aggressive than for nonaggressive children (Ladd & Burgess, 2001). Also under-researched is the premise that the effects of risky child behaviors on maladjustment are mediated through adverse peer relationships, although preliminary support for this premise has begun to accrue. In a short-term longitudinal study, Panak and Garber (1992) found that the association between aggression and depression was partially mediated by gains in children’s peer rejection. Boivin, Hymel, and Bukowski (1995) reported that the association between children’s withdrawn behavior (assessed in year 1) and two types of internalizing problems (assessed in year 2) was mediated by adverse peer experiences (peer rejection, victimization) that were measured concurrently with social withdrawal in year 1. Later, in a cross-sectional study, Boivin and Hymel (1997) found that the association between children’s behaviors (aggression, withdrawal) and feelings of loneliness were partially mediated by adverse peer experiences (rejection, victimization), but not by positive peer affiliations. Additionally, Ladd, Birch, and Buhs (1999) conducted a short-term prospective longitudinal study with kindergarten samples and obtained support for the premise that peer rejection mediates the effects of aggressive behavior on children’s scholastic maladjustment. The question of whether these risk models hold for other behavioral dispositions (e.g., solitary or withdrawn behavior) has not been examined extensively. Renshaw and Brown (1993) assessed grade-school children’s withdrawn behavior and low peer acceptance over a one-year interval and found that both predictors were additively associated (both concurrently and predictively) with loneliness. A similar pattern of concurrent linkages was reported by Boivin and Hymel (1997). In a five-year prospective longitudinal study, Gazelle and Ladd (2003) used growth-curve analyses to examine how a particular subclass of solitary behavior (i.e., anxious-solitary behavior) and specific form of relational risk (i.e., exclusion by peers) were linked with children’s trajectories toward depression. They found that anxious-solitary behavior and peer exclusion co-occurred in children soon after kindergarten entry, and that anxious-solitary children who were excluded early on, in comparison with their nonexcluded anxious-solitary counterparts, exhibited more stable patterns of anxious-solitary behavior over time. Moreover, evidence of a moderated association was found between the behavioral and relational risk factors and children’s adjustment in that the joint influence of anxious-solitude and exclusion predicted the most


77

Fig. 1. Depression trajectories for groups differing on anxious-solitary behavior and peer exclusion from kindergarten through grade 4 as estimated by Gazelle and Ladd (2003). Note: fK ¼ fall of kindergarten; sK ¼ spring of kindergarten; 1st through 4th ¼ grades 1 to 4. Reprinted with permission from the Society for Research in Child Development.

elevated depressive symptom trajectories. As can be seen in Figure 1, children who were prone toward anxious-solitary behavior during the early school years were much more likely to manifest and maintain depressive symptoms if they had been subjected to higher rather than lower levels of peer exclusion (i.e., the solitary anxious-high excluded vs. the solitary anxious-low excluded groups, respectively). Furthermore, as illustrated by the trajectories found for the two high exclusion groups (i.e., solitary anxious-high excluded and solitary anxious-low excluded externalizing groups), the propensity for solitary-anxious excluded children to exhibit depressive symptoms apparently was not attributable to the co-occurrence of other risk factors, such as externalizing problems. In contrast, children whose risk status was defined by anxious-solitary behavior and low levels of peer exclusion had adjustment trajectories that originated near the mean of the sample and declined over time. A similar but even greater decline in depressive symptoms was found for children whose tendency to engage in anxious-solitary behavior decreased over the first few grade levels (i.e., the solitary offset group).

78

Gary W. Ladd

2. School Adjustment In this domain, research has been guided by the premise that children’s behavioral propensities in combination with their involvement in specific types of classroom relationships (and relationship experiences) have an important bearing on school adjustment (see Ladd, 1989, 1996, 1999). To a large extent, hypotheses about which child behaviors and relationship experiences are influential in this regard, and how these interpersonal factors affect children’s adaptation to school, have been extrapolated from past evidence—particularly findings indicating that early and later-emerging forms of maladjustment are forecasted by: (a) children’s aggressive and withdrawn behavioral dispositions, and (b) children’s participation in supportive versus antagonistic peer relationships. In two prospective longitudinal studies with kindergarten samples, Ladd, Birch, and Buhs (1999) found that children whose interactions were more prosocial during the first 10 weeks of kindergarten tended to develop mutual friends and higher levels of peer acceptance by week 14 (see Figure 2). In contrast, children whose interactions were characterized by aggressive behavior became more disliked by classmates, had fewer friends, and developed more conflictual relationships with teachers. Significant, direct paths were found between children’s classroom relationships and subsequent participation in classroom activities, even after controlling for various ‘‘entry’’ factors that have been shown to predict school engagement (e.g., family backgrounds, ethnicity, child’s gender, IQ). The strongest of these paths emanated from negative features of their relationships (i.e., peer rejection) lending support to the hypothesis that peer group rejection operates as an impediment to children’s engagement in classroom social and scholastic activities. Also, consistent with past research on the antecedents of scholastic progress, direct, positive pathways were found from classroom participation to achievement (see Finn, 1989, 1993; Wentzel, 1991). Overall, these findings were consistent with a mediated model of early school adjustment and suggested that, as children enter school, their initial behavioral orientations influenced the types of relationships they formed with peers. In particular, young children’s use of force or coercive tactics was directly associated with rejection by the peer group. Furthermore, once children were rejected by their classmates, they were less likely to participate in classroom activities, suggesting that this form of relational adversity (e.g., peer rejection) interferes with children’s involvement in learning activities and eventually impairs their achievement. My colleagues and I have conducted additional longitudinal studies to further elucidate how classroom peer and teacher relationships might affect the adjustment of children who are prone toward aggressive behavior. In one such investigation, Ladd and Burgess (2001) sought to determine


79

Fig. 2. One of the two child behavior and peer environment models estimated by Ladd, Birch, and Buhs (1999); see Figure 2, page 1383. Reprinted with permission from the Society for Research in Child Development.

whether aggressive children’s participation in different types of classroom relationships might increase (e.g., exacerbate) or decrease (i.e., compensate for) their probability of developing psychological and school adjustment problems. We assessed not only children’s aggressive dispositions but also relationship risks (i.e., classroom peer rejection, victimization) and resources (i.e., classroom peer acceptance, mutual friendships) over a two-year period, from kindergarten through grade 1. Most of the findings obtained in this investigation were consistent with an additive model, or the view that children’s classroom relationships alter the probability that they will suffer maladjustment regardless of whether they exhibit confrontive styles of aggression. One relational risk—peer group rejection—was associated with increases in children’s thought problems and decreases in their classroom participation, positive school attitudes, and achievement. Thus, peer-group

80

Gary W. Ladd

rejection appeared to exacerbate (i.e., add to) the adjustment difficulties of aggressive children. In contrast, relational resources tended to predict decreases in maladjustment independently of aggression, suggesting that these factors inhibit the development of maladjustment, regardless of children’s risk status for aggression. After accounting for children’s initial risk status for aggression, early peer acceptance was associated with relative declines in attention problems and misconduct, and relative gains in cooperative participation and school liking. This evidence suggests that acceptance by classmates provides children with a sense of belongingness or promotes inclusion in classroom activities that decreases the likelihood that they will engage in resistive behavior patterns, form negative school attitudes, and disengage from school tasks. However, not all of the findings reported by Ladd and Burgess (2001) conformed to this general pattern. Exceptions included instances in which the linkage between aggression and maladjustment were contingent upon classroom relational risks or resources, lending support to a moderated model of adaptation. For aggressive children in particular, exposure to peer victimization was associated with gains in thought problems. Taken together, these findings strengthen the credibility of models in which it is assumed that both behavioral risks and relational experiences affect early-emerging patterns of maladjustment. Moreover, this evidence contradicts the view that relational factors have little or no adaptive significance beyond that attributable to manifest behavioral risks (see Parker & Asher, 1987). B. CHILD BY ENVIRONMENT PREMISE 2: IN CONJUNCTION WITH BEHAVIORAL STYLES, ENDURING PEER RELATIONSHIPS HAVE GREATER EFFECTS ON CHILDREN’S ADJUSTMENT THAN BRIEF OR TRANSIENT PEER RELATIONSHIPS

Here, the first child by environment premise is elaborated by incorporating the hypothesis that, along with behavioral styles, the contributions of peer relationships to children’s adjustment depends not only on the functional properties of particular peer relationships (e.g., peer group rejection vs. friendship), but also upon the duration of children’s participation in these relationships (i.e., history of exposure to relationship risks or resources). In part, this logic originates within theories of psychological risk, stress, and support (Dohrenwend & Dohrenwend, 1981; Johnson, 1988; Lazarus, 1984) in which it is argued that the likelihood that children will become maladjusted is increased by chronic relational risks and decreased by sustained relational resources. Accordingly, it would be expected that prolonged rather than brief exposure to relational adversity


81

or deprivation (e.g., history of peer rejection, victimization, friendlessness) or relational resources or support (e.g., history of peer group acceptance) would have greater consequences for children’s psychological and school adjustment. Few researchers have investigated whether children’s future adjustment varies as a function of their behavior and their sustained versus transient participation in different types of peer relationships. Those investigators who have examined predictive links between children’s peer relationship histories and adjustment have typically done so without estimating the potential contributions of children’s behavioral styles (e.g., DeRosier et al., 1994; Kochenderfer-Ladd & Wardrop, 2001). Another shortcoming is that researchers have not designed longitudinal studies so as to determine whether a history of peer relationship risks or resources are influential in shaping children’s adjustment above and beyond the more immediate strains or supports that they may be experiencing in their contemporary peer relationships. Additionally, little is known about whether distinct peer relationship histories function as unique or redundant pathways to health or dysfunction (i.e., potential multifinality, equifinality; see Gottlieb, Wahlsten, & Lickliter, 1998). Sustained friendlessness, for example, may deprive children of emotional resources that are essential for healthy development, such as emotional support, instrumental aid, and companionship (Furman & Robbins, 1985; Ladd et al., 1996). In contrast, sustained peer group rejection or victimization may foster maladjustment by exposing children to chronic stressors such as exclusion and maltreatment (see Ladd et al., 1997). Ladd and colleagues (Ladd & Burgess, 2001; Ladd & Troop, in press) have attempted to address these limitations by investigating how children’s sustained participation in stressful peer relationships (e.g., chronic peer rejection, victimization) and/or supportive peer relationships (e.g., stable peer acceptance, friendships) interface with a known behavioral risk (aggressive dispositions) to influence early-emerging psychological and school adjustment. Ladd and Burgess (2001) used prospective longitudinal assessments to assess children’s risk status for aggression and stressful versus supportive peer relationships as they entered grade school (initial behavioral and relational status) and as they progressed through the first two years of grade school (enduring aggressive and relational status). Predictive associations between these measures and changes in children’s psychological and school adjustment were then examined. Results from variable-oriented analyses were largely consistent with expectations; compared to early onset indicators, measures of the chronicity of children’s aggressive risk status and duration of relational stressors/supports were stronger predictors of changes in maladjustment. Analyses of aggressive

82

Gary W. Ladd

risk status showed that, after adjusting for initial scores, the chronicity of children’s status on this dimension across grades predicted changes in a host of school adjustment criteria, including increases in attention problems, thought problems, and behavioral misconduct, and decreases in cooperative classroom participation, and academic achievement. In addition to children’s aggressive risk status, the duration of peer group rejection and peer group acceptance were examined as a relational stressor and a relational support, respectively. After controlling for initial peer group rejection and the chronicity of children’s aggressive risk status and other relational risks, the chronicity of peer group rejection predicted many of the same forms of school maladjustment that were associated with aggression. To be specific, chronic peer group rejection forecasted increases in attention problems, and decreases in cooperative classroom participation and academic achievement. In contrast, application of the same analytic strategy showed that sustained peer group acceptance predicted decreases in children’s attention problems and gains in cooperative classroom participation. Furthermore, person-oriented analyses revealed that when compared against children who were not aggressive and were not at risk, children who were chronically aggressive but had a higher ratio of relational stressors to supports showed significant gains in psychological and school maladjustment trajectories as they progressed through the early grades. These children evidenced increases in attention problems, thought problems, and behavioral misconduct. Perhaps even more intriguing was the finding that children who were chronically aggressive but had a higher ratio of supports to stressors evidenced significant decrements in maladjustment over the same period. Included among these were decreases in attention problems, thought problems, and behavioral misconduct, and increases in cooperative classroom participation, and feelings of liking toward school (see Figure 3). These findings corroborated the inference that a powerful behavioral risk (aggressiveness) can be exacerbated by chronic relational risks but buffered by stable relational supports, further illustrating the importance of children’s relationship histories. Additional analyses, conducted by ethnic groups, suggested that African-American children, who were typically a minority among their European-American classmates, were more likely to experience particular stressors (e.g., chronic peer rejection), and were less likely to be afforded some forms of support (e.g., stable friendships). However, the predictive linkages found between the relational risk/ protective factors and later maladjustment did not differ substantially by ethnicity or socioeconomic status. A follow-up prospective longitudinal investigation conducted from kindergarten to grade 4 (Ladd & Troop, in press) overcame many

A Child by Environment Perspective 83

Fig. 3. Data used to construct this figure was taken from Ladd & Burgess (2001). The plots shown depict group means over time for children who were chronically aggressive but had a higher ratio of relational stressors to supports (i.e., the aggressive and relational risks, or ARR group), children who were chronically aggressive but had a higher ratio of supports to stressors (i.e., the aggressive and relational supports, or ARS group) and children who were not aggressive and were not at risk (i.e., the risk-free, or RF group) on measures of psychological and school adjustment from fall of kindergarten to spring of grade 1. Reprinted with permission from the Society for Research in Child Development.

84

Gary W. Ladd

limitations typical of past research by testing novel premises about the contributions of aggressive and anxious behavioral dispositions and histories of multiple forms of peer relationship adversity (i.e., chronic peer group rejection, chronic peer victimization, and chronic friendlessness). Central to this investigation was the hypothesis that children who participate in adverse peer relationships over longer periods of time have greater exposure to negative relational processes or learning experiences (e.g., sustained exclusion, abuse, lack of dyadic emotional support), and that the accumulation of such experiences—essentially a form of dysfunctional socialization—is a more powerful risk factor than are the strains or adversities present in their contemporary peer relationships. The evidence obtained in this investigation revealed that children prone toward anxious-withdrawal tended to exhibit later internalizing problems, but this disposition was not a significant antecedent of chronic peer relationship problems. In contrast, children disposed toward aggressive behavior were likely to experience peer relationship problems and, for these children, chronic more than current relational adversity forecasted later maladjustment. In particular, children with aggressive dispositions tended to: (a) remain friendless which, in addition to their behavioral disposition, forecasted later internalizing problems, and (b) experience persistent peer rejection and victimization which, in addition to their disposition, predicted later externalizing problems. Estimation of these linkages via structural equations analyses of both hypothesized and alternative models consistently showed that chronic friendlessness, rejection, and victimization, were positively and directly linked with later forms of maladjustment. Because these paths were adjusted for children’s concurrent peer relationships, the results of this investigation were consistent with the hypothesis that chronic peer relationship adversity, more than the strains of contemporary peer relationships, antecede later maladjustment. Moreover, these findings suggested that this was uniquely the case for each of the investigated forms of chronic relational adversity.

IV. Summary and Conclusions Accumulating evidence is perhaps best understood when we can evaluate the extent to which it conforms to (or fails to corroborate) differing theoretical premises. At issue in this area of investigation is the extent to which evidence supports differential premises concerning the role of children’s behavioral styles and relational risks and resources as antecedents of psychological and school adjustment.


85

A. THEORY AND EVIDENCE: GOODNESS OF FIT?

At the broadest level of analysis, the evidence that has accumulated thus far can be interpreted as corroborating multiple theoretical positions. First, a substantial body of evidence indicates that children’s early behavioral dispositions are moderately stable and predictive of later maladjustment. These findings are consistent with ‘‘main effects’’ models, in which it is argued that the child’s early-emerging behavioral dispositions contribute to later maladjustment. Second, a basic tenet of environmental or socialization perspectives has been substantiated by evidence indicating that children’s involvement in particular types of peer relationships is moderately stable and predictive of future maladjustment. These findings are also consistent with a ‘‘main effects’’ perspective. However, research guided by ‘‘child and environment’’ models has also produced discoveries that qualify some of the inferences that investigators have drawn from studies based on ‘‘main effects’’ models. These discoveries suggest that, over the course of development, children’s behavioral styles and their participation in peer relationships essentially co-determine their success in adapting to life- and school-based challenges. In particular, the largely correlational evidence reviewed in this chapter lends itself to several heuristic, although admittedly speculative, conclusions. First, although children’s behavioral dispositions and features of their peer relationships are significant antecedents of later adjustment, the predictive power of either factor alone appears to be less than their additive, contingent, or mediated contributions. For example, children’s early behavioral dispositions, especially those that make children prone toward aggressive interactions with peers may affect the nature of the relational ecology (i.e., form and nature of relationships) they develop within the school context. Especially as children enter new peer groups (e.g., at school entry), the risks posed by aggressive dispositions may be partially mediated through the nature of the relationships (e.g., relational risks and resources) they form in peer group settings such as classrooms. Second, along with children’s behavioral styles, exposure to enduring relationship adversity (e.g., peer rejection), deprivation (e.g., friendlessness), or support (e.g., peer acceptance) is more closely associated with children’s adjustment trajectories than are more transient or proximal experiences within these same relationship domains. In this sense, extant findings not only illustrate the adaptive significance of children’s peer relationships, but also suggest that sustained relational adversity affects children in ways that are consistent with chronic stress rather than acute, life-events perspectives (e.g., see DeRosier et al., 1994; Johnson, 1988).

86

Gary W. Ladd

Third, risks posed by children’s behavioral dispositions may be exacerbated by enduring relationship adversity (e.g., chronic victimization), and buffered by stable relationship advantage (e.g., a history of peer acceptance). These findings are consistent with moderator models in which it is argued that children’s peer experiences affect their adjustment by altering the continuity of their behavioral styles. B. CONCEPTUAL AND EMPIRICAL AGENDAS

Child and environment models, as applied to the study of psychological and school adjustment, provide a framework for conceptualizing the forces that affect (i.e., promote vs. interfere) children’s adaptation to developmental and ecological challenges. Inherent in this model are fundamental assumptions about: (a) the locus of these forces (i.e., originating within the child, the environment, or both), (b) the observable manifestations of these forces (e.g., the child’s underlying constitution, regulatory abilities, early learning experiences, and the like, are expressed in specific behavioral styles; principles of social life and group dynamics create interpersonal contingencies and relationship processes that impinge on the individual child), and (c) the means by which these forces combine to affect the individual’s adaptive success and ensuing health or dysfunction (e.g., the effects of child and environmental factors on adjustment can be understood as additive, moderated, or mediated contributions). Although all three assumptions represent a point of departure for hypothesis generation and testing, the last (i.e., assumption ‘‘c’’) deserves attention because it epitomizes the child and environment model and distinguishes it from main effects perspectives. At present, four categories of frameworks—termed continuity, additive, moderated, mediated models— have been advanced and investigated as a means of elucidating how child and environmental factors operate together to influence children’s psychological or school adjustment. Unfortunately, little effort has been made to delineate the differing theoretical positions that are represented in these models, nor to consider the implications of these positions for future research. Such an assessment is worth undertaking because it has the potential to sharpen the distinctions between competing frameworks, distill existing knowledge, and guide subsequent model development and evaluation. 1. Continuity Models In a behavior-continuity model, it is assumed that the social environment does little more than maintain the child’s behavior style. Experience in the interpersonal environment (e.g., peer interactions and relationships) is considered to be essential because it creates consequences that sustain (e.g.,


87

reinforce) children’s pre-existing dispositions, but it does not alter children’s behavioral dispositions nor make a distinct contribution to their adjustment. Investigators who embrace this perspective (e.g., Caspi, Elder, & Bem, 1987, 1988; Caspi, 1998) tend to work from the assumption that children’s behavioral dispositions are based on constitutional factors (e.g., differences in genes, temperament, regulatory capacities) or early learning experiences (e.g., secure/insecure attachments, intrusive or coercive parent–child relations), and that these dispositions consistently orient children toward social contexts that perpetuate those dispositions (i.e., ‘‘cumulative continuity’’; Caspi, Elder, & Bem, 1987; p. 308). Within these social contexts, children’s behavioral styles elicit interactions and cultivate relationships that reinforce and sustain these dispositions (i.e., ‘‘interactional continuity’’; Caspi, Elder, & Bem, 1987; p. 309). Thus, specific types of input from the social environment are seen as necessary for maintaining children’s behavioral styles (e.g., interaction patterns), and continuity in these styles, in turn, promotes and perpetuates maladjustment. Conversely, the absence of such input would reduce continuity in children’s behavioral styles and make it less likely that they would become maladjusted. As an illustration, Caspi, Elder, and Bem (1988) proposed that shy children tend to seek solitary contexts, and by doing so, reduce the probability that they will have to interact with others and, thereby, sustain their passive dispositions. Caspi, Elder, and Bem (1988) further contended that, even if these children were placed in social contexts, their behavioral style would tend to elicit nonresponsive reactions from others (e.g., being overlooked or ignored) and, therefore, sustain their passive dispositions. 2. Additive Models In an additive model, the social environment’s contribution to children’s adjustment is seen as distinct from that made by their behavioral styles. It is assumed that children’s experiences in the peer environment can have positive or negative effects on adjustment, but such effects are ‘‘additive’’ when they occur together with other risk factors, such as the child’s behavioral style. Ladd and Burgess (2001), for example, have proposed that children with aggressive behavioral dispositions may develop peer relationships that, depending on their form, expose them to higher levels of interpersonal stress or support. When participating in adversarial peer contexts, aggressive children are likely to experience interpersonal stress (e.g., peer group rejection, peer victimization) whereas in cooperative peer settings (or contexts in which aggressive behavior is more the norm; see DeRosier et al., 1994; Wright, Giammarino, & Parad, 1986), they are likely to experience interpersonal support (e.g., peer group acceptance, friendships with other

88

Gary W. Ladd

aggressive children). From the perspective of an additive model, whether children are aggressive or not, stressful peer relations would be seen as promoting maladjustment whereas supportive peer relations would be viewed as facilitating health (or reducing maladjustment). Thus, in the presence of one risk factor (i.e., the child’s aggressive disposition), the effect of a second risk factor (i.e., participation in stressful peer relationships) ‘‘adds’’ to the effect of the child’s disposition (i.e., exacerbate an existing risk) or ‘‘subtracts’’ from the effect of the child’s disposition (e.g., compensate for an existing risk). Theoretically, the logic of an additive model could be extended to say that children’s likelihood of becoming maladjusted is equal to the sum of the risk factors minus the sum of the resources that are operative in the child and the child’s environment. 3. Moderator Models The central premise of a moderator model is that the extent to which children’s dispositions determine whether they become healthy or maladjusted is contingent upon the presence of particular peer experiences. In this model, the joint effects of the child’s disposition and peer experience may create more extreme adjustment consequences than would the additive effects of the individual components. Ladd and colleagues (e.g., 1996, 1999; Gazelle and Ladd, 1999; 2003) have postulated that peer experience does more than maintain children’s dispositions, it also has the capacity to alter (i.e., amplify or modulate) their behavioral styles, thereby increasing or decreasing risk for later maladjustment. Gazelle and Ladd (2003), for example, have argued that many anxious withdrawn children tend to have social fears that make them reserved in social situations and, thus, render them especially vulnerable to the effects of peer exclusion. Some of these children will be targeted for peer exclusion, depending on how they behaviorally manifest these fears, and the dynamics or norms of their peer context. Those who are excluded will have a social experience that is likely to confirm or intensify their anxieties or social fears. By substantiating anxious withdrawn children’s fears (e.g., fear of mistreatment), peer exclusion is likely to increase rather than maintain the stability of their solitary tendencies, and place them at greater risk for dysfunction. In contrast, anxious withdrawn children whose peer groups are less prone to exclude them will not acquire an experiential basis for their social fears and, thus, be less prone to isolate themselves or develop corresponding forms of dysfunction. 4. Mediator Models In this model, the links from children’s behavioral dispositions to peer relationships, and from their peer experiences to adjustment, are


89

conceptualized as a series of cause-effect linkages. That is, behavioral dispositions are seen as the principal cause of peer relationships and, once formed, peer relationships become the proximal cause of adjustment. As an illustration, Ladd, Birch, and Buhs (1999) proposed a series of propositions about the behavioral and relational antecedents of kindergarten children’s achievement that followed the logic of a mediator model. First, they proposed that entry factors, or child and environmental influences that operate prior to school entry (e.g., child’s IQ, family SES) affect kindergarten achievement indirectly because they are mediated through child and environmental factors that are present in the school setting, including children’s classroom behavioral styles, and peer relationships. This proposition was predicated on the premise that organismic and background variables affect the way that children behave in the classroom milieu, and develop resources in this context (i.e., supportive vs. stressful peer and teacher relationships). The second proposition was that children’s behavioral styles in the classroom directly influence the types of relationships they form with peers and teachers. It was expected that children with antisocial behavioral styles would tend to become friendless and rejected by classmates because their style of relating (e.g., use of coercive, forceful behavioral styles) supplies partners with a higher ratio of costs (i.e., negative consequences) relative to benefits. Greater success at forming friendships, peer group acceptance, and close relationships with teachers was anticipated for children with prosocial orientations because their behaviors (e.g., cooperating, helping) provide partners with a larger proportion of positive or equitable consequences. Next, it was proposed that, once formed, classroom relationships would facilitate or impede children’s participation in classroom activities. It was hypothesized that classroom friendships offer children a sense of support or security, and that these provisions promote adaptive behaviors such as participation and exploration in the classroom. Because disliked children tend to be avoided by peers and denied access to peer activities, it was expected that peer group rejection would negatively impact classroom participation by undermining children’s opportunities to engage in scholastic tasks with peers. The final proposition was that a higher level of classroom participation contributes to children’s achievement. Thus, in this model, children’s behavioral dispositions were construed as the proximal cause of their classroom relationships. Once formed, these relationships, rather than children’s behaviors, were seen as creating conditions (i.e., social resources or constraints) that either empowered or restrained children from participating in classroom activities.

90

Gary W. Ladd

C. IMPLICATIONS FOR FUTURE RESEARCH

Each of these models embodies a different view of how children’s behavioral dispositions and peer experiences influence each other and/or children’s trajectories toward health and maladjustment. Yet, rarely have these views been compared with an eye toward identifying novel, and potentially testable hypotheses about the contributions of either child or environmental factors to children’s adjustment. In one respect, these models portray several variations on the social environment’s role as a potential precursor of children’s adjustment. In continuity models, the social environment’s role is more reactive than proactive in relation to the child’s behavioral style. Children are seen as active in the sense that their dispositions select them into distinct peer niches that supply responses that are uniquely suited to maintaining their dispositions. Because the social environment essentially supports risky behavior patterns, it is postulated that children consistently respond to developmental challenges in maladaptive ways that promote and maintain dysfunction over the life cycle. This view is in contrast to a more proactive view of the social environment that is typically represented within the other three models. In these paradigms, the peer environment tends to be portrayed as a ubiquitous socialization context that all children pass through and, within which, children are exposed to different forms of peer influence (e.g., periods of having friends and not having friends; brief or chronic periods of peer rejection, victimization). Unfortunately, in much of the research based on continuity models (e.g., Caspi et al., 1987, 1988), the social environment and its relation to children’s behavioral styles has not been assessed empirically, or prospectively over time. The absence of such data has made it difficult to evaluate one of the model’s basic premises—that is, whether children with particular dispositions participate in functionally similar peer environments over time, and whether these experiences are closely linked to children’s behavior but have little effect on their adjustment. In future studies, it may be possible to investigate the extent to which the features of children’s peer interactions and relationships remain selectively matched to (or largely interdependent on) their behavioral dispositions over extended periods of time, and across major developmental transitions. Evidence to the contrary could be seen as lending support to alternative perspectives. Data indicating that there is less interdependence between children’s behavior and the nature of their peer experiences might shift investigation toward models in which the social environment is portrayed as a more active contributor to children’s adjustment. Among the alternatives would be models in which the contributions of children’s peer experiences to


91

adjustment are construed as independent of children’s behavioral styles (an additive perspective), contingent on their behavioral styles (a moderator perspective), or intervening between the child’s behavioral styles and adjustment (a mediator perspective). As used in research on children’s psychological and school adjustment, additive, moderator, and mediator models diverge from continuity perspectives in that different assumptions are made about: (1) how closely children’s peer experiences are linked with their behavioral styles, and (2) how influential children’s peer experiences are in shaping their adjustment. Additive versus mediated models portray the most divergent views as to how closely children’s behavior is associated with the nature of their peer experiences. In additive models, it is not necessarily assumed that children’s behavioral dispositions determine the nature of their peer relations and, thus, such experiences may alter children’s development in ways that are distinct from their behavioral dispositions. Consistent with this view is the proposition that the peer milieu, regardless of the behavioral styles children bring to it, creates relatively ubiquitous challenges that have the potential to transform children’s development or adjustment. It might be argued, for example, that nearly all children have arguments with agemates (e.g., over rules for play activities) and thus learn basic moral principles (e.g., Piaget, 1965), or that nearly all children are exposed to teasing and thus are forced to experience peer-induced stress and learn about conflict management (Green, 1933; Jersild & Markey, 1935; Shantz, 1987). In contrast, mediated models resemble continuity models in relation to the first assumption but not the second assumption. As in continuity models, it is assumed that children’s behavioral dispositions shape the nature of their peer experiences and so, it is expected that these two factors will be highly correlated. However, on the second assumption, the logic of mediated perspectives departs substantially from that found in continuity models. Although peer experiences are expected to be highly associated with children’s behavior, these experiences are construed as a proxy for children’s behavioral dispositions, one that intervenes between children’s behavior and adjustment and is principally responsible for changes in children’s adjustment. Although this position is among the least well investigated of the various models, there is some evidence to suggest that the effects of children’s behavior on their motivation to engage in scholastic activities is largely indirect, or transmitted through the quality of the peer- and teacherbased support systems they develop in the classroom (e.g., Buhs & Ladd, 2001; Ladd, Birch, & Buhs, 1999). The validity of these different models, and their utility as paradigms likely to produce important new empirical discoveries, may be debated in light of extant empirical evidence. For example, there is considerable evidence in the

92

Gary W. Ladd

peer relations literature to indicate that children’s behavioral styles are predictive of the relationships they form with peers. Many of these findings suggest that aggressive children are more likely to become rejected by their classmates, have fewer friends, and in some cases, be at risk for peer victimization. Other findings show that children who are prone toward solitary or withdrawn behavior are more likely to be friendless and targeted for peer victimization (e.g., see reviews by Coie, Dodge, & Kupersmidt, 1990; Graham & Juvonen, 1998a; Ladd, 1999). This kind of evidence would seem to corroborate the first assumption of continuity and mediator models—that is, children’s behavioral styles are closely associated with the nature of their peer experiences. Yet, it is also clear from this literature that not all aggressive children are rejected by their classmates, nor are all withdrawn children friendless or victimized by peers (Bierman, Smoot, & Aumiller, 1993; Coie et al., 1991; Hymel, Bowker, & Woody, 1993; Perry, Kusel, & Perry, 1988). For example, Coie et al. (1991) estimated that no more than half of aggressive children are rejected by their peers. Thus, although some association has been documented between children’s behavioral styles and their peer relations, the magnitude of this linkage appears to be modest rather than robust. These findings, although somewhat equivocal, would appear to elevate the status of continuity, mediator, and moderator models as potential paradigms for future research. However, when the logic of these models is also weighed against evidence pertinent to the second assumption—that is, findings from studies in which both child and environmental predictors have been examined—all four perspectives appear to receive some degree of empirical support (although much of the evidence thus far would appear to be consistent with an additive interpretation). Given that there is insufficient evidence against which to judge the validity of differing child and environment models, it may be that the contemporary worth of these perspectives lies more in their ability to spark scholarly debate than it does in their ‘‘truthfulness’’ as scientific explanations. The process of comparing models, and their implications for the study of children’s adjustment, will encourage social scientists to argue the merits of differing theoretical positions and, in this process, generate controversies that will raise novel questions and stimulate new research. For example, a comparison of how the social environment is viewed within continuity and moderator models raises an under-researched question: Is it that the social environment acts to maintain behavior styles over time and it is the continuity of children’s behavioral styles that continues to create adjustment problems (Caspi et al., 1987, 1988), or is it that children’s dispositions make them vulnerable to the effects of certain kinds of peer experience, and the interaction of these factors is what causes maladjustment? In the former


93

case, one would expect to find that children’s behavioral styles consistently predicted dysfunction across development and that, once this factor had been accounted for, measures of children’s peer experiences would have little or no predictive power. Conversely, the former perspective could be interpreted as suggesting that children’s behavioral dispositions are but one facet of a larger syndrome or array of child characteristics (e.g., associated cognitive, affective propensities), and that some peer experiences tend to accentuate these vulnerabilities more than others. This logic is similar to that found in diathesis-stress frameworks in which it is argued that individuals with certain vulnerabilities are particularly susceptible to the effects of stress. Investigations based on this latter model might make it possible to differentiate between those social conditions that have little effect on children’s adjustment beyond their behavioral styles (in which case, only children’s behavioral styles would predict dysfunction) from those that tend to activate children’s vulnerabilities in ways that exaggerate their behavioral dispositions and risk for maladjustment. Although some findings are consistent with this hypothesis (e.g., Gazelle & Ladd, 2003), much remains to be learned about whether certain behavioral dispositions and accompanying propensities make children especially vulnerable to particular forms of peer influence. Furthermore, we need to know more of the validity and generalizability of continuity versus moderator models (and other competing perspectives) by evaluating them in model-comparison analyses rather than by assessing the predictive efficacy of each model separately with different samples, studies, and analytic tools (as has been typical in past research). Also at issue is the question of whether children’s behavior and the experiences they have in the peer environment have lasting or temporary effects on their adjustment. For example, there is some evidence to suggest that the nature and forms of peer experience that children encounter in classrooms change dramatically across developmental transitions (e.g., from preschool to grade school; from late childhood into adolescence; see Berndt & Keefe, 1995; Cairns & Cairns, 1994; Ladd, 1990; Ladd & Price, 1987). Such findings might lead us to expect that children’s peer experiences differ qualitatively across transitions and, thus, have effects on adjustment that are limited to particular developmental periods. Evidence of this type would argue against the idea that children’s peer experiences remain aligned with their behavioral dispositions across the life cycle and, instead, lead us to suspect that there may be developmental discontinuities or ‘‘turning points’’ (Rutter, 1996) where children’s peer experiences no longer support their dispositions and, therefore, permit or promote changes in their behavior. If the nature and form of children’s peer experiences shift over time, then it may be important to conceptualize the role of the social environment as a

94

Gary W. Ladd

dynamic force in shaping children’s adjustment. Development and socialization regularly expose children to challenges and periods of vulnerability and, during these transitions (see Ladd, 1996), the adaptive significance of the child’s attributes and ecological/environmental systems may vary. Similarly, early-emerging forms of maladjustment may not be static, but rather evolve into other (and possibly multiple) forms of dysfunction over the life cycle (see Coie et al., 1993). If this is the case, then the adaptive value of the child’s characteristics or environmental inputs may not be enduring; rather, their effects may change with age and experience (see Parker et al., 1995). Alternatively, some behavioral or environmental influences that appear transient may actually be enduring because the modes through which these forces are expressed may change as children grow older. Thus, a fundamental question is whether children’s peer experiences, in combination with their behavior, have effects on adjustment that endure across development, or are limited to particular life stages. On the one hand, as some have suggested (Bowlby, 1973; Freud & Dann, 1951; Rutter, 1979), early childhood may be a sensitive period for social development and certain types of peer experiences during this stage have lasting effects on children’s behavior or adjustment. Early behavioral styles and peer experiences may crystallize into enduring patterns that promote stable or even escalating maladjustment trajectories. On the other hand, as children’s peer experiences change, the effects of these experiences on their behavioral dispositions and adjustment may change as well. Transitions across developmental periods may destabilize established maladaptive behavioral tendencies and relationships, allowing for a progression toward more or less adaptive developmental outcomes. Thus, although we need to focus on delineating mechanisms that underlie continuity over time, we also must identify predictors of changes in developmental trajectories as children move from childhood through adolescence. In sum, research that has been developed from a child by environment perspective is at an early stage and the evidence that has accumulated thus far provides, at best, a preliminary look at a highly complex phenomenon. In all likelihood, multiple risks and resources come to bear upon children from many sources, including the intrapersonal and interpersonal realms that have been touched upon in this article. Moreover, further investigation may reveal that the predictive utility of the various child by environment models differs across adjustment domains. Should this occur, multiple, disorder-specific models may be needed to plot the pathways that children follow toward different forms of health or dysfunction. Furthermore, many of the personal and interpersonal risks and resources that were considered here may not be stable across developmental transitions and, thus, represent dynamic factors that operate and affect each other differently over the


95

course of the child’s development. As a result, it will be important for researchers to discover not only how these influences change over time, but also when during development these factors orient children toward different adjustment trajectories. Ultimately, the progress of this line of investigation will depend, in part, on the extent to which existing, evolving, or newlyemerging frameworks are able to aid researchers in uncovering significant empirical discoveries.

REFERENCES Achenbach, T. M., & Edelbrock, C. S. (1981). Behavioral problems and competencies reported by parents of normal and disturbed children aged four through sixteen. Monographs of the Society for Research in Child Development, 46 (Serial No. 188). Alexander, K., & Entwisle, D. R. (1988). Achievement in the first two years of school: Patterns and processes. Monographs of the Society of Research in Child Development, 53 (2, Serial No. 218). Alsaker, F. (1993, March). Bully/victim problems in day-care centers, measurement issues and associations with children’s psychosocial health. Paper presented at the biennial meeting of the Society for Research in Child Development, New Orleans, LA. Asendorpf, J. B. (1993). Beyond temperament: A two-factorial coping model of the development of inhibition during childhood. In K. H. Rubin & J. B. Asendorpf (Eds.), Social withdrawal, inhibition, and shyness in childhood (pp. 265–289). Hillsdale, NJ: Lawrence Erlbaum. Asher, S. R., & Gottman, J. M. (1981). The development of children’s friendships. New York: Cambridge University Press. Asher, S. R., Hymel, S., & Renshaw, P. D. (1984). Loneliness in children. Child Development, 55, 1456–1464. Asher, S. R., Singleton, L. C., Tinsley, B. R., & Hymel, S. (1979). A reliable sociometric measure for preschool children. Developmental Psychology, 15, 443–444. Bagwell, C. L., Newcomb, A. F., & Bukowski, W. M. (1998). Early adolescent friendship and peer rejection as predictors of adult adjustment. Child Development, 69, 140–153. Berndt, T. J. (1996). Exploring the effects of friendship quality on social development. In W. M. Bukowski, A. F. Newcomb, & W. W. Hartup (Eds.), The company they keep: Friendship in childhood and adolescence (pp. 346–365). Cambridge University Press. Berndt, T. J., & Burgy, L. (1996). Social self-concept. In B. A. Bracken (Ed.), Handbook of selfconcept: Developmental, social, and clinical considerations (pp. 171–209). New York: John Wiley & Sons. Berndt, T. J., & Keefe, K. (1995). Friends’ influence on adolescents’ adjustment to school. Child Development, 66, 1312–1319. Berndt, T. J., & Ladd, G. W. (1989). Peer relationships in child development. New York, NY: John Wiley. Bierman, K., Smoot, D. L., & Aumiller, K. (1993). Characteristics of aggressive-rejected, aggressive (nonrejected), and rejected (nonaggressive) boys. Child Development, 64, 139–151. Birch, S., & Ladd, G. W. (1996). Contributions of teachers and peers to children’s early school adjustment. In K. Wentzel & J. Juvonen (Eds.), Social Motivation: Understanding children’s school adjustment (pp. 199–225). New York, NY: Cambridge University Press.

96

Gary W. Ladd

Bjko¨rkqvist, K., Ekman, K., & Lagerspetz, K. (1982). Bullies and victims: Their ego picture, ideal ego picture and normative ego picture. Scandinavian Journal of Psychiatry, 23, 307–313. Boivin, M., & Hymel, S. (1997). Peer experiences and social self-perceptions: A sequential model. Developmental Psychology, 33, 135–145. Boivin, M., Hymel, S., & Bukowski, W. M. (1995). The roles of social withdrawal, peer rejection, and victimization by peers in predicting loneliness and depressed mood in childhood. Development and Psychopathology, 7, 765–785. Boulton, M. J., & Smith, P. K. (1994). Bully/victim problems in middle-school children: Stability, self-perceived competence, peer perceptions, and peer acceptance. British Journal of Developmental Psychology, 12, 315–329. Boulton, M. J., & Underwood, K. (1992). Bully/victim problems among middle school children. British Journal of Educational Psychology, 62, 73–87. Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation. New York: Basic Books. Buhrmester, D., & Furman, W. (1986). The changing functions of friends in childhood: A NeoSullivanian perspective. In V. J. Derlega & B. A. Winstead (Eds.), Friendship and social interaction (pp. 41–61). New York: Springer-Verlag. Buhs, E. & Ladd, G. W. (2001). Peer rejection in kindergarten: Relational processes mediating academic and emotional outcomes. Developmental Psychology, 37, 550–560. Bukowski, W. M., & Hoza, B. (1989). Popularity and friendship: Issues in theory, measurement, and outcome. In T. J. Berndt & G. W. Ladd (Eds.), Peer relationships in child development (pp. 15–45). New York: John Wiley & Sons. Bukowski, W. M., Newcomb, A. F., & Hartup, W. W. (1996). The company they keep: Friendship in childhood and adolescence. Cambridge University Press. Cairns, R. B., & Cairns, B. D. (1994). Lifelines and risks:Pathways of youth in our time. New York: Cambridge University Press. Cairns, R. B., Cairns, B. D., & Neckerman, H. J. (1989). Early school dropout: Configurations and determinants. Child Development, 60, 1437–1452. Cairns, R. B., Cairns, B. D., Neckerman, H. J., Ferguson, L. L., & Gariepy, J. L. (1989). Growth and aggression I: Childhood to early adolescence. Developmental Psychology, 25, 320–330. Caspi, A. (1998). Personality development across the life course. In W. Damon (Series Editor), R. M. Lerner (Volume Editor), Handbook of Child Psychology (5th Edition, Vol. 1, pp. 311–388), New York, NY: Wiley. Caspi, A., Elder, G. H., & Bem, D. J. (1987). Moving against the world: Life-course patterns of explosive children. Developmental Psychology, 23, 308–313. Caspi, A., Elder, G. H., & Bem, D. J. (1988). Moving away from the world: Life-course patterns of shy children. Developmental Psychology, 24, 824–831. Cassidy, J., & Asher, S. R. (1992). Loneliness and peer relations in young children. Child Development, 63, 350–365. Cillessen, A. H. N., Bukowski, W. M., & Haselager, G. J. T. (2000). Stability of sociometric categories. In A. H. N. Cillessen, & W. M. Bukowski (Eds.), Recent advances in the measurement of acceptance and rejection in the peer system (pp. 75–93). San Francisco, CA: Josey-Bass. Coie, J. D., Christopoulos, C., Terry, R., Dodge, K. A., & Lochman, J. E. (1989). Types of aggressive relationships, peer rejection, and developmental consequences. In B. Schneider, G. Attili, J. Nadel, & R. Weissberg (Eds.), Social competence in developmental perspective (pp. 223–238). Amsterdam: Kluwer Academic Publishers. Coie, J. D., & Dodge, K. A. (1983). Continuities and changes in children’s social status: A fiveyear longitudinal study. Merrill-Palmer Quarterly, 29, 261–282.


97

Coie, J. D., & Dodge, K. A. (1998). Aggression and antisocial behavior. In W. Damon (Series Editor) and N. Eisenberg (Volume Editor), Handbook of child psychology (Vol. 3) (pp. 779–862). New York: John Wiley. Coie, J. D., Dodge, K. A., & Coppotelli, H. (1982). Dimensions and types of social status: A cross-age perspective. Developmental Psychology, 18, 557–570. Coie, J. D., Dodge, K. A., & Kupersmidt, J. B. (1990). Peer group behavior and social status. In S. R. Asher & J. D. Coie (Eds.), Peer rejection in childhood (pp. 17–59). New York: Cambridge University Press. Coie, J. D., Dodge, K. A., Terry, R., & Wright, V. (1991). The role of aggression in peer relations: An analysis of aggression episodes in boys’ play groups. Child Development, 62, 812–826. Coie, J. D., Lochman, J. E., Terry, R., & Lee, C. (1987, November). Aggression and peer rejection as predictors of school adjustment during a transition to middle school. Paper presented at the annual meeting of the Association for Advancement of Behavior Therapy, Boston, MA. Coie, J. D., Lochman, J. E., Terry, R., & Hyman, C. (1992). Predicting adolescent disorder from childhood rejection and peer rejection. Journal of Consulting and Clinical Psychology, 60, 783–792. Coie, J. D., Watt, N. F., West, S. G., Hawkins, J. D., Asarnow, J. R., Markman, H. J., Ramey, S. L., Shure, M. B., & Long, B. (1993). The science of prevention: A conceptual framework and some directions for a national research program. American Psychologist, 48, 1013–1022. Cole, D. A., Peeke, L. G., Martin, J. M., Truglio, R., & Seroczynski, A. D. (1998). A longitudinal look at the relation between depression and anxiety in children and adolescents. Journal of Consulting & Clinical Psychology, 66, 451–460. Conger, J. J., & Miller, W. C. (1966). Personality, social class, and delinquency. New York: John Wiley. Connell, J. P., Spencer, M. B., & Aber, J. L. (1994). Educational risk and resilience in AfricanAmerican youth: Context, self, action, and outcomes in school. Child Development, 65, 493–506. Coplan, R. J. (2000). Assessing nonsocial play in early childhood: Conceptual and methodological approaches. In K. Gitlin-Weiner, A. Sandgrund, & C. Schaefer (Eds.), Play diagnosis and assessment (2nd ed., pp. 563–598). New York: Wiley. Coplan, R. J., Gavinski-Molina, M. H., Lagace-Seguin, D. G., & Wichmann, C. (2001). When girls and boys play alone: Nonsocial play and adjustment in kindergarten. Developmental Psychology, 37, 464–474. Coplan, R. J., & Rubin, K. H. (1998). Exploring and assessing nonsocial play in the preschool: The development and validation of the Preschool Play Behavior Scale. Social Development, 7, 73–91. Coplan, R. J., Rubin, K. H., Fox, N. A., Calkins, S. A., & Stewart, S. L. (1994). Being alone, playing alone, and acting alone: Distinguishing among reticence, and passive- and activesolitude in young children. Child Development, 65, 129–137. Cowen, E. L., Pedersen, A., Bagigian, H., Izzo, L. D., & Trost, M. A. (1973). Longterm followup of early detected vulnerable children. Journal of Consulting and Clinical Psychology, 41, 438–446. Crick, N. R. (1996). The role of overt aggression, relational aggression, and prosocial behavior in the prediction of children’s future social adjustment. Child Development, 67, 2317–2327. Crick, N. R., & Grotpeter, J. K. (1995). Relational aggression, gender, and social-psychological adjustment. Child Development, 66, 710–722.

98

Gary W. Ladd

Crick, N. R., & Ladd, G. W. (1993). Children’s perceptions of their peer experiences: Attributions, social anxiety, and social avoidance. Developmental Psychology, 29, 244–254. DeRosier, M. E., Cillessen, T., Coie, J. D., & Dodge, K. A. (1994). Group social context and children’s aggressive behavior. Child Development, 65, 1068–1079. DeRosier, M. E., Kupersmidt, J. B., & Patterson, C. J. (1994). Children’s academic and behavioral adjustment as a function of the chronicity and proximity of peer rejection. Child Development, 65, 1799–1813. Dodge, K. A. (1991). The structure and function of reactive and proactive aggression. In D. J. Pepler & K. H. Rubin (Eds.), The development and treatment of childhood aggression (pp. 201–218). Hillsdale, NJ: Erlbaum Associates. Dodge, K. A., & Coie, J. D. (1987). Social-information-processing factors in reactive and proactive aggression in children’s peer groups. Journal of Personality and Social Psychology, 53, 1146–1158. Dohrenwend, B. P., & Dohrenwend, B. S. (1981). Socioenvironmental factors, stress, and psychopathology. American Journal of Community Psychology, 9, 128–164. Dunn, J. (1993). Young children’s close relationships: Beyond attachment. London, England: Sage Publishers. Eckerman, C. O., & Stein, M. R. (1982). The toddler’s emerging interactive skills. In K. H. Rubin & H. S. Ross (Eds.), Peer relationships and social skills in childhood (pp. 41–71). New York: Springer-Verlag. Egan, S. E., & Perry, D. G. (1998). Does low self-regard invite victimization? Developmental Psychology, 34, 299–309. Ensminger, M. C., Kellam, S. G., & Rubin, B. R. (1983). School and family origins of delinquency: Comparisons by sex. In K. T. Van Dusen & S. A. Mednick (Eds.), Prospective studies of crime and delinquency (pp. 73–97). Hingham, MA: Kluwer-Nijhoff. Ensminger, M. E., & Slusarcick, A. L. (1992). Paths to high school graduation or dropout: A longitudinal study of a first-grade cohort. Sociology of Education, 65, 95–113. Farrington, D. P. (1979). Longitudinal research on crime and delinquency. In N. Morris & M. Toney (Eds.), Criminal justice: An annual review of research (Vol. 1, pp. 171–200). Chicago: University of Chicago Press. Feldhusen, J. F., Thurston, J. R., & Benning, J. J. (1973). Aggressive classroom behavior and school achievement. Journal of Special Education, 4, 431–439. Finn, J. D. (1989). Withdrawing from school. Review of Educational Research, 59, 117–142. Finn, J. D. (1993). School engagement and students at risk. Washington DC: Department of Education, National Center for Educational Statistics (ERIC Document Reproduction Service No. ED 362 322). Flemming, D., & Ricks, D. F. (1970). Emotions of children before schizophrenia and before character disorder. In M. Roff & D. F. Ricks (Eds.), Life history research in psychopathology (Vol. 1, pp. 240–264). Minneapolis: University of Minnesota Press. Frazee, H. E. (1953). Children who later become schizophrenic. Smith College Studies in Social Work, 23, 125–149. Freud, A., & Dann, S. (1951). An experiment in group upbringing. In R. Eisler, A. Freud, H. Hartmann, & E. Kris (Eds.), Psychoanalytic study of the child (Vol. 6, pp. 127–168). New York: International University Press. Furman, W., & Robbins, P. (1985). What’s the point? Issues in the selection of treatment objectives. In B. Schneider, K. H. Rubin, & J. E. Ledingham (Eds.), Children’s peer relations: Issues in assessment and intervention (pp. 41–54). New York: Springer-Verlag. Galen, B. R., & Underwood, M. K. (1997). A developmental investigation of social aggression among children. Developmental Psychology, 33, 589–600.


99

Garmezy, N., Masten, A. S., & Tellegen, A. (1984). The study of stress and competence in children: A building block for developmental psychopathology. Child Development, 55, 97–111. Gazelle, H., & Ladd, G. W. (1999, April). Solitude and interpersonal maladjustment in late childhood: Is peer perceived motivation for solitude related to relational difficulties? Paper presented at the biennial meetings of the Society for Research in Child Development, Albuquerque, NM. Gazelle, H., & Ladd, G. W. (2003). Anxious solitude and peer exclusion: A diathesis-stress model of internalizing trajectories in childhood. Child Development, 74, 257–278. Gottlieb, G., Wahlsten, D., & Lickliter, R. (1998). The significance of biology for human development: A developmental psychobiological systems view. In W. Damon (Series Editor) and Richard Lerner (Volume Editor), Handbook of child psychology (Vol. 1, pp. 233–274). New York: John Wiley. Graham, S., & Juvonen, J. (1998a). A social cognitive perspective on peer aggression and victimization. Annals of Child Development (pp. 23–70). London: Jessica Kingsley. Graham, S., & Juvonen, J. (1998b). Self-blame and peer victimization in middle school: An attributional analysis. Developmental Psychology, 34, 587–599. Green, E. (1933). Friendships and quarrels among preschool children. Child Development, 4, 237–252. Harrist, A. W., Zaia, A. F., Bates, J. E., Dodge, K. A., & Pettit, G. S. (1997). Subtypes of social withdrawal in early childhood: Sociometric status and social-cognitive differences across four years. Child Development, 68, 278–294. Hartup, W. (1970). Peer interaction and social organization. In P. H. Mussen (Ed.), Carmichael’s manual of child psychology, (Vol. 2, 3rd ed., pp. 360–456). New York: Wiley & Sons. Havighurst, R. J., Bowman, P. H., Liddle, G. P., Mathews, C. V., & Pierce, J. V. (1962). Growing up in River City. New York: John Wiley and Sons. Hawker, D. S. J. (1997). Socioemotional maladjustment among victims of different forms of peer aggression. Unpublished doctoral dissertation at Keele University, UK. Hinde, R. A., Titmus, G., Easton, D., & Tamplin, A. (1985). Incidence of ‘‘friendship’’ and behavior toward strong associates versus nonassociates in preschoolers. Child Development, 56, 234–245. Hoover, J. H., & Hazler, R. J. (1991). Bullies and victims. Elementary School Guidance & Counseling, 25, 212–219. Howes, C. (1983). Patterns of friendship. Child Development, 54, 1041–1053. Howes, C. (1988). Peer interaction of young children. Monographs of the Society for Research in Child Development, 53 (1, Serial No. 217). Howes, C., & Phillipsen, L. C. (1992). Gender and friendship: Relationships within peer groups of young childern. Social Development, 1, 231–242. Hymel, S., Bowker, A., & Woody, E. (1993). Aggressive versus withdrawn unpopular children: Variations in peer and self-perceptions in multiple domains. Child Development, 64, 879–896. Hymel, S., Rubin, K. H., Rowden, L., & LeMare, L. (1990). Children’s peer relationships: Longitudinal prediction of internalizing and externalizing problems from middle to late childhood. Child Development, 61, 2004–2021. Janes, C. L., & Hesselbrock, V. M. (1978). Problem children’s adult adjustment predicted from teachers’ ratings. American Journal of Orthopsychiatry, 48, 300–309. Janes, C. L., Hesselbrock, V. M., Meyers, D. G., & Penniman, J. H. (1979). Problem boys in young adulthood: Teachers’ ratings and twelve-year follow-up. Journal of Youth and Adolescence, 8, 453–472.

100

Gary W. Ladd

Jersild, A., & Markey, F. (1935). Conflicts between preschool children. Child Development Monographs: Teachers College Columbia University, No. 21, 1–181. John, R. S., Mednick, S. A., & Schulsinger, F. (1982). Teacher reports as a predictor of schizophrenia and borderline schizophrenia: A Bayesian decision analyses. Journal of Abnormal Psychology, 6, 399–413. Johnson, J. H. (1988). Life events as stressors in childhood and adolescence. Newbury Park, CA: Sage. Juvonen, J., & Graham, S. (2001). Peer harassment in School: The plight of the vulnerable and victimized. New York: The Guildford Press. Keefe, K., & Berndt, T. J. (1996). Relations of friendship quality to self-esteem in early adolescence. Journal of Early Adolescence, 16, 110–129. Kerr, M., Lambert, W. W., & Bem, D. (1996). Life course sequelae of childhood shyness in Sweden: Comparison with the United States. Developmental Psychology, 32, 1100–1105. Kochenderfer, B. J., & Ladd, G. W. (1996a). Peer victimization: Cause or consequence of school maladjustment? Child Development, 67, 1305–1317. Kochenderfer, B.J., & Ladd, G. W. (1996b). Peer victimization: Manifestations and relations to school adjustment. Journal of School Psychology, 34, 267–284. Kochenderfer-Ladd, B., & Ladd, G. W. (2001). Variations in peer victimization: Relations to children’s maladjustment. In J. Juvonen & S. Graham (Eds.), Peer harassment in school (pp. 25–48). New York: Guilford Press. Kochenderfer-Ladd, B., & Wardrop, J. (2001). Chronicity and instability in children’s peer victimization experiences as predictors of loneliness and social satisfaction trajectories. Child Development, 72, 134–151. Kohlberg, L., LaCrosse, J., & Ricks, D. (1972). The predictability of adult mental health from childhood. In B. Wolman (Ed.), Manual of child psychopathology (pp. 1217–1284). New York: McGraw-Hill. Kuhlen, R., & Collister, E. G. (1952). Sociometric status of sixth- and ninth-graders who fail to finish high school. Educational and Psychological Measurement, 12, 632–637. Kupersmidt, J. B., & Coie, J. D. (1990). Preadolescent peer status, aggression, and school adjustment as predictors of externalizing problems in adolescence. Child Development, 61, 1350–1362. Kupersmidt, J., Dodge, K. A., & Coie, J. D. (1990). The role of peer relationships in the development of disorder. In S. R. Asher & J. D. Coie (Eds.), Peer rejection in childhood (pp. 274–305). Cambridge University Press. Ladd, G. W. (1988). Friendship patterns and peer status during early and middle childhood. Journal of Developmental and Behavioral Pediatrics, 9, 229–238. Ladd, G. W. (1989). Children’s social competence and social supports: Precursors of early school adjustment? In B. Schneider, G. Attili, J. Nadel, & R. Weissberg (Eds.), Social competence in developmental perspective (pp. 277–292). Amsterdam: Kluwer Academic Publishers. Ladd, G. W. (1990). Having friends, keeping friends, making friends, and being liked by peers in the classroom: Predictors of children’s early school adjustment? Child Development, 61, 1081–1100. Ladd, G. W. (1996). Shifting ecologies during the 5–7 year period: Predicting children’s adjustment to grade school. In A. Sameroff & M. Haith (Eds.), The five to seven year shift (pp. 363–386). Univ. of Chicago Press. Ladd, G. W. (1999). Peer relationships and social competence during early and middle childhood. Annual Review of Psychology (Vol. 50, pp. 333–359). Palo Alto, CA: Annual Reviews Inc.


101

Ladd, G. W., Birch, S. H., & Buhs, E. (1999). Children’s social and scholastic lives in kindergarten: Related spheres of influence? Child Development, 70, 1373–1400. Ladd, G. W., Buhs, E., & Seid, M. (2000). Children’s initial sentiments about kindergarten: Is school liking an antecedent of early classroom participation and achievement? Merrill-Palmer Quarterly, 46, 255–279. Ladd, G. W., & Burgess, K. B. (1999). Charting the relationship trajectories of aggressive, withdrawn, and aggressive/withdrawn children during early grade school. Child Development, 70, 910–929. Ladd, G. W., & Burgess, K. B. (2001). Do relational risks and protective factors moderate the linkages between childhood aggression and early psychological and school adjustment? Child Development, 72, 1579–1601. Ladd, G. W., & Emerson, E. S. (1984). Shared knowledge in children’s friendships. Developmental Psychology, 20, 932–940. Ladd, G. W., & Kochenderfer-Ladd, B. J. (2002). Identifying victims of peer aggression from early to middle childhood: Analysis of cross-informant data for concordance, estimation of relational adjustment, prevalence of victimization, and characteristics of identified victims. Psychological Assessment, 14, 74–96. Ladd, G. W., Kochenderfer, B. J., & Coleman, C. (1996). Friendship quality as a predictor of young children’s early school adjustment. Child Development, 67, 1103–1118. Ladd, G. W., Kochenderfer, B. J., & Coleman, C. C. (1997). Classroom peer acceptance, friendship, and victimization: Distinct relational systems that contribute uniquely to children’s school adjustment? Child Development, 68, 1181–1197. Ladd, G. W., & Mars, K. T. (1986). Reliability and validity of preschoolers’ perceptions of peer behavior. Journal of Clinical Child Psychology, 15, 16–25. Ladd, G. W., & Price, J. M. (1987). Predicting children’s social and school adjustment following the transition from preschool to kindergarten. Child Development, 58, 1168–1189. Ladd, G. W., & Troop, W. P. (in press). The role of chronic peer difficulties in the development of children’s psychological adjustment problems. Child Development. Lagerspetz, K. M. J., Bjorkqvist, K., & Peltonen, T. (1988). Is indirect aggression typical of females? Gender differences in aggressiveness in 11- to 12-year-old children. Aggressive Behavior, 14, 403–414. Lahey, B. B., Loeber, R., Quay, H. C., Applegate, B., Shaffer, D., Waldman, I., Hart, E. L., McBurnett, K., Frick, P. J., Jensen, P. S., Dulcan, M. K., Canino, G., & Bird, H. R. (1998). Validity of DSM-4 subtypes of conduct disorder based on age of onset. Journal of the American Academy of Child and Adolescent Psychiatry, 37, 435–442. Lambert, N. A. (1972). Intellectual and nonintellectual predictors of high school status. Journal of Scholastic Psychology, 6, 247–259. Lazarus, R. S. (1984). The stress and coping paradigm. In J. M. Joffe, G. W. Albee, & L. C. Kelly (Eds.), Readings in primary prevention of psychopathology (pp. 131–156). Hanover, NH: University Press of New England. Ledingham, J. E. (1981). Developmental patterns of aggressive and withdrawn behavior in childhood: a possible method for identifying preschizophrenics. Journal of Abnormal Child Psychology, 9, 1–22. Ledingham, J. E., & Schwartzman, A. E. (1984). A 3-year follow-up of aggressive and withdrawn behavior in childhood: Preliminary findings. Journal of Abnormal Child Psychology, 12, 157–168. Lochman, J. E., & Wayland, K. K. (1994). Aggression, social acceptance and race as predictors of negative adolescent outcomes. Journal of the American Academy of Child and Adolescent Psychiatry, 33, 1026–1035.

102

Gary W. Ladd

Loeber, R. (1990). Development and risk factors of juvenile antisocial behavior and delinquency. Clinical Psychology Review, 10, 1–41. Loeber, R., & Farrington, D. P. (2000). Young children who commit crime: Epidemiology, developmental origins, risk factors, early interventions, and policy implications. Development and Psychopathology, 12, 737–762. MacDougall, P., Hymel, S., Vaillancourt, T., & Mercer, L. (2001). The consequences of childhood peer rejection. In M. R. Leary (Ed.), Interpersonal rejection (pp. 213–247). Oxford, UK: Oxford University Press. Magnussen, D., Stattin, H., & Duner, A. (1983). Aggression and criminality in a longitudinal perspective. In K. T. Van Dusen & S. R. Mednick (Eds.), Prospective studies of crime and delinquency (pp. 227–301). Hingham, MA: Kluwer-Nijhoff. Michael, C. M., Morris, D. P., & Soroker, E. (1957). Follow-up studies of shy, withdrawn children II. Relative incidence of schizophrenia. American Journal of Orthopsychiatry, 27, 331–337. Moffit, T. E. (1993). Adolescence-limited and life course persistent antisocial behavior: A developmental taxonomy. Psychological Review, 100, 674–701. Moore, D. R., Chamberlain, P., & Mukai, L. H. (1979). Children at risk for delinquency: A follow-up comparison of aggressive children and children who steal. Journal of Abnormal Child Psychology, 7, 345–355. Morris, D. P., Soroker, E., & Burruss, G. (1954). Follow-up studies of shy, withdrawn children I: Evaluation of later adjustment. American Journal of Orthopsychiatry, 24, 743–754. Morrison, P., & Masten, A. (1991). Peer reputation in middle childhood as a predictor of adaptation in adolescence: A seven-year follow-up. Child Development, 62, 991–1007. Moss, H. A., & Susman, E. J. (1980). Longitudinal study of personality development. In O. G. Brim Jr. & J. Kagan (Eds.), Constancy and change in human development (pp. 530–595). Cambridge, MA: Harvard University Press. Mulligan, G., Douglas, J. W. B., Hammond, W. A., & Tizard, J. (1963). Delinquency and symptoms of maladjustment: The findings of a longitudinal study. Proceedings of the Royal Society of Medicine, 56, 183–186. Newcomb, A. F., & Bukowski, W. M. (1983). Social impact and social preference as determinants of children’s peer group status. Developmental Psychology, 19, 856–867. Newcomb, A. F., & Bukowski, W. M. (1984). A longitudinal study of the utility of social preference and social impact sociometric classification schemes. Child Development, 55, 1434–1447. Ollendick, T. H., Green, R. W., Weist, M. D., & Oswald, D. P. (1990). The predictive validity of teacher nominations: A five-year follow-up of at risk youth. Journal of Abnormal Child Psychology, 18, 699–713. Ollendick, T. H., Weist, M. D., Borden, M. G., & Green, R. W. (1992). Sociometric status and academic, behavioral, and psychological adjustment: A five-year longitudinal study. Journal of Consulting and Clinical Psychology, 60, 80–87. Olweus, D. (1978). Aggression in the schools: Bullies and whipping boys. Washington, DC: Hemisphere. Olweus, D. (1979). Stability of aggressive reaction patterns in males: A review. Psychological Bulletin, 86, 852–875. Olweus, D. (1993). Victimization by peers: Antecedents and long-term outcomes. In K. H. Rubin & J. B. Asendorf (Eds.), Social withdrawal, inhibition, and shyness in childhood (pp. 315–342). Hillsdale, NJ: Erlbaum. Panak, W. F., & Garber, J. (1992). Role of aggression, rejection, and attributions in the prediction of depression in children. Development and Psychopathology, 4, 145–165.


103

Parker, J. G., & Asher, S. R. (1987). Peer acceptance and later interpersonal adjustment: Are low-accepted children at risk? Psychological Bulletin, 102, 357–389. Parker, J. G., & Asher, S. R. (1993). Friendship and friendship quality in middle childhood: Links with peer group acceptance and feelings of loneliness and social dissatisfaction. Developmental Psychology, 29, 611–621. Parker, J. G., Rubin, K. H., Price, J. M., & DeRosier, M. (1995). Peer relationships, child development, and adjustment: A developmental psychopathology perspective. In D. Cicchetti & D. Cohen (Eds.), Developmental psychopathology: Vol. 2: Risk, disorder and adaptation (pp. 96–161). New York: John Wiley & Sons. Perry, D. G., Kusel, S. J., & Perry, L. C. (1988). Victims of peer aggression. Developmental Psychology, 24, 807–814. Perry, K. E., & Weinstein, R. S. (1998). The social context of early schooling and children’s school adjustment. Educational Psychologist, 33, 177–194. Piaget, J. (1965). The moral judgment of the child. New York: Free Press. Poteat, G. M., Ironsmith, M., & Bullock, J. (1986). The classification of preschool children’s sociometric status. Early Childhood Research Quarterly, 1, 349–360. Renshaw, P. D. (1981). The roots of current peer interaction research: A historical analysis of the 1930’s. In S. R. Asher & J. M. Gottman (Eds.), The development of children’s friendships. Cambridge: Cambridge University Press. Renshaw, P. D., & Brown, P. J. (1993). Loneliness in middle childhood: Concurrent and longitudinal predictors. Child Development, 64, 1271–1284. Reznick, J. S., Kagan, J., Snidman, N., Gersten, N., Baak, K., & Rosenberg, A. (1986). Inhibited and uninhibited behavior: A follow-up study. Child Development, 57, 660–680. Ricks, D. F., & Berry, J. C. (1970). Family and symptom patterns that precede schizophrenia. In M. Roff & D. F. Ricks (Eds.), Life history research in psychopathology (Vol. 1, pp. 31–39). Minneapolis: University of Minnesota Press. Robbins, L. N. (1966). Deviant children grow up. Baltimore, MD: Williams & Wilkins Press. Roff, M. (1961). Childhood social interaction and young adult bad conduct. Journal of Abnormal Social Psychology, 63, 333–337. Roff, M. (1963). Childhood social interactions and young adult psychosis. Journal of Clinical Psychology, 19, 152–157. Roff, J. D., & Wirt, D. (1984). Childhood aggression and social adjustment as antecedents of delinquency. Journal of Abnormal Child Psychology, 12, 111–126. Rubin, K. H. (1982). Nonsocial play in preschoolers: Necessary evil? Child Development, 53, 651–657. Rubin, K. H. (1985). Socially withdrawn children: An ‘‘at risk’’ population? In B. Schneider, K. Rubin, & J. Ledingham (Eds.), Peer relationships and social skills in childhood: Issues in assessment and training (pp. 125–140), New York, NY: Springer-Verlag. Rubin, K. H. (1993). The Waterloo Longitudinal Project: Correlates and consequences of social withdrawal from childhood to adolescence. In K. H. Rubin & J. B. Asendorpf (Eds.), Social withdrawal, inhibition, and shyness in childhood (pp. 291–314). Hillsdale, NJ: Lawrence Erlbaum. Rubin, K. H., & Borwick, D. (1984). The communication skills of children who vary with regard to sociability. In H. Sypher & J. Applegate (Eds.), Social cognition and communication (pp. 152–170). Hillsdale, NJ: Erlbaum. Rubin, K. H., Burgess, K. B., & Hastings, P. D. (2002). Stability and social-behavioral consequences of toddlers’ inhibited temperament and parenting behaviors. Child Development, 73, 483–495. Rubin, K. H., Daniels-Bierness, T., & Bream, L. (1984). Social isolation and social problemsolving: A longitudinal study. Journal of Consulting and Clinical Psychology, 52, 17–25.

104

Gary W. Ladd

Rubin, K. H., LeMare, L., & Lollis, S. (1990). Social withdrawal in childhood: Developmental pathways to peer rejection. In S. R. Asher & J. D. Coie (Eds.), Peer rejection in childhood (pp. 217–249). New York: Cambridge University Press. Rushton, J. P., Brainerd, C. J., & Pressley, M. (1983). Behavioral development and construct validity: The principal of aggregation. Psychological Bulletin, 94, 18–38. Rutter, M. (1979). Maternal deprivation, 1972–1978: New findings, new concepts, new approaches. Child Development, 50, 283–305. Rutter, M. (1990). Commentary: Some focus and process considerations regarding effects of parental depression on children. Developmental Psychology, 26, 60–67. Rutter, M. (1996). Transitions and turning points in developmental psychopathology: As applied to the age span between childhood and mid-adulthood. International Journal of Behavioral Development, 19, 603–626. Savin-Williams, R. C., & Berndt, T. J. (1990). Friendship and peer relations. In S. S. Feldman & G. R. Elliott (Eds.), At the threshold:The developing adolescent (pp. 277–307). Cambridge, MA: Harvard University Press. Shantz, C. U. (1987). Conflicts between children. Child Development, 58, 283–305. Slee, P. T. (1994). Situational and interpersonal correlates of anxiety associated with peer victimization. Child Psychiatry and Human Development, 25, 97–107. Sullivan, H. S. (1953). The interpersonal theory of psychiatry. New York: W.W. Norton & Company. Tolan, P. H., & Gorman-Smith, D. (1998). Development of serious and violent offending careers. In R. Loeber & D. P. Farrington (Eds.), Serious and violent juvenile offenders: Risk factors and successful interventions (pp. 68–85). Thousand Oaks, CA: Sage Publications. Tremblay, R. E., Pihl, R. O., Vitaro, F., & Dobkin, P. L. (1994). Predicting early onset of male antisocial behavior from preschool behavior. Archives of General Psychiatry, 51, 732–739. Vandell, D. L., & Hembree, S. E. (1994). Peer social status and friendship: Independent contributors to children’s social and academic adjustment. Merrill-Palmer Quarterly, 40, 461–477. Wanlass, R. L., & Prinz, R. J. (1982). Methodological issues in conceptualizing and treating childhood social isolation. Psychological Bulletin, 92, 39–55. Wentzel, K. R. (1991). Social competence at school: Relation between social responsibility and academic achievement. Review of Educational Research, 61, 1–24. Werner, N. E., & Crick, N. R. (1999). Relational aggression and social-psychological adjustment in a college sample. Journal of Abnormal Psychology, 108, 615–623. White, J. L., Moffit, T. E., Earle, F., Robbins, L., & Silva, P. A. (1990). How early can we tell? Predictors of childhood conduct disorder and adolescent delinquency. Criminology, 28, 507–533. Wright, J. C., Giammarino, M., & Parad, H. W. (1986). Social status in small groups: Individual-group similarity and the social ‘‘misfit’’. Journal of Personality and Social Psychology, 50, 523–536.

THE ROLE OF LETTER NAMES IN THE ACQUISITION OF LITERACY

Rebecca Treiman and Brett Kessler DEPARTMENT OF PSYCHOLOGY WASHINGTON UNIVERSITY ST. LOUIS, MISSOURI 63130

I. INTRODUCTION II. LETTER NAMES IN ENGLISH AND OTHER ALPHABETIC WRITING SYSTEMS A. ICONICITY B. DISCRIMINABILITY C. PHONOLOGICAL PATTERNING D. LEGALITY III. CHILDREN’S LEARNING OF LETTER NAMES IV. CHILDREN’S LEARNING OF LETTER SOUNDS V. ROLE OF LETTER NAMES IN LEARNING TO READ WORDS VI. ROLE OF LETTER NAMES IN LEARNING TO SPELL WORDS VII. CONCLUSIONS REFERENCES

I. Introduction Children in the United States usually know a good deal about letters before formal reading instruction begins. Preschoolers learn to sing the alphabet song and see letters on children’s television programs and in alphabet books. Colorful displays of letters adorn the refrigerators of many homes. As a result of such experiences, most US kindergarten entrants know the names of a good many letters. In one study, with a group of children in Texas who had been in kindergarten for less than two months, the average child could label 18 or 19 of the 26 letters by name when shown each letter’s


105


106

Rebecca Treiman and Brett Kessler

upper-case and lower-case form (Treiman et al., 1998). Once children start kindergarten, much time is devoted to teaching them the names and the visual forms of the letters they do not yet know. This knowledge, it is assumed, will prepare them for the formal reading instruction that begins in first grade. The United States is not alone in the expectation that children will be familiar with letters before they start learning to read. For example, upper-middle class Israeli children often learn the names of letters at home before reading instruction begins (Levin et al., 2002). The belief, in the US and other societies, is that knowledge of letter names provides a foundation for early reading. True, a child’s ability to label the letters of the alphabet in kindergarten is an excellent predictor of how well he or she will succeed in learning to read once formal reading instruction starts in first grade (e.g., Snow, Burns, & Griffin, 1998). But is knowledge of letter names causally related to learning to read? Some have argued for a causal relation, proposing that knowledge of letter names helps children bridge the gap between print and speech (Durrell, 1980; Ehri, 1983; Levin et al., 2002). Others have argued against a causal relation and claimed that the observed correlation between letter-name knowledge and reading achievement reflects the fact that homes in which preschoolers master letter names are the same homes that stress achievement in school (Samuels, 1972). Indeed, some have suggested that ‘‘letter-name instruction in the preinitial or initial stages of beginning reading might be downright harmful’’ to children (Feitelson, 1988, p. 137). Research that has attempted to answer the question of whether letter-name knowledge helps children learn to read has been inconclusive. In studies by Johnson (1969) and Silberberg, Silberberg, and Iversen (1972), US children who received extra instruction about letter names before formal reading instruction began did not learn to read more easily than control children. As Ehri (1983) pointed out, however, the additional training may have had little impact because even the control children knew many letter names. One possible reason why the debates and research on letter names have not led to a clear resolution is that the issue has been examined too globally. In English and other languages, letter names vary in how they relate to the letters’ sounds. As a result, knowledge about the conventional names of letters may be helpful for some letters, less helpful for others, and perhaps even harmful for still others. In this chapter, we argue that researchers must look closely at the characteristics of letter names and the characteristics of the written and spoken language in order to understand the effects of letter names on the acquisition of literacy. We attempt to do so for English by discussing what children know about letters before they enter school and how this knowledge influences their early attempts to read and spell. We review research on letter names in languages other than English, where such

The Role of Letter Names in the Acquisition of Literacy

107

research exists, and consider how the effects of letter-name knowledge in other languages compare to those observed in English. Differences may reflect how letters are labeled in various languages and how the names mesh with the characteristics of the languages. An examination of children’s knowledge of letter names is also useful for the insights it may provide on more general issues. One such issue is the interaction between informally acquired knowledge and formal education, an interaction that occurs in learning to read as in the learning of other school subjects. The skills that children bring with them to school— knowledge of letters and their names, in this case—shape the children’s response to instruction. We examine how this occurs in the acquisition of literacy by asking how children use their knowledge of letter names to try to make sense of words’ spellings. As we will see, this perspective allows us to understand certain errors and patterns of performance that might otherwise appear bizarre. The study of children’s knowledge of letter names also provides an opportunity to examine the roles of rote memorization and principled learning in the acquisition of literacy. To what extent do children memorize printed letters and words and the pronunciations to which they correspond? To what extent do they take advantage of the principles that underlie the system, such as the fact that almost all printed words that start with p have a / / at the beginning of their pronunciation?1 Do letter names provide a bridge from the former approach to the latter?

1

Because spelling is not always an unambiguous guide to pronunciation, we represent phonemes (sounds) using the alphabet of the International Phonetic Association (1996, 1999). Spellings are given in italics and pronunciations in IPA symbols surrounded by slash marks, /. The values of most IPA symbols agree with those of the e.g., cat is pronounced / corresponding English letter, but the following require special attention. Usage reflects General American pronunciation. / / / / / / / / / / /

/ / / / / / / / / / /

aisle sauerkraut apple wand, car badge then Vegas edit casserole go machine

/ / / / / / / / / / /

/ / / / / / / / / / /

it hallelujah obey coin soup sure etch rude rude, with unrounded lips (Korean) ugly thick

The mark / / precedes a stressed syllable; stress is marked only for words of more than one syllable.

108


We begin, in Section II, with a discussion of the letter names of English and other languages and how the names relate to the letters’ sounds and shapes. Such information provides an important background for a discussion of the role of letter names in learning to read and spell, and has not been systematized previously. In Section III we present new data on the factors that make some pairs of letter shapes and letter names easier for children to learn than others. In Section IV, we review research showing that children use the names of the letters as a basis for learning about the sounds that the letters represent—something that is helpful for most English letters but that causes errors in a few cases. This discussion provides a foundation for considering, in Sections V and VI, how children use their knowledge of letter names in reading and spelling words. We review prior studies on this topic and present new information on the extent to which letter names can help in the reading and spelling of English. To conclude (Section VII), we return to the questions that motivate this chapter: Are letter names helpful or harmful to young children? Can linguistic analyses shed light on how letters are learned and used? Does the role of letter names in learning to read and spell differ across languages?

II. Letter Names in English and Other Alphabetic Writing Systems Children who are learning to read and write must learn the shapes and the sound values of the letters that are used in their language. Many languages also have formal labels for letters that include sounds other than those made by the letter. For example, the English label for b includes the phoneme / / as well as the phoneme that b represents, / /. As we have described, young children are expected to learn the conventional letter names in the US and a number of other countries. Why do letters have names that differ from their sounds, and why are letter names used with children? One answer is that sound-based labels are ambiguous for sounds that correspond to more than one letter, such as English / /, which may be spelled as k (e.g., kite) or c (e.g., cat). But formal letter names exist and are taught to children even in many languages with highly regular spelling-to-sound correspondences, suggesting that this is not a complete explanation. Four considerations help explain the structure of letter-name systems: iconicity, discriminability, phonological patterning, and legality. In the four sections below, we will show how and why these principles apply to letter names. Special attention will be given to the English system, but we will also touch on other languages.


109

A. ICONICITY

Iconic systems are those in which a sign has properties of its referent. Words that name sounds, such as tweet, are often iconic in that the word is similar to the named sound. Letter-name systems, where the named sound is actually a speech phoneme, are optimal candidates for such iconicity. All letter-name systems that we know of are iconic in that the names of most letters contain the phoneme that the letter represents. For example, the English name of s, / /, contains / /. The same is true for 24 of the 26 English letters. Only / /, the typical pronunciation of the name of the /, the name of w, are totally noniconic. letter h, and / Iconicity is such an important property of letter names that it is often introduced when lacking or deficient. For example, the name of h is / rather than / / in some parts of the Englishpronounced as / speaking world. Another example comes from Portuguese, where x typically /) lacked has the value / /. The traditional name of x (cf. Spanish / iconicity and has been replaced by / /. As a third example, Spanish has effectively increased the number of letters in the Roman alphabet by treating ch, ll, and n~ as separate letters and giving them their own iconic names. Thus ch, which has the value / /, is referred to by the single syllable / / rather than by the name for c followed by the name for h, as in English. Letter names tend to be fairly fluid because they are not usually spelled out in text and do not have the stabilizing influence of writing. The changes often reflect a drive toward iconicity. An important limitation of iconicity is that several letters have more than one sound, but letter names almost always use only one of those sounds. For example, the English names of the vowels consist of their historically long sounds: The name of a is / /, which iconically represents the letter’s sound in words like bake but is not iconic with respect to the sounds in cat, wall, and so forth. The Korean system of letter names may be the only one that systematically addresses this issue. In Korean, the typical consonant name begins with the sound that the consonant has at the beginning of a syllable, and ends with the sound that the consonant has at the end of a syllable; these sounds may be somewhat different. Another potential problem with iconicity is that it may not be clear which part of the name constitutes the letter’s sound. In English, the name of k begins with the letter’s sound, but the name of l ends with the letter’s sound. The generalization is that, if the name contains a consonant, that consonant is the value of the letter. However, that generalization fails for u / /, where the whole name is the value of the letter (u cannot spell the consonant / / /, where the vowel is the value of the letter (y cannot spell alone), and y / / /, but it can spell / /). In some languages, the locus of iconicity may be

110


more or less regular than in English. In Spanish and Portuguese, for example, the names of some consonants, such as s, contain the sound in the /). In contrast, in many systems, such as those of middle (e.g., Spanish / Hebrew, Greek, and Turkish, the names of all letters contain the letter’s sound at the beginning. Iconicity contributes to the fact that letter names are fairly short in most languages: Any phoneme besides the sound of the letter detracts from the overall iconicity of the letter name. English is typical in having oneor two-phoneme names for almost all its letters. Hebrew and Greek, which have some of the longest letter names, rarely go beyond a syllable / and Greek / /, contain or two. Exceptions, such as English / adjectives that qualify more basic letter names (double u, little o). Iconicity is a particular instance of linguistic motivation. The relatively few letter names that are not iconic are almost always motivated in the sense that the name has a transparent explanation. Some common types of letter names that are motivated but not iconic are names that describe the letter’s form, function, or origin. English w takes its name from its shape, for example.2 In French and Spanish, y is called ‘‘Greek i,’’ a name that tells not only its sound but also its origin. The iconicity of letter names should not be confused with the iconicity of the letter shapes themselves. The two types of iconicities did interact, though, in the development of the alphabet. In the Sinaitic script that is the ancestor of most alphabets, letter shapes were pictures of objects whose names began with the sound denoted by the letter. Letter names were the same as the names of those objects. For example, the letter that spelled / / /, which meant ‘‘door,’’ and this letter took the form of a was named / stylized picture of a door. Such confluence of iconicities does not exist in modern scripts because the shapes of the letters have changed so much as to be completely unrecognizable as pictures of natural objects. The names themselves, however, are still used in more or less altered forms in many modern languages, such as Hebrew and Greek. That is why their letter names are not as short as would be predicted by pure letter-sound iconicity, as discussed above: They were also influenced by an ancient letter-shape iconicity. B. DISCRIMINABILITY

The principle of iconicity may lead one to think that the ideal letter names should comprise only the sound of the letter. Indeed, some educators 2

i.e., ‘‘double u.’’ ‘‘Double v’’ might appear more appropriate, but originally v was just a variant of u and shared its name.


111

advocate that children should name letters by their sound alone. But in practice, it is fairly unnatural to pronounce consonants in isolation, and listeners can have trouble telling what letter is meant. Adding a vowel to the consonant sound adds greatly in making the name both pronounceable and discriminable. This is perhaps more important for the stop consonants (in English, / /, / /, / /, / /, / /, / /) than for the continuants (such as / /, / /, / /, / /, / /, / /);3 but in all cases, adding a vowel helps quite a bit. All traditional letter-name systems add a vowel to the names of consonants, as in English / / for k. In principle, the vowel in a consonant letter name could be added before the consonant or after it. As mentioned earlier, English offers examples of both (cf. k, l). The same is true for other languages that borrowed their letter names from Latin, with some of these languages (e.g., Spanish) also including letter names with the sound in the middle. Latin is unusual in its use of a vowel–consonant structure for a substantial number of consonant names. In most other systems, the names of consonants begin with the sound of the consonant. This observation suggests that it is more natural to put the more important element, the sound of the named letter, first. For the listener, a decided advantage is that consonants are easier to discriminate before vowels than after vowels. Korean, as mentioned earlier, has chosen to put the consonant sound both at the beginning and at the end of the consonant name. This increases discriminability. Overall, however, the discriminability of most letter-name systems is rather low. It is easy to mishear letter names, such as to mistake a b for a d or an n for an m. Even in Korean, not many acoustic cues can /, the name for n, from / /, the name for help one distinguish / m. The problem is well attested by the fact that many organizations that depend on accurate spoken letter understanding have adopted official sets of letter names designed for higher discriminability. Pilots, for example, use the names Mike and November for the letters m and n. Research dating back at least to Conrad (1964) has confirmed that excessive similarity can cause confusions in short-term memory. The languages that have maintained the ancient letter names essentially intact, principally Hebrew and Greek, make the job of the listener / to much easier: Compare the distinctiveness of Hebrew / / vs. / that of / / vs. / /.

3 Stops are so called because the flow of air is blocked for a while when producing these sounds; air flow is continuous for continuants. The fact that the main part of a stop is air blockage makes it particularly difficult to generate a discriminable stop in isolation.

112


C. PHONOLOGICAL PATTERNING

Phonological patterning refers to the tendency for letter names to be similar to one another. For example, above and beyond the fact that they include their sounds, the English names of the consonants l, m, and n all consist of / / followed by the consonant. There are several reasons why letter names tend to be similar to each other. One is that letter names for a new writing system are all invented at the same time. People tend to give similar names when naming similar items simultaneously, as when parents select rhyming names for twins. Likewise, it is natural to choose the same vowel to fill out the names of consonant letters. A second reason why letter names tend to be similar to one another is that, even when the names are not invented simultaneously, they may be patterned on existing names for related items. For example, parents often give girls names ending in a, even when inventing new names, because that vowel appears at the end of many other female names. Similarly, when the consonant v was added to the alphabet, it was named / / because many other consonant names ended in / / (b, c, d, etc.). A third reason for phonological patterning in letter names is that they are often recited in a series, in the same fixed order. In such cases, people tend to anticipate the sound of the next item in the series. For example, the / / in the English word four probably comes from anticipating the initial consonant of the following number, five. In like manner, the name /, took its vowel from the name of the following letter, k of the letter j, / / /. The original name, still heard occasionally in Scotland, was / /. The nature and extent of phonological patterning varies between languages. The English system is of intermediate consistency. All the vowel letters are named by their historically long value. Virtually all consonants have two-phoneme names, usually either an / / before the consonant or an / / after it. Even the selection of those two vowel extensions has a phonological patterning: / / is used before a continuant ( f, l, m, n, s) and / / after a stop (b, d, p, t). However, the English system has several perturbations due to its long historical development. The letter c is named with an / / because originally (in Latin) it was always a stop. After the consonant changed to a continuant, the original pattern was obscured, and the new letter names for v and (in America) z were built with / / instead of / / , even though they are continuants. The names for k and q used different vowels, to distinguish them from c. The name of r reflects the fact that / / / for z, used in many sequences changed to / / in English. The name / parts of the English-speaking world, comes from the Greek name zeta. The irregular names for h, j, and w have already been mentioned. The middling degree of patterning found in English is typical of languages that adopted the Latin letter-name system. Phonological patterning is, on


113

the whole, even lower in languages that retained the old Semitic letter names. In Hebrew, for example, the names of consonants all begin with the sound of the letter, but speakers of the language who are not aware of the language’s history and of related letter-name systems cannot predict the other phonemes in the letter name. In contrast, several other systems have a much higher degree of phonological patterning than English. In Turkish, for example, the names of consonants comprise the consonant sound, plus / /, with thoroughgoing regularity. The same system is used in many Balkan countries, and has recently come into use in Portugal. A high degree of patterning should make it simpler to learn letter names, recognize new names as letter names, and determine the sound value of the letter that has a given name. In contrast, strong phonological patterning hurts discriminability, because many letter names become extremely similar to one another. D. LEGALITY

Legality means that letter names should meet the phonological requirements placed on all words of the language. Letter names must draw from the same set of phonemes as any other word, and they must follow the same rules for the arrangement of the phonemes. For example, one might think that a schwa (/ /) would be the ideal vowel to add to a consonant sound to make a name (e.g., / /) because it is a very neutral sound. Although some languages, such as Sanskrit, do that, this option is not available in the many languages that do not have a schwa phoneme. English does have a schwa phoneme, but this phoneme cannot occur at the ends of stressed syllables. Thus, b cannot be named / /. The need to avoid illegal names has the further consequence that English vowels are labeled not with their ‘‘short’’ sounds, which would be illegal (e.g., / / for e), but with their ‘‘long’’ sounds (e.g., / / ). Although the traditional letter-name systems of English and other languages generally conform to the legality requirements of their respective languages, teaching methodologies sometimes ignore legality in favor of other criteria. Referring to letters as / /, / /, / /, and so on may be desirable from the points of view of iconicity and phonological patterning, but these are not legal words of English.

III. Children’s Learning of Letter Names Our discussion of letter names and their relation to the letters’ sounds and shapes provides a background for examining how children learn this information. Children in the United States often learn about the

114


phonological forms of certain letter names before they learn the letters’ shapes. Through the alphabet song and other experiences, preschoolers may learn that / /, / /, / / and so on belong to a special set before they learn the visual appearance of a, b, and c. As we have discussed, letter names share certain phonological characteristics (e.g., most English letter names are short, many consist of a consonant followed by / /), making them differ phonologically from other words in the language. Phonological cues may help children distinguish between letters and members of other categories, such as numbers, just as phonological cues apparently help children distinguish between male first names and female first names (Cassidy, Kelly, & Sharoni, 1999). To test the idea that some children pick up the phonological patterns in letter names at a young age, Treiman, Tincoff, and Richmond-Welty (1997) asked preschoolers (mean age 4 years, 9 months) whether various syllables were real letters. The children sometimes responded that consonant-/ / syllables such as / / were letters, producing false positive errors 11% of the time to such syllables. The error rate was significantly lower, 7%, for consonant–vowel syllables that ended with other vowels, such as / /, and vowel–consonant syllables that began with / /, such as / /. Letter names that follow the English system have been observed in production as well as recognition, as when a bright 3½-year-old child of our acquaintance said that Fred starts with the ‘‘letter’’ / / and that little starts with / /. Preschoolers’ tacit appreciation of the phonological patterns of letter names may explain why the letter names invented by Dr. Seuss (1955), such as yuzz, are funny. An important step in learning about letters is mastering the associations between the letters’ upper- and lower-case forms and their names. With its 26 letters, most of which appear rather different in upper and lower case, and with its largely arbitrary pairings between visual forms and names, this is a major undertaking for English. To study the factors that affect children’s learning of shape–name associations, we examined several sets of previously collected data on preschoolers’ ability to name upper-case and lower-case letters. As Table I shows, the data were gathered in the US and Australia, where young children’s experiences with letters are similar to those of US preschoolers. The children in each study were shown the letters of the alphabet in a scrambled order and were asked to say the name of each letter. The children in these studies had not yet entered kindergarten, where systematic instruction about letter names is typically provided in the US and where better knowledge of some letters than others is likely to reflect, in large part, the sequence in which the letters are taught. For each study, we calculated the percentage of correct responses to each letter of the alphabet. The mean values are shown in Table I.

115

The Role of Letter Names in the Acquisition of Literacy TABLE I Studies of English-speaking Preschoolers’ Ability to Produce Names for Letters Presented Visually Location Case of of letters study

Mean age of childrena

Number of children

Source of data

Mean (and SD) percent correct responses

Upper

California 3 years, 0 months California 4 years, 0 months Detroit 4 years, 10 months

38 35 57

Worden & Boettcher, 1990 Worden & Boettcher, 1990 Treiman et al., 1998

17.2 (5.7) 54.2 (10.3) 74.4 (8.4)

Lower

California 3 years, 0 months California 4 years, 0 months Australia 4 years, 8 months

38 35 77

Worden & Boettcher, 1990 Worden & Boettcher, 1990 Byrne, 1992

11.4 (8.7) 39.2 (13.5) 23.3 (12.9)

a In all but the study by Treiman et al. (1998), mean ages are estimated from information provided in the reports.

Before examining the factors that make some visual form–name pairs easier to learn than others, it is important to determine whether the letters that cause difficulty in one study also cause difficulty in other studies. We thus calculated correlations between the results for individual letters in all pairs of studies. The mean correlation for studies involving upper-case letters was a modest .39 using Pearson correlation coefficients (.23 when Spearman rank correlation coefficients were used). The figure was substantially higher for lower-case letters, .79 (.81 using rank correlation coefficients). Performance on upper-case and lower-case letters overlapped to some extent, with a mean correlation of .46 (.33 using rank correlation coefficients). Differences among children’s experiences with letters and other factors thus cause some differences across studies and age groups. Despite these differences, there are enough similarities in letters which children know to encourage a search for factors that may explain them. One factor that may affect the learning of letter names is the extent to which the visual form of the target letter looks like those of other letters. Letters with distinctive shapes, such as lower-case s, may have an advantage over letters whose shapes are similar to those of many other letters, such as lower-case d. Another potentially important factor is whether the shape of a letter is the same in upper- and lower-case, as with o and c. Children in many English-speaking countries tend to learn upper-case letters before lower-case letters (as shown, for example, by the data in Table I), and so a lower-case letter that is a smaller version of a previously known upper-case letter may be easier to recognize than one that appears quite different. In addition, the phonological commonalities among the names of letters may influence

116


learning. The effect could be positive, as when a child who remembers that a letter name begins with / / but forgets the following vowel guesses / /, the most common vowel in English letter names. The phonological similarities among letter names might alternatively hurt performance. A child who remembers only the final vowel of a letter name and who selects from among the set of known names has more possibilities from which to choose, and thus more chance of being wrong, when the vowel is / / than when it is / /. Table II shows the definitions of visual similarity, case similarity, and phonological similarity that we used for the analyses reported here, together with the mean value of each variable. Regression analyses were carried out to predict performance on letter naming, and the results are shown in Table III. For upper-case letters, only phonological similarity was consistently related to letter naming. The effect was negative, with letters whose names were phonologically similar to the names of many other letters yielding poorer performance than letters whose names were more distinctive. In all three data sets of lowercase letters, children performed significantly better on letters that resembled their upper-case forms than on letters that showed less resemblance. A reliable effect of visual similarity was found in two of the three data sets of lower-case letters, such that children tended to do well on letters that were visually dissimilar to other letters. Phonological similarity had a

TABLE II Predictor Variables for Analyses of Letter Naming Variable

Definition

Mean (SD)

Visual similarity

No. of other letters whose visual forms share 50% or more of strokes in target letter’s form in same position or whose overall visual form is identical to target letter when rotated or flipped

3.30 (3.77) for upper-case California study, 2.70 (3.44) for upper-case Detroit study, 2.80 (2.14) for lower-case California study, 2.90 (2.04) for lower-case Australia study

font in font in font in font in

Case similarity

1 if visual form of letter alike in upper and lower case, 0 if not

.35 (.49)

Phonological similarity

No. of other letters whose names share 50% or more of phonemes in target letter’s name in same position, counting affricates and diphthongs as single phonemes

4.50 (3.18) for US pronunciations of letters, 4.20 (2.81) for Australian pronunciations of letters

117

The Role of Letter Names in the Acquisition of Literacy TABLE III Results of Regression Analyses Predicting Performance in Letter Naming Study

R2 (adjusted R2)

p

Upper

California, age 3 California, age 4 Detroit

.36 (.27) .24 (.13) .30 (.21)

.019 .109 .044

.32 .11 .22

p .082 .574 .257

.13 .22 .31

p .454 .253 .099

.42 .40 .50

p .025 .049 .017

Lower

California, age 3 California, age 4 Australia

.66 (.61) .52 (.43) .63 (.58)

.000 .001 .000

.24 .38 .32

.096 .027 .048

.62 .44 .49

.001 .010 .004

.30 .20 .35

.029 .214 .015

Case of letters

Visual similarity

Case similarity

Phonological similarity

significant impact in two of the three data sets of lower-case letters, with letters whose names were phonologically similar to the names of many other letters yielding relatively poor performance. Together, the three factors explained over half the variance in each of the data sets with lower-case letters. The results of the regression analyses must be interpreted with some caution, as certain of the variables deviated from a normal distribution. Also, our goal is not to predict performance on new items, as is usually the case in regression, but to understand the factors that affect performance on the existing alphabet. To supplement the regression analyses, we examined the letters that yielded consistently good or consistently poor performance across studies. Upper-case letters showed a good deal of variability across studies, as mentioned earlier. However, A and O were in the top third of letters in all three upper-case data sets, and V was always in the bottom third. The names of A and O are below the median in phonological similarity, and their shapes are below the median in visual similarity. The fact that A is at the beginning of the alphabet may help performance. With O, performance may be facilitated by the fact that a circle is a basic shape, easily identified by children and often referred to as an O. V, the letter that was in the bottom third in all three data sets, is above the median in both visual and phonological similarity to other letters. Although phonological and visual similarity appear to affect performance on upper-case letters, the order of learning of these letters is rather variable and personal. Children’s experiences with the first letters of proper names likely contribute importantly to their knowledge of upper-case letters (e.g., Treiman & Broderick, 1998), and these experiences differ from one child to another depending on the child’s own name and the names of friends and family members.

118


With lower-case letters, i, o, s, w, and x were in the top third in all the studies and d, g, h, l, and q were in the bottom third. Of the five easiest letters, three are below the median in visual similarity, with the remaining two (i and x) above. Four are below the median in phonological similarity, and one (s) is at the median. Four of the easy-to-name letters, all but i, have identical shapes in lower and upper case. This is not true of any of the five letters in the bottom third in lower-case naming. All the difficult letters are above the median in their visual similarity to other letters, and three (all but h and q) are above the median in phonological similarity. These observations lend support to the results of the regression analyses and help explain why performance on lower-case letters is more similar across studies than is performance on upper-case letters. Children learn lower-case letters later than upper-case ones, and a major determinant of performance on lower-case letters is whether they have the same shapes as their upper-case counterparts. One would expect children to perform better on letters they have seen many times than on less common letters. The finding that children do better on lower-case letters that are identical in shape to their upper-case counterparts may be considered a kind of frequency effect, in that performance on a lower-case letter is aided by previous experience with its upper-case version if the two are alike in form. However, several other measures that might be expected to reflect children’s experience with specific letters did not account for significant additional variance in our analyses. One of these measures was based on McBride-Chang’s (1999) suggestion that the beginning letters of the alphabet are stressed more than the end letters in informal learning. In none of our preschool data sets, however, did position contribute significant additional variance when coded as first half or second half of the alphabet. We suggested that children may do relatively well on upper-case A because it is the first letter of the alphabet, but we do not find evidence for order effects beyond the first position. Another possible index of preschoolers’ exposure to letter names may be the frequency of letters in words in written materials designed for young children. For this measure, we examined the words that appear with a U value (frequency per million words adjusted for variation in distribution of words across content areas) of 1 or more in written materials at the kindergarten and first-grade levels in Zeno et al. (1995). The analyses were limited to words that also appeared in a second source (CMU Pronouncing Dictionary, 1998), eliminating 54 entries, mostly erratic or low-frequency words. The remaining set of 6,232 words comprises words that young children would have a chance of seeing, although a given child would certainly not see every word. How often each letter occurred in this corpus, weighting words by their frequencies, did not account for significant


119

additional variance in any of the regressions. Because young children may focus primarily on the initial letters of words, we also examined each letter’s frequency in this position. This variable did not add significantly to the regressions either. A measure of letter frequency that does influence preschoolers’ knowledge of letter names is whether the letter appears in the child’s own first name. A child’s name is often the first printed word that he or she learns to recognize and write, often as early as age 3 (Hildreth, 1936; Villaume & Wilson, 1989). Treiman and Broderick (1998) found that, in an upper-case letter naming task and a letter printing task, children performed better with the first letter of their own name than with other letters. These differences probably reflect the child’s greater exposure to the first letter of his or her name and the personal importance that this letter develops for the child. Treiman and Broderick found that the letters of the child’s name beyond the first showed somewhat elevated levels of performance in the naming task, although these effects were not significant. Children’s learning of the pairings between the visual forms of letters and their names has the hallmarks of paired-associate learning. It is influenced by stimulus similarity and response similarity and, apparently, by those measures of frequency that capture young children’s experiences with letters. Indeed, rote memorization of shape–name pairs is the only option with languages like English, where the shapes of almost all letters are, from the child’s point of view, arbitrary. To our knowledge, no data are available on children’s learning of shape–name pairs in writing systems with many motivated letter shapes. For example, Korean consonant letters were designed to graphically depict the location in the vocal tract where the consonant is made, but it is not known whether this helps Korean children learn the letters.

IV. Children’s Learning of Letter Sounds Among US children, the ability to provide sounds for individual letters, which we refer to as letter-sound knowledge, lags behind the ability to provide the letters’ names. One reason for the discrepancy is the common belief that children can learn letter sounds at school, whereas they should know most letter names before they start school. Another reason, not culturally determined, is that consonant letter sounds, as pronounced without a vowel, are difficult to discriminate and produce and are not legal words of the language. Nevertheless, many children learn the sounds of some letters before they enter school. The California 4-year-olds studied by Worden and Boettcher (1990) responded correctly to 21% of letters when

120


asked to provide their sounds. (The letter-sound task of this study used the case that had yielded best performance in the letter-name task for each child, which was usually upper case.) The Detroit preschoolers studied by Treiman et al. (1998) responded correctly to 33% of letters when shown the uppercase forms and asked to supply the sounds. As discussed earlier, the names of letters in English and other languages tend to be iconic, in that they include their sounds. Do children benefit from this iconicity, using the fact that / / is heard at the beginning of b’s name to learn and remember that b represents / /? Alternatively, do children treat name–sound pairings as arbitrary, using rote memorization to learn these pairs as they do to learn shape–name pairs in English? Children in the US have been found to benefit from the iconicity of letter names. Evidence for name-to-sound facilitation comes from the finding (McBride-Chang, 1999; Treiman et al., 1998) that children perform best on letter-sound tasks for letters whose sounds are at the beginnings of their names, such as b. Performance tends to be poor for letters whose sounds are not heard in their conventional names, such as w. Intermediate levels of performance are found for letters whose sounds are at the ends of their names, such as l. The / / of a syllable like / /, which is called the onset, is more salient and accessible than the / / of a syllable like / /. This, together with the fact that consonant–vowel letter names are more common in English than vowel– consonant letter names, makes the letter name more useful in the former case than the latter case. When we examined the two sets of data on preschoolers’ knowledge of upper-case letter-sound correspondences that were mentioned earlier, we found that B, G, K, O, P, T, and Z ranked in the top third of letters in both studies. For all these letters, at least one of its common sounds appears at the beginning of the letter’s name. The letters that ranked in the bottom third in both studies of letter-sound knowledge were L, R, Q, W, Y, and X. For these letters, the sounds that are taught to children and scored as correct in tests of letter-sound knowledge are not found at the beginning of the letter’s name.4 Although the results presented above support the idea that the position of a letter’s sound in its name is important for letter-sound learning, raw knowledge of letter sounds may not provide the best way to assess the idea that children use the names of letters to help learn and remember their sounds. A high or low ranking in sound knowledge could reflect high or low familiarity with a letter’s shape in addition to ease or difficulty of deriving

4

The sound of Q is popularly considered to be / /. at the beginning of the name /

/, only the first phoneme of which appears


121

the sound from the name. For example, O is among the easiest letters for shape–name learning, as mentioned earlier, and this could partially explain children’s good knowledge of its sound. The difference between name knowledge and sound knowledge may better reflect the ease with which children can derive the sound of a letter from its name. The difference should be relatively small when a letter’s sound can be generated easily from its name and large when this is not the case. In both of the preschool studies, the consonant / / letters D, G, K, and V ranked in the lowest third of all letters in terms of name–sound discrepancies. H, Q, R, W, Y, and X ranked in the top third. These two sets of letters clearly differ in the ease of deriving the sound from the name. Studies in which children are taught different types of letter–sound pairs provide direct evidence that the name–sound relation influences children’s learning of letter sounds. In a study by Treiman et al. (1998), preschoolers (mean age 4 years, 11 months) were taught the sounds of 10 letters. Children were selected who knew the names of the critical letters but few or none of their sounds. Over several sessions, children were told the sound of each letter and were asked to select, by pointing to one of the ten letters, the letter that corresponded to a sound spoken by the experimenter. Children showed substantial improvement over the course of the study for / /–d and / /–v, the two taught pairs for which the sound appears at the beginning of the letter’s name. Some improvement, although not as much, was seen for pairs such as / /–l and / /–m, which have their sounds at the ends of their names. Little or no improvement occurred for sounds that are not found in the name of the letter that is used to spell them, as with / /–w and / /–y. If children learned the sound–form pairs solely in a rote, paired-associate fashion, such differences would not have been expected. A number of US preschoolers, like those who participated in the training study just described, know many letter names but few letter sounds. Few culturally mainstream and cognitively normal 4-year-olds in the US know neither letter names nor sounds. Such children would provide a useful control group in testing the hypothesis that the learning of letter sounds is facilitated by prior knowledge of letter names. Share (1999) included such a control group in a study of Israeli children who did not know English. The experimental children learned English names (or slight modifications thereof ) for six letter-like symbols before they learned the letters’ sounds. The sound was included in the name in four cases (e.g., / / in / /), but not in the other two (e.g., / / in / /). The control group learned real-word names that were phonologically unrelated to the letters’ sounds. The experimental group performed significantly better than the control group when learning the letters’ sounds, and this superiority was especially marked for letters with the sound in the name. These findings support the view that

122


prior knowledge of letter names helps children learn letters’ sounds. The control group gained familiarity with the letters’ shapes before learning the sounds, as did the experimental group, so the superior performance of the experimental group is not due to better knowledge of the letters’ visual forms. The results of the training studies confirm that the knowledge of letter names that most US children bring with them to the task of learning letter sounds influences their performance. For the majority of English letters, such as b and v, the effects are positive. The sounds of the letters are heard in their names, and this helps children associate the sound with the name. For a few letters, the effects are negative. Children who know that y is called / probably have a harder time learning that it corresponds to / / when / used as a consonant than children who do not know the name of y. Supporting the idea that negative effects occur for some letters, kindergartners sometimes say that y makes the sound / / and that w makes the sound / / (Treiman, Weatherston, & Berch, 1994). These errors, which might at first seem bizarre, are regularizations of the kind that have been observed elsewhere in language learning. They reflect the generalization that the sound of a letter is often the first phoneme of the letter’s name, just as goed for went reflects a generalization about the form of the past tense. Errors in which children give the first phoneme in the name of letters such as s and r as the sound have not been systematically documented, but we have observed them in some English-speaking children. Such errors may reflect children’s experiences with English, where the sounds of letters are more often at the beginning of letters’ names than at the ends, or a more universal tendency to look to the beginning of the syllable for the sound of the named letter. Venezky (1975) argued that English letter names cannot provide much help in learning the letters’ sounds because only 9 of the 26 letters—b, d, j, k, p, t, q, v, and z—begin with the sound that is traditionally introduced first for the letter in reading programs. Although we would argue that q is not as useful as the other names just mentioned, as it begins with / / rather than the / / that is taught in schools, we believe that the number of helpful letter names is much larger than Venezky claimed. The data reviewed here show that young children benefit from letter names that end with the sound, although less than from letter names that begin with the sound. This means that f, l, m, n, r, s, and x are letter names that are at least somewhat helpful. Although the ‘‘hard’’ sounds of c and g, / / and / /, are typically taught first to children, the ‘‘soft’’ sounds / / and / /, which are at the beginning of the letter names, are reasonably common. With vowels, the short sounds are usually taught first but the long sounds, which are the letters’ names, are often found in words. Children are typically taught that y symbolizes / /


123

when used as a consonant, as in yellow, and / / when used as a vowel, as in baby. However, y corresponds to the / / at the end of its name in words like by and my. This leaves h and w as the only two letters whose American English names are arguably useless in learning their sounds. And h is called / / by some English speakers, as mentioned earlier, taking it off the list for them. Thus, although English letter names are surely less helpful in learning the letters’ sounds than are the names of Turkish or Korean, a large majority of English letter names provide cues that children can and do use. The fact that the names of most English letters are helpful in learning their sounds has implications for the teaching of letter–sound associations. Many US kindergartens adopt a ‘‘letter-of-the-week’’ method in which children spend a week learning about the visual form, name, and sound of each letter. The training studies show, however, that children need more time to learn the sounds of some letters than others. The sound of y is one that is difficult to learn, with the kindergartners tested by Treiman et al. (1994) often continuing to claim that y makes the sound / / even after they had spent the allotted week learning about y. Information about the ease and difficulty of various letter–sound pairs can help educators allocate classroom time effectively and respond appropriately to children’s errors. Most existing studies have focused on differences among letters rather than differences among children. Little is known about whether some children derive more benefit than others from the cues to sound that letter names generally offer and, if so, about the causes or consequences of such differences (but see Share, 1999, for some preliminary evidence). Failure to take advantage of the iconicity of letter names may be an early indicator of reading problems, reflecting poor phonological skills and a tendency to treat print–speech relations as arbitrary rather than principled. Another issue for future research is name-to-sound facilitation in languages other than English. We have suggested that the highly systematic letter names of languages like Turkish should be quite helpful in learning their sounds. Children may derive special benefit from the consonant letter names of Korean, which include the consonant both before and after a vowel and so demonstrate its sound in both positions. Although much research remains to be done, knowing the names of letters clearly facilitates the learning of associations between letters and sounds in many cases.

V. Role of Letter Names in Learning to Read Words So far, we have discussed children’s knowledge about the conventional labels and sounds of individual letters. What about the reading of entire words? Researchers have distinguished two methods of learning to

124


read words in alphabetic systems. One method involves rote learning of printed words’ pronunciations and/or meanings, much like rote learning of the largely arbitrary associations between English letter shapes and letter names. A second method, which takes full advantage of the alphabetic system, involves relating each letter in a printed word to a phoneme in the word’s pronunciation. The first method is poor at securing words in memory and allowing generalization to new words. For example, a child who has linked the red semicircle in the Crest logo to the word’s meaning may remember the word as toothpaste or brush teeth and may read other words that contain a semicircle in the same way. A child who has connected the C, / knows the r, e, s, and t of Crest to the / /, / /, / /, / /, and / / of / word’s precise pronunciation and can likely decipher words such as rest and set. Full alphabetic connections, with their benefits for precision and generalization, take time and effort to master. Children who would form such connections must treat spoken words as sequences of phonemes and must link the phonemes and the letters. Before children’s phonological skills and knowledge of letter sounds are advanced enough to do this, they are thought to rely on rote, paired-associate learning (Byrne, 1992; Ehri, 1998; Frith, 1985; Gough & Hillinger, 1980). In this section, we suggest a third option, one that takes more advantage of the alphabetic system than rote memorization but that requires less skill and knowledge than use of letter sounds. This method involves connections between print and speech that are based on letter names. For example, a child may use the fact that the names of b and e are heard in the pronunciation of bead to help learn and remember this word’s spelling. Such a child does not connect the a and d of bead to phonemes in the word’s pronunciation, and so may misremember the word as beach or beat. Having learned bead, the child may have difficulty deciphering new words in which b is followed by a vowel other than / /. This lack of precision and generalization make links based on letter names a transitional method at best. However, letter names may play an important role in the early acquisition of literacy by helping children understand that words’ spellings are not arbitrary. At least some of the letters in words’ spellings reflect the words’ pronunciations. Evidence for a role of letter names in linking print and speech comes from a study in which children were taught to pronounce artificial words that offered different types of connections between spellings and pronunciations (Treiman & Rodriguez, 1999). In the name condition, the entire name of the word’s first letter could be heard in the pronunciation, as when BD was taught as a spelling for bead. In the sound condition, the phoneme corresponding to the first letter was heard in the word’s pronunciation but the complete letter name was not. For example, BD was pronounced as bud


125

in this condition. In a third condition, the pronunciation did not correspond to either the names or sounds of the words’ letters, as with wine for BD. To avoid repetition, children learned a different set of words in each of the three conditions; the children were told that these were words in a ‘‘made-up’’ language. Children who were not able to read simple words (mean age 5 years, 0 months) performed significantly better in the name condition than in the sound condition or the condition with arbitrary spellings. Only children who had begun to read were able to benefit from sound-based links. In another study, younger preschoolers (mean age 4 years, 3 months) also showed some ability to form letter-name connections for consonants at the beginnings of words (Treiman, Sotak, & Bowman, 2001). Children find it harder to form these connections for consonants at the ends of words than consonants at the beginnings of words. In a study by Bowman and Treiman (2002), for example, prereaders benefited from the R at the beginning of RT for art but showed no significant benefit from the R at the end of TR for tar. The use of letter names to connect print and speech does not disappear once children become able to form connections based on letter sounds. Even adults use letter names to facilitate word learning (Bowman & Treiman, 2002; Treiman, Sotak, & Bowman, 2001). For older children and adults, letter-name connections may allow for a degree of redundancy that facilitates memory (Perfetti, 1992). For example, knowledge of letter names permits the / / in the spoken form of bead to link not only to the ea but also to the b. If readers use letter names in a direct way to connect print and speech, how much could this help in learning to read English? To find out, we examined the 6,232 words described earlier that occur in reading materials at the kindergarten and first-grade levels (Zeno et al., 1995). Only rarely, as with OK and CBS, can one read a word simply by concatenating the names of its symbols. This helps explain why some children who can benefit from letter names when learning to read artificial words cannot read common real words, where letter names do not permit a fully correct pronunciation. It also helps explain why the syllabic symbols of Japanese, the kana, are easy for children to read. Concatenation of names works well in this case, because the names of kana symbols almost always match the sounds they represent. Although English-speaking children cannot just say the letters’ names in sequence to read words, letter names are useful for some letters of some words. Of the words in our set, 43.1% contain in their spelling at least one letter whose American English name appears in the pronunciation. This letter is sometimes a consonant, as with the r of start, but is much more often a vowel, as with the a of take. Another way to look at the matter is to ask what percentage of letters in words’ spellings are associated, in the same word, with the corresponding letter name. For example, only one letter of the four in take is heard, the a.

126

Rebecca Treiman and Brett Kessler TABLE IV

Reliability of Letter Names in Reading and Spelling in Set of 6,232 Words Found in Reading Materials for Kindergarten and First-grade Children Reading

Letter A B C D E F G H I J K L M N O P Q R S T U V W X Y Z All

Spelling

No. of words with letter in spelling

Percent of those words with letter name in pronunciation

No. of words with letter name in pronunciation

Percent of those words with letter in spelling

2834 777 1364 1702 4319 559 1220 1169 2429 122 596 2018 963 2418 2168 1170 38 2645 3054 2222 1109 286 642 78 665 90 36657

21.7 6.2 1.6 4.8 25.1 1.3 .8 .1 19.9 8.2 2.5 8.3 4.8 8.5 24.2 5.4 .0 8.0 4.1 4.2 5.8 3.1 .0 10.3 .6 14.4 10.7

590 38 62 71 1030 7 12 1 472 10 44 114 37 160 476 52 9 183 75 78 61 9 0 11 34 22 3658

97.1 100.0 35.5 98.6 68.1 71.4 75.0 100.0 91.5 100.0 34.1 100.0 100.0 100.0 99.2 100.0 .0 100.0 98.7 98.7 90.2 100.0 – 72.7 11.8 45.5 85.6

Table IV shows the reliability of each letter in reading, meaning the percentage of cases in which words with this letter in their spelling have the corresponding letter name in their pronunciation. Over all letters, the average reading reliability is 10.7%. These counts include words where the letter name in the spelling does not actually symbolize the letter in question. For example, cherry contains an e in its spelling and an / / in its pronunciation, but the e does not symbolize the final / /. The lack of alignment is less relevant to young children than to adults, as children do


127

not at first know which letters represent which phonemes. The percentage of words that contain at least one letter whose name appears in the pronunciation falls to 37.5% for monosyllabic words, where misalignments like the one with cherry are unlikely to arise. The results in Table IV show that letters differ markedly in how reliably they predict the presence of names in pronunciations. Vowel letters are generally the best predictors. However, even the highest ranking vowel, e, has a reliability of only about one in four. Letters with their sounds (or one of their sounds) at the beginning of their names are somewhat less reliable than those with the sounds at the ends of their names, 3.8% as compared to 6.2%. Thus, US children could make some connections from print to speech if they use letter names, but only for a minority of the letters they encounter in words. Such letters may nevertheless play a special role in learning to read, helping children understand that the spellings of words are systematically related to their pronunciations and helping them form their first partial connections between printed and spoken words. Educators could take advantage of children’s tendency to link print and speech on the basis of letter names by including words such as OK, jail, and eat in early instruction. The teaching of such words may be especially valuable for children who are having difficulty grasping the idea that print is systematically related to speech. Research has not examined the possible relation between the letters in a child’s own name and the child’s ability to use those letters in connecting print and speech, but a link may well exist. Children may most readily form connections for words that start with the first letter of their own name and that have the letter name in their spoken form, as with jail for Joe. Children learning to read in languages other than English can also form connections from print to speech that are based on letter names. Brazilian prereaders who are familiar with letter names have been found to learn spellings like CBL for cebola (onion) more easily than those like HMN for cebola (Abreu & Cardoso-Martins, 1998). These children appear to benefit from the fact that the pronunciation of cebola starts with / /, the Portuguese name of the first letter in CBL. Prereaders with little knowledge of letter names do not learn the motivated spellings more easily than the arbitrary ones. What makes letter names particularly useful in Portuguese is that many words in this language have one or more letters in their spelling, especially vowels, that correspond to letter names in the pronunciation. Indeed, 51 of the 56 most common content words in a list of words in books for Brazilian kindergartners contain at least one vowel letter whose name appears in the corresponding spoken word (Cardoso-Martins, Resende, & Rodrigues, 2001). For a number of words, such as bola (ball), the names of all of the vowel letters are heard in the pronunciation.

128


Although comprehensive analyses have not been carried out, spellings that make sense on the basis of letter names appear to be more common in Portuguese than English. Young speakers of Hebrew also take advantage of letter names in and asked recognizing printed words. When presented with the word / (concrete) or / / (rustic), the Israeli whether that spelled / kindergartners tested by Levin et al. (2002) could usually judge correctly /. / / and / / are the names of Hebrew letters, and that it was / / helps seeing the letter at the beginning (right side) of the word / children identify the word. The children did not perform as well on words whose pronunciations do not begin with the name of the letter that their spelling begins with. The utility of matching full letter names may be restricted by the fact that relatively few Hebrew words contain an entire letter name, in part because the names of Hebrew letters are relatively long (averaging 3.5 phonemes). However, Levin et al. found that children’s performance also improves to a lesser extent when only some of the phonemes of the letter name are present in the word. Because partial matches are more frequent in the vocabulary, they may help Hebrewspeaking children connect print and speech. Direct use of letter names to connect print and speech, in the sense of mappings that link a letter to its entire spoken name, should be available in all or almost all languages with alphabetic writing systems. All languages show iconicity in their letter names; most have letter names with no ‘‘extra’’ phonemes besides the letter’s sound (in the case of vowels) or just one (in the case of consonants); and the ‘‘extra’’ phoneme, if present, tends to be reasonably common in the language. As a result, some words contain letter names in their spoken forms and the corresponding letters in their printed forms. The number of such words probably differs across languages, as discussed above, making the tendency to use letter names more useful for reading in some languages than others. Cross-linguistic studies are needed to quantify these differences and examine their effects on early reading. Letter-name knowledge also helps reading indirectly, in the sense that knowledge of letter names helps children learn letter sounds and knowledge of letter sounds in turn benefits reading. For example, children who use b’s name to help learn that b corresponds to / / can connect the b of a word like bone to the / / in the word’s spoken form. Cardoso-Martins, Resende, and Rodrigues (2001) have argued that Brazilian Portuguese-speaking children can make such sound-based connections from an early age, using their knowledge of letter names to learn simplified spellings in which the letters correspond to phonemes, not letter names. Thus, letter names can aid reading directly, for words in which letters correspond to entire names, and indirectly, by helping children form connections based on sounds.


129

VI. Role of Letter Names in Learning to Spell Words When children spell, they sometimes symbolize letter names in spoken words with the corresponding letters. This tendency is found for vowels, where it leads young speakers of English to produce misspellings like con for cone and mul for mule, omitting the final e that appears in the conventional spellings of these words (e.g., Read, 1975; Reece & Treiman, 2001; Treiman, 1993). The tendency is also found for consonants, where it leads young children to produce spellings like kr for car and hlp for help (e.g., Treiman, 1993, 1994). Children are more likely to make such errors on consonants with vowel–consonant names than consonants with consonant–vowel names. The errors are most common for r and next most common for l. When a vowel is followed by / / or / /, the sequence is quite cohesive and difficult for children to divide into its component phonemes. This cohesiveness probably reflects the fact that / / and / / are similar in many respects to vowels. We saw evidence for a special difficulty with / / earlier in the large discrepancy between letter-name knowledge and letter-sound knowledge for r. Errors like those described above are common among US kindergartners and first graders who are trying to spell whole words. Similar phenomena occur among younger children in simpler spelling tasks. In one study, US preschoolers (mean age 5 years, 5 months) who were asked to orally provide the first letters of words spoken to them by an experimenter were more likely to say that beech starts with b than that bone starts with b (Treiman, Tincoff, & Richmond-Welty, 1996). The children were also more accurate at saying that a word like deaf ends with f than that loaf ends with f. Similar phenomena occur among young learners of Hebrew (Levin et al., 2002). In the examples given so far, use of letter names aids spelling. However, letter names sometimes cause errors in both languages, as when English-speaking children think that seed starts with c rather than s. Of the 6,232 words in the list of kindergarten and first-grade words described earlier, 2,919 contain a letter name. Of these, 92.1% have that letter in their spelling. Thus, children who symbolize a letter name with that letter will usually produce a letter that is found in the conventional spelling of the word. Balanced against this is the fact that over half of the words in our list do not contain a letter name in their pronunciation. Table IV shows the reliability of the individual letter names in the pronunciation-to-spelling direction. Only a few letters drop below 50%—c, because / / is usually spelled with s; k and q, because most / / is spelled with c; z, because most / / is spelled with s; and y, because / / is usually spelled with i. For consonants, children who represent a letter name in a spoken word with the corresponding letter will usually produce a correct consonant letter but

130


omit the vowel. The resulting spellings are wrong, as with hlp for help, but comprehensible to parents and teachers. With vowels, spelling by letter names also fails to give a fully correct spelling in many cases. Children who use this method will often omit one letter of a digraph, as in et for eat, or a final e, as in con for cone. These are errors, but again understandable ones. Relatively rarely, as when children symbolize the first two phonemes of wife with y, are spellings based on letter names far afield. Languages differ in how many of their spoken words contain letter names. Portuguese and Spanish appear to contain more such words, especially those with letter-name vowels, than English and Hebrew. In English, as mentioned earlier, the names of vowel letters are the vowel’s long sound in order to make the name legal. However, the short forms of vowels are more common in English words. Few Hebrew words contain letter names because the names in this language are longer than those of other languages. Languages also differ in how many of the words with letter names in their spoken form contain the corresponding letter in their spelling. The percentage of such words is high in English, as we have seen, and is surely higher in many other writing systems. Languages also differ in how often use of letter names leads to a fully correct spelling. In all languages, letter-name spelling of consonants usually yields one letter of the correct spelling but, in the absence of other spelling strategies, causes spellers to omit vowels. For vowels, letter names probably permit more fully correct spellings in languages such as Spanish, which does not use digraphs to spell simple vowels, than in languages such as English, with its many vowel digraphs and silent es. Differences among languages like those just described may help explain why children’s early spellings sometimes differ in character from one language to another. For example, Ferreiro and Teberosky (1982) reported spellings such as ao for sapo (toad) and ao for palo (stick) among young Spanish-speaking children. The matter has not been studied quantitatively, but we believe that all-vowel spellings are uncommon among young English speakers. Ferreiro and Teberosky attributed spellings such as ao for sapo to children’s initial belief that print represents speech at a syllabic level. However, an alternative hypothesis is that these spellings reflect children’s reliance on letter names. Both vowels in the Spanish words sapo and palo are the names of letters, and the large number of such words in this and other languages may lead to a large number of all-vowel spellings. As with reading, letter names influence spelling both directly and indirectly. Direct influences occur when children symbolize an entire letter name with the corresponding letter. Indirect influences occur when children use the name of a letter in determining how to spell a phoneme. Such indirect influences emerge rather early. For example, Treiman, Weatherston,


131

and Berch (1994) asked preschoolers (mean age 5 years, 2 months) and kindergartners (mean age 5 years, 11 months) to state the consonant letter that would be used in spelling various consonant–vowel and vowel– consonant syllables. The children performed better with phonemes such as / / and / / than with phonemes such as / /, / /, and / /. Children probably perform relatively well on phonemes such as / / because this sound occurs at the beginning of a letter’s name. Performance is poorer on phonemes such as / /, which occurs at the end of a letter name, and / /, which does not occur in a letter name at all. Consistent with results reviewed earlier, phonemes that appear in a misleading letter name, such as / /, also caused difficulty for the children in this study. For example, the children sometimes misspelled / / as y. Phoneme frequency cannot be a complete explanation for the findings, as children were worse at spelling / / than / / even though / / is the more common phoneme. In English, a number of common consonant and vowel phonemes do not occur in the name of any letter. As mentioned, / / is one such consonant phoneme; / / and / / are others. Of the vowels, / / and / /, among others, do not occur in a letter name. Indeed, 29% of all English consonant phonemes (7/24) and 42% of all non-diphthongal English vowel phonemes (5/12) are not heard in the American English name of a letter.5 These facts could contribute to the difficulty of learning to spell in English. They could also help explain why young English-speaking spellers have more difficulty with vowels than consonants (e.g., Treiman, 1993). The inconsistency of English spelling is probably one contributor: vowel phonemes tend to have a number of different possible spellings and are more inconsistent than consonants. However, the fact that a number of vowels do not occur in the name of a letter may be another contributor. Brazilian Portuguese and French also have a fair number of phonemes that do not occur in a letter name. These phonemes are more often vowels than consonants, reflecting in part the nasalized vowels that occur in these languages. In certain other languages, virtually all phonemes occur in a letter name. For example, trilled / / is the only unambiguously missing phoneme in Spanish, and that gap is filled by unofficial versions of the alphabet that include an additional letter, rr, whose name does include that sound. Research is needed to examine the effects of these factors on spelling development and the role that they may play in different patterns of performance across languages.

5 /, and / /, are not included in these counts The most diphthongized vowels, / /, / because children may spell them on the basis of their components rather than as units. For the same reason, the discussions of other languages do not consider diphthongs.

132


Even those researchers who adopt a stage theory of literacy development, postulating an initial stage of reading development in which children rely on paired-associate learning, believe that children go beyond rote memorization earlier in spelling than in reading (e.g., Frith, 1985). From a young age, children attempt to construct the spellings of many words from the words’ phonological forms. The data we have presented show that letter names can facilitate this process. Children who rely solely on letter names cannot be good spellers, but letter names may help them move forward.

VII. Conclusions Discussions of alphabetic writing systems have concentrated on the regularity of print–speech relations. When differences in performance are found among children learning different writing systems, they are often attributed to differences in the regularity of the writing systems (e.g., Wimmer & Goswami, 1994). For example, systems such as Finnish have been characterized as predictable and easy to learn, whereas systems such as English have been characterized as less consistent and hence more difficult. In this chapter, we have focused on an aspect of writing systems that researchers have often ignored—the names that are given to the letters. This aspect of writing systems is important because children in literate societies often become very familiar with the names of letters before they start to read and write. This knowledge, typically acquired informally before systematic reading instruction begins, shapes how children learn to read and spell and how they respond to classroom instruction. Children use their knowledge of letter names to try to make sense of why words are written the way they are, and they do so in ways that teachers might not always suspect. Differences in performance among children learning different writing systems may reflect, in part, differences in the properties of the letter names and how the names mesh with the characteristics of the spoken language. The relation between the names and shapes of letters is largely arbitrary in English, although a number of motivated letter shapes exist in certain other alphabets. As a result, English-speaking children have no choice but to memorize the links between letters’ shapes and names in a rote fashion. Importantly, the relation between the names and the sounds of letters is not arbitrary in English or any other alphabetic system. English-speaking children take advantage of the fact that most letter names contain their sounds, and this makes learning of letter sounds quite different from learning of letter names. Children who know the names of letters can take advantage of certain relations between printed and spoken words, such as the link between the e in the printed eat and the / / in its spoken form. This


133

helps children move from treating printed words as arbitrary visual patterns to treating them as maps of linguistic structure. As we described earlier, some researchers have suggested that letter names may be harmful to children and should not be taught (Feitelson, 1988). The research reviewed here shows, on the contrary, that letter names are more helpful than harmful to children who are learning English and other languages. There is some truth to the belief, deeply held in the US and a number of other countries, that letter names provide an important foundation for reading instruction. The challenge for educators is to take advantage of the knowledge that children bring with them to school, showing children how letter names can often aid them in reading and spelling words and pointing out cases in which letter names mislead. The letter names of English, although generally iconic, are not as systematic as those of certain other languages. If we were to design a new system of English letter names to maximize pedagogical utility, we might make different choices. For example, it might be more useful to label d as / / than as / / if there are more words that contain / / than / /, as in fact there are in the sample of kindergarten and first-grade words examined here. It might be helpful to label ch as / / followed by a vowel rather than the current c, h to show how it sounds in words. As mentioned earlier, Spanish follows this approach. Efforts to reform the English spelling system have borne little fruit, however. We doubt that efforts to reform the letter names in general use would fare substantially better. Although some research on children’s use of letter names has been done in languages other than English, most of the research has examined English. One of our goals in writing this chapter has been to encourage crosslinguistic research on this topic. Letter names, like other aspects of language, show some underlying similarities across languages but some noticeable surface differences. The extent to which letter names provide direct and indirect aid in reading and spelling also differs across languages. A combination of behavioral studies and language statistics should be valuable in studying how young speakers of different languages learn about letter names and how they use this information in learning to read and spell.

ACKNOWLEDGMENTS Preparation of this chapter was supported, in part, by grants from the National Science Foundation and the March of Dimes Birth Defects Research Foundation. Thanks to Iris Levin and Margo Bowman for comments on a draft.

134


REFERENCES Abreu, M. D. de, & Cardoso-Martins, C. (1998). Alphabetic access route in beginning reading acquisition in Portuguese: The role of letter-name knowledge. Reading and Writing, 10, 85–104. Bowman, M., & Treiman, R. (2002). Relating print and speech: The effects of letter names and word position on reading and spelling performance. Journal of Experimental Child Psychology, 82, 305–340. Byrne, B. (1992). Studies in the acquisition procedure for reading: Rationale, hypotheses, and data. In P. B. Gough, L. C. Ehri, & R. Treiman (Eds.), Reading acquisition (pp. 1–34). Hillsdale, NJ: Erlbaum. Cardoso-Martins, C., Resende, S. M., & Rodrigues, L. A. (2001). Letter name knowledge and the ability to learn to read by processing letter-phoneme relations in words: Evidence from Brazilian Portuguese-speaking children. Reading and Writing, 10, 10–20. Cassidy, K. W., Kelly, M. H., & Sharoni, L. J. (1999). Inferring gender from name phonology. Journal of Experimental Psychology: General, 128, 362–381. CMU Pronouncing Dictionary (version 0.6) [Data file]. (1998). Carnegie Mellon University, Speech at CMU Web site: fttp://ftp.cs.cmu/project/speech/dict/cmudict.0.6. Conrad, R. (1964). Acoustic confusions in immediate memory. British Journal of Psychology, 55, 75–84. Durrell, D. D. (1980). Letter-name values in reading and spelling. Reading Research Quarterly, 16, 159–163. Ehri, L. C. (1983). A critique of five studies related to letter-name knowledge and learning to read. In I. L. M. Gentile, M. L. Kamil, & J. Blanchard (Eds.), Reading research revisited (pp. 143–153). Columbus, OH: C. E. Merrill. Ehri, L. C. (1998). Grapheme-phoneme knowledge is essential for learning to read words in English. In L. C. Ehri & J. L. Metsala (Eds.), Word recognition in beginning literacy (pp. 3–40). Mahwah, NJ: Erlbaum. Feitelson, D. (1988). Facts and fads in beginning reading: A cross-language perspective. Norwood, NJ: Ablex. Ferreiro, E., & Teberosky, A. (1982). Literacy before schooling. New York: Heinemann. Frith, U. (1985). Beneath the surface of developmental dyslexia. In K. E. Patterson, J. C. Marshall, & M. Coltheart (Eds.), Surface dyslexia: Neuropsychological and cognitive studies of phonological reading (pp. 301–330). Hove, England: Erlbaum. Gough, P. B., & Hillinger, M. L. (1980). Learning to read: An unnatural act. Bulletin of the Orton Society, 30, 179–196. Hildreth, G. (1936). Developmental sequences in name writing. Child Development, 7, 291–303. International Phonetic Association. (1996). Reproduction of The International Phonetic Alphabet. Retrieved from http://www2.arts.gla.ac.uk/IPA/ipachart.html. International Phonetic Association. (1999). Handbook of the International Phonetic Association: A guide to the use of the International Phonetic Alphabet. Cambridge, UK: Cambridge University Press. Johnson, R. J. (1969). The effect of training in letter names on success in beginning reading for children of differing abilities. Unpublished doctoral dissertation, University of Minnesota. Levin, I., Patel, S., Margalit, T., & Barad, N. (2002). Letter names: Effect on letter saying, spelling, and word recognition in Hebrew. Applied Psycholinguistics, 23, 269–300. McBride-Chang, C. (1999). The ABC’s of the ABC’s: The development of letter-name and letter-sound knowledge. Merrill-Palmer Quarterly, 45, 285–308.


135

Perfetti, C. A. (1992). The representation problem in reading acquisition. In P. B. Gough, L. C. Ehri, & R. Treiman (Eds.), Reading acquisition (pp. 145–174). Hillsdale, NJ: Erlbaum. Read, C. (1975). Children’s categorization of speech sounds in English (NCTE Research Report No. 17). Urbana, IL: National Council of Teachers of English. Reece, C., & Treiman, R. (2001). Children’s spelling of syllabic /r/ and of letter-name vowels: Broadening the study of spelling development. Applied Psycholinguistics, 22, 139–165. Samuels, S. J. (1972). The effect of letter-name knowledge on learning to read. American Educational Research Journal, 9, 65–74. Seuss, D. (1955). On beyond zebra. New York: Random House. Share, D. (1999). What’s in a name? On the relationship between preschoolers’ knowledge of letter names and early reading. Unpublished manuscript, University of Haifa, Israel. Silberberg, N. E., Silberberg, M. C., & Iversen, I. A. (1972). The effects of kindergarten instruction in alphabet and numbers on first grade reading. Journal of Learning Disabilities, 5, 254–261. Snow, C. E., Burns, M. S., & Griffin, P. (Eds.). (1998). Preventing reading difficulties in young children. Washington DC: National Academy Press. Treiman, R. (1993). Beginning to spell: A study of first-grade children. New York: Oxford University Press. Treiman, R. (1994). Use of consonant letter names in beginning spelling. Developmental Psychology, 30, 567–580. Treiman, R., & Broderick, V. (1998). What’s in a name: Children’s knowledge about the letters in their own names. Journal of Experimental Child Psychology, 70, 97–116. Treiman, R., & Rodriguez, K. (1999). Young children use letter names in learning to read words. Psychological Science, 10, 334–338. Treiman, R., Sotak, L., & Bowman, M. (2001). The roles of letter names and letter sounds in connecting print and speech. Memory & Cognition, 29, 860–873. Treiman, R., Tincoff, R., & Richmond-Welty, E. D. (1996). Letter names help children to connect print and speech. Developmental Psychology, 32, 505–514. Treiman, R., Tincoff, R., & Richmond-Welty, E. D. (1997). Beyond zebra: Preschoolers’ knowledge about letters. Applied Psycholinguistics, 18, 391–409. Treiman, R., Tincoff, R., Rodriguez, K., Mouzaki, A., & Francis, D. J. (1998). The foundations of literacy: Learning the sounds of letters. Child Development, 69, 1524–1540. Treiman, R., Weatherston, S., & Berch, D. (1994). The role of letter names in children’s learning of phoneme–grapheme relations. Applied Psycholinguistics, 15, 97–122. Venezky, R. L. (1975). The curious role of letter names in reading instruction. Visible Language, 9, 7–23. Villaume, S. K., & Wilson, L. C. (1989). Preschool children’s explorations of letters in their own names. Applied Psycholinguistics, 10, 283–300. Wimmer, H., & Goswami, U. (1994). The influence of orthographic consistency on reading development: Word recognition in English and German children. Cognition, 51, 91–103. Worden, P. E., & Boettcher, W. (1990). Young children’s acquisition of alphabet knowledge. Journal of Reading Behavior, 22, 277–295. Zeno, S. M., Ivenz, S. H., Millard, R. T., & Duvvuri, R. (1995). Educator’s word frequency guide. Brewster, NY: Touchstone Applied Science Associates.


EARLY UNDERSTANDINGS OF EMOTION, MORALITY, AND SELF: DEVELOPING A WORKING MODEL

Ross A. Thompson DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF NEBRASKA LINCOLN, NEBRASKA 68588

Deborah J. Laible DEPARTMENT OF PSYCHOLOGY SOUTHERN METHODIST UNIVERSITY DALLAS, TEXAS 75275

Lenna L. Ontai DEPARTMENT OF HUMAN AND COMMUNITY DEVELOPMENT UNIVERSITY OF CALIFORNIA DAVIS, CALIFORNIA 95616

I. INTERNAL WORKING MODELS AND RELATIONSHIPS II. INTERNAL WORKING MODELS AND COGNITIVE GROWTH A. WORKING MODELS AS DEVELOPING REPRESENTATIONS B. IMPORTANCE OF LANGUAGE AND CONVERSATION C. VARIATIONS IN PARENTAL CONVERSATIONAL STYLE III. DEVELOPING A WORKING MODEL A. EMOTION UNDERSTANDING AND ATTACHMENT SECURITY B. PARENT–CHILD CONVERSATION, ATTACHMENT, AND EMOTION UNDERSTANDING C. CONSCIENCE AND EMOTION D. FAMILY CONFLICT: LESSONS IN EMOTION, MORALITY, AND RELATIONSHIPS IV. CONCLUSION: THE IMPACT OF RELATIONSHIPS ON EMOTION, MORALITY—AND THE SELF REFERENCES


137


138

Ross A. Thompson et al.

The minds of young children amaze. After decades of describing their thinking as egocentric and lacking understanding, developmental scientists are now discovering how insightfully young children distill lessons about the physical world, people, and themselves from everyday experience. Infants and preschoolers constantly exploit their everyday transactions with objects and people to understand how physical events occur, what goes on in peoples’ minds, and who they are and what they can do. With respect to the physical world, scientists are discovering how powerfully young minds deduce understanding from simple observation of ordinary events and their own playful experimentation. With respect to the psychological world, scientists are discovering how early young children develop expectations for the behavior and thoughts of others that are based on a rudimentary understanding of human intentions and feelings. It is as if a young child’s readiness to learn is juxtaposed with abundant natural learning opportunities in daily experience. Children’s relationships with people are central to these learning opportunities. This realization has arisen from a convergence of scientific progress in two broad areas of developmental psychology. On the one hand, researchers studying cognitive development have elucidated the growth of the mind in their studies of developing event representation, memory, theory of mind, language, and other topics. They have shown that these intellectual achievements arise not only from the young mind’s surprising capacities for inducing understanding from everyday observation, but also from the ways that caregivers scaffold understanding through the structure of daily routines or by how they talk with the child about recent events. On the other hand, researchers studying social development have shown how young children’s experiences in close relationships provide avenues for psychological understanding of self and others. They have shown that the quality of communication shared by a young child and an adult partner, and the trust of their relationship, together influence children’s earliest understandings of who they are, what people are like, and the nature of human relationships. Taken together, these dual areas of developmental study are converging on the view that children’s relationships with caregivers provide powerful forums for learning about mental realities, especially as language provides avenues for making explicit the hidden (and sometimes confusing) psychological world that underlies behavior, relationships, and self-understanding. Our research has drawn on both fields of developmental study as we have sought to explore children’s earliest understandings of emotion, morality, self, and other psychological realities. In this chapter, we profile our work. We begin with the intriguing view from attachment theory that based on the security of their attachments with caregivers, young children create mental ‘‘working models’’ of these relationships that underlie their understandings

Early Understandings of Emotion, Morality, and Self

139

of other people, relationships, and themselves. But what are ‘‘working models’’ and how do they develop? For better understanding of this question, we turn to the work of cognitive developmentalists and their research on the growth of early representations of events, people, and the self—which are, in short, core elements of the working models described by attachment theorists. We identify several conclusions from this research that have formed the basis for our program of research, which we describe subsequently. In a concluding section, we identify multiple ways that young children’s experience in close relationships guides their understanding of the psychological world, and some of the implications of these conclusions for future research.

I. Internal Working Models and Relationships One of the most important and provocative ideas emerging from contemporary attachment theory is that early experiences with caregivers lead young children to develop mental representations of the caregiver’s sensitivity and responsiveness, and their deservingness of care. These social expectations contribute to the security (or insecurity) of the parent–infant relationship and affect how young children respond to their caregivers, especially in circumstances of stress or uncertainty (Lamb, 1981). This is, in a sense, what is meant by a secure or insecure attachment relationship. But to Bowlby (1973, 1980, 1988), as these core expectations become consolidated over time, they permit not only immediate forecasts of the caregiver’s behavior, but also guide expectations for relationships with other people, self-referential beliefs, and assumptions about people’s motives and intentions. Borrowing from Craik’s (1943) portrayal of mental models, classic object relations theory, and other theoretical sources, Bowlby described these ‘‘internal working models’’ as dynamic representational structures that are derived from relational experience and color the child’s responsiveness to partners. He believed that over time, working models become interpretive filters by which children reconstruct new experiences and relationships in ways that are consistent with past experiences and expectations, and which also provide implicit decision rules for relating to others. As a consequence, children with secure or insecure attachment histories respond to others based on expectations of warmth and intimacy that may cause them, for better or worse, to evoke the kinds of responses from others that confirm their initial expectations. Insecurely attached children may, for example, so anticipate a new partner’s unfriendliness or unreliability that they remain distant and unengaged and, in so doing, evoke the kind of disinterested response they expect from that person. A securely attached child may, by contrast, evoke a

140


much different response from the same partner, guided by a prior relationship history to respond more positively and thus contribute to creating a warmer, more intimate relationship with that person. In Bowlby’s view, therefore, internal working models constitute the bridge between early experiences of sensitive or insensitive care and personality development. As internal working models are maintained over time, in other words, they color a person’s experiences in new relationships and create self-confirming expectations of how other people will respond to them. These representations influence social dispositions and attributions of the motives, intentions, and emotions of others, as well as influencing selfreferential beliefs, such as about one’s competence as a social partner. In a broader view, these working models are also believed to shape parenting practices, leading to current research interest in adult attachment representations and their association with the sensitivity or insensitivity of parental care (see, for example, van Ijzendoorn, 1995). In this view, secure or insecure parental working models constitute the basis for the responsiveness of care that leads offspring to develop secure or insecure attachments to these parents, contributing to the ‘‘intergenerational transmission’’ of attachment security. Bretherton (1990, 1991; Bretherton & Munholland, 1999) elaborated Bowlby’s formulation by emphasizing how the development of working models is influenced by the quality of a young child’s shared communication with the caregiver. The working models of securely attached children, she proposed, are shaped by the open, fluent, and candid sharing of feelings and viewpoints that can occur in relationships of trust and confidence. Bretherton argued that, by contrast, insecure relationships are characterized by limited emotional sharing, especially of negative or disturbing feelings that either or both partners may find threatening and which may elicit defensive exclusion. Drawing on Schank’s (1982) theory of dynamic memory and Johnson-Laird’s (1983) portrayal of flexible mental models, Bretherton also described internal working models as a system of hierarchically organized representational systems that involve different levels of generalizability and are relevant to various broader belief systems. Taken together, the view that early relational experience provokes the development of wide-ranging mental representations related to sociability, understanding of people, and self-concept is an heuristically powerful theoretical formulation. It has led to a new generation of studies of the association between attachment security and various kinds of mental representations, including children’s developing self-concept, conceptions of friendship, memory for emotional issues, conscience, and a variety of other features of social and emotional understanding that are believed to be related to the working models associated with secure or insecure


141

attachments (see Thompson, 1999, for a review). Working models have also been enlisted to explain individual differences in preattentive processing, selective attention, defensive exclusion, and a variety of other psychological processes from infancy to adulthood. In a sense, the breadth of representational influences attributed to internal working models attests to the heuristic power of this concept. The problem, however, is that ‘‘in the very power of such a model lies a trap: it can too easily explain anything’’ (Hinde, 1988, p. 378). As a broad, inclusive formulation, working models potentially serve as a ‘‘catch-all, posthoc explanation’’ for an almost limitless variety of research findings on the correlates of attachment security (Belsky & Cassidy, 1994; see also Rutter & O’Connor, 1999). The problem is that Bowlby’s concept of the internal working model is a conceptual metaphor, not a systematically defined theoretical construct, and thus it lacks the specificity required to guide and constrain its theoretical applications. This not only contributes to expansive theoretical applications, but makes it difficult to create conceptually consistent measures of working models for developmental research (Thompson & Raikes, in press). In addition, central questions concerning consistency and change in internal working models over time, conscious and unconscious features of the functioning of working models, and how these representations are associated with other social–cognitive constructs (such as relational schemas and attributional processes) remain unclarified. Perhaps most important, little is known about the development of working models apart from Bretherton’s provocative formulations concerning openness of communication. Understanding the nature of developmental changes in these mental representations, however, and how working models relate to other aspects of cognitive growth, would contribute to a clearer portrayal of the nature of working models and how they are likely to influence sociopersonality functioning in children of different ages. Such understanding would also enable attachment researchers to conceptualize more precisely the correlates of secure or insecure attachments at different stages of development. In the end, the usefulness of the internal working models concept hinges on the capacity of attachment theorists to define it with greater clarity and precision, and to link it to other developmental accomplishments of the growing mind. This is a significant challenge for the future of attachment theory (Thompson & Raikes, in press).

II. Internal Working Models and Cognitive Growth Fortunately, at the same time that attachment theorists became interested in the mental representations associated with attachment security, cognitive

142


developmental researchers began significantly expanding their appreciation of mental development and its social influences. Their research has addressed different kinds of mental representations: the development of theory of mind, event representation, episodic memory, social scripts and schemas, autobiographical memory, and the concept of shared social reality (the latter drawn from neo-Vygotskian theory). But underlying different conceptual rubrics, these cognitively oriented scientists have been concerned with features of understanding that are central to attachment theorists’ concept of the working model: how young children achieve insight into peoples’ intentions, motives, and emotions; how children conceptualize the specific events and experiences of their lives; and the influences on an emerging sense of psychological individuality and self-concept. Given that the developing mind that concerns attachment researchers is the same developing mind of interest to cognitive researchers, scientists in each field are likely studying allied mental processes relevant to the encoding, interpretation, and representation of social experience. Moreover, in light of Bretherton’s cogent argument that internal working models are hierarchically organized representational systems, understanding the development of working models in terms of growth in the mental representations studied by cognitive developmentalists should not only add clarity and specificity to the internal working models concept, but also begin to identify the broader representational systems with which working models are associated. In several papers, one of us has reviewed the cognitive developmental literature that might be relevant to the growth of working models (e.g., Thompson, 1998, 2000). This review has yielded three conclusions that can provide insight into the growth of working models and their developmental influences. A. WORKING MODELS AS DEVELOPING REPRESENTATIONS

Although attachment theorists have devoted considerable attention to how working models function as fully developed mental representations, these representations clearly develop over an extended period of time. Working models grow and change not only as the result of new experience in close relationships, but also as children acquire new representational capacities that lead them to see other people, themselves, and relationships differently than before (Ainsworth, 1989; Marvin & Britner, 1999). Indeed, this conclusion is inevitable given that infants and young children are incapable of the sophisticated self-referent beliefs and social inferences assumed to be inherent in the functioning of fully developed working models in adults. Working models are thus developing representations that change significantly in scope and sophistication throughout


143

childhood and adolescence. As new conceptual and thinking skills develop, there may be growth and consolidation of earlier-developed working models at a new level of representational sophistication. Furthermore, new representational capacities may provoke a reorganization and revision of earlier working models as children acquire new ways of understanding past and present experiences (Stern, 1989). Indeed, when new representational advances occur, they may also be associated with changes in attachment security because of how new forms of psychological understanding provoke new ways of regarding familiar relationships. At least two representational advances are relevant to understanding the growth of working models. The first occurs in early childhood. Owing largely to developing language ability, young children begin to conceptualize in more sophisticated and memorially more enduring ways the experiences of care that were earlier represented using the simpler perceptual-affective schemas of infancy. Research on the growth of theory of mind indicates, for example, that between the ages of 3 and 5 years, a preschooler’s appreciation of the feelings, thoughts, and beliefs of another expands considerably (Bartsch & Wellman, 1995; Flavell & Miller, 1998; Wellman, Cross, & Watson, 2001). This contributes to a developing appreciation of the motives and intentions of attachment figures, especially as they differ from the child’s own, and expands the child’s capacity to understand the reciprocal, complementary nature of relational interaction, such as conflict resolution through compromise or negotiation (Harris, 1996). During the same period, young children begin to retain more coherent recollections of past events that are built on their developing appreciation of the structure of routines in daily experience (Hudson, 1993; Nelson, 1989; Nelson & Gruendel, 1981). This contributes to a young child’s ability to begin to forecast the caregiver’s behavior based on mental models of familiar events, especially those involving close relational partners. Research on autobiographical memory suggests that after the age of 3, young children become capable of thinking about, and remembering, past events in a more consistently self-referential manner that contributes to, and draws on, the child’s developing capacities for self-reflection (Hudson, 1990; Nelson, 1993a; Welch-Ross, 1995). This contributes to the growth of psychological self-awareness and to the initial construction of an ongoing narrative account of events of personal significance, including those involving attachment figures, that contributes to self-understanding (Miller, 1994). In these and other ways, the conceptual advances of early childhood suggest the significant changes in working models that are likely to be occurring during this period. The second representational transition (extensive discussion of which is beyond the scope of this chapter) occurs in early adolescence with the

144


emergence of metarepresentational capacities and abstract thought (Ainsworth, 1989; Kobak & Cole, 1994; Marvin & Britner, 1999). This enables young people to reflect more competently on their own mental processes, especially as they acquire more sophisticated forms of interpersonal understanding and self-awareness. Moreover, as youth strive for internally coherent self-referent beliefs, earlier working models may become subject to deliberate reexamination as young people reflect on their consistency with other beliefs and with current relational experience. Consequently, many features of the working models of childhood may be subject to elaboration, enrichment, and revision. Understanding working models as developing representations contributes to the realization that representations of relational experience are very different at different ages, and thus have different associations with social behavior and emotional responding. This, in turn, has important implications for studying the association between attachment and sociopersonality functioning. A secure attachment may not be comparably predictive of different aspects of personality growth at various stages of development. More specifically, attachment security may be developmentally most influential when the working models with which it is associated have matured sufficiently to influence the specific features of psychosocial functioning that are emerging at the same time. If early childhood is when an integrated, enduring capacity for self-understanding begins to take shape, for example, then a secure attachment may be most strongly associated with developing self-concept during this period rather than before (see, for example, Verschueren & Marcoen, 1999). Similarly, the associations between attachment and emotional understanding (Laible & Thompson, 1998; Ontai & Thompson, 2002) and conscience development (Kochanska, 1993, 1995; Kochanska & Thompson, 1997) may be most apparent in young preschoolers, when theory of mind is rapidly developing and when internal working models are most likely to influence the emergence of emotional and moral understanding and self-regulation (see Meins et al., 1998). More speculatively, hypothesized associations between attachment security and developing identity or sexuality may be most apparent in adolescence when these features of psychosocial development are taking shape. In short, as working models grow in sophistication and scope throughout childhood, a secure attachment is influential in different ways at different ages, based on the pivotal psychosocial accomplishments of each stage of psychological growth. Traditionally researchers have assessed attachment in infancy (using the Strange Situation) to explore how attachment security predicts later sociopersonality development. But the most interesting hypothesized psychological outcomes of secure or insecure attachment relationships—including


145

self-understanding, capacities for developing satisfying close relationships, and psychological understanding of other people—are developmental achievements of early childhood or later, when the internal working models associated with attachment security have also become more sophisticated. Consequently, early childhood is when the internal working models associated with a secure or insecure attachment may be developmentally more influential than in infancy. Research that examines the contemporaneous or predictive correlates of attachment security in early childhood may yield associations that are obscured when Strange Situation assessments in infancy are used. For other hypothesized correlates of secure attachment, moreover, researchers may be wise to study attachment influences at even later ages. B. IMPORTANCE OF LANGUAGE AND CONVERSATION

Attachment theorists emphasize, of course, how secure or insecure working models derive from the sensitivity of parental care. Children’s direct experiences of an adult’s warmth, responsiveness, and reliability constitute core components of the development of working models. But as young children become more sophisticated in thinking about and sharing their experiences with others, conversational discourse becomes another powerful influence on their understanding of relational experiences. During the preschool years, therefore, working models are shaped not only by children’s direct representations of parental care, but also by the secondary representations of experience mediated by language. Bretherton (1990, 1991) noted, for example, that the open or restricted communicative style of the parent–child dyad significantly guides how easily young children can share troubled or conflicting emotions and learn from the adult about their feelings and experiences. Bowlby (1980, 1988) also believed that psychological difficulty was exacerbated when children were forced to choose between markedly inconsistent direct and indirect representations of care, such as when a parent acted coldly but spoke affectionately, or when a parent acted harshly but avowed benign intentions. In these cases, he reasoned, the risks of defensive exclusion and of distorted perceptions of relational experience were enhanced. Cognitive-developmental theorists agree that language significantly influences early representations of experience (see, for example, Nelson, 1996; Rogoff, 1990; Tomasello, 1999). This is especially true of aspects of relational experience that are, in a sense, invisible: the nature of people’s thoughts and intentions, the values and rules governing conduct, the personality characteristics that predict people’s individual propensities, and, of course, the child’s own internal dispositions and attributes. Although

146


these features of human functioning are indirectly revealed in behavior, understanding the associations between these internal characteristics (thoughts, beliefs, feelings, dispositions) and specific behavior is a considerable conceptual challenge for young children. Conversations with adults contribute clarity to the mysteries of why people act as they do by articulating in language the influence of these internal motivators. Moreover, conversations do so by embedding interpersonal understanding within the broader values of the culture, and belief systems of the adult. Language contributes to this representational advance in psychological understanding relevant to working models in several ways. First, it permits the reconstruction of the toddler’s earlier, implicit understanding of mental states into more explicit knowledge that can be represented using words. A baby who feels anxious in unfamiliar situations without the caregiver becomes a preschooler who can label her feelings and talk about them for reassurance and understanding. Second, because the child can exchange information with others through language, direct representations of personal experience can be compared—even altered—with the secondary representations of another who has shared or witnessed that experience. At times, this contributes to clarifying the young child’s understanding of personal experiences, such as why another person acted or felt as she did. On occasion, discourse with another becomes a tutorial in divergent mental states when the child and the caregiver have different representations of the experience they shared. The awareness of divergent mental states is conceptually provocative for young children because it confronts the child with the realization that different people have different understandings of shared events, and motivates efforts to understand why. Third, and most significantly, language becomes the implicit conduit for the child’s appropriation of values, beliefs, and a sense of personhood that comes from being a cultural member as these are incorporated into the structure and content of language. This is what Nelson (1996) means by ‘‘the collaborative construction of the mediated mind’’ that occurs with language development, a view that is consistent with Vygotskian and neo-Vygotskian views (e.g., Rogoff, 1990). In all these ways, language acquisition provides the basis in early childhood for working models becoming more complex, explicit, consciously-accessible, and influenced by shared conversations with others. Studies of the conversations of parents with young children confirm some of these theoretical views. They indicate that mothers’ emotion language in shared conversation predicts preschoolers’ emotion understanding and affective competence (e.g., Brown & Dunn, 1996; Denham, Zoller, & Couchoud, 1994; Dunn, Brown, & Beardsall, 1991; Fivush, 1993). Mother– child conversation about behavioral standards also influences emergent


147

conscience and moral understanding (Dunn, Brown, & Maguire, 1995; Kochanska, 1995). Welch-Ross (1997) found that mother–child conversations about past events contributed to the theory of mind understanding in young children as mothers clarified the thoughts, feelings, and motives of people, including themselves (see also Dunn, Brown, & Beardsall, 1991). Moreover, as caregivers establish an interpretive framework for shared experiences that clarifies the significance of personal events to a young child, mother–child discourse has been linked to the organization of event representation (Nelson, 1989, 1993b) and early autobiographical reference (Hudson, 1990; Miller, Fung, & Mintz, 1996; Nelson, 1993b). In a sense, adults convey in their everyday conversations with young children varied lessons about people’s emotions and thoughts, the bases of cooperation and conflict, and the child herself that children are quick to appropriate into their own representations of shared events. As an example, consider the following conversation between a 21-monthold and his mother about an event earlier in the morning (from Dunn & Brown, 1991, p. 97): Child: Mother: Child:

Eat my Weetabix. Eat my Weetabix. Crying. Crying, weren’t you? We had quite a battle. ‘‘One more mouthful, Michael.’’ And what did you do? You spat it out! (pretends to cry)

This shared recollection is, in some respects, simply a recounting of the morning’s confrontation over breakfast cereal. But incorporated into the mother’s rendition is a sequential structure of events and a causal representation for the outcome (i.e., the child’s emotional reaction). It is easy to see how her description of the morning’s events would contribute to the organization and coherence of her young child’s subsequent memory for that event. In a sense, the mother has provided both a model for remembering and guidance concerning how and what to remember from personal experiences. On closer examination, moreover, other lessons for the child are apparent in the mother’s recounting. There are lessons about emotion and morality: crying is associated with misbehavior and lack of cooperation (rather than resulting from having to eat unpleasant breakfast cereal). There are lessons about the self: good boys cooperate and thus do not get upset, but Michael was uncooperative and that is why he cried. There are lessons about relationships: battles arise when young children are uncooperative (rather than battles occurring when parents insist on children eating Weetabix!). In addition to providing a memorable representation of the event, therefore, the parent has also interpreted it within a framework of assumptions, attributions, beliefs, and values. Although it is unclear how many of

148


these lessons are likely to be internalized by a 2-year-old from a single conversation, as conversations like these become part of the landscape of parent–child interaction in early childhood, the lessons about events, people, relationships, emotions, self, and other psychological realities embedded in these conversations are likely to become part of the young child’s psychological understanding. In our research, we have recorded many of these kinds of conversations that are representative of the kinds that are shared informally by young children with their caregivers. Such simple verbal exchanges are especially influential for young children whose direct representations of experience may be inchoate or incomplete, and for whom the articulate verbal structure of maternal discourse provides clarity and organization. This is especially true for understanding those aspects of everyday experience that are conceptually most elusive (and most interesting) for young children, such as people’s motives, feelings, and thoughts. This means that subsequent to such conversations, the explicit and implicit interpretations embedded in the caregiver’s verbal representation of events may supplant the child’s direct representations of personal experience, or at least significantly alter them. A young child like Michael would probably remember the morning’s confrontation much differently after this conversation with his mother compared with before. A young child’s earliest understandings of events and the psychological world are thus guided by how parents help to interpret, organize, and construct understanding through simple conversations. Developing working models are, in a sense, a co-construction of the parent and the young child. C. VARIATIONS IN PARENTAL CONVERSATIONAL STYLE

Parents vary, of course, in how they talk to their young children. Some— like Michael’s mother—provide a richly elaborative description of events that includes attributions of motives and emotion. Others provide a narrative structure that is more sparse or incomplete, or which incorporates a more impoverished variety of psychological attributions for the child. As Bowlby observed, moreover, some parents provide a narrative representation of events that strikingly conflicts with the child’s direct experience. Attachment theorists like Bretherton argued that these variations in parental conversational style are significant influences on young children’s developing working models, primarily owing to the open or restricted sharing of feelings and perspectives. But what is meant by ‘‘open’’ or fluent communication between parent and child, and what are its consequences for early psychological understanding and the working models that emerge in early childhood?


149

Cognitive-developmental scientists who study parent–child discourse about shared events have also found that parents vary in the ways they converse with offspring (see Fivush & Fromhoff, 1988; Hudson, 1990; Reese & Fivush, 1993). Some parents (labeled ‘‘pragmatic’’ or ‘‘repetitive’’) use short, directive conversations centered on specific events or questions that invite a simple ‘‘yes’’ or ‘‘no’’ response. Others (labeled ‘‘elaborative’’) provide rich background and contextual detail, and enlist questions that probe the child for further information, in a shared retelling of past events. In these studies, the offspring of elaborative mothers have a more complete and sophisticated representation of their past experiences that results from the direct impact of parent discourse style on memory for events. Children of elaborative mothers remember more also because of the child’s appropriation of the adult’s narrative approach. Young children learn to remember more, and remember differently, as a consequence of the mother’s narrative style (Fivush, 1993; McCabe & Peterson, 1991). These findings are consistent with those of Dunn, Denham, and their colleagues described earlier, who found that mothers who talk more about feelings and their causes have children who are subsequently more advanced in emotion understanding and theory of mind. In each case, the richness and elaboration of maternal discourse contributes to more complex, sophisticated representations of personal experiences by offspring. This may be part of what is meant by more open communication. But the distinction between elaborative and pragmatic conversational style may be only the beginning. Parents vary in other ways in their conversations with young offspring. What they choose to emphasize or recall in their conversations about past events and the details they omit may be important influences on how young children represent their experiences (especially because, as Levine, Stein, & Liwag, 1999, have noted, parents often remember different features of a shared experience than do young children). The emotions parents attribute to others and to the child, their moral evaluations of behavior, their attributions of culpability and causality for the consequences of actions, and the characteristics they attribute to the child may each be appropriated by the children who listen to them. Miller and her colleagues observed, for example, that Chinese and ChineseAmerican mothers emphasize moralistic themes and the shame inherent in misbehavior in their shared recounting of their 2-year-olds’ experiences, but Anglo-American mothers de-emphasize misbehavior and attribute bad conduct instead to the child’s spunk or mischievousness (Miller et al., 1990; Miller, Fung, & Mintz, 1996). It is easy to see how personal and cultural belief systems become internalized by young children through conversations like these. In this sense, language underlies the intergenerational transmission of parental beliefs and values.

150


Not only the adult’s beliefs and implicit causal theories but also the pragmatics and emotional tone of the caregiver’s discourse may be influential. Shared conversations about everyday events can occur in the relaxed context of smiles, hugs, and other expressions of warmth, or may instead be didactic exchanges with a more critical emotional tone, or there may be heated exchanges in the context of conflict. The impact of these conversations likely varies according to their emotional, pragmatic context. Moreover, the relational context of parent–child conversations is also likely to be important. Individual differences in parental narrative style may be related to broader variability in the caregiver’s sensitivity, defensiveness, emotional style, or to the adult’s personal representations of past events. These issues remain to be explored in future research. For the present, however, the realization that variations in parental narrative style, belief systems, communicative pragmatics, and other qualities may significantly influence the lessons appropriated by offspring in discourse contributes significantly to elucidating the nature of the open or restricted communication patterns distinguishing secure from insecure attachment relationships.

III. Developing a Working Model This review of the cognitive research relevant to the concept of internal working models yields several conclusions that have guided our program of research on the growth of early psychological understanding. First, early childhood may be a particularly important period for the development of working models and for the influence of attachment security on psychosocial development (the other important period, as noted earlier, is early adolescence). Because of the representational advances afforded by language and concurrent growth in event representation, theory of mind, autobiographical memory, emotion understanding, and other conceptual achievements, a secure or insecure attachment may be especially provocative in guiding children’s emergent understandings of people, relationships, and themselves at this time. Second, psychological understanding is shaped by the conceptual advances of early childhood and the conceptual catalysts of parent–child discourse. The latter helps to clarify the invisible qualities of psychological experience that young children seek to understand. How parents converse with children about shared experiences is as important as what they say, and this requires sensitive and detailed assessments of discourse style as it is relevant to specific features of emergent psychological understanding. Third, consistent with attachment theory, discourse is influential in a relational context. Close relationships are one of the primary motivators


151

for growth in psychological understanding, and relational experiences are one of the primary forums for learning about others and the self. The quality of relationships is thus important, and a parent’s elaborative narrative style may have a different influence in the context of a conflicted or insecure parent–child relationship than in a warm or secure relationship, for example, because relational quality is likely to influence the content and style of what the parent says and how the child responds to it. In our research program, therefore, we are exploring the development of psychological understanding in early childhood by enlisting formulations of attachment theory and the insights of cognitive researchers. We have not, however, sought to study working models directly. The conceptual uncertainties about how to define working models and their development, together with the empirical challenges of directly assessing working models through semiprojective stories, self-report, or other methods, each pose significant obstacles to research in this area (Oppenheim & Waters, 1995; Thompson & Raikes, in press). Consequently, we have focused on the growth of emotion and moral understanding in young children because these elements of psychological understanding are central to relationships, and are also crucial to the working models young children develop about themselves in relation to others. In a sense, therefore, our research is an effort to understand the development of conceptual processes related to the growth of working models—in other words, to develop a working model of the growth of working models. A. EMOTION UNDERSTANDING AND ATTACHMENT SECURITY

In an initial study, our goal was to understand whether understanding of emotions was one of the facets of psychological understanding influenced by a secure or insecure attachment relationship (Laible & Thompson, 1998). The more open or fluent communication styles that securely attached dyads are thought to share should permit greater discussion and mutual understanding of emotions, for example, and this may contribute to the greater social competence observed in securely attached children (see Thompson, 1999). However, this study was the first to explore a direct association between attachment and emotion understanding in young children as a prelude to examining the parent–child communication patterns that may contribute to emotion understanding. A sample of 40 preschoolers (mean age of 4 years) and their mothers participated. The security of attachment was assessed through the Attachment Q-Sort (AQS) (Waters & Deane, 1985), which mothers completed at home under the guidance of a trained research assistant. Although disagreement exists concerning whether trained observers or

152


mothers provide the most valid information via the AQS (mothers have more extensive and representative experience of the child but may be biased; observers can be more objective but have more limited information about the child), Teti and McGourty (1996) have shown that mothers can be valid informants when certain procedures are followed. These procedures include sending mothers the AQS items in advance and asking them to think about them in relation to the child, providing a standard set of instructions when mothers complete the sort, and ensuring that a well-trained research assistant is available to answer any questions. In this study and the others described in this chapter, we strictly followed the Teti and McGourty procedures. Two measures were used to assess preschoolers’ understanding of emotion. First, children participated in an affective perspective-taking task developed by Denham (1986), in which children were presented with 20 short stories enacted by a research assistant using words and puppets. The child’s task was to identify how the story character felt at the conclusion of each story by attaching a felt face to the puppet that the child had identified earlier as representing either sad, happy, angry, or fearful emotion. Children’s performance on this task was assessed as their accuracy in attributing the correct emotion to the story character, and the accuracy of their prior identification of the facial expressions of emotion. Second, children’s spontaneous understanding of the emotions of their peers in preschool or child-care centers was assessed, using a procedure adapted from Fabes et al. (1988). Research assistants watched for overt expressions of emotion from any child who could be observed by the target child. After noting the emotion and its cause, the assistant then approached the target child and asked ‘‘How does [other child] feel?’’ and ‘‘Why does [other child] feel that way?’’ and recorded the target child’s responses verbatim. We gathered five to ten such interviews for each child in the study over a 4-week period. Independent judges later evaluated the agreement between the reports of the target child and the research assistant concerning emotion and its causes, resulting in accuracy scores for each child for emotion understanding and identification of the causes of the peer’s emotions. The measures of emotion understanding derived from the affective perspectivetaking task and the on-site interviews were highly correlated (supporting the validity of each), so they were converted to standard scores and summed. Consistent with our expectations, there was a significant association between attachment security and emotion understanding in hierarchical regressions with age and gender controlled. Children with higher AQS scores also performed better on the assessments of emotion understanding, suggesting that a secure attachment may contribute to greater appreciation of emotional experiences in others. In separate hierarchical regressions, we also found that secure children earned higher scores for their understanding


153

of the causes of their peers’ emotions in the preschool and child-care on-site interviews. We took these analyses a step further, however, to explore whether the valence of the emotion was important. We were intrigued by a report by Belsky, Spritz, and Crnic (1996) that securely attached 3-year-olds were more likely to remember the positive events depicted in a puppet show they had previously observed, whereas insecurely attached children were more likely to remember negative events in the puppet show. The authors interpreted this as resulting from the heightened sensitivity of insecure children to negative emotion and its outcomes, perhaps because prior relational experiences involving ambivalent or negative feelings contributed to biases in affective-cognitive information-processing, consistent with the working models formulation. This suggested that memory for emotional events—and perhaps aspects of emotion understanding—varied by attachment security and the valence of the emotion. To explore this, children’s responses to the affective perspective-taking task and the on-site interviews were distinguished according to whether positive or negative emotions were concerned. Accuracy scores for each task were recalculated for each child and, because scores on the two tasks were significantly correlated for both positive valence and negative valence, scores were again converted to standard scores and then combined. In hierarchical regression analyses with age and gender controlled, there were no significant effects for attachment security for positive emotion understanding. Only for negative emotion understanding did children with higher AQS scores obtain significantly higher scores. These findings are somewhat different from those of Belsky, Spritz, and Crnic (1996), but are consistent with the expectations of attachment theory. Although insecurely attached children may have greater memory for negative events, enhanced understanding of those events requires opportunities to share, discuss, and learn about emotional experiences in the everyday contexts in which they are encountered, such as discipline encounters, sibling bickering, and even marital conflict. This is especially true for negative emotions because feelings like anger, fear, and sadness are motivationally and socially complex, far more than positive emotions which can be comprehended more easily and with much less personal threat. If securely attached children have greater opportunities to talk with their caregivers about events entailing negative emotion that they experience at home and elsewhere, it could significantly enhance their comprehension of the causes and outcomes of these feelings when compared with insecurely attached young children. This view is consistent with the findings of this study, and also the findings of Belsky, Spritz, and Crnic (1996).

154


B. PARENT–CHILD CONVERSATION, ATTACHMENT, AND EMOTION UNDERSTANDING

The problem with this conclusion, however, is that we did not directly observe mother–child conversations related to emotion. Our second study in this series, therefore, was designed to examine how mothers of young children converse with their offspring about emotional issues, and to see whether their conversational style was related to differences in children’s emotion understanding (Ontai & Thompson, 2002). In doing so, we hoped to better understand the nature of the more open, fluid communication that is believed to characterize the conversations of securely attached dyads. We were particularly interested in discovering whether parental discourse style and attachment security interact in predicting children’s emotion understanding. In other words, would more elaborative parent–child discourse be especially beneficial to the development of emotion understanding in the context of a secure parent–child relationship? Such a conclusion would suggest that how parents communicate about emotion and the broader relational context of their conversations are each important to developing psychological understanding. We studied 52 3-year-olds and their mothers. We studied a younger sample of children than in the previous study because of our interest in examining understanding of emotion during a period when developing representations of others’ mental states are still very basic, and when relational influences could shape children’s emerging conceptions of emotion. Mothers and children were observed in two situations that elicited shared conversations about emotion. In the first, mothers read and discussed five one-page stories with their children, each taken from a children’s storybook entitled Feelings (Brandenburg, 1984). Each story involved emotional themes, such as a boy who was happy about receiving a birthday party invitation, or a girl who was scared about her first day in a new class. Mothers were asked to ‘‘read them as you would normally read a story with your child.’’ The stories (and book) were chosen because they have minimal narration and rely primarily on pictures to convey the story, prompting considerable spontaneous mother–child discussion of the stories. Next, mothers were asked to talk to their children about an event that had occurred within the past week during which the child had displayed negative emotion. We focused on negative emotion in light of the findings of the previous study, and because negative emotions are conceptually more challenging for young children to understand, we focused on a recent personal event because this captures young children’s attention and interest. Mothers were asked to ‘‘recall the event as you normally would recall an event with your child’’ to elicit the child’s memory and how the child felt at


155

the time. Mothers were invited to take as long as necessary for each conversation, which was tape-recorded for later transcription. As in the previous study, Denham’s affective perspective-taking task was used as the measure of emotion understanding, and mothers completed the AQS during a second visit to assess attachment security. The transcribed conversations about emotion issues (storybook and past event) were coded in detail for characteristics of maternal speech specifically related to emotional states (see Ontai & Thompson, 2002). We focused on emotion state language following criteria identified by Dunn and colleagues to include feeling state terms but excluding references to other nonfeeling internal state terms, such as those referring to thinking and believing (see Dunn, Brown, & Beardsall, 1991). These emotion-related utterances were then coded for several characteristics believed to be related to children’s developing emotion understanding (Cervantes & Callanan, 1998; Dunn et al., 1991; Fivush & Fromhoff, 1988), including: . . . . .

. . .

The frequency of references to emotion by the mother (total emotion references) References to the causes of emotion References to the outcomes (or behavioral results) of emotion Definitions of emotion (e.g., ‘‘Furious is when you are really, really angry’’) Linking events: attempts to help the child comprehend the situation by linking the emotion to an event in the child’s life in which similar feelings were experienced (e.g., ‘‘He’s angry like when you were angry at sissy for hitting you this morning’’) Requests for information from the child related to emotion Directives about the proper behavior in response to emotional arousal (e.g., ‘‘You don’t hit when you get mad’’) Mothers’ confirmations, negations, or repetition of the child’s emotionrelated utterances.

We also conducted a more global rating of the mother’s overall elaborative narrative style using the criteria earlier described. Our first question was whether these discrete codings would reveal more general, coherent conversational styles in the mothers we studied. They did, as revealed in two analyses. First, a simple display of the correlational associations among these discourse codes revealed a strong network among the features of maternal speech that would be expected to be most provocative of children’s emotion understanding (see Figure 1). The frequency of references to the causes of emotion was significantly associated with references to the outcomes of emotions, mothers’ efforts to define

156


Fig. 1. Direct correlational associations between maternal discourse measures (from Ontai & Thompson, 2002). Variables are defined in the text.

or explain emotion, to link events in the child’s life to emotion concepts, and her requests for further information about emotion from the child. These behaviors were also significantly associated with the total frequency of emotion references in the mother’s discourse, suggesting that mothers were using their emotion language to enrich the child’s understanding rather than merely to confirm or repeat the child’s utterances, or to provide directives. Moreover, each of the features of maternal discourse identified in Figure 1 was significantly associated with global ratings of the mother’s elaborative conversational style. This was the first investigation to associate elaborative conversational style with more specific features of maternal discourse about emotion, and these findings suggest that elaborative discourse benefits socioemotional understanding as well as memory and event representation in young children. Mothers who talk more elaboratively during their shared conversations with young offspring embellish the child’s understanding of emotional themes, and in doing so they contribute to the development of emotion understanding. These conclusions were supported by a second analysis in which a subset of the discourse codes were submitted to a principal components analysis. Two factors emerged from this analysis. The primary factor had strong, positive loadings for maternal references to causes, linking events, requests for information, and the outcomes of emotion and, consistent with the foregoing findings, we labeled this factor elaborative style. The second factor had strong, primary loadings for maternal confirmations, directives, negation, and repetition, and we labeled this factor pragmatic style.


157

In light of these findings, however, we were surprised to discover that neither of the maternal conversational style factors nor attachment security had the predicted relations to children’s emotion understanding at age 3. Indeed, the only significant association was opposite to what we expected: in hierarchical regressions with age and gender controlled, the maternal pragmatic style factor was significantly and positively associated with emotion understanding. In attempting to make sense of these unexpected results—and looking at other patterns in the data—we began to conclude that by studying these relational influences at age 3, our investigation had started too early. In our effort to understand relational influences on emerging understanding of emotion, we began with children who were too young for the secondary representations of parent–child conversations, or the security of the parent–child relationship, to yet have a significant influence on emerging representations of people’s feelings. Indeed, theory of mind understanding is still very rudimentary at this age, and although the belief-desire psychology of 3-year-olds is likely to sensitize them to others’ feelings, it may not permit generous understanding of emotions in others, especially negative emotions. If it is true that we started our inquiry too early, this suggests that the timing of studies in this field is crucial to accurately portray social influences on developing psychological understanding in early childhood. To explore this possibility further, therefore, we recontacted as many of the children in this study as possible for another assessment of attachment and emotion understanding, using the same procedures as had been used earlier. By this time, the children were 5 years old. Consistent with our expectations (but contrary to the age 3 results), hierarchical regression analyses revealed that attachment security at age 5 significantly predicted emotion understanding at the same age, with age and gender (and attachment security at age 3) controlled. Children with more secure relationships at age 5 obtained higher scores on the emotion understanding measure at the same age. Moreover, follow-up analyses showed that securely attached 5-year-olds were especially strong in understanding negative emotions, compared with insecurely attached children. These findings replicated those of our earlier study. The maternal conversational style factors at age 3 did not directly predict children’s emotion understanding at age 5. But we also explored the interaction between attachment and conversational style at age 3 in predicting later emotion understanding. In other words, would the quality of maternal discourse in interaction with the security of attachment at age 3 significantly predict children’s competence in emotion knowledge at age 5? Although the interaction was not significant with respect to overall emotion understanding or for negative emotions, a significant interaction was

158


obtained for the measure of positive emotion understanding. More specifically, securely attached children whose mothers had used a more elaborative narrative style at age 3 were more advanced in their understanding of positive emotions at age 5 (Ontai & Thompson, 2002). These findings are consistent with those of our other studies. Laible (2002a) also coded maternal elaborative discourse during shared recall of the child’s past behavior, and obtained measures of attachment security (via maternal-report AQS) and children’s emotion understanding (via the Denham task). In this study, emotion understanding of these 4-year-olds was significantly predicted by secure attachment, and also by maternal elaborative speech and discussion of positive emotions during conversations about the child’s past behavior. Mothers who were more elaborative and talked more about positive emotions, in the context of secure relationships with their offspring, had children who showed greater emotion understanding, consistent with the foregoing results. In a separate investigation, Laible (2002b) also found that maternal elaborative style in parent–child conversations when children were 30 months old significantly predicted children’s emotion understanding 6 months later (attachment security was not measured in this study). Remarkably, Laible’s (2002b) findings were for 3-year-olds. Taken together, findings from these three studies suggest that the quality of parent–child conversational discourse and the overall quality of the parent–child relationship are each influential in the growth of emotion understanding. Elaborative discourse in the context of a secure relationship provides an especially rich forum for developing emotion understanding, particularly as children become capable of participating more extensively in the rich give-and-take of conversational exchanges with their caregivers, and as their representational capacities for understanding mental experience in others begins to mature more fully. Although the security of attachment alone is a significant predictor of children’s understanding of negative emotions—consistent with the benefits of more open communication with a trusted caregiver about feelings that may be disturbing or confusing—it is the security of attachment as it interacts with parental conversational style that predicts understanding of positive emotions. This may arise from the greater opportunities affording secure dyads to enjoy relaxed, positive times together that provide a forum for talking about feelings, their causes and consequences. C. CONSCIENCE AND EMOTION

As young children acquire greater understanding of people’s emotions, their motivation for cooperation and compliance expands. Concern about


159

the effects of their actions on others’ feelings begins to emerge as a motivator of moral conduct to supplement their concern over punishment and reward. Early childhood witnesses not only the growth of emotion understanding but also the origins of conscience, as young children begin to understand behavioral expectations, become capable of monitoring their actions according to such standards, and are motivated to cooperate and comply (Kochanska & Thompson, 1997). Early advances in conscience and moral understanding are fostered by the conversations that children share with their caregivers who, like Michael’s mother and the Weetabix breakfast confrontation, convey their expectations and the consequences of violating them in their everyday interactions with offspring. The importance of parent–child conversations to moral development is not a new idea. Developmental scientists like Hoffman (1983) have long emphasized that the quality of parental communication in the discipline encounter is a crucial catalyst for moral internalization, and that whether parents engage in a reasoned discussion of human needs or emotionally laden threats influences whether offspring will adopt parental values as their own. There are two ways that our interest in parent–child conversations and the development of conscience differs from earlier approaches, however. First, we became interested in studying the influence of parent–child discourse outside of the discipline encounter. The heightened emotional arousal that is characteristic of discipline encounters may make it difficult for children to fully comprehend and reflect on the parent’s message when parents and children confront each other, especially during the early years (Thompson, 1998). By contrast, as suggested by our prior findings, when parents and children converse later about past misbehavior—as they often do—the less confrontational atmosphere of shared recall may foster a fuller processing of the parent’s message about expectations and values, and contribute better to the child’s understanding. This might be especially true if the parent discusses standards and compliance using a richer, more elaborative conversational style that contributes to children’s deeper conceptualizations of cooperation and compliance. We became interested in exploring whether this was true, and whether other features of parental discourse would have also influenced the growth of conscience. Second, whereas the traditional research literature on parental discipline focuses on school-age children, research in our laboratory and by Kochanska (1995; Kochanska & Aksan, 1995; Kochanska & Thompson, 1997) caused us to emphasize early childhood as an important developmental period for conscience development. Kochanska’s research has focused, like ours, on how the general quality of parent–child relationships (conceptualized by her as variations in ‘‘shared positive affect’’ as well

160


as attachment security) and specific parental practices contribute to the internalization of values by preschoolers. In light of the findings of our previous studies suggesting that attachment security and shared discourse together influence emotion understanding, we sought to extend this work by exploring how relationship quality and the style of parental discourse together influence developing conscience in young children (Laible & Thompson, 2000). A sample of 42 4-year-olds and their mothers participated in a 45-minute laboratory session involving several activities. First, after 15 minutes of free play, mothers were asked to enlist their children in cleaning up the toys to permit observations of child compliance (following Kochanska & Aksan, 1995). Second, mothers conversed with their children as they shared recall of two events from the recent past—one in which the child misbehaved, and another in which the child behaved well—following procedures of our previous research, in order to observe variability in maternal conversational style. Next, the child was given a dull sorting task while mothers went into a nearby room to complete a questionnaire, after cautioning the child not to touch a nearby shelf of highly attractive toys. This resistance to temptation task, drawn from Kochanska and Aksan (1995), lasted 8 minutes, during which the child’s behavior was observed for cooperation with the mother’s request. Several measures were derived from these procedures. The central assessment of conscience development, called behavioral internalization, was a factor score derived from a principal components analysis of detailed ratings of the child’s behavior during the resistance to temptation task, and had high loadings for ratings of deviance (i.e., touching the forbidden toys) and on-task behavior (i.e., sorting activity). A second conscience measure, called committed compliance, was based on ratings of the child’s behavior during the clean-up task. Maternal discourse during the shared recall activity with the child was evaluated in several ways. In addition to a summary rating of maternal elaborative style, detailed codings of maternal references to feelings, rules (social, moral, and family), the consequences of behavior, and evaluative statements were made consistently with the previous studies. These yielded two general factors: maternal references to feelings and evaluatives and maternal references to rules and consequences. Similar factor scores were derived from coding children’s verbal behavior during these conversations. Finally, two relational measures were enlisted into this study. As in earlier research, mothers completed the AQS during an independent home visit to index attachment security. In addition, detailed time-sampled coding of maternal and child affect throughout the laboratory procedure yielded a summary measure of their shared positive affect.


161

Consistent with the expectations of attachment theory, the security of attachment was associated with several features of maternal discourse, including a marginally significant association with elaborative style and a significant relation to maternal references to feelings and evaluatives. Several other researchers have also found an association between attachment security and maternal elaborative style, suggesting that this association is robust (see Reese, 2002, for a review). In our study, securely attached children also referred more often to feelings than did insecurely attached children (and they were higher on most conscience measures than insecurely attached children). These findings broaden our understanding of the more ‘‘open’’ communication of secure parent–child dyads, and suggest that one of the characteristics of the relationships they share is more elaborative conversations about recent events that incorporate more expanded discussion of emotion. These features of parent–child discourse would be expected to contribute to working models of relationships, people, and self that are illuminated by the child’s enhanced psychological understanding. Consistent with the conclusions of the previous investigation, mothers with a more elaborative conversational style also made more frequent references to feelings and evaluatives, affirming the ways that an elaborative style enhances the lessons that children can learn about psychological processes in other people. Equally important, mothers who were high in elaborative style had offspring who were marginally significantly higher in the measure of behavioral internalization. This finding was replicated by Laible (2002b) in a longitudinal study using similar measures. Maternal elaborative style and references to emotion during shared conversation with offspring at 30 months were significant predictors of children’s behavioral internalization during the resistance to temptation task 6 months later. Both studies show that elaborative speech, and frequent references to people’s feelings, together contribute to conscience development in early childhood owing, most likely, to their richer portrayal of the needs and feelings of other people. Our broader goal was to understand how the overall quality of the parent–child relationship and parent discourse together predicted children’s conscience development. To examine this, regression models were constructed to predict behavioral internalization, with predictors including the security of attachment, shared positive affect, maternal references to feelings and evaluatives, and maternal references to rules and consequences, along with the child’s age and gender (because of the collinearity of elaborative style with other predictors, elaborative style was not included in these regressions). A significant overall predictive model resulted, with marginally significant contributions from shared positive

162


affect and maternal references to feelings and evaluatives. The marginal significance of these predictors may have resulted from the collinearity between attachment security and the maternal references to feelings factor, and when attachment security was eliminated in a second, reduced regression model, the resulting model accounted for as much variance as the full model and each predictor was significant (see Laible & Thompson, 2000). Moreover, in a subsequent longitudinal study (Laible & Thompson, 2002), maternal references to emotions in conflict episodes with offspring at 30 months was a significant predictor of children’s behavioral internalization in the resistance to temptation task at 36 months. Each of these studies suggests that children in relationships with high amounts of shared positivity, and whose mothers make more frequent references to people’s feelings, are higher in conscience. Both relationship quality and the style of discourse seem to be important, as we had expected and as previous findings suggested. It is striking that emotion-laden discourse was most predictive of early conscience in these studies. By contrast, the frequency of maternal references to rules and the consequences of violating them never predicted conscience in either study. Taken together, this suggests that one of the most important features of maternal discourse about misbehavior and good behavior is how it puts a human face on cooperation and compliance. A caregiver’s frequent references to people’s feelings, and moral statements framed in the form of evaluatives (e.g., ‘‘that was a nice thing to do’’), especially in the context of a generally elaborative discourse style, probably enables young children to connect the challenges of behavioral compliance with the needs and emotions of others. This suggests that a humanistic orientation toward morality may have surprisingly early roots in moral development, with origins in how caregivers discuss everyday issues of compliance with their young offspring. Such an orientation capitalizes on young children’s capacities for empathy and shared emotion in their interactions with others (Dunn, 1987). Consistent with this view is the second general conclusion of this research: a warm, secure relationship is an important context for the development of early conscience. Both attachment security and shared positive affect were important to conscience development. Shared positive affect was also the primary contributor to a significant regression model predicting children’s committed compliance in this study. The conclusion that relational quality is important to conscience is consistent with the views of developmental theorists, like Maccoby (1984) and Kochanska (1993), who have argued that a mutually responsive, harmonious parent–child relationship contributes to a child’s willingness to embrace parental messages and values. This is especially likely to be true early in life owing to young children’s emotional


163

dependency on their caregivers. Taken together, the findings from these studies suggest that when caregivers create relationships of security and shared positivity with their offspring and, in those relationships, communicate with rich elaboration the values and human needs that underlie behavioral standards, young children are most likely to adopt parental values as their own. D. FAMILY CONFLICT: LESSONS IN EMOTION, MORALITY, AND RELATIONSHIPS

Our studies have thus far focused on the features of parent–child discourse that are shared during relaxed conversation or reading a story together. Of course, relaxed conversation is not the only forum for developing conscience, emotion knowledge, and self-understanding. In addition, young children and their caregivers engage in heated exchanges during conflict over misbehavior, preferences, or intentions. Conflict not only is a ubiquitous feature of early parent–child interaction but may also be especially provocative of psychological understanding in young children. Nothing focuses a preschooler’s attention like the awareness that disagreement exists with another. This realization affords important opportunities for young children to explore the origins of conflicting mental states and the differing feelings, expectations, and intentions that cause them. This is especially so when parental discourse during conflict episodes offers children insight into the thoughts and emotions that relate to disagreement, another’s understanding of the child’s intentions and motives (which may or may not be accurate), and efforts that might occur to resolve conflict in psychologically sophisticated ways, such as through negotiation, justifications, compromise or bargaining (Thompson, 2000). Parent–child conflict has been rarely studied in the early years, however, and least of all by attachment researchers who instead focus on the warmth and sensitivity of the parent–child relationship. Thus our next study in this series was a prospective longitudinal investigation of mother–child conflict and its relations to the child’s later understanding of emotion, morality, and relationships (Laible, 2002b; Laible & Thompson, 2002). This study involved assessments at 30 and 36 months. At 30 months, parent–child conversations about previous episodes of the child’s misbehavior and good behavior were recorded and coded in a manner similar to the previous studies. In addition, however, the mother and child participated in a series of frustrating laboratory tasks that were designed to elicit conflict between them. Furthermore, one-and-a-half hours of unstructured parent–child interaction were audiorecorded at home during the period prior to and during dinner, which is a time of frequent mother–child

164


conflict. From the laboratory and home observations, each episode of mother–child conflict was identified, transcribed, and carefully coded according to several features of parent and child discourse. In addition to references to feelings, evaluatives, rules and their consequences (as in previous studies), we also coded conflict themes (e.g., over possessions, rules, aggression), conflict resolution strategy of either partner (e.g., mitigating conflict through compromise or bargaining; justification of one’s views through reasoning; aggravation and other uses of teasing or threats), and who took the initiative in resolving the conflict (adapted from Dunn & Munn, 1987; Herrera & Dunn, 1997). At 36 months, outcome measures of emotion understanding (using the Denham task), conscience (using the behavioral internalization factor from the resistance to temptation activity), and other measures were obtained. We have described earlier some of the findings of this investigation—for example, that conscience development at 36 months was significantly predicted by maternal references to feelings during conflict episodes 6 months earlier, consistent with the results of earlier studies. What this study added to the previous research, however, is how mothers’ strategies for resolving conflicts also contributed to children’s developing understanding of emotions and morality (Laible, 2002b; Laible & Thompson, 2002). In hierarchical regressions predicting emotion understanding, for example, the mother’s initiative in resolving conflict, her use of justifications to clarify and explain her expectations, and her low amounts of aggravation during conflict episodes at home were each significant predictors. Similarly, low maternal aggravation and high justification during lab conflict at 30 months significantly predicted conscience at 36 months. These characteristics of maternal conflict-relevant discourse are important, and maternal justification and low aggravation are together likely to provide young children with a richer understanding of the causes and consequences of interpersonal conflict without unduly arousing the child’s feelings of defensiveness or threat. Maternal justifications usually offer many lessons in psychological understanding as mothers constructively explain their expectations, convey their feelings, and clarify their perceptions of the situation. In short, adults who take the initiative in resolving conflict with their offspring and provide rational explanations for doing so appear to foster greater emotion understanding and conscience development in young children. These conclusions are consistent with the well-documented effects of inductive discipline practices on moral internalization with older children, and suggest that how parents convey their behavioral expectations in the context of family conflict—often before offspring have misbehaved—is comparably influential on conscience development at younger ages.


165

IV. Conclusion: The Impact of Relationships on Emotion, Morality—and the Self How do the working models that arise from attachment relationships develop? Although we have not sought to directly measure working models in these studies, these findings are nevertheless relevant to attachment theory because understanding how people feel, how to get along with others, and how to find approval in the eyes of people who matter are each relevant to the representations of self, others, and relationships that the working models construct embodies. If Bretherton (1991; Bretherton & Munholland, 1999) is correct that working models comprise a system of hierarchically organized representations that are tied to other belief systems, then a young mind that is striving to understand the psychological realities underlying relationships and social interaction enlists that knowledge into working models of the characteristics of attachment figures, the rules of intimacy, and the value of the self. What do the findings of these studies tell us, then, about the development of working models? First, they are shaped by the emotional tenor of parent– child relationships that are defined not only by security but also by shared positivity, conflict resolution style, and by other characteristics. Attachment theorists are correct that the security of attachment is a significant influence on developing psychological understanding in early childhood, especially as it provides a confident haven in which to share and understand distressing feelings or confusing experiences. But it is not the only definition of relational quality. The shared positive affect enjoyed by parents and children is another important influence on developing conscience, for example, although variations in this relational measure are not strongly correlated with attachment security (Kochanska & Thompson, 1997; Laible & Thompson, 2000). How parents and offspring mutually resolve conflict is yet another definer of their relationship quality that also affects psychological understanding, especially as young children learn about differing mental states and how they can be constructively harmonized. Close relationships are, in short, multidimensional. Early psychological understanding has origins in these broader emotional qualities of the parent–child relationship that teach young children what to expect from others. Second, working models arise not only from a young child’s direct experience in such relationships but also from the secondary representations that arise from parent–child conversation. Because of how language articulates the hidden psychological world underlying human behavior, there are many lessons embedded in conversations between parents and offspring about the day’s events, while reading a storybook, or when

166


arguing during dinnertime. A mother’s discussions of people’s feelings is, for example, a catalyst to conscience development. The most important feature of parental conversational style, however, appears to be the extent to which the adult richly elaborates the child’s understanding of everyday events through background details, provocative questions, and retelling. Our studies indicate that in addition to its benefits for event representation and memory, elaborative discourse also fosters enhanced understanding of emotion and morality through the richer portrayal it provides of the psychological characteristics of other people. These features of parental discourse help to clarify the nature of the more open, fluid communicative style shared by parents and children in secure attachment relationships, and suggests that parental sensitivity influences working models partly through a parent’s style of discussing everyday events with offspring. Third, and most speculatively, parent–child conversation teaches psychological understanding and also humanistic concern, and both are incorporated into developing working models. In two studies, maternal references to feelings predicted children’s conscience development, suggesting that frequent comments about people’s emotions may sensitize young children to the human needs underlying behavioral standards and the human costs of misbehavior. This is one of the reasons that inductive discipline is believed to be influential in moral internalization for older children, and the evidence of our research and other studies increasingly suggests that similar processes are relevant to conscience development in early childhood. If so, this suggests not only that traditional theories of moral development require updating, but also that the working models that arise from the experiences of early childhood may incorporate developing concern for the needs and interests of others that arise, in part, from processes of parent–child communication. These conclusions provide a basis for future studies in this area, because a number of questions remain to be explored. One concerns the origins of variations in parental conversational style, particularly elaborative discourse. Is this merely another manifestation of parental sensitivity? How is conversational style associated with variations in education, parenting involvement, beliefs about children, and other characteristics that may also contribute to parental practices that foster emotion understanding, conscience, and other aspects of psychological understanding in offspring? Better conceptualizing the network of parental influences that contribute to the development of working models, especially through conversational style, is one of the important tasks of attachment researchers and others who are interested in sociopersonality growth during the preschool years. Another question for further study concerns other elements of parent– child conversation that may also guide emerging psychological


167

understanding. As noted earlier, maternal representations of shared events incorporate motivational and emotional inferences, attributions of causality, moral evaluations of behavior, and references to character and personality (especially of the child) that are likely to influence how young children think about people and themselves. Moreover, typical conversations are accompanied by a rich nonverbal language that embeds words in facial expressions, vocal tone, affective gestures, and postural cues that add to (or change) the significance of what is said to young children. Our coding systems have only begun to capture the richness and variety of these multifaceted dimensions of parent–child conversation, but understanding this variability is essential to comprehending the effects of discourse on young children’s representations of experience. How is elaborative discourse related, for example, to nonverbal cues of emotion used by a parent during conversational exchanges? These and other questions also remain for future research. Moreover, conversations about past events, storybook reading, and conflict episodes encompass only a small portion of the circumstances in which conversation can influence how young children understand the psychological world. More difficult to study—but perhaps equally significant for developing representations—is what caregivers say during the course of shared activity, or in anticipation of events that will occur later in the day. These and other kinds of shared discourse may be especially important in shaping young children’s understanding of experience by providing concurrent or predictive conceptual structures within which to organize event representation. We are particularly interested in how these conceptual structures may attune young children’s expectations for the psychological responses of other people. Finally, the self. Developing conscience and emotion understanding contribute to a child’s conceptions of human relationships, psychological motivation, and the self in relation to others. The skills that young children exhibit in Denham’s puppet task are relevant to how children conceptualize their own feelings, and children’s capacities to resist the forbidden toys in the lab contribute to feelings of self-worth as compliance is exhibited at home also. Consistent with attachment theory, the relational experiences that we have studied—shared conversation about everyday events, conflict resolution at home, storybook reading, shared positive affect in casual interaction—provide catalysts for how young children begin to regard themselves as well as others. Even so, another new direction for our research program is to examine how these features of parent–child interaction contribute to young children’s earliest forms of psychological self-understanding. We are particularly interested in how mothers’ representations of the personalities of their offspring are reflected in their conversations

168


with children about shared experiences, and whether these attributions are also reflected in how children describe themselves to a research assistant. In doing so, we are beginning to probe the working models of the self and the intergenerational influences that shape developing self-awareness and psychological understanding.

REFERENCES Ainsworth, M. D. S. (1989). Attachments beyond infancy. American Psychologist, 44, 709–716. Bartsch, K., & Wellman, H. M. (1995). Children talk about the mind. New York: Oxford University Press. Belsky, J., & Cassidy, J. (1994). Attachment: Theory and evidence. In M. Rutter & D. Hay (Eds.), Development through life (pp. 373–402). Oxford: Blackwell. Belsky, J., Spritz, B., & Crnic, K. (1996). Infant attachment security and affective-cognitive information processing at age 3. Psychological Science, 7, 111–114. Bowlby, J. (1973). Attachment and loss. Vol. 2. Separation: Anxiety and anger. New York: Basic. Bowlby, J. (1980). Attachment and loss. Vol. 3. Loss: Sadness and depression. New York: Basic. Bowlby, J. (1988). A secure base: Parent–child attachment and healthy human development. New York: Basic. Brandenburg, A. (1984). Feelings. New York: Greenwillow Books. Bretherton, I. (1990). Open communication and internal working models: Their role in the development of attachment relationships. In R. A. Thompson (Ed.), Socioemotional development. Nebraska Symposium on Motivation, Vol. 36 (pp. 57–113). Lincoln: University of Nebraska Press. Bretherton, I. (1991). Pouring new wine into old bottles: The social self as internal working model. In M. R. Gunnar & L. A. Sroufe (Eds.), Self processes and development. Minnesota Symposia on Child Psychology, Vol. 23 (pp. 1–41). Hillsdale, NJ: Erlbaum. Bretherton, I., & Munholland, K. A. (1999). Internal working models in attachment relationships: A construct revisited. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment (pp. 89–111). New York: Guilford. Brown, J. R., & Dunn, J. (1996). Continuities in emotional understanding from 3 to 6. Child Development, 67, 789–802. Cervantes, C. A., & Callanan, M. A. (1998). Labels and explanations in mother–child emotion talk: Age and gender differences. Developmental Psychology, 34, 88–98. Craik, K. (1943). The nature of explanation. Cambridge, England: Cambridge University Press. Denham, S. (1986). Social cognition, prosocial behavior, and emotion in preschoolers: Contextual validation. Child Development, 57, 194–201. Denham, S. A., Zoller, D., & Couchoud, E. A. (1994). Socialization of preschoolers’ emotion understanding. Developmental Psychology, 30, 928–936. Dunn, J. (1987). The beginnings of moral understanding: Development in the second year. In J. Kagan & S. Lamb (Eds.), The emergence of morality in young children (pp. 91–112). Chicago: University of Chicago Press. Dunn, J., & Brown, J. (1991). Relationships, talk about feelings, and the development of affect regulation in early childhood. In J. Garber & K. A. Dodge (Eds.), The development of emotion regulation and dysregulation (pp. 89–108). Cambridge: Cambridge University Press. Dunn, J., Brown, J., & Beardsall, L. (1991). Family talk about feeling states and children’s later understanding of others’ emotions. Developmental Psychology, 27, 448–455.


169

Dunn, J., Brown, J., & Maguire, M. (1995). The development of children’s moral sensibility: Individual differences and emotional understanding. Developmental Psychology, 31, 649–659. Dunn, J., Brown, J., Slomkowski, C., Tesla, C., & Youngblade, L. (1991). Young children’s understanding of other people’s feelings and beliefs: Individual differences and their antecedents. Child Development, 62, 1352–1366. Dunn, J., & Munn, P. (1987). Development of justification in disputes with mother and sibling. Developmental Psychology, 23, 791–798. Fabes, R. A., Eisenberg, N., McCormick, S. E., & Wilson, M. S. (1988). Young children’s appraisals of others’ spontaneous emotional reactions. Developmental Psychology, 27, 858–866. Fivush, R. (1993). Emotional content of parent–child conversations about the past. In C. A. Nelson (Ed.), Memory and affect in development. Minnesota Symposia on Child Psychology, Vol. 23 (pp. 39–77). Hillsdale, NJ: Erlbaum. Fivush, R., & Fromhoff, F. A. (1988). Style and structure in mother–child conversations about the past. Discourse Processes, 8, 177–204. Flavell, J. H., & Miller, P. H. (1998). Social cognition. In W. Damon (Ed.), Handbook of child psychology (5th Ed.), Vol. 2. Cognition, perception, and language (D. Kuhn & R. S. Sigler, Vol. Eds.) (pp. 851–898). New York: Wiley. Harris, P. L. (1996). Between Strange Situation and false beliefs: Working models and theories of mind. In W. Koops, J. Hoeksma, & D. van den Boom (Eds.), Early mother– child interaction and attachment: Old and new approaches (pp. 187–199). Amsterdam: Elsevier. Herrera, C., & Dunn, J. (1997). Early experiences with family conflict: Implications for arguments with a close friend. Developmental Psychology, 33, 869–881. Hinde, R. A. (1988). Continuities and discontinuities: Conceptual issues and methodological considerations. In M. Rutter (Ed.), Studies of psychosocial risk: The power of longitudinal data (pp. 367–383). Cambridge: Cambridge University Press. Hoffman, M. L. (1983). Affective and cognitive processes in moral internalization. In E. T. Higgins, D. Ruble, & W. Hartup (Eds.), Social cognition and social development (pp. 236–274). New York: Cambridge University Press. Hudson, J. A. (1990). The emergence of autobiographical memory in mother–child conversation. In R. Fivush & J. A. Hudson (Eds.), Knowing and remembering in young children (pp. 166–196). Cambridge: Cambridge University Press. Hudson, J. A. (1993). Understanding events: The development of script knowledge. In M. Bennett (Ed.), The child as psychologist: An introduction to the development of social cognition (pp. 142–167). New York: Harvester Wheatsheaf. Johnson-Laird, P. N. (1983). Mental models. Cambridge: Harvard University Press. Kobak, R. R., & Cole, H. I. (1994). Attachment and meta-monitoring: Implications for adolescent autonomy and psychopathology. In D. Cicchetti & S. L. Toth (Eds.), Disorders and dysfunctions of the self: Rochester Symposium on Developmental Psychopathology, Vol. 5 (pp. 267–297). Rochester, NY: University of Rochester Press. Kochanska, G. (1993). Toward a synthesis of parental socialization and child temperament in early development of conscience. Child Development, 62, 325–347. Kochanska, G. (1995). Children’s temperament, mothers’ discipline, and security of attachment: Multiple pathways to emerging internalization. Child Development, 66, 597–615. Kochanska, G., & Aksan, N. (1995). Mother–child mutually positive affect, the quality of child compliance to requests and prohibitions, and maternal control as correlates of early internalization. Child Development, 66, 236–254.

170


Kochanska, G., & Thompson, R. A. (1997). The emergence and development of conscience in toddlerhood and early childhood. In J. E. Grusec & L. Kuczynski (Eds.), Parenting and children’s internalization of values (pp. 53–77). New York: Wiley. Laible, D. J. (2002a). Mother–child discourse in two contexts: Factors that predict differences in the quality and emotional content of the discourse and the consequences of those differences for socioemotional development. Unpublished manuscript, Southern Methodist University. Laible, D. J. (2002b). Mother–child discourse surrounding a child’s past behavior at 30 months: Links to emotional understanding and early conscience development at 36 months. Unpublished manuscript, Southern Methodist University. Laible, D. J., & Thompson, R. A. (1998). Attachment and emotional understanding in preschool children. Developmental Psychology, 34, 1038–1045. Laible, D. J., & Thompson, R. A. (2000). Mother–child discourse, attachment security, shared positive affect, and early conscience development. Child Development, 71, 1424–1440. Laible, D. J., & Thompson, R. A. (2002). Mother–child conflict in the toddler years: Lessons in emotion, morality, and relationships. Child Development, 73, 1187–1203. Lamb, M. E. (1981). Developing trust and perceived effectance in infancy. In L. Lipsitt (Ed.), Advances in infancy research, Vol. 1 (pp. 101–127). New York: Ablex. Levine, L. J., Stein, N. L., & Liwag, M. D. (1999). Remembering children’s emotions: Sources of concordant and discordant accounts between parents and children. Developmental Psychology, 35, 790–801. Maccoby, E. E. (1984). Socialization and developmental change. Child Development, 55, 317–328. Marvin, R. S., & Britner, P. A. (1999). Normative development: The ontogeny of attachment. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment (pp. 44–67). New York: Guilford. McCabe, A., & Peterson, C. (1991). Getting the story: A longitudinal study of parental styles in eliciting narratives and developing narrative skill. In A. McCabe & C. Peterson (Eds.), Developing narrative structure (pp. 217–253). Hillsdale, NJ: Erlbaum. Meins, E., Fernyhough, C., Russell, J., & Clark-Carter, D. (1998). Security of attachment as a predictor of symbolic and mentalising abilities: A longitudinal study. Social Development, 7, 1–24. Miller, P. J. (1994). Narrative practices: Their role in socialization and self-construction. In U. Neisser & R. Fivush (Eds.), The remembering self: Construction and accuracy in the self-narrative (pp. 158–179). Cambridge: Cambridge University Press. Miller, P. J., Fung, H., & Mintz, J. (1996). Self-construction through narrative practices: A Chinese and American comparison of early socialization. Ethos, 24, 237–280. Miller, P. J., Potts, R., Fung, H., Hoogstra, L., & Mintz, J. (1990). Narrative practices and the social construction of self in childhood. American Ethnologist, 17, 292–311. Nelson, K. (Ed.) (1989). Narratives from the crib. Cambridge: Harvard University Press. Nelson, K. (1993a). The psychological and social origins of autobiographical memory. Psychological Science, 4, 7–14. Nelson, K. (1993b). Events, narratives, memory: What develops? In C. A. Nelson (Ed.), Memory and affect in development. Minnesota Symposia in Child Psychology, Vol. 26 (pp. 1–24). Hillsdale, NJ: Erlbaum. Nelson, K. (1996). Language in cognitive development: The emergence of the mediated mind. New York: Cambridge University Press. Nelson, K., & Gruendel, J. (1981). Generalized event representations: Basic building blocks of cognitive development. In M. Lamb & A. Brown (Eds.), Advances in developmental psychology (pp. 131–158). Hillsdale, NJ: Erlbaum. Ontai, L. L., & Thompson, R. A. (2002). Patterns of attachment and maternal discourse effects on children’s emotion understanding from 3 to 5 years of age. Social Development, 12, 657–675.


171

Oppenheim, D., & Waters, H. A. (1995). Narrative processes and attachment representations: Issues of development and assessment. In E. Waters, B. E. Vaughn, G. Posada, & K. Kondo-Ikemura (Eds.), Caregiving, cultural, and cognitive perspectives on secure-base behavior and working models: New growing points in attachment theory and research. Monographs of the Society for Research in Child Development, 60 (Serial no. 244), 197–215. Reese, E. (2002). Social factors in the development of autobiographical memory: The state of the art. Social Development, 11, 124–142. Reese, E., & Fivush, R. (1993). Parental styles of talking about the past. Developmental Psychology, 29, 596–606. Rogoff, B. (1990). Apprenticeship in thinking. New York: Oxford University Press. Rutter, M., & O’Connor, T. G. (1999). Implications of attachment theory for child care policies. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment (pp. 823–844). New York: Guilford. Schank, R. C. (1982). Dynamic memory: A theory of reminding and learning in computers and people. Cambridge: Cambridge University Press. Stern, D. N. (1989). The representation of relational patterns: Developmental considerations. In A. J. Sameroff & R. N. Emde (Eds.), Relationship disturbances in early childhood (pp. 52–69). New York: Basic. Teti, D., & McGourty, S. (1996). Using mothers vs. trained observers in assessing children’s secure base behavior: Theoretical and methodological considerations. Child Development, 67, 597–605. Thompson, R. A. (1998). Early sociopersonality development. In W. Damon (Ed.), Handbook of child psychology (5th Ed.), Vol. 3. Social, emotional, and personality development (N. Eisenberg, Vol. Ed.) (pp. 25–104). New York: Wiley. Thompson, R. A. (1999). Early attachment and later development. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment (pp. 265–286). New York: Guilford. Thompson, R. A. (2000). The legacy of early attachments. Child Development, 71, 145–152. Thompson, R. A., & Raikes, H. A. (in press). Toward the next quarter-century: Conceptual and methodological challenges for attachment theory. Development and Psychopathology. Tomasello, M. (1999). The cultural origins of human cognition. Cambridge: Harvard University Press. van Ijzendoorn, M. H. (1995). The association between adult attachment representations and infant attachment, parental responsiveness, and clinical status: A meta-analysis of the predictive validity of the Adult Attachment Interview. Psychological Bulletin, 113, 404–410. Verschueren, K., & Marcoen, A. (1999). Representation of self and socioemotional competence in kindergartners: Differential and combined effects of attachment to mother and to father. Child Development, 70, 183–201. Waters, E., & Deane, K. E. (1985). Defining and assessing individual differences in attachment relationships: Q-methodology and the organization of behavior in infancy and early childhood. In I. Bretherton & E. Waters (Eds.), Growing points of attachment theory and research. Monographs of the Society for Research in Child Development, 50 (1/2, Serial no. 209), 41–65. Welch-Ross, M. K. (1995). An integrative model of the development of autobiographical memory. Developmental Review, 15, 338–365. Welch-Ross, M. K. (1997). Mother–child participation in conversation about the past: Relationships to preschoolers’ theory of mind. Developmental Psychology, 33, 618–629. Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind development: The truth about false belief. Child Development, 72, 655–684.


WORKING MEMORY IN INFANCY

Kevin A. Pelphrey and J. Steven Reznick DEPARTMENT OF PSYCHIATRY UNIVERSITY OF NORTH CAROLINA SCHOOL OF MEDICINE CHAPEL HILL, NORTH CAROLINA 27599

I. INTRODUCTION II. TOWARD A DEFINITION OF WM IN INFANTS A. WORKING MEMORY AS SHORT-TERM STORAGE B. PUTTING THE WORK INTO WM C. WORKING MEMORY IN COGNITIVE NEUROSCIENCE AND NEUROBIOLOGY III. RESEARCH ON WM IN INFANTS A. ESTABLISHING CRITERIA FOR EVALUATING RESEARCH ON WM B. IDENTIFYING RELEVANT PARADIGMS C. RESEARCH FINDINGS D. CONCLUSIONS REGARDING WM IN INFANCY IV. FRONTIERS OF RESEARCH ON WM IN INFANTS A. METHODOLOGICAL CHALLENGES B. DEVELOPMENTAL COGNITIVE NEUROSCIENCE AND WM IN INFANTS C. WHAT DRIVES THE DEVELOPMENT OF WM? D. HOW DOES THE EMERGENCE OF WM CHANGE THE INFANT? REFERENCES

I. Introduction Memory refers to the representation, storage, control, maintenance, retrieval, and use of various kinds of information including, but not limited to, aspects of experience. This sweeping definition subsumes a vast array of behavioral and psychological phenomena. Folk psychology often treats memory as a unitary phenomenon, as in expressions like ‘‘she has a good memory’’ or ‘‘he has lost his memory,’’ but a century of psychological research underscores the utility of identifying more or less specific and


173


174

Kevin A. Pelphrey and J. Steven Reznick

separable aspects of memory (e.g., Nadel, 1992; Nelson, 1995; Sherry & Schacter, 1987; Tulving, 1985). Because researchers define specific types of memory across a subset of possible dimensions, definitions of types of memory are inherently complex, and no single procedure or paradigm can adequately capture a unique aspect of memory. For example, if we focus on memory for the specific details of an event (often called episodic memory), we must define this aspect of memory across different durations of storage or retrieval contexts and tap it using procedures that reflect this range of instantiations. If we focus on memory across a particular duration of storage (e.g., short-term memory), we must talk about storage of many kinds of information using various storage strategies, and we must be prepared to find different operating characteristics for each type of information and for each retrieval context. The problematic nature of simplistic, single-operation definitions of types of memory seems obvious, but a leitmotif of this chapter will be that it is a mistake to ignore memory’s complexity. In this review, we focus on an aspect or type of memory that has been labeled working memory (WM) with attention to topics such as theoretical and operational definitions of WM, the relation of WM to other cognitive abilities, and the development of WM during human infancy. To establish a general context for this discussion, we note that WM refers to the processes that allow an organism to store temporarily and manipulate a limited amount of information pursuant to the performance of various cognitive and behavioral tasks. WM allows one to remember a phone number long enough to dial it, keep track of the locations that have been searched unsuccessfully for misplaced car keys, or construct a strategy while listening to the details of a problem. WM is tapped in tasks such as delayed-response (DR) in which a response is designated as correct but the participant must delay responding for some specified interval and must avoid repeating a previous response when a different response is required; in the n-back task in which the participant must maintain information presented across trials and use this information to guide behavior on each new trial (e.g., remembering the final word on each of n previous sentences); or in a foraging task in which the participant must remember which locations in an array have already been searched. We will devote considerable attention to issues related to the measurement of WM in infants because we see methodology as a major obstacle impeding progress on this topic. As an initial disclaimer, we note that WM is an extraordinarily popular topic within psychology, with well over 2,000 published articles listing WM as a primary descriptor. Obviously, we can review but a small fraction of this burgeoning literature; our goal here is to focus on several vitally important questions regarding the development of WM during human

Working Memory in Infancy

175

infancy. How should we define the construct of WM in infants? On what basis could we claim that an infant is or is not using WM? How does WM change during the first years of life? Definitive answers to these questions will not be forthcoming at present, but if we are successful, we will have helped establish the legitimacy and underscore the importance of ongoing efforts to answer them. A second leitmotif of this chapter is that although infants cannot perform many of the tasks that have become synonymous with WM in adults, it would be a serious mistake to simply assume that they do not have WM. This would be a flawed assumption on theoretical and empirical grounds. Theoretically, explaining many common infant behaviors (e.g., goaloriented behavior) without positing some underlying WM would be difficult. Empirically, many studies suggest an emergence of WM at around 6 months of age with considerable improvement thereafter. We begin our review by defining WM in a way that highlights its relevance for informing our understanding of infant cognitive development, keeping in mind the need for a complex definition that will be relevant across the lifespan.

II. Toward a Definition of WM in Infants A. WORKING MEMORY AS SHORT-TERM STORAGE

1. Early Ideas about WM The concept of short-term information storage has appeared in various forms in the psychological literature of the past 100 years. James (1890/ 1950) distinguished between primary memory that holds information in conscious awareness and secondary memory or ‘‘memory proper’’ that brings to the mind a former state of knowledge or an event and gives rise to the conscious awareness of having thought about or experienced the knowledge or event previously. According to James, primary memory serves as a limited-capacity store for information that may or may not become secondary memory. In James’ words, ‘‘for a state of mind to survive in memory it must have endured [in primary memory] for a certain length of time’’ (Chapter XVI, p. 643). James distinguished between substantive states of mind, which become part of secondary memory, and transitive states, which are attended to briefly but are lost before long-term storage. A second function of primary memory is to provide the subjective experience of awareness, at least for those concepts that attain the status of a substantive state. James saw primary memory as serving the important function of providing momentary eddies of reflective processing in our stream of

176


thought, thereby allowing us to hold an idea or concept in awareness for post-perceptual processing. These two functions of James’ primary memory system capture most of the essential aspects of the later proposed WM system. Soon after James introduced the primary–secondary memory distinction, Baldwin (1895) applied the concept of a limited capacity short-term storage system as an aspect of development. Baldwin used the term ‘‘attention span’’ to refer to the maximum number of ideas or concepts that a child could consider at any one moment. He asserted that attention span increased over ontogeny, allowing more extensive processing of information, and that this process was mediated by cortical development. Baldwin’s perspective established the notion of behavioral development resulting from changes in a finite-capacity storage system linked to brain development. Moreover, Baldwin’s notion of development in attention span presaged modern proposals by neo-Piagetian theorists who identify short-term storage as a critical resource in the development of thinking and reasoning (e.g., Case, 1978, 1991; Pascual-Leone, 1970). 2. Short-term Storage in Information-processing Models Short-term storage was a central feature in early information processing models. For example, Broadbent (1958, 1963) proposed a dichotomization of memory into separate long- and short-term components. Similarly, in developing their multistore model of information processing, Atkinson and Shiffrin (1968) identified a sensory register, a short-term store, and a long-term store. Much contemporary theory and research can be traced to a tripartite model of WM, proposed by Baddeley and Hitch (1974), that included a central executive and two specialized storage systems: the visuo-spatial sketchpad and the phonological loop. The central executive is responsible for reasoning, decision making, and for coordinating the activities of the slave systems. The phonological loop allows for the temporary retention and manipulation of verbal material; the visuo-spatial sketchpad temporarily stores and manipulates visual and spatial information. The Baddeley and Hitch model has been generative and has thus evoked various criticisms. For example, Nairne (2002) questioned the assumption that activation in WM decays spontaneously in the absence of rehearsal, and posited instead that short-term retention is cue driven (like long-term memory) and not well characterized by processes like rehearsal or decay. In a retrospective, Baddeley (2001) noted that although the model’s phonological and spatial subsystems have been the subject of extensive investigation and their existence generally supported (with some important caveats and debates), the central executive remains the least understood


177

component of the model. Furthermore, he proposed that the central executive is analogous to a limited capacity attentional system, aided by a newly postulated fourth system, the episodic buffer. The Baddeley model’s capacity for change in response to challenge reflects either increasing veridicality or reduced parsimony, but in either case, the Baddeley model continues to serve as the primary organizational framework for describing WM. B. PUTTING THE WORK INTO WM

The term ‘‘working memory’’ was introduced by Miller, Galanter, and Pribram (1960) in their claim that ‘‘[the] most forward portion of the primate frontal lobe appears to us to serve as a ‘working memory’ where Plans can be retained temporarily when they are being formed, or transformed, or executed’’ (Miller, Galanter, & Pribram, 1960, p. 207). Thus, WM implies both temporary storage and manipulation, with both aspects contributing to the performance of cognitive tasks such as language comprehension, reasoning, decision-making, problem-solving, and mental arithmetic (Baddeley, 1983, 1992; Baddeley & Hitch, 1974). Moscovitch’s (1994) expression ‘‘working-with-memory’’ emphasizes this facet of WM, as does Fuster’s (1995, 1997) contention that WM is part ‘‘memory for the future.’’ Theorists have conceptualized the working aspect of WM in different ways; we review several popular perspectives as a basis for formulating what working might mean in the context of WM in infants. 1. Multi-component Models of WM Short-term memory (STM) differs from WM in that STM entails temporary storage and retrieval of a limited amount of information, whereas WM incorporates the manipulation and/or transformation of the information held in memory (Baddeley & Hitch, 1974; Case, 1978; Cowan, 1999; Engle, Kane, & Tuholski, 1999; Kail & Hall, 2001; Schneider & Pressley, 1997). Kail and Hall (2001) examined the STM versus WM issue in a large sample of 7- to 13-year-old children. They administered various STM and WM tasks. Exploratory and confirmatory factor analyses revealed both similarities (e.g., performance on both was associated with age-related increases in processing speed) and differences between the two constructs (e.g., WM performance was related to a reading task, whereas STM performance was not). In the Baddeley and Hitch model (1974; Baddeley, 1986), the central executive actively regulates the distribution of limited attentional resources and coordinates information storage and processing within the limited capacity memory stores. This emphasis on active processing as opposed to

178


passive storage is also salient for Pennington (1994, 1997) who argues that the central executive represents WM per se, and that the phonological and visuo-spatial slave systems represent distinct and relatively passive shortterm stores. In this model, because WM is a limited capacity system, inhibition of information irrelevant to the task is a consequence of WM’s operation; WM both maintains some information on-line and inhibits other information. 2. Working Memory, Controlled Attention, and Inhibition Cowan’s embedded-processes model defines WM as the set of mnemonic processes that maintain information in the highly accessible or activated state needed to carry out cognitive tasks (Cowan, 1988, 1995, 1999). Cowan argues against WM as a possibly reified theoretical entity but uses the term instead as a practical, task-oriented label for a set of processes with particular functional characteristics. From this perspective, WM involves an embedded assortment of information relevant for completion of a particular cognitive task. This information includes items in the focus of attention, items activated but not currently in the focus of attention, and aspects of long-term memory that are not currently in WM but that have pertinent retrieval cues. The information in WM is arranged in a nested organization: attention is focused on a subset of items in WM, which is in turn a subset of items in long-term memory. Cowan emphasizes linkages between WM and attention, noting that the subset of active items that are receiving attention will be those most completely active in WM. Attention thus plays a dual role of enhancing the processing of task-relevant data and suppressing or inhibiting processing of task-irrelevant or interfering information (i.e., attention can be directed toward or away from items in WM). From this perspective, inhibitory processes are an emergent part of the WM system. Cowan proposes that attention may be automatically recruited by stimuli with special significance to an individual (e.g., faces, which are salient even to young infants), thus requiring less central executive engagement for directing the focus of attention, and thereby freeing up WM capacity. In discussing potential processing limits of WM, Cowan suggested that different aspects of WM are limited in different ways. Two are particularly relevant: the number of items to which one can attend in a given period is limited, and the activation of any one item in WM is time limited. Thus, holding all else constant, a finite number of items in WM will remain active for a limited duration before they are lost from memory. Cowan’s embedded-processes model of WM, particularly by emphasizing two kinds of capacity limitations, has much to offer. As we discuss in more detail later, the literature on WM in infants has relied primarily upon tasks


179

that explore duration (or durability) of storage. Relatively little research has been conducted to explore WM capacity defined as the amount of information held for a given period per se. Thus, we know very little about the earliest developments in this aspect of the WM system. Cowan’s theory explicitly incorporates the idea of inhibition as an element of WM by including the control of attentional focus as an aspect of WM. This suggests the need to look for tasks that require both memory for relevant information and inhibition of irrelevant information in operationalizing WM in infants. In a model that emphasizes the role of controlled attention in WM, Engle, Kane, and Tuholski define WM as ‘‘a system consisting of those long-term memory traces active above threshold, the procedures and skills necessary to achieve and maintain that activation, and limited-capacity controlled attention’’ (1999, p. 102). As defined by this model, controlled attention is similar to the Baddeley and Hitch (1974) central executive. Because individual differences in the capacity for controlled processing are prominent in this perspective, it may be especially valuable for understanding individual differences in infant development and variation in developmental pathways. In particular, it may help to explain infant performance on tasks, which will be discussed in sections to follow, that require both holding relevant information on-line and inhibition of prepotent responses. Engle and colleagues suggest that limited-capacity controlled processing is required for maintaining temporary goals in the face of distraction and interference and for blocking, gating, and/or suppressing distracting events. They further propose that capacity limitations (and developments in WM capacity) reflect the status of one aspect of the system, controlled attention in the presence of interference or distraction (including suppression of competing representations) and not limits on the storage of information in memory per se. In summary, the ‘‘working’’ in WM, can assume a variety of meanings including inhibition of prepotent responses, control of attention, use of information to accomplish a cognitive task, and/or simultaneous storage and manipulation of information. Each of these meanings is relevant to our understanding of WM in infants.

C. WORKING MEMORY IN COGNITIVE NEUROSCIENCE AND NEUROBIOLOGY

WM was originally defined in the context of brain systems supporting task-relevant storage and processing functions (Miller, Galanter, &

180


Pribram, 1960). Since then, scientists have generated an extensive body of research linking WM processes to the functioning of a distributed network of brain regions, with areas in prefrontal cortex playing a central role. This research has involved lesion studies and single-cell recordings in neurobiology and functional neuroimaging techniques in cognitive neuroscience. We review a limited subset of this research here, with emphasis on the findings most relevant to understanding WM in infancy. 1. Animal Studies Neurophysiological studies in nonhuman primates have identified a network of brain regions involved in WM tasks such as DR, with preeminent involvement of dorsolateral prefrontal cortex (see Fuster, 1997; GoldmanRakic, 1987, for reviews). In one of the first demonstrations of the sensitivity of the DR task to prefrontal cortex functioning, Jacobsen (1936) trained monkeys to perform a two-location DR task, prefrontal cortex was then surgically lesioned, and the animals were again tested on the DR task. The resulting selective deficit on tests of spatial DR was interpreted as arising from the loss of immediate memory. This work influenced subsequent efforts to develop an animal model of WM and suggested the utility of the DR procedure as a marker task for the functioning of prefrontal cortex (Fuster, 1980; Fuster & Alexander, 1971; Fuster & Bauer, 1974; Goldman & Rosvold, 1970; Gross & Weiskrantz, 1964; Meyer, Harlow, & Settlage, 1951; Pribram, 1950; Rosvold et al., 1961). Fuster and Alexander (1971), and Kubota and Niki (1971) recorded electrical activity from individual neurons in monkeys as they performed variations of the DR task. These recordings revealed neurons in prefrontal cortex that became active during the delay, suggesting that these neurons were involved in WM storage. Goldman-Rakic and colleagues strengthened and extended this conclusion by recording neuronal activity in prefrontal cortex in nonhuman primates during performance of an oculomotor analog of the DR task (e.g., Funahashi, Bruce, & Goldman-Rakic, 1989, 1990; Goldman-Rakic, 1987). Monkeys saw a stimulus in a peripheral location and then fixated on a central stimulus point and maintained this fixation during a brief delay. After the delay, a gaze shift to the place where the target had appeared produced food reinforcement. Recordings of single unit activity revealed sharp increases in neuronal activity that began immediately after the target stimulus appeared and ended immediately before execution of a response. In addition to their sensitivity to the delay interval, individual neurons were differentially responsive to the spatial location of the target stimulus. These findings suggest that prefrontal cortex neurons play a role both in holding information on-line and in coding the spatial location of a stimulus.


181

2. Functional Neuroimaging of Human WM Functional neuroimaging research with human adults supports the conclusion that prefrontal cortex is involved in WM (e.g., Belger et al., 1998; Cohen et al., 1997; Courtney et al., 1997; D’Esposito et al., 1995; Jha & McCarthy, 2000; McCarthy et al., 1994; Petrides et al., 1993; Smith & Jonides, 1998; Smith, 2000). For example, in an early functional neuroimaging study of WM, McCarthy et al. (1994) had participants perform a modified DR task in which they judged whether the location occupied by a current stimulus had been occupied during a recent sequence of stimuli. Activation in Brodmann’s Area 46 of prefrontal cortex (the middle frontal gyrus and some adjacent cortex) was higher during the DR task than during a control task that required participants to detect the color or presence of a stimulus. Similarly, Petrides et al. (1993) used imaging techniques to measure activation within prefrontal cortex during a WM task that was analogous to a foraging task. Adults saw a series of unique designs and were required to construct a sequence that did not use the same design twice. Thus, participants were required to maintain an ongoing record of the stimuli that they had already selected and to monitor this record in preparation for the next response. Activation within mid-dorsolateral prefrontal cortex was higher during performance of the DR task than during control activities. Other neuroimaging work highlights the importance of considering both storage and processing functions of WM. For example, D’Esposito et al. (1995) monitored brain areas while participants performed a semanticjudgment task, a spatial-rotation task, or both. Prefrontal cortex was involved when participants performed both tasks simultaneously but not when either task was performed alone. Similarly, work by Cohen et al. (1997) and Jonides et al. (1997) revealed activation in prefrontal cortex coupled with the mnemonic demands of the task. During n-back tasks in which participants see a sequence of letters and report the immediate target or a letter that was one, two, or three back (Cohen et al., 1997) or judge whether the immediate letter matches a letter that was one, two, or three back (Jonides et al., 1997), prefrontal cortex activation increases as a function of memory demand. Whereas the bulk of the neuroimaging investigations of WM have involved adults, some investigators have extended this research to children. In a functional imaging study, 9- to 11-year-olds pressed a button when the letter in the current trial matched the letter seen two trials before (Casey et al., 1995). Levels of activation in child prefrontal cortex were similar to the patterns found in adult participants. Nelson et al. (2000) observed adultlike activations in areas of dorsolateral prefrontal cortex in 8- to 11-yearolds while they attempted to remember the location of a dot presented on

182


previous trials. Klingberg, Forssberg, and Westerberg (2002) measured brain activity in 9- to 18-year-olds as they performed a spatial WM task that required memory for the location of an item presented in a grid. Levels of activity in frontal and parietal cortical areas positively correlated with levels of performance on the WM task. Thus, children and adolescents show a pattern of neuroanatomical localized activation in a spatial WM task similar to that of adults and nonhuman primates. Additional developmental neuroimaging work will be important for our understanding of how developments in prefrontal cortex and other brain areas relate to WM development. We discuss possible directions later.

III. Research on WM in Infants A. ESTABLISHING CRITERIA FOR EVALUATING RESEARCH ON WM

Based upon the focal concept underlying various definitions of WM and relevant memory models, we define WM as a finite-capacity system that enables storage and processing of task-relevant information on-line in the service of cognition and/or action. To summarize the literature on WM in infants with this definition in mind, we have adopted several criteria for considering a particular study to be an examination of WM in infants. We based these, in part, on Diamond’s (1990a) discussion of the characteristics of memory dependent upon dorsolateral prefrontal cortex. Some criteria are obvious in the sense of being entailed by all descriptions of WM (e.g., all WM procedures require short-term information storage followed by a response based on the stored information). Other criteria are more subtle but relevant to the goal of establishing general standards for defining WM in infants. For example, a convincing demonstration of WM in infants should demonstrate that storage is temporary (i.e., information is stored, used, and then discarded). This criterion is problematic in some sense because many theories regard WM as a processing step that can lead to the establishment of long-term memories (or, indeed, simply as an aspect of long-term memory per se). Thus, some of the information processed in WM could be retained for extremely long periods. To separate processing that is long-term from processing that is inherently short-term, we focus on procedures in which WM is used for the storage of information that has only temporary usefulness and thus would likely be discarded quite rapidly. For example, in a memory procedure in which one of two alternatives is the correct response, any isolated trial might reveal either long-term or temporary storage. Only over a set of trials in which the


183

correct location varies randomly (with replacement) does this sort of procedure tap temporary storage. A second subtlety is the sense in which a WM procedure (for infants) evokes the working aspect of memory. As noted previously, ‘‘working’’ in this context can have various meanings including inhibition of prepotent responses, control of attention, use of information to accomplish a cognitive task, and/or simultaneous storage and manipulation of information. In the procedures used to tap WM in adults, working is usually instantiated by instructing the participant to perform some relevant ongoing task while attempting to retain information (e.g., perform arithmetic operations while trying to remember specific numbers; read and comprehend sentences while attempting to remember specific words). Unfortunately, these stratagems do not work with nonverbal infants. From some perspectives, the lack of explicit ongoing distraction might undermine confidence in any procedure that claims to measure WM for infants. As stated previously, we do not accept this argument for several reasons. More important, it is possible to constrain infant tasks such that the infant must inhibit prepotent responses and control his or her attention. For example, by providing multiple trials on which the correct response varies randomly with replacement across the various alternatives, infants must inhibit prepotent responses. Furthermore, by distracting infants during the delay interval such that they must not only process some additional stimulus but also redirect their attention away from the correct response, we become more confident that these operationalizations of WM evoke an appropriate amount of work. A third subtlety is the context in which an infant responds. Some demonstrations of WM require recognition rather than recall. That is, the participant sees a stimulus (e.g., a particular arrangement of dots in a matrix) and then sees this stimulus or an alternative stimulus after a delay and must indicate whether the stimulus is familiar (e.g., Casey et al., 1995). An animal model highlights the importance of the distinction between recognition and recall in WM. Petrides (1991) designed two conditions to contrast recall of a previous choice with recognition of a previously presented stimulus. In the recall condition, a monkey saw three objects and selected one of the objects. On a subsequent trial, the monkey saw two of the three original objects and received reinforcement for repeating the original selection. Because both objects were familiar, recognition per se would not allow a correct choice. The second condition had the same initial trial as the first condition, but the test trial pitted the original selection and a novel object. In this condition, the monkey could merely recognize a stimulus as being familiar. Prefrontal cortex lesions, which affect WM, altered performance in the first condition but not in the second.

184


We suggest that the recognition memory context is tapping WM in a weak sense: if the participant can recognize the previously presented stimulus then it is possible that the stimulus was in WM. However, there is a stronger sense of WM that entails the active processing of information (e.g., through rehearsal). In this sense, WM refers to the availability of information that is dynamic, ongoing, and present. It is true that WM might be used to make a recognition judgment, but this sort of recognition might also occur after a delay of hours, if not years, which certainly belies any sense of ongoing availability of information. Thus, we are less interested in demonstrations of WM that could be accomplished by recognition of a previously presented stimulus alone. Related, the to-be-remembered information on a particular trial should be presented once, briefly. Infants might have WM for a stimulus that has been presented many times or for a stimulus presented in the context of a sequence or pattern, but an interpretation that this reflects WM could be gratuitous because other explanations are feasible. In the case of multiple presentations of a stimulus, over-learning facilitates the transfer of information to long-term memory and this could serve as the basis for a response on any particular trial. In the case of a pattern or sequence, infants might detect or infer a rule (e.g., reach to the previously rewarded location or reach to the object that does not seem familiar) and use this as a basis for their response rather than trial-specific information held in WM. Although either a learning context or a rule-use context could reflect the engagement of WM, we find neither context particularly compelling (for a similar argument, see McDonough, 1999). B. IDENTIFYING RELEVANT PARADIGMS

There are many procedures for assessing WM in older children and adults; researchers can modify some to assess WM in infants. Other procedures that tap various aspects of infant cognitive development can be adapted to assess WM. We describe the paradigms that we believe to be most relevant for describing the development of WM in infants. 1. DR and the A-not-B Paradigm The DR task is the most prominent measure of WM. Hunter (1913) trained rats, dogs, raccoons, and young children to approach the location marked by a light to escape from confinement or to obtain a food reward. The correct location was randomized across trials, thus requiring the participant to learn to approach the location marked by the light regardless of its position. During subsequent test trials, the participant was restrained in the start area while the light was turned on briefly and extinguished before


185

a response could be emitted, thus forcing the participant to respond on the basis of memory for the correct location. Hunter assessed WM durability by incrementing the length of confinement in the starting box (i.e., the retention interval) until the participant could no longer remember which of the locations had been active. Hunter’s results revealed substantial differences among the species he tested. Rats and dogs could delay their responses for a maximum of 10 sec and 5 min, respectively, but only if they maintained a direct orientation toward the active location. In contrast, raccoons and 2.5- to 8-year-old children (M ¼ 5.7 years) were able to delay their reactions for delays ranging from 3 to 25 sec and from 50 sec to 25 min, respectively, without maintaining their orientation toward the correct location. In a subsequent longitudinal study of a single 13-month-old infant, Hunter (1917) refined his procedure, eliminating the need for preliminary training by using a desired object (e.g., an attractive toy) as both the discriminative stimulus and the reward. Hunter hid a small toy in one of three identical locations and allowed the infant to search for the toy following a brief delay. We discuss this study’s findings in detail later. A specific variation of the DR task is familiar to most developmental psychologists through Piaget’s (1954/1937) observation that infants between 6 and 9 months of age who retrieve an object at an initial location (A), will often perseverate in responding to the A location when the object is subsequently hidden at another location (B), hence the A-not-B or perseverative error. Piaget interpreted incorrect search in this context as a sign of the infant’s incomplete understanding of object permanence. An extensive literature has developed around investigations of infant performance within the A-not-B protocol (e.g., Harris, 1987; Marcovitch & Zelazo, 1999; Wellman, Cross, & Bartsch, 1987). The canonical A-not-B procedure closely resembles the first few trials of the standard DR protocol, and the developmental trajectories of infant performance on these two tasks are identical (Diamond, 1985; Diamond & Doar, 1989). However, although the A-not-B protocol certainly engages WM, it is not a straightforward measure of WM ability. The primary focus within the A-not-B protocol is on the infant’s behavior during the trial on which the location of the object is shifted from a series of hidings at location A to a new hiding at location B, not on the infant’s general ability to remember the location of a hidden object. Consequently, the A-not-B protocol does not counterbalance the hiding locations across a sequence of trials. The A-not-B paradigm is more analogous to a training and transfer task in which the infant is trained to reach to the A location and is subsequently transferred to the B location. Infant performance on

186


isolated A trials is extremely robust (e.g., McDonough, 1999, reported successful visual search by 7.5-month-olds after a 90-sec delay). Indeed, one might expect that following repeated trials at the A location, the infant will search at the A location after a delay of hours, if not days! In contrast, in a randomized protocol, the change of locations over trials forces the infant to maintain (we presume in WM) a representation of the current location of a stimulus over a short delay and then, after each stimulus presentation, to update WM with new information reflecting a new hiding location. The percentage of correct responses across a set of trials with hidings in randomized locations can be an index of the infant’s WM ability (e.g., Diamond & Doar, 1989; Schwartz & Reznick, 1999).1 2. Memory for Previous Responses Several procedures tap WM by requiring the participant to keep track of previous responses in the context of potential responses. For example, the radial arm maze, which was developed to study spatial memory in the rat, typically consists of eight or more elevated arms, each about a meter long, radiating from a central starting platform (Olton, 1978; Olton, Collison, & Werz, 1977; Olton & Samuelson, 1976). The maze is in a lighted laboratory room with furniture, windows, and doors to provide the rat with cues regarding its location on the maze. The experimenter places a rat on the central platform, and the rat searchers for food concealed at the end of each arm. Revisits to previously selected arms are errors because the food at these sites had been eaten. Researchers operationalize WM capacity as the number of correct choices before the rat revisited a previously depleted location. In trials involving a maze with eight arms, rats made few errors and chose an average of more than seven different arms within the first eight choices. Even when tested in a more challenging 17-arm maze, the rats chose, on average, more than 14 different arms in the first 17 choices. Radial maze tasks have been adapted for use with human adults (e.g., Aadland, Beatty, & Maki, 1985; Glassman, Lenick, & Haegerich, 1998; O’Connor & Glassman, 1993) and children (e.g., Aadland, Beatty, & Maki, 1985; Foreman, Arber, & Savage, 1984; Overman et al., 1996). Human adults and laboratory rats perform quite well and at comparable levels,

1

Performance on the B trial of the A-not-B paradigm meets several criteria for WM including memory for a specific location and the need to inhibit a prepotent response to A. However, to obtain an effective index of WM, researchers would need to combine data from various ‘B’ trials across locations.


187

whereas children below the age of approximately 5 years perform poorly and even older children do not reach adult proficiency before the age of 12 years.2 Naturalistic foraging contexts are another example of tasks that tap WM by requiring memory for previous responses in the context of potential responses. For example, Menzel (1973) showed juvenile chimpanzees 18 randomly hidden foods in an outdoor pen and then observed their search for the food. Menzel noted that chimpanzees rarely rechecked a location where the food had already been eaten. These results have been replicated with several other nonhuman primate species including yellow-nosed monkeys (MacDonald & Wilkie, 1990), gorillas (MacDonald, 1994), and marmosets (MacDonald, Pang, & Gibeault, 1994). The naturalistic foraging task is particularly interesting in that it taps an ecologically valid use of WM: we assume that primates have a long history of using WM to plan the most efficient route for food gathering and to avoid revisiting depleted locations. In addition, researchers have incorporated this general approach to WM in a wide range of related procedures. For example, Petrides (1995) used a visual self-ordered task in which animals see a set of stimuli on successive trials and must select a different stimulus on each trial until all stimuli have been selected. Stimulus location changes across trials so that selection is based on WM for stimulus identity rather than location. Similarly, using various way-finding tasks, Cornell, Heth, and colleagues investigated how children and adults learn to navigate within natural environments and remember relevant environmental features (Cornell, Heth, & Alberts, 1994; Cornell, Heth, & Rowat, 1992). For example, Cornell, Heth, and Alberts (1994) guided 8-, 12-, and 25-year-olds during a first walk across a university campus. During the return trips, the experimenter sometimes led participants away from the original route. They stopped at intersections on and off the original route to obtain various estimates of place recognition accuracy. Eight-year-olds were less accurate than either 12- or 25-year-olds, with the latter two age groups equivalent in accuracy. 2 The radial maze task has not been linked to PFC activity. Rather, it is most frequently discussed in relation to hippocampal function in animals (Olton & Papas, 1979; Olton, Walker, & Gage, 1978). Although the task probably recruits WM, its neurobiological substrates may be different from the classic DR tasks. Diamond (1990b) suggests various explanations including the fact that lower animals have less developed cortices. In addition, animals such as rats and birds typically studied in radial maze tasks are generally moving around in space (a hippocampal function), whereas, during DR, the animal or child usually remains in a constant location. Finally, the radial maze task is more complex than the DR task in that the participant must keep track of searched and unsearched locations.

188


3. Naturalistic Hide-and-Seek Tasks DeLoache and her colleagues have investigated the spatial memory capabilities of 1.5- to 2.5-year-old children in the context of a naturalistic hide-and-seek game. The child watches as a researcher hides a toy in a room (e.g., under a sofa cushion, behind a chair, and so on). After a delay, the child is encouraged to search for the toy. This task is effective for children as young as 18 months, and performance is usually quite accurate even after protracted delays (e.g., DeLoache & Brown, 1979, 1983, 1984; DeLoache, Cassidy, & Brown, 1985). However, aspects of the paradigm make a WM interpretation problematic. For example, the task does not involve an intervening distraction (e.g., the child is not taken from the room during the delay). As a result, the child uses various memory strategies, and thus the task may tap strategic memory rather than WM per se. However, we note that with an appropriate distraction between hiding and finding, and a series of trials requiring the updating of WM, the naturalistic hide-and-seek task could be an effective procedure for assessing toddler WM. Indeed, if one were to hide toys in several locations and then allow the child to search, this task would be very similar to the radial maze tasks described earlier. Newcombe and her colleagues have created a protocol for examining the development of young children’s abilities to represent the spatial location of a hidden object (Huttenlocher, Newcombe, & Sandberg, 1994; Newcombe et al., 1998). The child searches for a familiar toy hidden in a long rectangular sandbox.3 During the hiding phase of the procedure, the child sits in his or her parent’s lap approximately 18 inches from the sandbox and equidistant from the two ends of the box. As the child looks on, the experimenter buries a single toy in the sand. After hiding the toy, the experimenter distracts the child’s attention from the surface of the sand and imposes a brief delay (1–2 sec). Following the short delay, the child is encouraged to search for the toy, and the experimenter records the distance of the child’s response from the buried toy’s actual position. A testing session typically consists of nine trials with different toys buried, one at a time, in randomly varying locations. Thirteen- to 16-month-olds perform quite well on this task (Huttenlocher, Newcombe, & Sandberg, 1994) with 16-month-olds selecting the correct location most of the time. Even the youngest infant’s errors typically involved reaches that miss the correct location by only a few inches, indicating a level of performance comparable to that of older children and adults (e.g., Newcombe et al., 1998). Moreover, even at brief delays (2–3 sec), there were no age differences in performance, which suggests that infants are performing at optimal levels (Huttenlocher, Newcombe, & Sandberg, 1994). 3

This paradigm has also been used to examine spatial working memory in rodents (e.g., Cheng, 1986; Cheng & Gallistel, 1984).


189

Although the focus of investigations involving this paradigm has not been on WM per se, researchers could reengineer this task in ways that would strengthen its effectiveness as an assessment of WM in infants. For example, the demands on WM durability could be increased by lengthening the delay interval between hiding and search and perhaps by decreasing the delay between successive trials. As with the hide-and-seek tasks described earlier, WM capacity could also be assessed by hiding several toys in various locations on a given trial. Following a delay with distraction, the child could dig for one of the toys. After successful retrieval, the distraction and delay could be repeated, and the child would again be allowed to search among the original locations. This process would continue until the child found all the toys or made an error involving revisiting a previous location or digging in a location far from a buried toy. In this manner, the child would be required to maintain an ongoing record of the hiding locations and the locations already searched, and then monitor this record in preparation for the next excavation. 4. Spatial-orientation Tasks Researchers have examined the development of spatial localization and orientation abilities during the first year of human infancy. For example, Acredolo (1978) trained infants to expect the appearance of a female experimenter to the left or right of center after the onset of a buzzer. After this expectation was established, the infant’s view of the room was reversed through a change in position. Direction of infant gaze from the new position in anticipation of the event can be interpreted as an indication of whether the infant has learned the objective location in the room where the event occurred and has kept track of his or her movements in relation to it. Longitudinal data from 24 infants tested at 6, 11, and 16 months indicated a shift from egocentric responses (repeating the previous head turn) at 6 and 11 months to allocentric responses (tracking current position relative to the correct window) at 16 months. Similarly, McKenzie and colleagues trained infants to expect an experimenter to appear at some point along a circular opaque enclosure and then rotated the infant 90 and observed the infant’s expectations (McKenzie, Day, & Ihsen, 1984). The results indicated that infants as young as 6-months-old could successfully anticipate the experimenter’s reappearance following brief delays. The focus of these investigations has not been on WM per se. However, these tasks could be reconfigured in ways that assess WM in infancy. For example, lengthening the delay interval between hiding and search would impose demands on WM durability. Furthermore, after training the infant to anticipate the return of the experimenter at the location where she last

190


appeared, a series of trials could be administered in which the location of hiding is randomized. To respond correctly, the infant would be required to update WM with information concerning the new hiding location at the beginning of each trial and maintain a representation of this location over the subsequent delay. 5. Possible and Impossible Events Baillargeon and her colleagues have developed a procedure that seems similar to DR. Infants see a toy hidden at one of two locations and then the experimenter retrieves the toy from where it was hidden (the ‘‘possible’’ condition) or where it was not hidden (the ‘‘impossible’’ condition). In several investigations, 8-month-olds look longer at the impossible event after delays of up to 70 sec (e.g., Baillargeon, DeVos, & Graber, 1989; Baillargeon & Graber, 1988). Similarly, Wilcox, Nadel, and Rosser (1996) reported reliable fixation toward the unexpected test event for 2½-montholds tested after a 5-sec delay. Ahmed and Ruffman (1998) replicated the general finding with 8- to 12-month-olds at a 15-sec delay and instituted several control conditions to strengthen the claim that infants use memory to perform in this nonsearch task. Newcomb, Huttenlocher, and Learmonth (1999) used the sandbox task (described earlier) in a possible/impossible event format. Five-month-olds saw a toy hidden in the sand and retrieved for four trials and then saw a test condition in which the toy is retrieved from a novel location (or other variations). These young infants showed heightened visual interest in the impossible event after delays of up to 50 sec. These findings could suggest a robust WM. However, this interpretation is problematic. Infant responses on a single, isolated trial could reflect longterm associative memory accrued from experience gained on the familiarization trial rather than WM per se. The extremely long delay reported by Baillargeon and colleagues (1989; 1988) and by Newcombe, Huttenlocher, and Learmonth (1999), and the results from extremely young infants reported by Wilcox et al. (1996) reduce our confidence that WM is being assessed here. Differential responding to possible and impossible conditions would be more impressive if, as in the DR paradigm, it occurred across a sequence of trials in which the hiding location was randomized. More important, the possible and impossible event procedure seems likely to be tapping a form of recognition memory whereas the DR task clearly involves a form of recall ability. That is, successful performance in this paradigm requires the infant to simply recognize that something is wrong. Indeed, Ahmed and Ruffman (1998) compared performance on a typical search task (albeit using the A-not-B format) and a nonsearch task configured comparably, and found strong correlations with age in the former but not


191

the latter. This inconsistency undermines confidence in the impossible event procedure as a measure of WM. 6. Short-term Visual Recognition Some paradigms assess WM in a retrieval context that is oriented toward recognition rather than recall. For example, Jha and McCarthy (2000) presented adults with an array consisting of one to three faces. After a 24-sec delay period, the participants saw a one-face probe and were required to indicate whether the probe stimulus was part of the memory array. Rose, Feldman, and Jankowski (2001) presented 5-, 7-, and 12-month-olds with up to four items in succession and tested for recognition in a series of trials that paired each familiar item with a novel one. Infants performed less accurately when presented with more items, and older infants responded correctly to more items. WM implies active, short-term representation. It is possible that recognition procedures could tap WM. That is, adults might maintain an active representation of the faces in an array, and infants might maintain an active representation of familiar items. We are skeptical regarding this inference because recognition could also occur based on recognition memory per se rather than WM. That is, rather than maintain an active short-term representation, adults or infants could respond differentially to a previously presented stimulus merely on the basis of a general sense of familiarity. This suspicion is compelling because comparable performance would be expected across intervals much too long to be considered WM. For example, recognition of a passerby as a former classmate from the second grade or an item as a long-lost relic from one’s childhood implies some sort of enduring representation but surely does not imply that a representation survived in WM for the intervening decades. Comparable performance in a short-term visual recognition task is an ambiguous assessment of WM at best. C. RESEARCH FINDINGS

Table I summarizes a collection of investigations that meet criteria for assessments of WM in infants and thus, are relevant for describing the development of WM in infants. The table is organized by assessment technique and publication date. Notably, each of these studies uses behavioral paradigms adopted from the comparative literature, including DR, oculomotor DR, and naturalistic ‘‘foraging’’ tasks. Of these paradigms, researchers have used DR most extensively. In some sense, the hegemony of the DR procedure reflects historical influences (it emerged first) and the wide adoption of the DR procedure among neuroscientists interested in the

192

TABLE I Brief Summaries of Selected Studies on Infant Working Memory Relevant Parameters Participants and Assessment Schedule

Response Modality

Length of Delay(s)

Number of Locations

Number of Hidden Items

Type of Stimulus

Longitudinal; 1 infant; aged 13–16 months Cross-sectional; 100 infants; aged 12 months Cross-sectional; 40 infants; aged 14 & 16 months Cross-sectional; infants aged 9.5–10 months Cross-sectional; 74 infants; aged 8, 12, & 16 months Compilation of 4 longitudinal & cross-sectional investigations at multiple testing sites; 152 infants; aged 13– 36 months Sequential; 12 infants tested longitudinally from 6–12 months; 36 infants tested cross-sectionally at 8, 10, & 12 months

Reaching

0–160 sec

3

1

Toy

Reaching

10–160 sec

3

1

Toy

Reaching

5, 10, or 15 sec

3

1

Toy

Reaching

0 or 5 sec

2

1

Toy

Reaching

250 msec, 3, 6, or 9 sec 1, 5, or 10 sec

3

1

2, 4, 6, or 8

1

Light Show Food

2

1

Toy

Delayed Response Hunter (1917) Allen (1931) Webb et al. (1972) Harris (1973) Brody (1981) Kagan (1981)

Diamond & Doar (1989)

Reaching

Reaching

0, 3, 5, 8, 10, or 12 sec


Assessment Technique and Citation

Hofstadter & Reznick (1996) Reznick et al. (1997)

Reznick et al. (1998) Schwartz & Reznick (1999) Pelphrey et al. (2003)

Reaching & gaze

3 sec

2

1

Toy

Reaching

1, 5, or 10 sec

2, 4, 6, or 8

1

Toy

Reaching

3 & 10–20 sec

3

1

Toy

Gaze

10 & 20 sec

2

1

Person

Gaze and Reach

2, 6, or 10 sec

2, 3, or 4

1

Person or toy

Cross-sectional; 28 infants; aged 6 months

Gaze

0–5 sec

2

1

Abstract design

Cross-sectional; 60 infants; aged 12 & 18 months Cross-sectional; 60 infants; aged 26–32 months

Reaching

3 sec

4

4

Food

Reaching

5 sec

3

3

Stickers

Oculomotor Delayed Response Gilmore & Johnson (1995) Naturalistic Foraging Tasks Reznick & Fieselman (1999) Pelphrey et al. (2002)


Cross-sectional; 132 infants; aged 5, 7, 9, & 11 months 252 pairs of twins; tested longitudinally at 14, 20, & 24 months Cross-sectional; 64 infants; aged 7 & 9 months Cross-sectional; 92 infants; aged 9 months Cross-sectional; 66 infants; aged 6, 8, 10, & 12 months

193

194


neural aspects of WM. Unfortunately, the popularity of DR may also reflect the field’s complacency regarding measurement. That is, using an existing measure is easier than developing a new one, and accruing data that is comparable across laboratories has many advantages (e.g., the possibility of performing meta-analysis). However, as we noted in the introduction, any single measurement procedure can provide only a limited perspective on a complex phenomenon such as WM. Another limitation is that all the studies in Table I focus on spatial WM. Distinctions can be made between spatial and object WM, and between spatial and auditory WM, and these distinctions will be discussed later. Very little work has been done to address object WM or on auditory WM in infants, which obviously has strong implications for our knowledge of WM development—we know nothing about these two important kinds of WM in infants. In the paragraphs that follow, we flesh out the entries in Table I. A number of DR studies involving infants and young children appeared early in the 20th century, and research involving this paradigm has continued to the present. Hunter (1917) inaugurated the examination of infant memory for locations with the DR paradigm with an intensive longitudinal observation of his 13-month-old daughter—Thayer—as she performed a three-location DR task. Hunter used three identical boxes with hinged covers placed in a triangular pattern on a card table. As Thayer looked on, a Hunter hid the toy in one of the three open locations. After concealing the toy, Hunter disrupted his daughter’s spatial orientation by turning her around or covering her eyes and imposed a delay that ranged from 3 to 35 sec. After the delay, Thayer was repositioned and allowed to search for the hidden toy. During each testing session, several trials were administered sequentially, with the correct location and length of delay counterbalanced across trials. WM durability, defined by Thayer’s ability to tolerate increasing delays before selecting the location of the hidden toy, increased between 13 and 16 months. At 13 months, Thayer correctly located the toy on approximately 72% of trials involving 8- and 12-sec delays. At 16 months, she responded correctly on approximately 82% of trials involving the same delays and correctly on roughly 70% of trials with delays of 15–20 sec. Allen (1931) expanded Hunter’s seminal contribution using a threelocation DR task to investigate memory durability in 100 12-month-olds. While the infant watched, a toy was tapped on one of the three open boxes, placed inside, and all three boxes were closed simultaneously. After concealment of the toy, the parent turned the infant to disrupt his or her spatial orientation for a 10- to 165-sec delay. Accuracy declined as delay increased, with mean accuracy of 64, 61, 48 and 28% after delays of 10, 20, 30, and 45 sec, respectively.


195

Diamond and Doar (1989) conducted an investigation of WM durability in infants, employing a sequential assessment design. Durability was operationalized as the ability to tolerate increasing delays while selecting the correct location across 16 trials of a two-location direct DR task. One cohort of 12 infants was tested longitudinally every two weeks from 6 to 12 months, and a second cohort of 36 infants (12 each at 8, 10, and 12 months) was tested cross-sectionally. The longitudinal sample revealed a striking developmental progression in performance. Ability to span a delay increased linearly at a rate of 2 sec per month from approximately 2 sec at 7.5 months to 12 sec at 12 months. The shape of the developmental function was confirmed in the cross-sectional cohort, where infants tolerated delays of 0, 3, and 8 sec at 8, 10, and 12 months, respectively.4 Other researchers report results consistent with this pattern. Harris (1973) tested 9.5- and 10-month-olds on four trials of a two-location DR task at delays of 0 or 5 sec. Infants performed significantly more accurately at delays of 0 than 5 sec. Brody (1981) trained 8-, 12-, and 16-month-olds to discriminate between two locations by reaching to the one location marked by a light, with correct selection resulting in visual reinforcement. On subsequent test trials, infants were required to reach to the location marked by the light after delays ranging from 250 msec to 9 sec. Eightmonth-olds recalled the location of the discriminative stimulus after delays of 250 msec and 12- and 16-month-olds recalled the stimulus location at delays of 3, 6, and 9 sec. Gilmore and Johnson (1995) developed an oculomotor DR procedure based on the work of Funahashi and colleagues (Funahashi, Bruce, & Goldman-Rakic, 1989, 1990) with nonhuman primates. Infants learn to delay their saccade to a spatial cue for several seconds, or to delay their saccade to a spatial location cued by an abstract central stimulus. For example, infants saw a central fixation stimulus, and then a cue was presented to the left or right of center for 150 msec. Targets appeared to the left and right of the central fixation point after a 50– 5000 ms delay. A correct response was a gaze to the target that appeared in the same position as the cue. Results from two experiments indicated that 6-month-olds oriented more to the correct target stimulus after delays of at least 5 sec (Gilmore & Johnson, 1995).

4

Although the linear form of the developmental trajectory was confirmed across the longitudinal and cross-sectional samples, the longitudinal sample attained higher levels of performance at some ages. This might suggest that practice (or perhaps general experience with the lab environment) is an important factor in DR task performance. Researchers should address the possibility in future work. Such research could provide insight into the relative roles of experiential factors in WM development.

196


In contrast, some studies suggest greater WM durability. For example, Allen (1931) reported that 12-month-olds could tolerate delays longer than the 10-sec delays suggested by Diamond and Doar and others, with performance greater than chance even at delays of 20 sec. This estimate of memory durability is closer to the results of a study by Reznick, Fueser, and Bosquet (1998) in which 9-month-olds searched for a toy hidden in one of three possible locations. Infants who searched incorrectly were allowed to select from the two remaining wells after delays of 10–20 sec, and they were correct on approximately 62% of their responses, more than what would be expected by chance. In addition to the canonical manual DR task, researchers have used various modified DR tasks to assess infant WM. For example, Schwartz and Reznick’s (1999) windows and curtains DR procedure reflects a modification of the DR task designed to use a visual response and to allow infants to search for an attractive human face. The infant sat on the parent’s lap facing a large screen with two small windows. An experimenter appeared in one of the two windows, and after engaging the infant’s attention, lowered curtains to block both windows. The parent rotated a swivel chair so that the infant no longer faced the screen, and then after a delay, reoriented the infant toward the screen. The curtains rose to reveal empty windows, and an observer recorded the direction of the infant’s gaze. Performance of the 9-month-olds was above chance, with percentage correct scores at 69% and 58% after delays of 10 and 20 sec, respectively. The windows and curtains procedure has various aspects that may boost WM performance in infants. A human face is an extremely salient stimulus and the procedure is essentially a peek-a-boo game and thus is naturalistic and engaging. Infants enjoy riding in the pivoting chair and thus can tolerate a relatively long delay, maintaining a high level of engagement across a relatively large number of trials. As we discuss in detail in a later section, a visual response might be more likely to be correct than a reaching response. Finally, the opening of the curtains after the delay captures the infant’s attention and usually evokes a gaze toward one of the windows. D. CONCLUSIONS REGARDING WM IN INFANCY

1. Changes in WM Ability Much research on WM in infancy has relied on the DR procedure and has focused on length of delay for memory for locations. Within these limited parameters, a consensus is emerging regarding the developmental function of WM during early infancy. We summarize some of this evidence in Figure 1, which reflects a selection of studies conducted in several different laboratories that have examined the development of WM. The approximate


197

Fig. 1. A summary scatterplot of data drawn from a selection of studies that have examined WM in infants. Duration (in seconds) of the highest delay at which infants performed successfully on the DR task is plotted against age in months.

duration (in seconds) of the highest tolerable delay (typically defined as the length of delay at which infants perform significantly above chance or meet some other specified criterion of percentage correct at the delay as suggested by the original authors) is shown here as a function of age. Two aspects of Figure 1 are noteworthy. First, there is substantial between-study variability in the estimates of durability at any one level of age. This variability reflects, in part, differences in assessment technique including the choice of response modality (reach or gaze) and the type of the object hidden in the to-be-remembered location (person or a toy), both of which we discuss subsequently. This variability notwithstanding, the obvious linear developmental trend in performance observed across studies is equally noteworthy. That is, the examination of central tendency in this figure strongly suggests an onset of measurable WM at around 5–6 months and a linear developmental trajectory thereafter across the first year. 2. Influence of Stimulus Type Several of the investigations listed in Table I have compared memory performance for the location of different categories of stimuli. Based on these results, infants tend to perform more accurately on DR tasks when the hidden stimulus is a person rather than an object. For example, Bell (1970) found that 8-, 11-, and 13-month-olds performed more accurately on

198


a two-location DR task when the hidden stimulus was a person than when it was a toy. Jackson, Campos, and Fischer (1978) did not find differential responses when comparing search for people and objects, but Legerstee (1994) and Bigelow, MacDonald, and MacDonald (1995) found that performance was more accurate when infants searched for a person rather than an object. Schwartz and Reznick (1999) examined the search behavior of 9-month-olds in a modified two-location DR procedure (described in detail previously). In the context of a naturalistic ‘‘peek-a-boo’’ game in which the hidden stimulus was an experimenter, performance was significantly above chance with percentage correct scores at 69% and 58% in delay conditions of 10 and 20 sec, respectively. Reznick and colleagues found comparable results for 5- to 6-month-olds in a delay condition of 1 sec (Reznick et al., 2003). However, in the Schwartz and Reznick (1999) and the Reznick et al. (2003) study, there was no object condition against which to compare performance for faces. Thus, although their findings suggest early WM for the location of hidden faces, the studies do not conclusively demonstrate a performance advantage for faces over hidden objects. Why is a person such a compelling stimulus in the DR paradigm? Obviously, the human face is an attractive stimulus because of intrinsic stimulus parameters as well as a dense history of association with reinforcement. Moreover, face-like patterns draw the attention of young infants (Goren, Sarty, & Wu, 1975; Johnson et al., 1991; Morton & Johnson, 1991), insuring both experience with and interest in faces. Cowan (1999) suggests that attention may be automatically recruited by stimulus categories with special significance to a subject (e.g., stimuli with exceptional social significance or for which we have developed substantial expertise). In these situations, fewer demands are made of the central executive for directing the focus of attention, thereby freeing up WM capacity. Given the evidence that faces form a unique stimulus category for infants, they are also excellent candidates for stimuli with special significance to the infant. Faces may automatically draw attention, thereby opening up limited processing resources for other task-relevant purposes. Another explanation emphasizes the dynamic nature of faces compared to the typically more static nature of objects. Objects can be, and usually are, made animate by the examiner in ways that capture an infant’s attention. The movement of eyes, the shape of the mouth, and the complex changes associated with facial expressions are regarded as a special category of motion—biological motion—that may carry particular significance for the infant (e.g., Booth, Pinto, & Bertenthal, 2002; Fox & McDaniel, 1982). Consistent with this explanation, functional neuroimaging studies in adults indicate that biological motion activates cortical regions that are not activated


199

by complex nonbiological motion, including the posterior superior temporal sulcus (e.g., Grossman et al., 2000; Pelphrey et al., 2003; Puce et al., 1998). In summary, the choice of stimulus affects estimates of WM durability in infants. Paradigms that make use of face-like stimuli are likely to evoke greater WM durability than paradigms based on other categories of stimuli. 3. Influence of Response Modality Hunter (1917) noted that the DR procedure and apparatus must be adapted to the subject’s mode of response, ‘‘be it walking, flying, or swimming.’’ This statement reflects the fact that estimates of WM ability may depend on the mode of response used as the dependent measure. A comparison of results from several of the studies summarized in Table I indicates that direction of gaze may be a more sensitive index of memory durability than reaching per se. Diamond (1985) noted anecdotally that infants sometimes look at the correct location even as they reach toward the incorrect one. Hofstadter and Reznick (1996) directly compared gaze and reach performance on a two-location DR task in samples of 5-, 7-, 9-, and 11-month-olds. Overall accuracy was 61% in the reach condition and 72% in the gaze condition. At a delay of 3 sec, performance was more often correct with gaze as the response, and when direction of gaze and reach differed (14% of all trials), direction of gaze was more likely to be correct (but see McDonough, 1999). Furthermore, at 5 months, when the infants were not capable of reaching efficiently, their gazes were correct more often than would be expected by chance, suggesting that WM is available before 7 months. With a variation of the DR task, Dunst, Brooks, and Doxey (1982) obtained similar findings. Infants gazed to the correct location at a rate significantly greater than chance, although they were unable to search manually. Three studies that have compared reach and gaze responses in DR or A-not-B tasks provide equivocal results. Whereas Hofstadter and Reznick (1996) found that infants perform more accurately in the visual domain, Matthews, Ellis, and Nelson (1996) and Bell and Adams (1999) did not find differences in performance as a function of response domain. A critical design difference among these studies may account for the variability in findings. Specifically, in the studies that did not find a performance advantage in the visual domain gaze and reach were compared across different DR tasks. In contrast, Hofstadter and Reznick (1996) directly compared infants’ gaze and reach responses within the same task. Task-related differences between DR procedures might overwhelm effects of modality, making the within-task design a more optimal paradigm for examining performance differences that are a direct result of response modality. Additional research is needed to verify differences between gaze and reach performance in DR tasks and, specifically, to identify the circumstances in

200


which a decalage does and does not occur. At one level of analysis, modality effects on WM could be of theoretical interest. At another level of analysis, a difference in results for gaze versus reach responses has methodological implications and underscores the advantages of a multiple measurement strategy. Why might gaze and reach differ as an index of WM in infants? One category of explanations refers to anatomical or neurological differences in the underlying systems. For example, given that physical growth and the development of motor control proceed from the head downward, the development of the visual system (and thus, the ability to make a visual response in a DR paradigm) is likely to precede visually guided reaching responses (Bertenthal & von Hofsten, 1998). Responses requiring only the earlier developing visual system could reveal competencies masked by the immaturity of the system subserving motor responses. A second category of explanation reflects intrinsic differences in the nature of a reaching response versus a gaze shift. For example, Bushnell (1985) notes that young infants who are just learning to execute a visually guided response must devote considerable processing resources to the act of reaching itself. This allocation of cognitive resources in the manual task could lead to the forgetting of the correct location, consistent with Baddeley and Hitch’s (1974) interpretation of subject performance within the classic dual-task paradigm. Diamond (1991) notes that reaching for a hidden object requires relating two or more actions in a behavioral sequence. DR tasks that require visual responses require only a gaze shift and thus, do not impose this extra-memorial requirement. Related, although both gaze and reach responses require inhibitory control of behavioral tendencies and prepotent responses, the two response modalities may be differentially susceptible to inhibitory processes. Data from a detour-reaching task suggest that infants between 5 and 7 months cannot easily inhibit reflexive motor reactions to objects they touch (Diamond, 1991). Finally, the nature of the DR task creates a greater match between encoding and responding for the gaze response. That is, in both conditions, infants initially gaze at an object and its hiding location. Then, after a delay, infants in a visual condition gaze to the previous location again but infants in a manual condition must launch a new type of response (i.e., a reach). Finally, a third category of interpretation posits effects due to the differences in the underlying representations that support gaze and reach responses. Some theories of human memory postulate a distinction between implicit and explicit memory (e.g., Roediger, 1990; Sherry & Schacter, 1987; Schacter & Tulving, 1994; Tulving, 1985). Explicit memory entails conscious recall of specific items of information. In contrast, when prior learning influences a current task, we infer no conscious effort to recall the earlier


201

experience, but implicit memory is implied. Modality effects on cognitive processes are often interpreted in terms of an implicit–explicit distinction. For example, Clements and Perner (1994) report that children succeed on false-belief tasks earlier when looking behavior is used as the dependent measure than when their verbal response is used. Similarly, 2-year-olds fail to display knowledge of the concepts of solidity and support in explicit (verbal) measures of performance, but 4- and 6-month-olds show looking behaviors consistent with knowledge of solidity and support (Berthier et al., 2000; Hood, Carey, & Prasada, 2000). From this perspective, a gaze-reach differential could emerge because gaze taps implicit cognitive systems while reaching taps explicit cognition. In a related view of representations, Munakata and colleagues have argued that different levels of representational completeness are required in the context of the reaching and looking versions of the A-not-B task (Munakata, 1998; Munakata et al., 1997). From this ‘‘graded representation’’ perspective, younger infants are able to represent the hidden object’s location, but the representation is so fragile that it can only support correct visual responses. With age, infants are able to form representations that are more robust (particularly in terms of competing with representations still in memory from previous responses) and are thus able to support more complicated (in the senses described earlier) reaching responses as well. There are various reasons why infant abilities might differ when assessed using different response modalities. One obvious conclusion is that investigators must avoid basing conclusions on results from a single response modality. Moreover, various findings suggest that visual responses may be more sensitive to WM in infancy than traditional motor responses, thus pointing up a promising direction for future research. 4. Inhibition of Perseverative Responses Perseveration refers to the tendency to repeat a previous response. Models of WM from the literatures of cognitive and comparative psychology emphasize that WM involves both the storage and processing of information (Baddeley, 1986). From this perspective, the updating of old information and the inhibition of incorrect responses reflect WM’s active processing, and perseveration can reveal a failure of this processing. This perspective is compatible with Pennington’s proposal (1994, 1997) that because WM is a limited capacity system, inhibition of irrelevant information is critical to WM’s functionality and is thus an aspect of WM ability (see also Engle, Kane, & Tuholski, 1999). Infants have a strong tendency to perseverate in the DR task. For example, Diamond found that infants in the A-not-B task tended to repeat previously reinforced responses, and thus performed correctly on DR trials

202


that required repetition of the previous response but erred on trials that required a reversal of the previously correct response. Moreover, when infants must find an object hidden among several potential locations, they err disproportionately in the direction of the previously correct location (Diamond, Cruttenden, & Neiderman, 1994). Even when infants can see that there is no toy hidden at the previous location, they still demonstrate considerable difficulty in inhibiting previous responses (Butterworth, 1977; Harris, 1974; Sophian & Yengo, 1985; Thelen & Smith, 1994). We do not have a comprehensive understanding of perseveration and its relation to WM, but one general conclusion is that motor probes are more vulnerable to the influence of perseveration than are gaze responses. For example, Hofstadter and Reznick (1996) compared 7-, 9-, and 11-montholds in gaze and reach conditions of a two-location DR task. Infants in the reach condition were more likely to respond perseveratively (68% of their trials) than were infants in the gaze condition (59% of their trials). As noted earlier, vulnerability of motor responses to perseveration is one reason why estimates of WM calculated on the basis of gaze responses might more accurately reflect underlying competency. At a theoretical level, perseveration of previous responses would seem to be an integral and salient aspect of WM. Previous research confirms that infants perseverate, but we know little about the nature of this perseveration or how it changes with development.

IV. Frontiers of Research on WM in Infants A. METHODOLOGICAL CHALLENGES

1. Alternative Definitions of WM Ability The literature on WM in infants has relied on span across a temporal delay as the principal measure of ability. This could be considered a substantial weakness when contrasted with the literature on WM in verbal children and adults in which ability is frequently defined on the basis of the amount of verbal or spatial information that can be stored and manipulated in WM. Most procedures used to operationalize WM in adults and verbal children have evaluated capacity (i.e., the quantity of stored information) as a measure of ability rather than durability (i.e., the length of delay over which a given amount of information can be stored). For example, WM ability defined as capacity is apparent in such tasks as a backward digit span, in which capacity is the number of digits retained and reported in correct order, and in procedures that require participants to recall the final word of each previous sentence, with capacity defined as the number of words reported.


203

A potential cause of the bias toward durability as the primary measure of ability in this literature may lie in the history of the DR task itself. Hunter (1913) viewed the delay over which the subject could recall the location of a discriminative stimulus as the most relevant manipulation for comparison of species at different points along the phylogenetic continuum (reviewed by Maier & Schneirla, 1935). Another reason for this bias may stem from the fact that the delay between hiding and search is an obvious and highly effective parameter in Piaget’s A-not-B paradigm (e.g., Bjork & Cummings, 1984; Cummings & Bjork, 1983; Diamond, 1985; Diamond, Cruttenden, & Neiderman, 1994). Indeed, two meta-analyses of the A-not-B error literature have identified length of delay as a primary predictor of infant performance (Marcovitch & Zelazo, 1999; Wellman, Cross, & Bartsch, 1987). Despite the emphasis on length of delay, a few investigations summarized in Table I have demonstrated age-related improvements in the ability to search for an object among increasing numbers of alternative locations. For example, Kagan and colleagues described a DR protocol in which a toy is hidden under one of 2, 4, 6, or 8 cups with delays of 1, 5, or 10 sec before the child was allowed to search (Kagan, 1981; Kagan, Kearsley, & Zelazo, 1978). A trial with an increased number of cups or an increased delay followed successful search. Results from cross-sectional and longitudinal investigations of 152 infants in the United States and Fiji indicated a clear developmental progression between 13 and 36 months, in which infants and toddlers could successfully locate the hidden toy among increasingly greater numbers of cups (Kagan, 1981). Similarly, Reznick, Corley, and Robinson (1997) reported that modal levels of performance were 4 cups at 1-sec, 6 cups at 1-sec, and 6 cups at 5-sec, for 14-, 20-, and 24-month-olds, respectively. Together, these studies illustrate the efficacy of number of locations as an alternative measure of WM ability. Some investigators exploring retrieval performance with multiple locations have reported the counterintuitive finding that infant performance in the DR paradigm improves when more than two locations are used (Cummings & Bjork, 1983; Sophian, 1985; Wellman, Cross, & Bartsch, 1987). Research by Diamond, Cruttenden, and Neiderman (1994) indicates that this counterintuitive finding arises from a subtle but critical procedural artifact. In the standard protocol for two-location DR studies, the experimenter hides the object in one of two locations and then covers both locations simultaneously. Some researchers who use more than two locations cover all of the locations at the start of each trial and then hide the object in one of the already covered locations. This makes the protocol easier to administer, but in a within-participants comparison of these two protocols, Diamond et al. (1997) found that infants perform more accurately when searching for an object hidden in an already covered

204


location in contrast to performance when searching for an object that is hidden and then covered. Alp (1994) used an imitation sorting task to estimate WM in very young children. Children saw a number of disparate objects sorted into two containers. The objects were then removed from the containers, and children were asked to reproduce the modeled sort. The largest number of objects that children could sort correctly (taken as an index of WM capacity) increased from 12 to 36 months. A follow-up session conducted 6 months later also revealed comparable rank ordering (r ¼ .55, controlling for age), suggesting considerable continuity for this aspect of WM. Another way to manipulate the quantity of information held in WM is to hide a number of items among several different locations on the same trial. Tasks employing this manipulation are adaptations of procedures developed in comparative psychology including the fully baited radial maze (e.g., Olton, Collison, & Werz, 1977; Olton & Samuelson, 1976), and the foraging tasks (MacDonald & Wilkie, 1990; MacDonald, 1994; MacDonald, Pang, & Gibeault, 1994; Menzel, 1973) described earlier. Participants are required to maintain an ongoing record of previously selected stimuli and monitor this record in preparation for the next response. WM capacity is defined as the number of times a participant visits a location that still contains a reward while avoiding those that have been searched before. Methods involving the manipulation of the number of hidden items have proved successful in several studies involving young children (e.g., Aadland, Beatty, & Maki, 1985; Foreman, Arber, & Savage, 1984; Foreman et al., 1989; Foreman, Warry, & Murray, 1990; Overman et al., 1996), and researchers are beginning to engineer tasks for infants that are based on this manipulation. For example, Reznick and Fieselman (1999) used a naturalistic foraging task with 12- and 18-month-olds. Their apparatus consisted of four small boxes with hinged tops. The infant sat on his or her parent’s lap and watched the experimenter place a piece of cereal or a raisin in each of the four boxes in random order. After a delay of 3 sec, the child was allowed to reach into one of the boxes. The first four reaches were designated as a trial. After each trial, the experimenter refilled the locations in a different random order and allowed the child to reach again. Twelve-month-olds made an average of 2–3 correct selections on a trial, whereas 18-month-olds performed significantly more accurately, making, on average, 3.5 correct responses per trial. Pelphrey et al. (2002) conducted an experiment to investigate toddler spatial WM using a novel ‘‘foraging’’ WM task. Three buckets were arranged on the floor in a triangular configuration. The child watched the experimenter hide a sticker under each bucket, before leaving the room for an average delay of 5 sec. The child returned and searched for one of the three stickers. When the child correctly located a sticker, the process was


205

repeated (i.e., leave the room, return and select a bucket) until the child found all three stickers or revisited a location. This constituted one trial out of three administered. The average percentage correct score per trial for a group of 40 toddlers (M ¼ 26-months-old: range ¼ 23–32 months) was 73%. A significant positive correlation (r ¼ .48) between age and percent correct per trial indicated a development trend in performance. This procedure taps WM in the sense that children must retain information about a specific location and use that information to select an appropriate response and inhibit an inappropriate response. Given the importance of both the durability to span a delay and the amount of information retained as aspects of WM, an operational definition that includes both attributes has advantages. For example, Pelphrey et al. (2003) used multiple DR measures of WM to operationalize ability as both amount and duration of storage. Six- to 12-month-olds were tested on four DR tasks. Two tasks challenged WM by varying the delay between hiding and search and two varied the number of locations among which infants searched. A regression model in which task by challenge-level combinations were used to predict age suggested little improvement in WM performance from 6 to 8 months and linear improvement in WM from 8 to 12 months, with WM tasks accounting for 66% of the variance in age (see Figure 2). Measures from individual tasks suggested a more varied pattern. Notably, improvement in spanning multiple locations lagged behind improvement in Composite of Four Delayed Response Tasks

Fig. 2. Positive correlation between age in months and levels of overall percentage correct responding (corrected for chance) on a composite score representing the average across two DR measures of WM in infants using the length of delay as the measure of durability. Adapted from Pelphrey et al. (2003).

206


spanning longer durations. This finding suggests that two distinct aspects of WM have different developmental trajectories. Moreover, a composite measure provides a highly stable estimate of WM, more potent than any one task considered in isolation. This highlights the value of a multiple method or converging operations assessment strategy for measuring WM in infants. More broadly, WM ability might reflect processes such as efficiency of encoding, rehearsal strategy, some task-specific cognitive skill or control of attention. For example, Barrouillet and Camos (2001) found that delay interval and counting skill both influence performance on a counting span task in which children must hold calculated array values in WM. Engle (2002) argues that WM ability is not about memory per se, but rather reflects the ability to use attention to maintain or suppress (inhibit) information. On this argument, what appears to be greater WM ability is really greater ability to use attention to avoid distraction. The data that support this view are compelling but limited in the present context because they rely on procedures with complex instructions not suitable for children with limited language. Researchers who are interested in WM in infants must find ways to tap various aspects of ability. 2. Competence or Performance? Infants seem to have more WM ability in the DR paradigm when gaze is the mode of response and when the hidden stimulus is highly motivating. The primary implication of these performance effects is that accounts of the development of WM in infants based on a single response modality or a single type of hidden stimulus may be limited. This conclusion should not come as a surprise. One of the best-documented phenomena in cognitive development is that levels of performance on different tasks thought to measure the same underlying construct may vary widely and that performance on a single task may vary markedly from setting to setting (e.g., BakerWard, Ornstein, & Principe, 1997; Fischer, 1980; Fischer & Bidell, 1998; Flavell, 1985). It is almost a truism to claim that a single task or performance context seldom provides an adequate assessment of developmental function. One practical issue related to the competence–performance distinction in WM assessment is selection of the number of test trials administered. In research with rodents and nonhuman primates, testing on a WM task often continues for hours or days until the animal reaches asymptotic performance (e.g., Funahashi, Bruce, & Goldman-Rakic, 1989, 1990; Olton & Samuelson, 1976). This regime does not necessarily insure an accurate measure of competence, but it does promote reliability in measurement, which is an important component of measuring competence. That is, multiple trials administered across multiple testing occasions will tend to randomize local perturbations in performance, thus providing a better


207

estimate of competence. Human infants will tolerate only a limited number of trials before boredom or distraction intervenes. Thus, WM assessment of infants often taps performance rather than competence, and we have little reason to believe that scores are definitive or reliable. Analogous to signalto-noise ratios in neurophysiological and psychophysical studies, the ability to detect a relevant signal (i.e., infant competence) above noise (i.e., irrelevant trial-to-trial variability) improves as increasing numbers of trials are averaged. Because infants receive only a limited number of trials, researchers face a choice regarding testing strategy. Most researchers use a traditional approach in which each infant sees a predetermined sequence of trials of varying difficulty or a single block of trials at a single level of difficulty. An alternative strategy is to use a dynamic protocol, where the experimenter adjusts the trial sequence based on the infant’s performance. For example, Diamond et al. (1997) used a performance-based assessment strategy in studies using the A-not-B and DR tasks (e.g., Diamond, 1985; Diamond & Doar, 1989). In the Diamond and Doar (1989) study, when performance reached 88% correct at a given delay during a testing session, the delay used for the next session was incremented by 2–3 sec. Alternatively, if infants did not attain the 88% criterion during a session, but did reach correctly on 70% or more trials during two consecutive sessions at a particular delay, the delay was incremented 2–3 sec at the next session. The dynamic or infantcontrolled procedure has various advantages including the potential for providing a valid measure of WM ability with fewer trials relative to the traditional blocked approach. 3. WM Content The Baddeley and Hitch (1974) model of WM contains a visuo-spatial sketchpad and a phonological loop, reflecting the assumption that individuals process visual-spatial information and acoustic information separately in WM. Although much evidence has accrued regarding the early emergence of WM for visual information, we know little about the early emergence of WM for acoustic information. This gap is particularly problematic given the likely importance of acoustic WM for language acquisition. The main limitation on knowledge about WM for acoustic information is the lack of methods for assessing this aspect of WM in infants. Within WM for visual information, a distinction between spatial (where) and object (what) information can be made. Spatial WM captures the location of a particular target (e.g., item X, whatever it is, is in the well). Object WM refers to the identity of a particular target (e.g., item X is a blue rattle, although I might not know where it is hidden). Some evidence supports a neuroanatomical distinction between these two WM systems within

208


prefrontal cortex, with the organization being similar to that identified for the visual system (Ungerleider & Haxby, 1994); object WM is associated with ventrolateral prefrontal cortex and spatial WM with dorsolateral prefrontal cortex (e.g., Smith et al., 1995; Wilson, O’Scalaidhe, & Goldman-Rakic, 1993). The procedures used to examine WM in infants involve spatial (memory for where) rather than object (memory for what) WM. Accordingly, more is known about the development of WM in infants for location than is known about object WM. The reason for this lack of information is straightforward—researchers have yet to develop effective measures of object WM. However, as with the need to expand definitions of ability, a comprehensive description of WM in infancy will require further developments in assessment techniques. It seems likely that future research could be profitably directed at charting the potentially dissociable ontogenetic pathways of spatial and objects WM. Moreover, developmental change in each system is likely to have unique implications for the functions that WM serves in facilitating and constraining broader cognitive changes. B. DEVELOPMENTAL COGNITIVE NEUROSCIENCE AND WM IN INFANTS

Developmental cognitive neuroscientists study the behavioral and cognitive reorganizations that occur over human ontogeny using methods and research approaches that blend behavioral observation and experimental techniques with the tools of cognitive neuroscience and neuropsychology (e.g., functional neuroimaging, marker tasks, and electrophysiological recordings) (see Crnic & Pennington, 1987; Johnson, 1997, 1998, 2000; Johnson & Gilmore, 1996; Nelson & Luciana, 2001, for reviews). WM is an important area of research within the rapidly developing field of developmental cognitive neuroscience. There have been some efforts within the literature of developmental cognitive neuroscience to study WM in infants. Diamond (1990a; see also Diamond & Goldman-Rakic, 1989) compared the performance of human infants and normal infant rhesus monkeys, and examined the impact of lesions to prefrontal cortex on the performance of adult and infant rhesus monkeys with DR tasks involving both the A-not-B and DR protocols. Normal 2-month-old monkeys demonstrated levels of accuracy similar to levels observed in 7-month-old human infants, in that the monkeys failed to reach to the correct location after delays of more than 2 sec. Moreover, like their human counterparts, the normal infant rhesus monkeys demonstrated a clear (albeit accelerated) developmental progression in their ability to withstand increasingly longer delays on the A-not-B and DR tasks. By 4 months, infant monkeys were able to perform successfully with delays of 10 sec. Comparable levels of performance in


209

human infants were not achieved until 12 months of age, suggesting a developmental progression for DR performance in the rhesus monkey infants of approximately 4 sec per month. The two infant monkeys with prefrontal cortex lesions did not show this characteristic developmental progression in their performance on the DR tasks. One was able to successfully perform the DR task with a 2-sec delay but could not perform the task at longer delays. The other lesioned monkey could not pass the DR task when it involved any length of delay. These findings implicate the development of dorsolateral prefrontal cortex as an important mechanism in the early development of WM ability. We can draw similar conclusions regarding the role of prefrontal cortex development in the emergence of WM from longitudinal studies using electroencephalography (EEG) conducted with normal human infants between 6 and 12 months of age. For example, Bell and Fox (1992) reported that increases in frontal activity correlate strongly with increases in the ability to span longer delays during performance in A-not-B versions of the DR task. Moreover, Fox and Bell (1990) reported significant individual differences in both the developmental trajectories of infant performance on the A-not-B task and in age-related changes in frontal EEG activity. They identified two distinct patterns of change in DR performance and frontal EEG activity. One group of infants demonstrated a typical developmental progression from 7 to 12 months in their ability to withstand increasingly longer delays between object hiding and search. Associated with this developmental change was a concurrent progression in levels of frontal EEG activity. In contrast, a second group of infants showed neither the typical developmental progression in DR performance nor the associated increase in frontal EEG activity. Infants in this group solved the A-not-B task when there was no delay between hiding and search, but were unable to tolerate delays longer than 2 sec even at 12 months of age. Throughout this period of observation, their levels of frontal EEG activity remained constant. More recently, Bell and Fox (1994, 1997) reported that individual differences in baseline frontal EEG among 8-month-olds were related to individual differences in performance on the A-not-B task. For example, Bell (2001) recorded EEG from 8-month-olds as they performed a looking version of the A-not-B task and during a baseline period. EEG activity was correlated with task performance. Specifically, the experimenters divided the sample into higher and lower performing groups. Only the higherperforming infants exhibited increases in 6- to 9-Hz EEG power from baseline to task, measured at frontal (i.e., over the frontal lobes) and posterior (i.e., near occipital cortex) scalp locations. Increasing power values across age are often interpreted as a marker of brain maturation, suggesting that the higher-performing infants had more advanced brain development.

210


Moreover, these same high-performing infants exhibited lower EEG coherence values at right hemisphere frontal locations relative to the infants that performed less well. Lower coherence values were evident during baseline and task. All infants showed increased frontal-parietal coherence during the spatial WM task relative to baseline values. These findings are important because they establish the basic phenomenon (i.e., a relation between frontal brain electrical activity and performance on the A-not-B task) in a large sample of infants and demonstrate that effects are observable within individual participants during task performance. Nevertheless, methodological factors limit conclusions derived from these studies regarding the neural basis for WM in infants. Specifically, the spatial resolution of EEG is limited, making it difficult to localize the source of activity related to DR task performance. Moreover, because the designs have not been event-related, they tell us nothing about the temporal profile of activity during the task itself. The observed activity could reflect memory storage, efforts to inhibit prepotent responses, task-directed attention, or activity related to response preparation/execution. C. WHAT DRIVES THE DEVELOPMENT OF WM?

At least two aspects of WM change during the first two years: the durability to tolerate a delay (e.g., Diamond & Doar, 1989; Hunter, 1917; Reznick, Fueser, & Bosquet, 1998) and the capacity to remember the location of an object among several potential hiding locations (e.g., Kagan, 1981; Kagan, Kearsley, & Zelazo, 1978; Reznick, Corley, & Robinson, 1997). We might adduce various mechanisms for these developmental changes in WM. 1. Structural and Functional Brain Development Almost every aspect of the human nervous system continues to develop postnatally, with many dramatic changes occurring before the first birthday (see Herschkowitz, Kagan, & Zilles, 1997; Webb, Monk, & Nelson, 2001, for reviews). This vast array of change provides numerous candidates for mechanisms that could drive WM development, including dendritic arborization, synaptogenesis and subsequent pruning, developments in neurotransmitter systems, myelination, and gross developments in brain volume. For example, synaptic density in most areas of cortex reaches adult levels shortly after birth, plateaus between 1 and 2 years of age, and then declines during the later childhood years (Huttenlocher, 1979, 1990, 1994; Huttenlocher & Dabholkar, 1997). Synapse formation in prefrontal cortex, including the middle frontal gyrus, lags behind other areas of cortex, with peak density occurring around


211

3.5 years of age (Huttenlocher & Dabholkar, 1997). At 3 months, the density has reached 50% of its adult peak, but the synapses are still immature relative to the adult form. Between 6 and 24 months, the synapses become adult-like (Huttenlocher, 1979). Huttenlocher’s data suggests that the most dramatic increase in synaptic density occurs around 8 months of age and peaks at 2 years (reviewed in Webb, Monk, & Nelson, 2001). This period and pattern of development coincides with the developmental trajectory of WM as measured by DR. Imaging studies of cortical glucose metabolism have revealed a developmental rise-and-fall pattern similar to the pattern observed for synaptogenesis and synapse pruning, with glucose utilization rapidly increasing and reaching normal adult levels at about one year, increasing to a plateau at 4 years and remaining high throughout childhood, and then gradually decreasing to stable adult levels by 16–18 years of age (Chugani, 1993, 1994, 1998; Chugani, Phelps, & Mazziotta, 1987). As with synaptogenesis, within the prefrontal cortex, the rise in glucose metabolism lags behind the rest of the cortex by approximately 6–8 months. Excessive connectivity and glucose utilization during infancy could provide the anatomical substrate that supports neural plasticity and the development of such higher cognitive functions as WM (Huttenlocher, 1990, 1994; Johnson, 1997; Elman et al., 1998). Diamond et al. demonstrated that developments in the dopamine system are relevant to changes in WM ability (Diamond et al., 1997). Specifically, Diamond et al. (1997) reported reductions in the DR performance of infants treated for phenylketonuria, a metabolic disorder associated with reductions in the amino acid tyrosine, a precursor to dopamine. Similarly, researchers examining neurological conditions involving dysfunctions of the dopamine system (e.g., Parkinson’s disorder) in human patients have documented deficits in WM (Gabrieli et al., 1996). Consistent with these results, Sawaguchi, Matsumura, and Kubota (1988) demonstrated that local application of dopamine-enhanced neural activity in monkey prefrontal cortex was associated with the delay period of a modified DR task. A cogent developmental perspective requires detailed analysis of activity at more than one level of organization of the organism. In the case of WM in infants, the two most relevant levels appear to be behavior and brain function. Methodological gains within cognitive neuroscience will undoubtedly advance our abilities to study structural and functional brain development in children and to visualize brain function during task performance. It is thus important to maintain the perspective that no single method can fully address developmental questions such as determining the mechanisms underlying the development of WM in infants. Within neurophysiological recording techniques, methods differ dramatically in

212


temporal and spatial resolution, thus offering very different windows into neural structure and function. Also informative will be situations where deficits in WM are observed early in development secondary to a neurodevelopmental disorder. Such natural experiments will reveal more about the neural bases of WM development and how loss of WM affects development in multiple domains. A converging-methods approach is critical, both within attempts to localize function and characterize brain development, and for attempts to formulate theories concerning relations between behavioral and brain development. A converging-methods approach will allow us to narrow down the number of possible mechanisms underlying WM development. To summarize, although many neurological mechanisms could drive the development of WM, we do not know which of the potential mechanisms are necessary or sufficient for normal WM development. This situation provides an interesting perspective on the relation between psychological and biological approaches to development. When research on the psychological development of WM has accrued, and we have an increasingly accurate picture of the infant’s changing WM ability, it will then become possible to fully explore the specific behavioral implications of the various correlated neurological changes. 2. Cognitive Development Improvements in strategies for holding information in memory or processing efficiency could increase the functional durability or capacity of WM without accompanying structural change. For example, infants and toddlers may develop strategies for rehearsing, chunking, and sequencing information. These strategies would allow them to increase their speed and efficiency of information processing thus indirectly increasing functional WM ability. Research supporting this interpretation of WM development has been conducted primarily with children (e.g., Dempster, 1985; Hulme et al., 1984; Kail, 1992; Kail & Park, 1994). Increases in WM ability observed during childhood are due, in part, to more efficient processing of stimuli (e.g., faster recognition and speed of processing). Although these sorts of hypotheses have not been developed and tested in infants, this is a promising area for future research. For example, it seems highly unlikely that infants would have any recognizable form of rehearsal strategy, although they might have some form of imagery that maintains a representation of a specific location. From an alternative perspective, this could be viewed as an ability to maintain attention (albeit, covert attention) in the context of ongoing distraction. Finally, as discussed earlier, the ability to inhibit a prepotent response is an aspect of processing that would affect WM.


213

3. Experience Even a robust correlation between neural or cognitive development and WM abilities does not offer unambiguous support for a unidirectional maturational account in which neural or cognitive events are the primary cause of changes in WM. Consider, for example, the potential for developmental transactions among neural development, the growing behavioral repertoire of the infant, and the environment. It is easy to imagine a scenario in which parental interaction (e.g., peek-a-boo games or hiding-finding games) challenges the infant by placing demands on her or his WM and rewards the infant for further developments in memory durability and capacity. In addition, improvements in WM might evoke changes within the infant’s environment, leading the environment to accommodate to the infant’s growing behavioral repertoire in ways that further support changes in neural and behavioral activity. For example, as the infant develops greater WM, parents may change the substance of their interactions to accommodate the infant’s growing behavioral repertoire and further scaffold development in this cognitive system. A comparison of the performance of preterm and full-term infants on the A-not-B task highlights the potential role of experience in the development of WM in infancy (Matthews, Ellis, & Nelson, 1996): premature infants tolerated longer delays than did full-term infants based on age since conception but not based on age since birth. That is, after equating experience (i.e., amount of time since birth), the two groups of infants performed equally, despite presumably less development in brain structure in the preterm infants. These results suggest that experiences in the postnatal environment influence developments in brain structure and function that support successful A-not-B performance. In sum, a comprehensive explanation of changes in WM will likely attribute causal efficacy to experiential and interactional factors. D. HOW DOES THE EMERGENCE OF WM CHANGE THE INFANT?

WM increases substantially over the first year, and this increase may well enable other cognitive skills. Our current understanding of WM in infancy is too limited to support specific links between WM and cognitive development in infancy. However, we can speculate regarding the nature of some expected linkages. 1. WM and Language Acquisition The time course for early production and comprehension of words is well established (e.g., see Bates, Bretherton, & Snyder, 1988; Fenson et al.,

214


1994), but researchers know relatively little about the cognitive mechanisms that allow and support language acquisition. WM is one plausible component of the cognitive machinery that allows an infant to discover and use language. From this perspective, individual differences in aspects of infant WM may be particularly relevant to our understanding of individual differences in developmental trajectories for acquiring language. WM could affect multiple levels and stages of language development. For example, research on speech perception and auditory processing in infants has revealed remarkable capabilities of very young infants to segment the ongoing stream of language and to integrate auditory and visual information (see Aslin, Jusczyk, & Pisoni, 1998, for a review). Some WM seems necessary to allow infants to distinguish speech elements, to retain relevant categories of information for further processing, and to update this information with new data from the ongoing flow of speech in the environment. Furthermore, WM probably aids young infants in detecting new speech elements by allowing them to compare successive auditory stimuli to determine if they are the same or different. In older infants and toddlers, WM may support new word acquisition by allowing the infant to hold onto a newly encountered word long enough to map it onto the proper referent or to review a previous utterance when that utterance is corrected or extended. Researchers have not fully evaluated these hypotheses in research involving infants, but they are consistent with evidence from studies involving children and adults. Indeed, research indicates a link between WM and language comprehension in adults and children (see Gathercole & Baddeley, 1993, for a review). For example, Gathercole and colleagues (Gathercole & Baddeley, 1989; Gathercole et al., 1992) found that phonological WM was strongly related to receptive vocabulary scores, with correlations of .55, .52, .56, and .28 at 4, 5, 6, and 8 years, respectively. Furthermore, cross-lagged correlational techniques suggested a unidirectional pathway of influence, with better phonological WM abilities at 4 years predicting advances in language development at 5 years. Note, however, that some researchers have proposed a different view of the relation between WM and language acquisition. For example, from computer simulations, Newport (1991) and Elman (1993) each suggested that limits in children’s WM make the task of acquiring grammar manageable. They suggest that by holding smaller parcels of information on-line, an immature WM allows an immature language processing system to operate on more manageable units than would a more advanced (i.e., larger capacity) WM.


215

2. WM and Problem Solving Zelazo and colleagues (Zelazo et al., 1997; Zelazo, Reznick, & Spinazzola, 1998) proposed a problem-solving framework to explain the development of executive function. Their framework involves four temporally and functionally distinct phases of executive function: problem representation, planning, execution, and evaluation. Many interrelated factors are likely to underlie development in these four phases, but WM is one of the probable mechanisms underlying age-related changes in problem solving. For example, WM is the computational arena in which task-relevant information is retained and processed, thus allowing for the selection of actions appropriate for attaining specific goals. Accordingly, increases in WM may affect children’s executive function at every phase of the problem-solving process. WM is needed to construct an adequate representation of the problem and the range of possible alternative solutions. In the planning phase of problem solving, the processing aspect of WM is particularly relevant, in that an appropriate plan must be selected from a number of alternatives. Furthermore, a problem-solving plan is likely to include a sequence of temporally constrained actions. Thus, WM could be important for holding this sequence on-line for subsequent execution. Within the execution phase of problem solving, WM would be involved in keeping a plan in mind long enough for it to guide one’s actions. Finally, WM may be involved in the evaluation phase of problem solving, as the individual continues to keep in mind the actions he or she has completed, and their results, while completing the sub-phases of error detection and error correction. Research on infant problem solving suggests notable changes in ability during the time that WM is developing. For example, Willatts (1999) reports a longitudinal study of changes in how 6- to 8-month-olds pull a cloth to retrieve a toy, with increasing responsiveness to task constraints and intentionality in their actions. Production of means-ends behavior increased between 6 and 8 months. Six-month-olds failed to use a cloth, placed under a toy, to intentionally pull the toy within their reach. Eight-month-olds often used the cloth to retrieve the toy. More importantly, infants showed increasing levels of intentionality in their actions. At 6 months, retrieval of the toy appeared to be incidental to playing with the cloth. At 8 months, infants obviously used the cloth as a means to retrieve the toy, the real object of their interest. The relation between problem solving and WM will remain obscure until we have more precise and inclusive measurement techniques in both domains. However, we predict that there will be strong interconnections between these two constructs. Changes in problem solving might reflect developments in WM to the extent that infants use WM to hold in mind

216


temporal sequences of events as one step unfolds into the next. In addition, WM would be used to hold in mind a goal and a response. 3. The Phenomenology of WM Though highly speculative from both an empirical and a theoretical perspective, we conclude with some thoughts about the subjective experience of WM. WM is such a pervasive aspect of our cognitive landscape that to ask what it would be like to have no WM is akin to philosophical speculation about what it would be like to be a bat or a butterfly. Returning to James’ notion of WM as enabling the subjective experience of enduring memory, one can speculate that life without WM would be a life lived in a perpetual present. That is, the experience might be akin to a rushed montage of abruptly ended images, with limited coherence and contiguity across time. From this perspective, life without WM is life ‘‘in the moment,’’ with no mechanism for reflecting on previous moments or anticipating upcoming moments. An infant who has an appreciable WM should become able to recall things that have happened previously and to use these recollections to guide his or her behavior. This perspective allows the infant to extrapolate into the future, to predict what is going to happen next, and to form expectations about upcoming events, supporting such behaviors as using actions to attain goals (Mosier & Rogoff, 1994; Willatts, 1984), and communicating via gestures or words (Bates et al., 1979; Bruner, 1975; Harding & Golinkoff, 1979). From the parent’s perspective, these WM-supported changes allow the infant to engage in intentional behaviors and to seem ‘‘mindful’’ (Feldman & Reznick, 1996; Reznick, 1999; Zeedyk, 1996). If Brentano was correct in his claim that intentionality is the ‘‘mark of the mental’’ (Brentano, 1874/1973), it is hard to imagine an accomplishment more significant than either the infant’s emerging intentionality or a research goal more important than understanding the development of the WM that supports intentionality. AUTHORS’ NOTE Preparation of this manuscript was supported in part by grant NICHD HD30678 to J. S. Reznick. Dr. Pelphrey was supported by grant 1-T32-HD40127. The authors would like to thank Peter Ornstein, Jerome Kagan, and Lynne Baker-Ward for comments on earlier versions of this chapter and Ronald Viola for help in manuscript preparation. Dr. Pelphrey is now affiliated with the Department of Psychiatry at the University of North Carolina at Chapel Hill School of Medicine. Address correspondence concerning this chapter to J. Steven Reznick at the Department of Psychology, University of North Carolina at Chapel Hill, CB#3270, Chapel Hill, North Carolina 27599-3270 or to Kevin A. Pelphrey at the Department of Psychiatry, University of North Carolina at Chapel Hill School of Medicine, CB# 7160, Chapel Hill, North Carolina 27599-7160.


217

REFERENCES Aadland, J., Beatty, W. W., & Maki, R. H. (1985). Spatial memory of children and adults assessed in the radial maze. Developmental Psychobiology, 18, 163–172. Acredolo, L. (1978). Development of spatial orientation in infancy. Developmental Psychology, 14, 224–234. Ahmed, A., & Ruffman, T. (1998). Why do infants make A not B errors in a search task, yet show memory for the location of hidden objects in a nonsearch task? Developmental Psychology, 34, 441–453. Allen, C. N. (1931). Individual differences in the delayed reaction of infants: A study of sex differences in early retentiveness. Archives of Psychology, 19 (Whole No. 127). Alp, E. (1994). Measuring the size of working memory in very young children: The imitation sorting task. International Journal of Behavioral Development, 17, 125–141. Aslin, R. N., Jusczyk, P. W., & Pisoni, D. B. (1998). Speech and auditory processing during infancy: Constraints on and precursors to language. In W. Damon (Series Ed.), D. Kuhn, & R. S. Siegler (Volume Eds.), Handbook of child psychology, Vol. 2: Cognition, perception, and language (5th Edition, pp. 147–198). New York: John Wiley & Sons. Atkinson, R. C., & Shiffrin, R. M. (1968). Human memory: A proposed system and its control processes. In K. W. Spence & J. T. Spence (Eds.), The psychology of learning and motivation (Vol. 2, pp. 89–195). New York: Academic Press. Baddeley, A. D. (1983). Working memory. Philosophical Transactions Royal Society of London, Series B, 302, 311–324. Baddeley, A. D. (1986). Working Memory. Oxford: Oxford University Press. Baddeley, A. D. (1992). Working memory. Science, 255, 556–559. Baddeley A. D. (2001). Is working memory still working? American Psychologist, 56, 851–864. Baddeley, A. D., & Hitch, G. J. (1974). Working memory. In G. Bower (Ed.), The psychology of learning and motivation (Vol. 8, pp. 47–90). New York: Academic Press. Baillargeon, R., DeVos, J., & Graber, M. (1989). Location memory in 8-month-old infants in a non-search A-not-B task: Further evidence. Cognitive Development, 4, 345–367. Baillargeon, R., & Graber, M. (1988). Evidence of location memory in 8-month-old infants in a non-search A-not-B task. Developmental Psychology, 24, 502–511. Baker-Ward, L. E., Ornstein, P. A., & Principe, G. (1997). Revealing the representation: Evidence from children’s reports of events. In P. van den Broek, P. Bauer, & T. Bourg (Eds.), Developmental spans in event comprehension and representation: Bridging fictional and actual events (pp. 79–107). Hillsdale, New Jersey: Lawrence Erlbaum Associates. Baldwin, J. M. (1895). Mental development in the child and the race. New York: Macmillan. Barrouillet, P., & Camos, V. (2001). Developmental increase in working memory span: resource sharing or temporal decay? Journal of Memory and Language, 45, 1–20. Bates, E., Beninni, L., Bretherton, I., Camaioni, L., & Volterra, V. (1979). The emergence of symbols. New York: Academic. Bates, E., Bretherton, I., & Snyder, L. (1988). From first words to grammar. Cambridge, UK: Cambridge University Press. Belger, A., Puce, A., Krystal, J. H., Gore, J. C., Goldman-Rakic, P., & McCarthy, G. (1998). Dissociation of mnemonic and perceptual processes during spatial and nonspatial working memory using fMRI. Human Brain Mapping, 6, 14–32. Bell, S. M. (1970). The development of the concept of the object as related to infant–mother attachment. Child Development, 41, 291–311. Bell, M. A. (2001). Brain electrical activity associated with cognitive processing during a looking version of the A-not-B task. Infancy, 2, 311–330.

218


Bell, M. A., & Adams, S. E. (1999). Comparable performance on looking and reaching versions of the A-not-B task at 8 months of age. Infant Behavior and Development, 22, 221–235. Bell, M. A., & Fox, N. A. (1992). The relations between frontal brain electrical activity and cognitive development during infancy. Child Development, 63, 1142–1163. Bell, M. A., & Fox, N. A. (1994). Brain development over the first year of life: Relations between EEG frequency and coherence and cognitive and affective behaviors. In G. Dawson & K. Fischer (Eds.), Human behavior and the developing brain (pp. 314–345). New York: Guilford. Bell, M. A., & Fox, N. A. (1997). Individual differences in object permanence performance at 8 months: Locomotor experience and brain electrical activity. Developmental Psychobiology, 31, 287–297. Bertenthal, B. I., & Hofsten, C. von. (1998). Development of eye, head, and trunk control as prerequisites for reaching. Neuroscience and Biobehavioral Review, 22, 515–526. Berthier, N. E., DeBlois, S., Poirier, C., Novak, M., & Clifton, R. K. (2000). Where’s the ball? Two- and three-year-olds reason about unseen events. Developmental Psychology, 36, 394–401. Bigelow, A. E., MacDonald, D., & MacDonald, L. (1995). The development of infants’ search for their mothers, unfamiliar people, and objects. Merrill-Palmer Quarterly, 41, 191–208. Bjork, E., & Cummings, E. (1984). Infant search errors: Stage of concept development or stage of memory development. Memory & Cognition, 12, 1–19. Booth, A., Pinto, J., & Bertenthal, B. I. (2002). Perception of the symmetrical patterning of human gait by infants. Developmental Psychology, 38, 554–563. Brentano, F. (1874/1973). Psychology from an empirical standpoint (O. Kraus, Ed.; Schattle, & L. L. McAllister, trans.). London: Routledge & Kegan Paul. Broadbent, D. E. (1958). Perception and communication. London: Pergamon Press. Broadbent, D. E. (1963). Differences and interactions between stresses. Quarterly Journal of Experimental Psychology, 15, 205–211. Brody, L. R. (1981). Visual short-term cued recall memory in infancy. Child Development, 52, 242–250. Bruner, J. S. (1975). The ontogenesis of speech acts. Journal of Child Language, 2, 1–19. Bushnell, E. W. (1985). The decline of visually guided reaching during infancy. Infant Behavior and Development, 8, 139–155. Butterworth, G. (1977). Object disappearance and error in Piaget’s stage IV task. Journal of Experimental Child Psychology, 23, 391–401. Case, R. (1978). Intellectual development from birth to adulthood: A neo-Piagetian interpretation. In R. S. Siegler (Ed.), Children’s thinking: What develops? (pp. 37–72). Hillsdale, NJ: Erlbaum. Case, R. (1991). The mind’s staircase. Hillsdale, NJ: Erlbaum. Casey, B. J., Cohen, J. D., Jezzard, P., Turner, R., Noll, D. C., Trainor, R. J., Giedd, J., Kaysen, D., Hertz-Pannier, L., & Rapport, J. L. (1995). Activation of prefrontal cortex in children during a non-spatial working memory task with functional MRI. Neuroimage, 2, 221–229. Cheng, K. (1986). A purely geometric module in the rat’s spatial representation. Cognition, 23, 149–178. Cheng, K., & Gallistel, C. R. (1984). Testing the geometric power of a spatial representation. In H. L. Roitblat, H. S. Terrace, & T. G. Bever (Eds.), Animal cognition (pp. 409–423). Hillsdale, NJ: Erlbaum. Chugani, H. T. (1993). Positron emission tomography scanning: Applications in newborns. Clinical Perinatology, 20, 395–409.


219

Chugani, H. T. (1994). Development of regional brain glucose metabolism in relation to behavior and plasticity. In G. Dawson & K. Fischer (Eds.), Human behavior and the developing brain (pp. 153–175). New York: Guilford. Chugani, H. T. (1998). A critical period of brain development: Studies of cerebral glucose utilization with PET. Preventive Medicine, 27, 184–188. Chugani, H. T., Phelps, M. E., & Mazziotta, J. C. (1987). Positron emission tomography study of human brain functional development. Annals of Neurology, 22, 487–497. Clements, W., & Perner, J. (1994). Implicit understanding of belief. Cognitive Development, 9, 377–397. Cohen, J. D., Perlstein, W. M., Braver, T. D., Nystrom, L. E., Noll, D. C., Jonides, J., & Smith, E. E. (1997). Temporal dynamics of brain activation during a working memory task. Nature, 386, 604–608. Cornell, E. H., Heth, C. D., & Alberts, D. M. (1994). Place recognition and way finding by children and adults. Memory & Cognition, 22, 537–542. Cornell, E. H., Heth, C. D., & Rowat, W. L. (1992). Way finding by children and adults: Response to instructions to use look-back and retrace strategies. Developmental Psychology, 28, 328–336. Courtney, S. M., Ungerleider, L. G., Keil, K., & Haxby, J. V. (1997). Transient and sustained activity in a distributed neural system for human working memory. Nature, 386, 608–611. Cowan, N. (1988). Evolving conceptions of memory storage, selective attention, and their mutual constraints within the human information processing system. Psychological Bulletin, 104, 163–191. Cowan, N. (1995). Attention and memory: An integrated framework. New York: Oxford University Press. Cowan, N. (1999). An embedded-processes model of working memory. In A. Miyake & P. Shah (Eds.). Models of working memory: Mechanisms of active maintenance and executive control (pp. 62–101). Cambridge, UK: Cambridge University Press. Crnic, L. S., & Pennington, B. F. (1987). Developmental psychology and the neurosciences: An introduction. Child Development, 58, 533–538. Cummings, E. M., & Bjork, E. L. (1983). Search behavior on multi choice hiding tasks: Evidence for an objective conception of space in infancy. International Journal of Behavioral Development, 6, 71–87. DeLoache, J. S., & Brown, A. L. (1979). Looking for Big Bird: Studies of memory in very young children. Quarterly Newsletter of the Laboratory of Comparative Human Cognition, 1, 53–57. DeLoache, J. S., & Brown, A. L. (1983). Very young children’s memory for the location of objects in a large-scale environment. Child Development, 62, 736–752. DeLoache, J. S., & Brown, A. L. (1984). Where do I go next? Intelligent searching by very young children. Developmental Psychology, 20, 37–44. DeLoache, J. S., Cassidy, D. J., & Brown, A. L. (1985). Precursors of mnemonic strategies in very young children’s memory. Child Development, 56, 125–137. Dempster, F. N. (1985). Short-term memory development. In C. J. Brainerd & M. Pressley (Eds.). Basic process in memory development (pp. 208–248). NY: Springer-Verlag. D’Esposito, M., Detre, J. A., Alsop, D. C., Shin, R. K., Atlas, S., & Grossman, M. (1995). The neural basis of the central executive system of working memory. Nature, 378, 279–281. Diamond, A. (1985). Development of the ability to use recall to guide action, as indicated by infants’ performance on A-not-B. Child Development, 56, 868–883. Diamond, A. (1990a). The development and neural bases of memory functions as indexed by the A-not-B and delayed-response tasks in human infants and infant monkeys.

220


In A. Diamond (Ed.), Development and neural bases of higher cognitive functions (No. 608, pp. 267–317). New York: New York Academy of Sciences Press. Diamond, A. (1990b). Rate of maturation of the hippocampus and the developmental progression of children’s performance on the delayed non-matching to sample and visual paired comparison tasks. In A. Diamond (Ed.), Development and neural bases of higher cognitive functions (No. 608, pp. 394–433). New York: New York Academy of Sciences Press. Diamond, A., & Doar, B. (1989). The performance of human infants on a measure of frontal cortex function, the delayed-response task. Developmental Psychobiology, 22, 271–294. Diamond, A. (1991). Neuropsychological insights into the meaning of object concept development. In S. Carey & R. Gelman (Eds.), The epigenesis of mind: Essays on biology and knowledge (pp. 67–110). Hillsdale, NJ: Lawrence Erlbaum Associates. Diamond, A., Cruttenden, L., & Neiderman, D. (1994). A-not-B with multiple wells: 1. Why are multiple wells sometimes easier than two wells? 2. Memory or memory þ inhibition? Developmental Psychology, 30, 192–205. Diamond, A., & Goldman-Rakic, P. S. (1989). Comparison of human infants and rhesus monkeys on Piaget’s A-not-B task: Evidence for dependence on dorsolateral prefrontal cortex. Experimental Brain Research, 74, 24–40. Diamond, A., Prevor, M. B., Callender, G., & Druin, D. P. (1997). Prefrontal cortex cognitive deficits in children treated early and continuously for PKU. Monographs of the Society for Research in Child Development, 62 (2–3, Serial No. 240). Dunst, C. J., Brooks, P. H., & Doxey, P. A. (1982). Characteristics of hiding places and the transition to stage IV performance in object permanence tasks. Developmental Psychology, 18, 671–681. Elman, J. L. (1993). Learning and development in neural networks: The importance of starting small. Cognition, 48, 71–99. Elman, J., Bates, E., Johnson, M. H., Karmiloff-Smith, A., Parisi, D., & Plunkett, K. (1998). Rethinking innateness: A connectionist perspective on development. Cambridge, MA: MIT Press. Engle, R. W. (2002). Working memory capacity as executive attention. Current Directions in Psychological Science, 11, 19–23. Engle, R. W., Kane, M. J., & Tuholski, S. W. (1999). Individual differences in working memory capacity and what they tell us about controlled attention, general fluid intelligence, and functions of the prefrontal cortex. In A. Miyake & P. Shah (Eds.). Models of working memory: Mechanisms of active maintenance and executive control (pp. 102–134). Cambridge, UK: Cambridge University Press. Feldman, R., & Reznick, J. S. (1996). Maternal perception of infant intentionality at 4 and 8 months. Infant Behavior and Development, 19, 483–496. Fenson, L., Dale, P., Reznick, J. S., Bates, E., Thal, D., & Pethick, S. J. (1994). Variability in early communicative development. Monographs of the Society for Research in Child Development, 59 (5, Serial No. 242). Fischer, K. W. (1980). A theory of cognitive development: The control and construction of hierarchies of skills. Psychological Review, 87, 477–531. Fischer, K. W., & Bidell, T. R. (1998). Dynamic development of psychological structures in action and thought. In W. Damon (Series Ed.), D. Kuhn, & R. S. Siegler (Volume Eds.), Handbook of child psychology, Vol. 1: Theoretical models of human development (5th ed., pp. 468–561). New York: John Wiley & Sons. Flavell, J. H. (1985). Cognitive development (2nd ed.), Englewood Cliffs, NJ: Prentice-Hall.


221

Foreman, N., Arber, M., & Savage, J. (1984). Spatial memory in preschool infants. Developmental Psychobiology, 17, 129–137. Foreman, N., Foreman, D., Cummings, A., & Owens, S. (1989). Location, active choice, and spatial memory in children. The Journal of General Psychology, 117, 215–232. Foreman, N., Warry, R., & Murray, P. (1990). Development of reference and working spatial memory in preschool children. The Journal of General Psychology, 117, 267–276. Fox, N. A., & Bell, M. A. (1990). Electrophysiological indices of frontal lobe development. In A. Diamond (Ed.), Development and neural bases of higher cognitive functions (No. 608, pp. 677–698). New York: New York Academy of Sciences Press. Fox, R., & McDaniel, C. (1982). The perception of biological motion by human infants. Science, 218, 486–487. Funahashi, S., Bruce, C. J., & Goldman-Rakic, P. S. (1989). Mnemonic coding of visual space in the monkey’s dorsolateral prefrontal cortex. Journal of Neurophysiology, 61, 331–349. Funahashi, S., Bruce, C. J., & Goldman-Rakic, P. S. (1990). Visuospatial coding in primate prefrontal neurons revealed by oculomotor paradigms. Journal of Neurophysiology, 63, 814–831. Fuster, J. M. (1980). The prefrontal cortex. New York: Raven Press. Fuster, J. M. (1995). Memory in the cerebral cortex: An empirical approach to neural networks in the human and nonhuman primate. Cambridge, MA: MIT Press. Fuster, J. M. (1997). The prefrontal cortex. Philadelphia, Lippincott-Raven. Fuster, J. M., & Alexander, G. E. (1971). Neuron activity related to short-term memory. Science, 173, 652–654. Fuster, J. M., & Bauer, R. H. (1974). Visual short-term memory deficit from hypothermia of frontal cortex. Brain Research, 81, 393–400. Gabrieli, J. D. E., Singh, J., Stebbins, G. T., & Goetz, C. G. (1996). Reduced working memory span in Parkinson’s disease: Evidence for the role of a frontostriatal system in working and strategic memory. Neuropsychology, 10, 322–332. Gathercole, S. E., & Baddeley, A. D. (1989). Evaluation of the role of phonological STM in the development of vocabulary in children: A longitudinal study. Journal of Memory and Language, 28, 200–213. Gathercole, S. E., & Baddeley, A. D. (1993). Working memory and language. Hillsdale, NJ: Lawrence Erlbaum. Gathercole, S. E., Willis, C., Emslie, H., & Baddeley, A. (1992). Phonological memory and vocabulary development during the early school years: A longitudinal study. Developmental Psychology, 28, 887–898. Gilmore, R. O., & Johnson, M. H. (1995). Working memory in infancy: Six-month-olds’ performance on two versions of the oculomotor delayed response task. Journal of Experimental Child Psychology, 59, 397–418. Glassman, R. B., Leniek, K. M., & Haegerich, T. M. (1998). Human working memory capacity is 7 plus or minus 2 in a radial maze with distracting interruption: Possible implication for neural mechanisms of declarative and implicit long-term memory. Brain Research Bulletin, 47, 249–256. Goldman, P. S., & Rosvold, H. E. (1970). Localization of function within the dorsolateral prefrontal cortex of the rhesus monkey. Experimental Neurology, 27, 291–304. Goldman-Rakic, P. S. (1987). Circuitry of primate prefrontal cortex and regulation of behavior by representational memory. In F. Plum & V. Mountcastle (Eds.), Handbook of physiology (Vol. 5, pp. 373–417). Bethesda, MD: American Physiological Society. Goren, C. C., Sarty, M., & Wu, P. Y. K. (1975). Visual following and pattern discrimination of face-like stimuli by newborn infants. Pediatrics, 56, 544–549.

222


Gross, C. G., & Weiskrantz, L. (1964). Some changes in behavior produced by lateral frontal lesions in the macaque. In J. M. Warren & K. Akert (Eds.), The frontal granular cortex and behavior (pp. 74–98). New York: McGraw-Hill. Grossman, E., Donnelly, M., Price, R., Morgan, V., Pickens, D., Neighbor, G., & Blake, R. (2000). Brain areas involved in perception of biological motion. Journal of Cognitive Neuroscience, 12, 711–720. Harding, C. G., & Golinkoff, R. M. (1979). The origins of intentional vocalization in prelinguistic infants. Child Development, 50, 33–40. Harris, P. L. (1973). Perseverative errors in search by young infants. Child Development, 44, 29–33. Harris, P. L. (1974). Perseverative search at a visibly empty place by young infants. Journal of Experimental Child Psychology, 18, 535–542. Harris, P. L. (1987). The development of search. In P. Salapatek & L. B. Cohen (Eds.), Handbook of infant perception (Vol. 2, pp. 155–207). New York: Academic Press. Herschkowitz, N., Kagan, J., & Zilles, K. (1997). Neurobiological bases of behavioral development in the first year. Neuropediatrics, 28, 296–306. Hofstadter, M., & Reznick, J. S. (1996). Response modality affects human infant delayedresponse performance. Child Development, 67, 646–658. Hood, B. M., Carey, S., & Prasada, S. (2000). Do 2-year-olds predict physical events that young infants seem to know? Child Development, 71, 1540–1554. Hulme, C., Thomson, N., Muir, C., & Lawrence, A. L. (1984). Speech rate and the development of short-term memory span. Journal of Experimental Child Psychology, 38, 241–253. Hunter, W. S. (1913). The delayed reaction in animals and children. Behavior Monographs, 2, 1–86. Hunter, W. S. (1917). The delayed reaction in a child. Psychological Review, 24, 74–87. Huttenlocher, J., Newcombe, N., & Sandberg, E. (1994). The coding of spatial location in young children. Cognitive Psychology, 27, 115–147. Huttenlocher, P. R. (1979). Synaptic density in human frontal cortex: Developmental changes and effects of aging. Brain Research, 163, 195–205. Huttenlocher, P. R. (1990). Morphometric study of human cerebral cortex development. Neuropsychologia, 28, 517–527. Huttenlocher, P. R. (1994). Synaptogenesis, synapse elimination, and neural plasticity in human cerebral cortex. In C. A. Nelson (Ed.), Threats to optimal development: Integrating biological, psychological, and social risk factors (Minnesota Symposia on Child Psychology, Vol. 27, pp. 35–54). Hillsdale, NJ: Lawrence Erlbaum Associates. Huttenlocher, P. R., & Dabholkar, A. S. (1997). Developmental anatomy of prefrontal cortex. In N. A. Krasnegor, G. R. Lyon, P. S. Goldman-Rakic (Eds.), Development of the prefrontal cortex: Evolution, neurobiology, and behavior (pp. 69–83). Baltimore: Paul H. Brookes. Jackson, E., Campos, J., & Fischer, K. (1978). The question of decalage between object permanence and person permanence. Developmental Psychology, 14, 1–10. Jacobsen, C. F. (1936). Studies of cerebral function in primates. Comparative Psychology Monographs, 13, 1–68. James, W. (1890/1950). Principles of psychology (Vol. 1). New York: Holt, Rinehart, and Winston. Jha, A. P., & McCarthy, G. (2000). The influence of memory load upon delay interval activity in a working memory task: An event-related functional MRI Study. Journal of Cognitive Neuroscience, 12, 90–105.


223

Johnson, M. H. (1997). Developmental cognitive neuroscience: An introduction. Oxford, England: Blackwell. Johnson, M. H. (1998). The neural basis of cognitive development. In W. Damon (Series Ed.), D. Kuhn, & R. S. Siegler (Volume Eds.), Handbook of child psychology, Vol. 2: Cognition, perception, and language (5th ed., pp. 1–50). New York: John Wiley & Sons. Johnson, M. H. (2000). Functional brain development in infants: Elements of an interactive specialization framework. Child Development, 71, 75–81. Johnson, M. H., Dziurawic, S., Ellis, H., & Morton, J. (1991). Newborns’ preferential tracking of face-like stimuli and its subsequent decline. Cognition, 40, 1–19. Johnson, M. H., & Gilmore, R. (1996). Developmental cognitive neuroscience: A biological perspective on cognitive change. In R. Gelman & T. Kit-Fong Au (Eds.), Perceptual and cognitive development (pp. 333–372). New York: Academic Press. Jonides, J., Schumacher, E. H., Smith, E. E., Lauber, E. J., Awh, E., Minoshima, S., & Koeppe, R. A. (1997). Verbal working memory load affects regional brain activation as measured by PET. Journal of Cognitive Neuroscience, 9, 462–475. Kail, R. V. (1992). Processing speed, speech rate, and memory. Developmental Psychology, 28, 899–904. Kail, R., & Hall, L. K. (2001). Distinguishing short-term memory from working memory. Memory & Cognition, 29, 1–9. Kail, R. V., & Park, Y. (1994). Processing time, articulation time, and memory span. Journal of Experimental Child Psychology, 57, 281–291. Kagan, J. (1981). The second year. Cambridge, MA: Harvard University Press. Kagan, J., Kearsley, R., & Zelazo, P. R. (1978). Infancy: Its place in human development. Cambridge, MA: Harvard University Press. Klingberg, T., Forssberg, H., & Westerberg, H. (2002). Increased brain activity in frontal and parietal cortex underlies the development of visuospatial working memory capacity during childhood. Journal of Cognitive Neuroscience, 14, 1–10. Kubota, K., & Niki, H. (1971). Cortical unit activity and delayed alternation performance in monkeys. Journal of Neurophysiology, 34, 337–347. Legerstee, M. (1994). The role of familiarity and sound in the development of person and object permanence. British Journal of Developmental Psychology, 12, 455–468. MacDonald, S. E. (1994). Gorillas’ (Gorilla gorilla gorilla) spatial memory in a foraging task. Journal of Comparative Psychology, 108, 107–113. MacDonald, S. E., Pang, J. C., & Gibeault, S. (1994). Marmoset (Callithrix jacchus jacchus) spatial memory in a foraging task: Win-stay versus win-shift strategies. Journal of Comparative Psychology, 108, 328–335. MacDonald, S. E., & Wilkie, D. M. (1990). Yellow-nosed monkeys’ (Cercopithecus ascanius whitesidei) spatial memory in a simulated foraging environment. Journal of Comparative Psychology, 104, 382–387. Maier, N. R. F., & Schneirla, T. C. (1935). Principles of animal psychology. New York: McGraw-Hill. Marcovitch, S., & Zelazo, P. D. (1999). The A-not-B error: Results from a logistic metaanalysis. Child Development, 70, 1297–1313. Matthews, A., Ellis, A. E., & Nelson, C. A. (1996). Development of preterm and full-term infant ability on AB, recall memory, transparent barrier detour, and means-end tasks. Child Development, 67, 2658–2676. McCarthy, G., Blamire, A. M., Puce, A., Nobre, An C., Bloch, G., Hyder, F., Goldman-Rakic, P. S., & Schulman, R. G. (1994). Functional magnetic resonance imaging of human prefrontal cortex activation during a spatial working memory task. Proceedings of the National Academy of Science USA, 91, 8690–8694.

224


McDonough, L. (1999). Early declarative memory for location. British Journal of Developmental Psychology, 17, 381–402. McKenzie, B. E., Day, R. H., & Ihsen, E. (1984). Localization of events in space: Young infants are not always egocentric. British Journal of Developmental Psychology, 2, 1–9. Menzel, E. W. (1973). Chimpanzee spatial memory organization. Science, 182, 943–945. Meyer, D. H., Harlow, H. F., & Settlage, P. H. (1951). A survey of delayed response performance by normal and brain-damaged monkeys. Journal of Comparative Physiological Psychology, 44, 17–25. Miller, G. A., Galanter, E., & Pribram, K. H. (1960). Plans and the structure of behavior. New York: Holt, Rinehart, and Winston. Morton, J., & Johnson, M. H. (1991). CONSPEC and CONLERN: A two-process theory of infant face recognition. Psychological Review, 98, 164–181. Moscovitch, M. (1994). Memory and working with memory: Evaluation of a component process model and comparisons with other models. In D. L. Schacter & E. Tulving (Eds.), Memory systems (pp. 269–310). Cambridge, MA: MIT/Bradford Press. Mosier, C. E., & Rogoff, B. (1994). Infants’ instrumental use of their mothers to achieve their goals. Child Development, 65, 70–79. Munakata, Y. (1998). Infant perseveration and implications for object permanence theories: A PDP Model of the A-not-B task. Developmental Science, 1, 161–184. Munakata, Y., McClelland, J. L., Johnson, M. H., & Siegler. R. S. (1997). Rethinking infant knowledge: Toward an adaptive process account of successes and failures in object permanence tasks. Psychological Review, 104, 686–713. Nadel, L. (1992). Multiple memory systems: What and Why. Journal of Cognitive Neuroscience, 4, 179–188. Nairne, J. S. (2002). Remembering over the short-term: The case against the Standard Model. Annual Review of Psychology, 53, 53–81. Nelson, C. A. (1995). The ontogeny of human memory: A cognitive neuroscience perspective. Developmental Psychology, 31, 723–738. Nelson, C. A., & Luciana, M. (Eds.) (2001). Handbook of developmental cognitive neuroscience. Cambridge, MA: MIT Press. Nelson, C. A., Monk, C. S., Lin, J., Carver, L. J., Thomas, K. M., & Truwit, C. L. (2000). Functional neuroanatomy of spatial working memory in children. Developmental Psychology, 36, 109–116. Newcombe, N., Huttenlocher, J., Drummey, A. B., & Wiley, J. G. (1998). The development of spatial location coding: Place learning and dead reckoning in the second and third years. Cognitive Development, 13, 185–200. Newcombe, N., Huttenlocher, J., & Learmonth, A. (1999). Infants’ coding of location in continuous space. Infant Behavior and Development, 22, 483–510. Newport, E. L. (1991). Contrasting conceptions of the critical period for language (pp. 111–130). In S. Carey & R. Gelman (Eds.), The epigenesis of mind: Essays on biology and cognition. Hillsdale, NJ: Lawrence Erlbaum Associates. O’Connor, R. C., & Glassman, R. B. (1993). Human performance with a seventeen arm radial maze analog. Experimental Neurology, 125, 125–133. Olton, D. S. (1978). Characteristics of spatial memory. In S. H. Hulse, W. K. Honig, & H. Fowler (Eds.), Cognitive aspects of animal behavior (pp. 341–373). Hillsdale, NJ: Lawrence Erlbaum Associates. Olton, D. S., Collison, C., & Werz, M. A. (1977). Spatial memory and radial arm maze performance of rats. Learning and Motivation, 8, 289–314. Olton, D. S., & Papas, B. C. (1979). Spatial memory and hippocampal function. Neuropsychologia, 17, 669–682.


225

Olton, D. S., & Samuelson, R. J. (1976). Remembrance of places passed: Spatial memory in rats. Journal of Experimental Psychology: Animal Behavior Processes, 2, 97–116. Olton, D. S., Walker, J. A., & Gage, F. H. (1978). Hippocampal connections and spatial discrimination. Brain Research, 139, 295–308 Overman, W. H., Pate, B. J., Moore, K., & Peuster, A. (1996). Ontogeny of place learning in children as measured in the radial arm maze, Morris search task, and open field task. Behavioral Neuroscience, 110, 1205–1228. Pascual-Leone, J. (1970). A mathematical model for the transition rule in Piaget’s developmental stages. Acta Psychologia, 63, 301–345. Pelphrey, K. A., Reznick, J. S., Goldman, B. D., Sasson, N. J., Morrow, J., Donahoe, A., & Hodgson, K. (2003). Working memory in the first year. Unpublished manuscript. Pelphrey, K. A., Kroncke, A., Moore, T., Goldman, B. D., Morrow, J., Reznick, J. S. (2002, April). Toddler working memory: A foraging task. Poster presented at the annual meeting of the International Society for Infant Studies, Toronto, Canada. Pelphrey, K. A., Singerman, J. D., Allison, T., & McCarthy, G. (2003). Brain activation evoked by the perception of gaze shifts: The influence of context. Neuropsychologia, 41, 156–170. Pennington, B. F. (1994). The working memory function of the prefrontal cortices. In M. M. Haith, J. B. Benson, R. J. Roberts Jr., & B. F. Pennington (Eds.), The development of futureoriented processes (pp. 243–289). Chicago: University of Chicago Press. Pennington, B. F. (1997). Dimensions of executive function in normal and abnormal development. In N. A. Krasnegor, G. R. Lyon, P. S. Goldman-Rakic (Eds.), Development of the prefrontal cortex: Evolution, neurobiology, and behavior (pp. 265–281). Baltimore: Paul H. Brookes Publishing. Petrides, M. E., Alivisatos, B., Evans, A. C., & Meyer, E. (1993). Dissociation of human middorsolateral from posterior dorsolateral frontal cortex in memory processing. Proceedings of the National Academy of Science USA, 90, 873–877. Petrides, M. E. (1995). Impairments on nonspatial self-ordered and externally ordered working memory tasks after lesions of the mid-dorsal part of the lateral frontal cortex in the monkey. Journal of Neuroscience, 15, 359–375. Petrides, M. E. (1991). Monitoring of selections of visual stimuli and the primate frontal cortex. Proceedings of the Royal Society, London B, 246, 293–298. Piaget, J. (1954) [Original French edition, 1937]. The construction of reality in the child. New York: Basic. Pribram, K. H. (1950). Some physical and pharmacological factors affecting delayed response performance of baboons following frontal lobotomy. Journal of Neurophysiology, 13, 373–382. Puce, A., Allison, T., Bentin, S., Gore, J., & McCarthy, G. (1998). Temporal cortex activation in human subjects viewing eyes and mouth movements. Journal of Neuroscience, 18, 2188–2199. Reznick, J. S. (1999). Influences on maternal attribution of infant intentionality. In P. D. Zelazo & J. W. Astington (Eds.), Developing theories of intention: Social understanding and selfcontrol (pp. 243–267). Mahwah, NJ: Lawrence Erlbaum Associates. Reznick, J. S., Corley, R., & Robinson, J. (1997). A longitudinal twin study of intelligence in the second year. Monographs of the Society for Research in Child Development, 62 (1, Serial No. 249). Reznick, J. S., & Fieselman, S. D. (1999, April). A ‘‘foraging’’ task to assess infant working memory capacity. Poster presented at the meetings of the Society for Research in Child Development, Albuquerque, New Mexico. Reznick, J. S., Fueser, J. J., & Bosquet, M. (1998). Self-corrected reaching in a three-location delayed-response search task. Psychological Science, 9, 66–70.

226


Reznick, J. S., Morrow, J. D., Goldman, B. D., & Snyder, J. (2003). Onset of working memory in infants. Unpublished manuscript. Roediger, H. L. (1990). Implicit memory: retention without remembering. American Psychologist, 45, 1043–1056. Rose, S. A., Feldman, J. F., & Jankowski, J. J. (2001). Visual short-term memory in the first year of life: Capacity and recency effects. Developmental Psychology, 37, 539–549. Rosvold, H. E., Szwarcbart, M. K., Mirsky, A. F., & Mishkin, M. (1961). The effect of frontallobe damage on delayed response performance in chimpanzees. Journal of Comparative Physiological Psychology, 54, 368–374. Sawaguchi, T., Matsumura, M., & Kubota, K. (1988). Dopamine enhances the neuronal activity of a spatial short-term memory task in the primate prefrontal cortex. Neuroscience Research, 5, 465–473. Schacter, D. L., & Tulving, E. (1994). What are the memory systems of 1994? In D. L. Schacter & E. Tulving (Eds.), Memory systems (pp. 1–38). Cambridge, MA: MIT Press. Schneider, W., & Pressley, M. (1997). Memory development between two and twenty (2nd ed.). Mahwah, NJ: Erlbaum. Schwartz, B. B., & Reznick, J. S. (1999). Measuring infant spatial working memory with a windows and curtains delayed-response procedure. Memory, 7, 1–17. Sherry, D. F., & Schacter, D. L. (1987). The evolution of multiple memory systems. Psychological Review, 94, 439–454. Smith, E. E. (2000). Neural bases of human working memory. Current Directions in Psychological Science, 9, 45–49. Smith, E. E., & Jonides, J. (1998). Neuroimaging analyses of human working memory. Proceedings of the National Academy of Sciences USA, 95, 12061–12068. Smith, E. E., Jonides, J., Koeppe, R. A., Awh, E., Schumacher, E. H., & Minoshima, S. (1995). Spatial versus object working memory: PET investigations. Journal of Cognitive Neuroscience, 7, 337–356. Sophian, C. (1985). Perseveration and infants’ search: A comparison of 2- and 3-location tasks. Developmental Psychology, 21, 187–194. Sophian, C., & Yengo, L. (1985). Infants’ search for visible objects: Implications for the interpretation of early search errors. Journal of Experimental Child Psychology, 40, 260–278. Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the development of cognition and action. Cambridge, MA: Bradford Books/MIT Press Tulving, E. (1985). How many memory systems are there? American Psychologist, 40, 385–398. Ungerleider, L. G., & Haxby, J. V. (1994). ‘What’ and ‘where’ in the human brain. Current Opinion in Neurobiology, 4, 157–165. Webb, R. A., Massar, B., & Nadolny, T. (1972). Information and strategy in the young child’s search for hidden objects. Child Development, 43, 91–104. Webb, S. J., Monk, C. S., & Nelson, C. A. (2001). Mechanisms of postnatal neurobiological development: Implications for human development. Developmental Neuropsychology, 19, 147–171. Wellman, H., Cross, D., & Bartsch, K. (1987). Infant search and object permanence: A metaanalysis of the A-not-B error. Monographs of the Society for Research in Child Development, 51 (3, Serial No. 214). Wilcox, T., Nadel, L., & Rosser, R. (1996). Location memory in healthy preterm and full-term infants. Infant Behavior and Development, 19, 309–323. Willatts, P. (1984). Stages in the development of intentional search by young infants. Developmental Psychology, 20, 389–396.


227

Willatts, P. (1999). Development of means-end behavior in young infants: Pulling a support to retrieve a distant object. Developmental Psychology, 35, 651–667. Wilson, F. A. W., O’Scalaidhe, S. P., & Goldman-Rakic, P. S. (1993). Dissociation of object and spatial processing domains in primate prefrontal cortex. Science, 260, 1955–1958. Zeedyk, M. S. (1996). Developmental accounts of intentionality: Toward integration. Developmental Review, 16, 416–461. Zelazo, P. D., Carter, A., Reznick, J. S., & Frye, D. (1997). Early development of executive function: A problem-solving framework. General Psychology Review, 1, 198–226. Zelazo, P. D., Reznick, J. S., & Spinazzola, J. (1998). Representation flexibility and response control in a multi-step A-not-B search task. Developmental Psychology, 34, 203–214.


THE DEVELOPMENT OF A DIFFERENTIATED SENSE OF THE PAST AND THE FUTURE

William J. Friedman DEPARTMENT OF PSYCHOLOGY OBERLIN COLLEGE, OBERLIN, OHIO 44074

I. THE PAST AND FUTURE IN CHILDREN’S AND PARENTS’ LANGUAGE A. CHILDREN’S PRODUCTION AND COMPREHENSION B. PARENTS’ TALK ABOUT THE PAST AND THE FUTURE II. DIFFERENTIATION OF THE PAST AND OF THE FUTURE A. THE DEVELOPMENT OF A DIFFERENTIATED SENSE OF THE PAST B. THE DEVELOPMENT OF A DIFFERENTIATED SENSE OF THE FUTURE C. DIFFERENTIATION OF THE PAST AND OF THE FUTURE III. DIFFERENTIATION BETWEEN THE PAST AND THE FUTURE A. DIFFERENCES AND SIMILARITIES BETWEEN THE PAST AND THE FUTURE B. EVIDENCE OF PAST–FUTURE CONFUSION C. CONCEPTS OF THE PAST AND THE FUTURE IV. CONCLUSIONS AND DIRECTIONS FOR FUTURE RESEARCH A. CONCLUSIONS ABOUT CHILDREN’S SENSE OF THE PAST AND THE FUTURE B. CONCLUSIONS ABOUT COGNITIVE DEVELOPMENT C. FUTURE DIRECTIONS REFERENCES

Adults in modern societies view their lives as unfolding within the frameworks of the day, week, year, and other personal and social time patterns. We have a compelling sense of ‘‘where’’ we are in time and where past and future events are relative to the present. This sense of place in time is often referred to as ‘‘temporal orientation,’’ which can be defined, by analogy to spatial orientation (Kuipers, 1978), as ‘‘. . . the ability to determine the current time and relative times of other events with respect to some temporal framework’’ (Friedman, 1990b, p. 68). Although progress has been made in describing a number of aspects of children’s and adults’


229


230

William J. Friedman

experience of time, the process of temporal orientation remains poorly understood (Friedman, 1990b). In this chapter I focus on a part of this temporal orientation ability, our sense of the times of past and future events. I describe a series of studies on the development of the capacity to differentiate the times of events within the past and within the future and the ability to distinguish between these two categories of experience. By analyzing the development of children’s sense of the past and the future, it may be possible to disentangle some of the processes that contribute to adults’ sense of place in time. Physicists sometimes argue that the past–present–future distinction is a psychological illusion with no basis in the physical world (Teichmann, 2000). Instead, reality can be described as a set of before–after relations, with no special vantage point, the present. This way of describing time is quite alien to most adults in Western societies, who view the past–present– future distinction as a very basic aspect of the real world. But the distinction may be, at least in part, a social construction, rather than an objective feature of reality to be learned. Harner (1982a) analyzed the logical structure of this tripartite division in the Western view of time and pointed out that it poses substantial challenges to children’s understanding. Among these are the need to reconcile an unchanging order of events with constantly changing contents of the past, present, and future categories; and shifting bounds, with the present sometimes brief and other times extended. The likelihood that cultural learning plays a major role in this development is also supported by the observation that not all cultures make a sharp distinction between these categories of experience. According to Gell (2000), traditional societies blur the division, thinking instead of a ‘‘present coming out of a past and oriented to a future which are both included within it, as lags and anticipations. . .’’ (p. 262). This is like being ‘‘immersed in an evolving present, rather than passing from period to period according to an abstract scheme. . .’’ (p. 262). Nelson (1996) also argued for the social construction of the past–present–future distinction. In her view, tense and other features of language highlight for young children temporal relations that were previously unnoticed. If the past– present–future distinction is a social construction, we may be able to learn about the component processes by studying how Western children acquire the distinction. But a sense of the times of past and future events involves more than just assigning them to the two categories. The general view in modern societies is that distant future events are constantly drawing nearer and past events receding farther into the past. Furthermore, these distances are quite important: something that will happen tomorrow has a greater immediacy than something that will occur next year, just as an event

The Development of a Differentiated Sense of the Past and the Future

231

that just occurred has a different status from one that happened long ago. In modern societies schemes for achieving a highly differentiated sense of the past and of the future are provided by the clock and calendar systems, and children’s ability to differentiate the times of events within the past and within the future should be closely tied to learning these systems. These analyses suggest that the development of a sense of the times of nonpresent events is a matter of mastering linguistic and other conventional representations of time. However, developmental studies may also tell us about more fundamental ways of grasping the pastness or futurity of events and their distances within these categories. Basic cognitive and perceptual processes could contribute to the past–present–future distinctions as well. For example, perceptual and other characteristics of memories that are important in reality monitoring (Johnson et al., 1988) can distinguish them from events that are anticipated but have not yet been experienced. Basic cognitive processes may also permit some differentiation of times within the past and within the future. For example, automatic changes in the properties of memories with the passage of time could provide clues to the distances of events in the past. Children may even be able to distinguish near-future events and more distant events on the basis of the differential accessibility of memories of adults’ statements about upcoming events. Developmental studies can play an important part in elucidating the processes underlying adults’ sense of the past and the future. First, they allow us to examine the relation between the ages at which children learn about conventional representations of time and their ability to differentiate the times of life events. Second, developmental studies can shed light on the role of basic cognitive processes, because one can determine whether children are able to differentiate times before conventional systems are understood. In the following discussion, I summarize research on children’s differentiation of the past and the future. Section I is a review of studies by researchers who have investigated the acquisition of linguistic devices that mark temporal perspective. This section also includes a summary of work on references to past and future events in parent–child discourse. In the remaining sections, I describe other approaches to learning about the development of a sense of the past and the future. In Section II, I examine children’s differentiation of times within the past and within the future. In Section III, I focus on differentiation between the past and the future; I include studies of children’s ability to judge the past–future status of life events and studies of their understanding of the differences between the past and the future.

232

William J. Friedman

I. The Past and Future in Children’s and Parents’ Language A. CHILDREN’S PRODUCTION AND COMPREHENSION

An important source of information about children’s differentiation of the past and the future comes from studies of temporal language. Several authors have described the linguistic devices available to English-speaking children for indicating the relations between the time that speech takes place and other times (Bates, Elman, & Li, 1994; Harner, 1982a; Nelson, 1996; Weist, 1989), including tense and temporal adverbs (e.g., yesterday, tomorrow). Three principle methods have been used to learn about children’s understanding of these devices: measures of spontaneous use, tasks in which production of the forms is elicited, and comprehension tasks. Studies of language production show that by the second or third year of life children refer to events that are objectively in the past and in the future (Harner, 1982a; Nelson, 1996; Sachs, 1983; Weist, 1989). Depending on the particular language children learn, different tense or aspect forms begin to be used to refer to past and future events late in the second year or early in the third year of life. Reference to past and future times through tense precedes children’s use of temporal adverbs, and when adverbs like yesterday, last night, or this afternoon are produced by 2- and 3-year-olds they are seldom used correctly. In general, it is difficult to draw strong conclusions about children’s ability to differentiate the past and the future from the studies of spontaneous production. On the one hand, measures of tense use could overestimate knowledge of pastness or futurity if children often respond to tense cues in the preceding adult utterances. On the other hand, children may make temporal distinctions before they mark them in particular speech forms, leading to an underestimation of their understanding. This is especially likely if individual utterances, rather than connected discourse, are examined (Nelson, 1996). Another problem with speech samples is that it is very difficult to estimate the overall accuracy of children’s use. Examples of correct use do not tell us about the frequency of absent temporal markers in contexts in which they are obligatory for adults. In other studies researchers have measured the accuracy of children’s tense use by eliciting the speech forms. In these studies, children witnessed actions, including some that were completed and others that were impending, and were asked to describe them. Weist et al. (1984), who cued the appropriate tense in the question forms (e.g., ‘‘What was the boy doing first?’’), found that by 2½ years of age 92% and 66% of children in their Polish sample produced the past and future tenses, respectively, in the appropriate contexts. Somewhat lower percentages of English-speaking


233

3½-year-olds (the youngest age group tested) in Harner’s (1981) study produced the past and the future forms, but they did so in response to tenseneutral questions (e.g., ‘‘Tell me about this one.’’). The specificity of children’s productions in these studies shows that they distinguish between actions that have just occurred and those that are about to happen. Unfortunately, this method does not tell us about the ability to distinguish past and future events on longer timescales. A third method involves assessing the accuracy of children’s comprehension of past- and future-referring language (Harner, 1975, 1976, 1980, 1982b; Weist, 1983). I will focus on those studies in which the referent actions are in the real past or future relative to the testing time. Collectively, these studies have included three different linguistic devices to cue past and future times: before vs. after (e.g., ‘‘before this day’’), yesterday vs. tomorrow, and tense or aspect (e.g., ‘‘a toy you will play with’’). If we consider the linguistic devices that show the most accurate performance, 3-year-olds sometimes perform at levels significantly above chance and 4-year-olds are quite accurate. In addition, young children are more accurate when the future forms refer to the immediate future than the next day, whereas past forms are about equally accurate when referring to the previous day as the previous moment. Apparently, children first understand future-referring language when reference is to the immediate future. The comprehension studies, like those using spontaneous and elicited production, indicate that young children associate linguistic markers of pastness and futurity with times that adults consider to be past and future. But they may be limited in their ability to tell us about children’s capacity to discriminate the past versus future status of everyday events. The comprehension tasks were designed to measure linguistic rather than cognitive competence, and for this reason used unusually supportive contexts (including other language forms to cue children that toys were played with yesterday or would be played with tomorrow). Conclusions must also be limited to the brief timescales that were used in these studies, up to just one day in the past and the future. B. PARENTS’ TALK ABOUT THE PAST AND THE FUTURE

A number of studies of children’s temporal language production also provide information about parents’ conversations with young children about past and future events (Hudson, 2001, 2002; Lucariello & Nelson, 1987; Nelson, 1989; Nelson, 1996; Sachs, 1983; and see Benson, 1994; Benson, Talmi, & Haith, 2003 (in press), and Reese & Fivush, 1993). Such conversations must be the main source of information that young children

234

William J. Friedman

have about future events, and they are probably very important as well for the formation of autobiographical memories in young children (Nelson, 1993). Unfortunately, because the studies were designed for other purposes, their findings do not answer a number of important questions about the information provided to children that might aid their differentiation of times within the past and within the future. However, several general conclusions can be drawn. Perhaps most important, although most speech to young children focuses on the here-and-now, at least by the third year of life children hear references to events that have happened in the past and those that are anticipated in the future. Of course, parents use appropriate tense in these utterances, and they also make frequent use of temporal adverbs, such as tomorrow, last night, the other day, and later, even in speech to 2-yearolds. Samples of parent speech to young children also include references to remote birthdays, seasons, and holidays. Parents’ future references are usually to events that will occur in the immediate future, whereas statements about the past often refer to events ranging from earlier the same day to months in the past (findings that parallel the pattern shown in order of children’s productions). However, remote future events are discussed with young children, and, when they are, parents appear to show the same degree of elaboration about the events as for past events. Children initially find it more difficult to participate in discussions about remote future events than past events but by 4 years of age become active in discussions of both. The information conveyed by parents in the course of these discussions probably allows both past events (for which the child may also have specific memories) and future events (which are not yet anchored in direct experience) to become clusters of information in long-term memory. Parent–child discourse appears to be a sufficient foundation for children to acquire, by the end of early childhood, a basic grasp of the past–future distinction and an understanding that there are remote times, such as holidays and birthdays, which have names. But these references to specific times are presented in such a fragmentary way that it is difficult to see how young children could make any progress in working out the meaning of the time systems of which they are a part or use them to differentiate times within the past and within the future. Another potential source of information that could help children differentiate times within these categories (and one I have assumed in the following sections to be present) is adults’ statements about the distances of events in the past and the future. Parents and preschool teachers may well sometimes refer to distances in global terms, for example explaining that an event will happen soon or in a long time, or was recent or a long time ago. Unfortunately, there have been no studies that address this issue.


235

II. Differentiation of the Past and of the Future The studies of child language indicate that by about 4 years children distinguish the past and the future status of some events, at least when timescales are brief, and adults’ speech supports the distinction. But there is no clear indication in the literature on language development that young children differentiate times within these categories. What little information we have comes from young children’s production of temporal adverbs, such as yesterday and tomorrow, and the frequent errors in such use (e.g., Ames, 1946; Harner, 1982a; Nelson, 1996; Sachs, 1983; Weist, 1989) suggest a lack of differentiation among the times of events that happened in the past and among those that are expected in the future. In this section, I describe a series of studies using other methods for learning about the development of a differentiated sense of the past and of the future.

A. THE DEVELOPMENT OF A DIFFERENTIATED SENSE OF THE PAST

1. Adults’ Memory for the Times of Past Events Conventional time systems provide members of modern societies with ways of reckoning the times of past events with great precision. When written records or other sources of precise information are available, the possibility of differentiating past times relative to the present is unlimited. In everyday thinking about when past events occurred, however, records are usually not easily available, and we rely on other ways of recovering their times. Cognitive psychologists have posited a number of ways in which remembered events can be related to time and conducted a substantial number of studies to test these theories. In a review of this literature (Friedman, 1993), I distinguished three main types of information that could be used to remember the times of past events. First, the times could be gauged as temporal distances from the present, as in theories that appeal to the strength or vividness of memory traces. Changes in the memory that occur with the passage of time provide clues to the ages of the memories. Second, remembered events could be linked to locations in natural, personal or conventional time patterns, such as parts of a day or year or the period of time when one was in college. For example, in reconstructive theories the times of past events are judged by retrieving whatever information is associated with the event in memory and, where possible, making temporal inferences based on one’s general knowledge about time patterns. A third, logically independent, type of temporal information in memory is the order of two or more remembered events. According to order-code theories, the

236

William J. Friedman

order of some pairs of events is stored when the earlier event is retrieved at the time when the later one occurs. In my 1993 review, I discovered considerable support for the importance of location-based processes in adults’ memory for the times of past events, but little evidence for distance-based theories, and a number of findings that contradict their predictions. However, subsequent research (reviewed in Friedman, 1996, 2001a) has shown that information about distances also plays some role in humans’ sense of the times of past events. Adults use impressions of the ages of memories when location information is unavailable or when judgments must be made very rapidly. The contribution of the third type of temporal information, order, is supported by a small set of studies in which, as predicted, adults are more accurate in judging the order of semantically related than unrelated pairs of items (e.g., Tzeng & Cotton, 1980; Winograd & Soloway, 1985). 2. Developmental Studies In the 1970s and 1980s, a number of researchers tested children with methods similar to those used in laboratory studies of adults (Brown, 1973; Jackson, Michon, & Vermeeren, 1984; Mathews & Fozard, 1970; von Wright, 1973). In a single session, children were shown a sequence of pictures and periodically asked which of a pair of test items had been presented more recently. Most of the studies showed age improvements in accuracy between about 5 and 12 years, but some demonstrated that children as young as 5 years can discriminate recency at levels greater than those expected by chance. Similar skill is shown when items are much more widely separated in time. In a study by McCormack and Russell (1997), one set of pictures was shown initially and a second set on the following day. Immediately following the presentation of the second set, children judged whether pictures had been presented ‘‘today’’ or ‘‘yesterday.’’ Even 4-yearolds were 90% correct. From this and the earlier studies, by early childhood children clearly have access to information about the times of past events. In a number of studies, I have investigated children’s ability to differentiate the times of past events that are distributed over much longer intervals of time. The distinction between distance and location processes was tested in a first set of three experiments, in which children of 4, 6, and 8 years compared the recency of a 1- and a 7-week-old event and they estimated the time of occurrence of the older event on several different timescales (Friedman, 1991). The two events to be compared were very different in nature (e.g., a visitor videotaping the class at recess and the regular teacher lecturing the children about proper tooth-brushing technique), so it was very unlikely that order codes would have been formed when the second event occurred. The results repeated, with much longer


237

retention intervals, the findings of earlier studies: even 4-year-olds reliably discriminated which of the two events had occurred a longer time ago. The study also revealed substantial increases from 4 to 8 years in the ability to reconstruct the time when the older event had occurred, at least on month and season scales; 4-year-olds were at chance levels on these long timescales. The 4-year-olds’ success on the recency discrimination task, coupled with their inability to place the events in long-scale time patterns, supports the distinction between distance and location processes in memory for time. Furthermore, distance-based processes are present by early childhood, whereas the ability to reconstruct the temporal locations of remembered events changes substantially from early to middle childhood. Although the focus of this chapter is on the past and the future, also worth mentioning are findings from this study concerning another component of temporal orientation, children’s ability to identify the present time. In one of the experiments in the first study (Friedman, 1991, Experiment 3), children were asked what day and what month it is now, and they judged the time of day and season by pointing to a position along a row of four cards representing the four seasons and the daily activities waking, eating lunch, eating dinner, and going to bed. Children in the youngest group, 4-year-olds, were inaccurate on the season, month, and day-of-theweek scales. However, 95% of them were able to indicate that it was morning, the time that their nursery class met, by pointing to one or the other of the cards representing the nearest events, wake and lunch, or to the space in between them, a pattern that is far more accurate than chance responding would produce. Children in the older groups, first and third graders, were more accurate than would be expected by chance on all of the timescales. From these findings children evidently have a sense of the present time within the briefest cycle, parts of a day, by early childhood, and by about 6–7 years most children can report the current day, month, and season. In a second study (Friedman, 1992) on children’s memory for time, I found different developmental levels at which the locations of autobiographical events are understood. Children from 4 to 9 years of age were asked to retrieve memories from specific temporal locations in the past, including yesterday, last weekend, last summer, and a number of holidays from the past year. Four- and 5-year-olds were able to produce accurate memories for most of these locations. Many recollections were specific to the most recent weekend, summer, or holiday, suggesting that the locations were differentiated from one another and that updating processes are present at these ages. However, the ability to retrieve memories when cued by the name of a location does not imply that children know about the relative times of the locations. Only older children were able to order a set of cards representing the locations, and even the 8- and 9-year-olds had

238

William J. Friedman

difficulty judging which of a pair was a longer time ago. These findings indicate that locations are initially like isolated ‘‘islands of time.’’ Only during middle childhood do children develop representations of long-scale time patterns that allow them to achieve an integrated view of the times when past locations occurred. A third set of experiments (Friedman, Gardner, & Zubin, 1995) was designed to provide information about distance-based processes in children’s memory for the times of events that had occurred from the past weeks to nearly a year ago. Children from 3 to 12 years of age were asked to compare the recency of their birthday and a holiday. As in the Friedman (1992) study, errors were common through 9 years of age, showing that using locationbased processes to compare the recency of two events is difficult well into middle childhood. However, even children under 6 years of age discriminated the recency of pairs of events when the events were widely separated and one of them fell within the past several months. Taken together with the results of the Friedman (1991) study and a study of adults’ distance-based memory for time (Friedman & Huttenlocher, 1997), these findings indicate that the ability to discriminate recency on spans of time from the past weeks to many months in the past is a basic property of human memory that changes little with development, at least from the preschool years on. One other finding of this study was that children under 6 years of age showed a strong tendency to judge their birthday as a shorter time ago when it was actually only a few weeks in the future. This error seems to indicate confusion between the past and the future, a phenomenon that is explored in Section III.B. In a final developmental study, Friedman and Kemp (1998) investigated the function relating the subjective distances of events in the past to their objective ages. We also tested whether retrieving information about one of a pair of events prior to comparing their recency would decrease the subjective age of that event. Two manipulations, retrieval of details about an event and priming, were designed to evaluate strength theories of memory for time, according to which the weakening of trace strengths with the passage of time provides the information used to make distance-based judgments. Children from 4 to 7 years of age compared the distances of pairs of events or placed events on a continuum representing distances in the past. In two experiments these manipulations failed to influence the subjective recency of events. The failure of experimental manipulations to influence subjective recency is consistent with McCormack and Russell’s (1997) finding that repetition did not cause children to judge the presentation of a picture to be more recent. In both cases the source of information underlying impressions of the ages of memories apparently is not a unitary strength-based quantity, in which the characteristics of a memory that influence probability of recall also influence its subjective recency.


239

Friedman and Kemp’s (1998) experiments also provided information about the other issue—how subjective ages of events change with the passage of time. In one experiment, children compared pairs of events from the last year, as in Friedman, Gardner, and Zubin’s (1995) study. In the two others, a scaling method was used: children were asked to judge the times of individual events using a spatial representation of distances in the past (a kind of ruler, with recent events to be placed near the closer end and events from long ago to be placed near the far end). Even 4-year-olds were able to differentiate past times with this spatial representation, just as they were able to differentiate distances in the past using the pair-comparison procedure. The results of the pair-comparison and spatial-representation experiments also showed that subjective distance is a nonlinear function of objective distance. Plots from the scaling studies were similar to those from another study with adults (Friedman & Huttenlocher, 1997): all indicate that subjective distance increases at a decelerating rate with increases in the ages of memories and follows a power function with an exponent between about .20 and .35. This means that the ratio of two distances in the past is a strong determinant of the likelihood that recency discrimination will be accurate when children (and adults) rely on distance-based processes. Finally, the two scaling studies replicated Friedman, Gardner, and Zubin’s (1995) finding that children under about 6 years of age often judge events expected to occur in the coming weeks as belonging to the recent past. 3. Conclusions The studies summarized in this section provide a number of pieces of information about the development of a differentiated sense of the past that were not available from the studies of language development. First, even children as young as 4-year-olds have some sense of events being different distances in the past. The laboratory studies show differentiation of events from up to one day in the past, and my studies of holidays and life events show differentiation of the times of events that range up to many months in the past. However, young children’s ability to differentiate past times is quite limited. When the temporal distance of the nearer of two events to be compared is a small ratio of the distance to the farther event’s (e.g., 1 month vs. 9 months), preschool children are more accurate than would be expected by chance. But when the ratio is large (e.g., 8 vs. 9 months), performance falls to chance levels. Success on these tasks is not easily explained by the use of location-based processes or order codes, so one can conclude that distance-based processes are present by early childhood. Furthermore, the basic properties of these processes seem to change little with development. In contrast, a number of findings on children’s use of temporal locations show that young children’s sense of the past is quite different from adults’.

240

William J. Friedman

Most children cannot use conventional locations, such as the months of events, to judge relative distances in the past until at least 10 years of age. Through 8 and 9 years of age, children have difficulty judging which of two holidays was more recent, even when they know the months of the two events. The ability to use long-scale time patterns to judge distances in the past may depend on the development of image representations of these patterns (Friedman, 1986, 1989), which usually occurs between late middlechildhood and adolescence. Locations do play some role in children’s sense of the past from about 4 years of age onward. Typically, 4-year-olds use their knowledge of daily routines to infer that an event that had taken place at nursery school must have been in the morning. They can also recall things that happened at the locations yesterday, last weekend, and last summer. However, these locations are like ‘‘islands in time’’ for young children; evidently they do not form parts of a structured past. This developmental pattern suggests that adult’s sense of the past depends on both basic cognitive processes and representations of conventional time patterns. Basic memory processes provide clues to the ages of memories, and although this information is quite limited, it is sometimes sufficient to distinguish different distances in the past. Adults probably use this information as part of a back-up strategy, when no information about locations is available or they must respond rapidly (Friedman, 1996, 2001a). Location-based processes play a more important role in a mature sense of the past. Information recalled about an event can be used to reconstruct when, in conventional or other time patterns, it must have occurred. These processes depend on formal learning of cultural systems for reckoning time and the construction of effective representations by individual children (Friedman, 1986). They also involve control processes that allow the integration of remembered aspects of an event with general time knowledge. B. THE DEVELOPMENT OF A DIFFERENTIATED SENSE OF THE FUTURE

Of course, we cannot have memories of events that will occur in the future, which would seem to imply that a differentiated sense of the future must depend entirely on learning about time patterns. However, a number of simpler processes are available to young children that could make some anticipated events seem near (or far). We have seen in Section I.B. that children hear future-referring speech from an early age. If parents and other adults refer repeatedly to a future event, the strength of memory for such references may make that event seem near. Children may also hear and remember adults’ statements that an event of interest (e.g., one’s birthday or Christmas) is coming soon or is a long time away. The child’s own repeated


241

thoughts of an anticipated event can also contribute to its subjective proximity. Finally, there may be perceptual cues to approaching events, such as holiday decorations. For these reasons it is important to search for the origins of a differentiated sense of the future in early childhood. There has been surprisingly little research on children’s knowledge of the times of future events (Haith et al., 1994; Moore & Lemmon, 2001; but see Nurmi, 1991, for a review of the substantial literature on future orientation in adolescents). Most of what is known about pre-adolescents’ knowledge of the future comes from studies of infants’ anticipation in perceptual–motor tasks, measures of the ability to delay gratification, and tasks in which children are asked to plan familiar activities (Haith et al., 1994; Hudson, Shapiro, & Sosa, 1995; Thompson, Barresi, & Moore, 1997). Unfortunately, this research does not provide information about the development of children’s understanding of the times of future events beyond the next day or so, leaving a gap in our knowledge between the short timescales studied in infants and young children and the very long timescales examined in studies of adolescents’ future orientation. The one exception is a study of children’s differentiation of the times of daily and of annual events by Silverman (1996). In four separate tasks, one for each timescale and one each for the past and the future, children from grades 1 through 4 were asked to place cards on continua in which a horizontal line represented the flow of time, and a short vertical intersecting line represented the present. The past was the portion of the line to the left of the present marker, and the future was the portion to the right. In each task they were given five cards representing events on a scale (e.g., school lunch, or the child’s birthday) and asked to place them all where they belong on a Velcro strip covering the relevant half of the timeline. The results showed that even in the oldest group just under half of children were correct for the future times for the annual events. Surprisingly, performance on the briefer timescale, the daily events, was even poorer, and children were more accurate in placing future than past events. The pattern of errors suggests that many children did not use the present as a reference point, as they were expected to. Instead, they seemed to focus on when the events occur relative to one another. This pattern may have been induced by the requirement to place all of the cards at the same time. In the remainder of this section, I summarize my research on the development of a differentiated sense of the future from early to middle childhood. Like Silverman (1996) and Friedman and Kemp (1998), I used a spatial representation of time to learn about children’s ability to differentiate the times of events. In the studies summarized here, children were asked to judge the future distances of events by pointing to places on a picture of a road that began near the viewer and receded toward distant

242

William J. Friedman

mountains (Friedman, 2000, Figure 1). The first series of studies focused on events ranging from the coming days and weeks to those nearly a year in the future. The second set of studies was designed to shed light on children’s sense of future distances on a shorter timescale, the day. In a third set of studies, children judged the future distances of both daily activities and annual events. A comparison of judgments for events on the two timescales can show whether a differentiated sense of the future is a monolithic achievement or instead depends on specific representations available for particular time patterns. 1. Annual Events In the first study (Friedman, 2000, Study 1), 4-, 7-, and 10-year-old children were tested several weeks before and several weeks after Hallowe’en. Children judged the future distances of the events: dinner, Saturday, Hallowe’en, Thanksgiving, Christmas, Valentine’s Day, and summer. On individual trials they were asked to point to a spot on the road to indicate whether it was ‘‘very soon, a long time from now, or an in-between amount of time from now.’’ For most of these events, children were also asked the ?’’ and the yes–no question open-ended question, ‘‘How long is it until coming soon?’’ Finally, children were asked to report things that ‘‘Is would happen soon and things that would not happen for a long time. The results showed marked changes in children’s differentiation of the future during this age range. Four-year-olds’ judgments were largely undifferentiated. They gave about the same ratings on the picture-pointing task of events that would occur within the next few weeks as events that were many months in the future, and they showed a similar lack of differentiation on most of the verbal tasks. The only evidence that children of this age have access to information about the times of future events came from the questions asking for events that were coming soon and would not occur for a long time. About half the 4-year-olds responded to the comingsoon question with a holiday that actually would occur within about the next 2 months, and many also reported events that would not happen for a long time. This suggests that young children can retrieve from memory propositions that some events are near and some are distant. However, they apparently had difficulty using this categorical information on the picturepointing task and the other verbal tasks in this study. Like the 4-year-olds, 7-year-olds seemed to depend on categorical information about the times of future events. However, they appeared to have a greater amount of, or easier access to, such information. On the picture-pointing task, they distinguished near and far distance categories. For example, at the first testing time (a few weeks before Hallowe’en), dinner, Saturday, and Hallowe’en were given low ratings and Thanksgiving,


243

Christmas, Valentine’s Day and summer high ratings, with no significant differentiation within either category. Further evidence that the 7-year-olds relied on categorical information came from their answers to the questions, ?’’ For most questions half or more gave global ‘‘How long is it until descriptions, such as ‘‘a short time’’ or ‘‘a long time.’’ However, there were interesting exceptions: about half reported an approximately accurate number of days until the weekend, and many could give the number of months until their birthday with some accuracy (but only if their birthday was within the next 2 months). The results of the weekend question indicate that many 7-year-olds have access to future distances on this scale using their emerging representations of the days of the week (Friedman, 1986). This suggests that a differentiated sense of the future may emerge at different ages for different time patterns, an issue I will return to later in this section. Answers to the birthday question, however, probably reflect memory for parents’ statements that one’s birthday is a particular number of months away. The lack of accuracy for distant birthdays may be attributable to parents seldom mentioning the specific distances of birthdays when they are far in the future. In contrast to the 4- and 7-year-olds, 10-year-olds showed a highly differentiated sense of the future by all of the measures. Even their responses to the picture-pointing task, where the representation is novel, the units are arbitrary, and the regions were only described in general terms (e.g., ‘‘very soon’’) revealed a clear linear increase in distance ratings for events ranging from the coming days to those up to about one year in the future. It was evident from their responses to the ‘‘how-long’’ questions that 10-year-olds are able to judge future distances using their knowledge of the order of the months. Nearly all responded with distances in conventional units, and these responses were quite accurate. Thus, by about 10 years of age children’s sense of future distances within the year fundamentally resembles adults’. A follow-up study (Friedman, 2000, Study 2) was conducted to provide more information about the development of differentiation between 4 and 7 years of age. The picture-pointing task was administered to 4-, 5-, and 6-year-olds approximately a week before or a week after Valentine’s Day; children were asked to judge the future distances of this holiday, summer, Hallowe’en, Christmas, and their birthday. Again, the 4-year-olds failed to differentiate the future distances of events. However, 5-year-olds did distinguish two categories: Valentine’s Day (both before and after the holiday actually occurred) and summer were assigned to the relatively near-future and Hallowe’en and Christmas to the more distant future. The 6-year-olds’ performance was more advanced: they distinguished three future-distance categories, and they overcame 5-year-olds’ error of assigning Valentine’s Day to the near-future one week after the holiday had occurred.

244

William J. Friedman

Like the 7-year-olds in the previous study, however, even this oldest group appeared to rely on the retrieval of propositions to judge distances in the future. This conclusion is supported by the absence of differentiation beyond about the next several months on the picture-pointing task and by the fact that they seldom answered the ‘‘how-long’’ questions using conventional units such as months. The past–future confusion shown in 5-year-olds’ judgments of Valentine’s Day echoes the similar confusion shown in the studies of children’s sense of the past. In another study of children’s sense of the future distances of annual events (Friedman, 2000, Study 4), 8- to 10-year-olds judged future distances using the picture-pointing task, and they were also assessed on their general knowledge of the order of annual events. The inclusion of both the futuredistance task and tasks tapping knowledge of the order of the events from other reference times within the year made it possible to test the hypothesis that general representations of the order of annual events are responsible for the highly differentiated judgments that emerge in middle childhood. I correlated indices of future differentiation on the picture-pointing task with measures of children’s ability to judge the order of months, the ability to produce any correct order of summer and three holidays (linear order), and the ability to judge the order of summer and three holidays from various reference points within the year (relative order). An example of the latter task was asking the child to pretend it is summer and to tell whether Hallowe’en or Valentine’s Day would come next. Errors were common on all of the order tasks, showing that many of these 8-, 9-, and 10-year-olds were still in the process mastering the orders of the annual events and months. The most important finding was that the ability to judge the order of annual events from different reference points was significantly related to the degree of differentiation shown on the picture-pointing task (r ¼ .62). This supports the hypothesis that the acquisition of general representations of the times of annual events underlies the development of a highly differentiated sense of the future distances of these events. Furthermore, the correlation was more closely tied to the ability to judge the order of annual events from multiple reference points than the ability to produce a correct order from a single reference point that the child chose. This finding is consistent with the idea that judging the future distances of events relies on the specific ability to think about the times of events in a pattern in a flexible way, adopting various reference points. 2. Daily Activities Although a single word is used to refer to the future, and we often think of the future as a single line stretching from the present into the distant future, our sense of the future may actually be the product of different processes on


245

different timescales. This is especially likely if representations of time patterns play a prominent role in a differentiated sense of the future. The representations that underlie our knowledge of the parts of different time patterns (such as parts of the day, week, and year) differ in their nature and develop at different ages (Friedman, 1986, 1989, 1990a). For example, children can correctly order cards representing the main events within the waking day at substantially earlier ages than they can order cards representing holidays or seasons (Friedman, 1977, 1986, 1990a). If representations are important in our sense of the future, then an understanding of a differentiated sense of the future depends on studying additional contents. For this reason I conducted a number of studies of children’s knowledge of the future distances of daily activities. This content differs from annual events in a number of important ways. Obviously, children experience many more repetitions of the day than the year, and the spans of time included in the patterns are tremendously different. Both of these facts may contribute to children’s earlier development of representations of this pattern than of the week or year. The briefer timescale of the day also means that whatever information is used to judge future times must be updated much more frequently. Another difference is that because daily activities are, by definition, routine, on most days children probably do not hear references to approaching activities, such as dinner, until shortly before they occur. For this reason memory for adults’ statements about the future distances of events may play a less important role in young children’s sense of the future distances of daily activities than of annual events. Finally, although the day and year are both formally cyclic, the pattern of daily activities has a psychological breakpoint, the night, and the day is far more likely to be experienced as a linear order (from waking to going to bed at night). If children’s representations strongly incorporate this feature, they may find it difficult to judge the relative distances of daily activities from the vantage of the present: a tendency to think of distances from the start of the day might interfere. In the first of two studies (Friedman, 2001b), 4- and 5-year-old children were asked to judge the future distances of breakfast, lunch, dinner, taking a bath, and going to bed using the same picture-pointing task employed in the studies of annual events. Most children were tested in the morning, and I focus on their results. However, it is worth noting that children tested before and after lunch did not differ significantly in their future-distance ratings of lunch, suggesting the same sort of past–future confusion as 4- and 5-yearolds showed in the studies of annual events. The findings of main interest concern children’s ability to differentiate the future distances of daily activities from the vantage of the present. To tap this ability, children were separated into groups who apparently adopted a present reference time and

246

William J. Friedman

those who ignored the present reference and used distance from the start of the day. My approach was to divide children according to whether their score for the waking stimulus was below the midpoint of the scale, 12.5—the morning-reference group—or above the midpoint—the present-reference group. For children tested at school in the morning, waking is among the most distant daily activities if the present is used as the reference time. But if children adopt the start of the day as their vantage time, waking should receive their lowest rating. About half of the children fell in each group. The results for the morning-reference group show a general pattern of events’ means increasing from earlier to later in the day, with a leveling off after dinner. In contrast, children in the present-reference group tended to judge events that were more remote from the present to be more distant in the future, although they substantially underestimated the distance of the most recent activity, breakfast. The anomalous value for breakfast could be due to children’s confusion of the near-future and the recent past, a phenomenon noted earlier. The different patterns found for the morningand present-reference groups suggest that children’s representations of the pattern of daily activities can contribute to or interfere with the ability to accurately judge the distances of events in the future. The pattern in the morning-reference group appears to reflect a strong tendency to think about the times of daily activities from the vantage of the morning, a pattern also found in Silverman’s (1996) study. As we will see in a later study, in which measures were taken to counter this approach, the morning-reference response pattern is not easily eliminated in young children. In contrast, the pattern produced by children in the present-reference group indicates that many young children have access to information about the relative distances of daily activities from the vantage of the present. Because adults’ references to remote future activities probably cannot account for children’s updated temporal perspective, children in this group probably are using general representations of the pattern of daily activities. To solve this task, they probably need to be able to think about the times of daily activities from multiple vantages within the day, rather than from a fixed reference point. The next study tested the hypothesis that such flexibility in daily-activities representations underlies the ability to differentiate their future distances. I selected an age group in which, according to past findings (Friedman, 1990a), children were likely to be developing and solidifying the ability to think about the pattern of daily activities in flexible ways. In this study, 6-, 7-, and 8-year-olds were tested on the same procedure, and they were separated into morning- and present-reference groups according to the same criteria. In addition, children performed two tests of knowledge of the order of the activities that did not involve the present perspective. In one they were asked to order a set of cards from a starting point chosen by the


247

child (the linear-order task), and in the other they judged which of a pair of activities would come before the other from varying reference points specified by the experimenter (the relative-order task). These tasks are similar to the ones used in a study described earlier of children’s sense of the future distances of annual events. As in that study, I was interested in determining whether the ability to differentiate future times from the perspective of the present is related specifically to judgments based on general representations of the relative times of events in a time pattern. The morning-reference group showed a linear increase in their ratings of events according to their distances from the start of the day, just as one would expect if they ignored the present vantage point and used the beginning of the day as their reference point. Clearly, the tendency to think of times within the day from this privileged perspective is not limited to young children. In contrast, children in the present-reference group produced a generally linear pattern from the present. The 6-, 7-, and 8-year-olds in this group showed an even more differentiated sense of the future distances of daily activities from the vantage of the present than the present-reference group in the previous study. However, as for the presentreference group in that younger sample, the mean for breakfast was less than that for waking. The extent of this underestimation was less in the present sample than was the case with the 4- and 5-year-olds in the previous study (Friedman, 2001b), but the recent pastness of breakfast still seems to influence children’s judgments. The other main analyses were conducted to determine whether differentiation of future-distance judgments from the present is related to measures of children’s knowledge of the order of daily activities. Accuracy on the picture-pointing task was significantly correlated with accuracy on the relative-order task (r ¼ .28) but not the linear-order task, a pattern echoing the one found for the annual events in the earlier study. The significant correlation was even more specific, though: it was due to those relative-order problems where the correct answer lay across the night boundary (r ¼ .33) from the reference point in the problem, but not the remaining eight problems (r ¼ .13). Interestingly, the morning- and presentreference groups did not differ in their ability to judge relative order across the night boundary. This indicates that two separate problems often stand in the way of children’s ability to differentiate the future distances of daily activities. One is a strong tendency to think about the times from the perspective of the start of the day, and the other is difficulty contending with the continuity between one waking day and the next. In general, the correlational findings confirm that the development of representations of the pattern of daily activities is responsible for children’s ability to differentiate the future distances of events on this timescale.

248

William J. Friedman

3. Comparing Daily Activities and Annual Events A comparison of the studies focusing on annual events with those testing differentiation of the future distances of daily activities suggests that young children have a more differentiated sense of the future on the scale of the day than the year. It also indicates that different processes may underlie early differentiation on the two timescales. In two additional studies (Friedman, 2002), children were tested on both contents, making a more direct comparison possible. In addition, I modified the picture-pointing task to address the argument that the morning-reference pattern is really a result of misunderstanding the instructions rather than an intrinsic characteristic of representations: children showing this pattern might not have understood that they were expected to adopt the present perspective, thinking instead that the tester meant them to indicate order within a waking day. In the two new studies, greater emphasis was given to the fact that children were judging distances from ‘‘now.’’ In addition, during training children were given counter-examples to assigning early morning events to the nearest part of the road. They were shown a card depicting the sun rising, and told that it would be a mistake to point to the closest part of the road, ‘‘. . . because it’s a really long time until the sun rises again.’’ In the first of these studies (Friedman, 2002, Study 1), 4- and 5-year-old children were tested on daily-activities and annual-events picture-pointing task on separate days. Testing took place in the morning and during spring. Annual-event stimuli (in order of future distance) were: Easter, summer, Hallowe’en, Thanksgiving, Christmas, and Valentine’s Day. The dailyactivities stimuli were: lunch, dinner, going to bed, the middle of the night, wake, and breakfast. Despite the modifications to the procedure, about the same proportion of the children gave ratings for the waking stimulus that fell below the midpoint of the scale as in the earlier study of 4- and 5-year-olds’ dailyactivities judgments (where no counterexample or special emphasis on ‘‘now’’ was used). Children in this morning-reference group in the present study also produced future-distance judgments that showed a strong linear trend from the start of the day. The persistence of this approach to the task suggests that it is a product of children’s representations rather than of the testing procedure. Apparently, the representations of many children include a privileged temporal reference point, the morning, which is not easily abandoned. Not only do these representations fail to contribute to a differentiated sense of distances in the future, they actually seem to interfere with future differentiation. Children in the present-reference group showed some tendency to judge events to be more distant when they were more remote from the present, with the familiar underestimation of the most recent activity, breakfast.


249

To compare differentiation on the two timescales, the day and the year, I examined plots of the present-reference groups’ judgments of daily activities and annual events as a function of the events’ true distances in the future. On the daily-activities task, these children appeared to distinguish three distance categories (if we exclude breakfast, which seems to be subject to past–future confusion): lunch, a category of events from late in the day, and waking. The results for the annual-events task were similar to those of 5-year-olds in one of the earlier studies of this content. Three categories were distinguished: Easter, summer, and holidays from late in the year. Based on past findings on the development of representations of daily activities and annual events, I had expected that children would show greater differentiation of the future distances of daily activities than of annual events. The present results are at odds with the prediction; even the 4- and 5-year-olds in the present-reference group made similar differentiated judgments on the two tasks. These results suggest a more complex developmental pattern than a simple sequence. Differentiated information about the future distances of annual events (presumably based on memory for adults’ references to some of the events) coexists with some awareness of the distance of daily activities (presumably based on representations of this content). I investigated the development of future differentiation on the day and year scales in another study with older children, 6- to 8-year-olds (Friedman, 2002, Study 2). Previous studies, including one described earlier, indicate that many children at these ages are accurate at judging which of two daily activities will come next from various reference points within the day, and the study of 6- to 8-year-olds’ judgments of the future distances of daily activities showed considerable differentiation. But other research shows a limited ability of children at these ages to think about the order of annual events from varying reference points. Furthermore, we saw in one of the studies of future differentiation of annual events that second graders show considerable distortion, collapsing the distances of events when they are beyond the next month or so in the future. For these reasons it seemed that a study of future judgments of both daily activities and annual events with first- and second-graders might provide useful information about the unity or multiplicity of children’s ability to differentiate the times of future events. The same procedure used for the 4- and 5-year-olds (Friedman, 2002, Study 1), including the countersuggestion and the emphasis on judging distances from ‘‘now,’’ was used with 6- to 8-year-olds. The new procedure led to a significant decrease in the proportion of 6- to 8-year-olds producing the morning-reference pattern, from 59% in the old study to 27% in the new one. As we saw, the changes had no discernible effect on the proportion of 4- and 5-year-olds producing the morning-reference pattern. Together, these

250

William J. Friedman

findings suggest that the older children are better able to shift from a favored perspective on the times of daily activities to the perspective of the present. This developmental change is consistent with a trend toward increasingly flexible representations of the pattern of daily activities, a trend seen in earlier studies (Friedman, 1977, 1990b). As expected, those children who fell in the morning-reference group showed a linear increase in distance ratings from waking to the middle of the night, whereas those in the presentreference group produced a linear increase according to distances from the present (with some underestimation of the distance of breakfast). I compared future differentiation of daily activities and annual events for the present-reference group, as in the study with younger children. This time the daily-activities set was more clearly differentiated. Children judged lunch to be nearer than the remaining events (again, excluding breakfast); dinner to be closer than going to bed, the middle of the night, and waking; and bed to be closer than night and waking. In contrast to this pattern of differentiation over much of the day cycle, the same children appeared to distinguish only two categories of future distances for the annual events. The distances of Valentine’s Day, Easter, and summer were not distinguished from one another, but they were judged as nearer than events from late in the year, Hallowe’en, Thanksgiving, and Christmas. Christmas was assigned a greater future distance than Hallowe’en and Thanksgiving, suggesting a third category, but this was probably due to clear memories that Christmas had occurred recently (coupled with the knowledge that recent annual events will not occur again for a long time). These results are consistent with the earlier studies of annual events in showing that from 3 or 4 years through about age 8, children can distinguish a few future-distance categories on the scale of the year. Evidence from those studies indicated that they do so by accessing propositions that some events are coming soon or very soon and probably assigning other events, for which no such propositions are accessible, to a category of distant events. It is very difficult to see how such processes could account for the substantial degree of differentiation shown by 6- to 8-year-olds on the dailyactivities task or their ability to update the distances of multiple events from particular locations within the day. Instead, children of 4 years of age and older probably rely on their representations of the pattern of daily activities, just as in late middle childhood children can gauge the future distances of annual events using general representations of this much longer pattern. 4. Conclusions The findings of the studies summarized in this section indicate that the development of a differentiated sense of the future is not a monolithic achievement. The capacity to differentiate times throughout most of a


251

pattern appears at different ages for different contents. Although some degree of differentiation of both daily activities and annual events is present by early childhood, children can distinguish the future distances of daily activities up to nearly one day in the future long before they can differentiate the distances of events up to about one year in the future. Another example of content variation is children’s ability to use conventional representations to reckon future distances of parts of the week (e.g., the weekend) before they can be used for parts of a year. There is another way in which a sense of the future is manifold: multiple psychological processes underlie a differentiated sense of the future. For annual events these processes include memory for adults’ statements about the future distances of events. Even 4-year-olds can retrieve the information that Hallowe’en and other holidays are coming soon, and from at least 5 through 7 years of age information about a few categories of future proximity seems to be available in propositions that children can access. By 8–9 years of age, children can use another source of information to differentiate the future distances of annual events: representations of when some important events, such as holidays, fall within the year. By 10 years of age, they can determine the number of months between the present and an event. In the case of the cycle of daily activities, propositions are not involved in differentiating future distances. Instead, general representations of the pattern are probably used by children 4 years of age and older. The flexibility with which children can use dailyactivities representations increases with age. Although many 4- and 5-yearolds seem unable to abandon the perspective of the start of the day, most 6- to 8-year-olds can adopt a present perspective if the task makes this requirement especially clear. Even in this older group, however, some children have difficulty thinking about distances beyond the night boundary. It is evident from these findings that possessing a representation of a time pattern does not ensure the ability to differentiate future distances. Particular characteristics of representations may stand in the way of future differentiation. C. DIFFERENTIATION OF THE PAST AND OF THE FUTURE

1. The Order of Past and Future Differentiation We have seen that on timescales longer than the day, basic memory processes underlie young children’s differentiation of the past, whereas their ability to differentiate future distances is subject to the availability of specific propositions about the nearness of future events. Thus, we might expect that on long timescales young children’s sense of the past is more differentiated than their sense of the future. This prediction is supported by a comparison of data from two studies (Friedman, 2000, Study 3;

252

William J. Friedman

and Friedman & Kemp, 1998, Study 3). In the former study, 3-, 4-, and 5-year-olds were tested in October and asked to judge the future distances of their birthday, summer, and the holidays: Hallowe’en, Thanksgiving, Christmas, and Valentine’s Day. To increase the comparability of the two studies, children made their judgments by placing cards on the same wooden board that was used in Friedman and Kemp’s (1998) study (rather than the usual picture-pointing task used in the studies of future differentiation described in the previous section). The 3- and 4-year-olds’ did not discriminate the future distances of the events. Although they were tested an average of only 12 days before Hallowe’en, the mean for this event did not differ from the means of the more distant events. Comparison data came from age- and school-matched 3- and 4-year-old children in Friedman and Kemp’s (1998) Study 3, who made past-distance judgments of most of the same events during a testing session an average of 29 days after Christmas. These children showed a highly significant tendency to judge the nearest event, Christmas, to be closer than the remaining events. Despite the fact that Christmas was more than twice as far in the past in Friedman and Kemp’s (1998) study as Hallowe’en was in the future in Friedman’s (2000) study, only the former condition showed differentiation. By 5 years of age children in both conditions produced significantly differentiated judgments of the nearest versus farther events, showing that the problem was not the greater salience of Christmas or greater difficulty applying the scale to the future. Instead, basic memory processes evidently provide children less than 5 years of age with a more differentiated sense of the past. However, this difference does not hold throughout development. Not only has it weakened by the end of early childhood, but when children use representations to think about past and future times, they are better able to differentiate future distances. Research on children’s representations of parts of the day, days of the week, and months of the year (Friedman, 1986, 1990b) shows that children can mentally move forward through these time patterns at earlier ages than they can move backwards. This may explain the findings of several studies of past and future judgments from the perspective of the present. I noted earlier that children at least as old as 8 or 9 years are very poor at judging which of two annual events, whose months are known, was a longer time ago (Friedman, 1992), and Friedman, Gardner, and Zubin (1995) found many errors on similar comparisons of one’s birthday and Christmas well into late middle-childhood. In contrast, studies described in the section on the future show that 8- through 10-year-olds can judge the future distances of annual events with considerable accuracy (Friedman, 2000). Finally, in a study in which participants were asked to indicate past and future order on the same scale (Silverman, 1996), futureorder judgments were found to be easier than past-order judgments in


253

children as old as fourth graders. Together with the findings from young children, these data indicate that there is not a simple developmental sequence of differentiating the past and the future. Both show substantial change from early childhood through late middle childhood. 2. The Processes Underlying Past and Future Differentiation The studies that were reviewed in the previous sections show multiple levels in the ability to differentiate the times of past events and of future events, levels based on different processes. In early childhood a differentiated sense of the past is rooted in distance-based memory processes, whereas representations of time patterns play an increasingly important role in middle childhood. In the case of future differentiation, memory for adults’ statements about future events provides some information about their times by early childhood. Again, representations of time patterns assume increasing importance from early to middle childhood. Representations of the pattern of daily activities, though often limited in flexibility, supply some information about future distances in children younger than 5 years. Representations of the days of the week can be used to distinguish future distances by 7 or 8 years, and within a year or two thereafter, representations of the times of annual events allow future differentiation on the scale of the year. There is a general trend toward representation-based differentiation of both the past and the future, and these processes are probably the most important ones in adults’ sense of the times of past and the future events. By adulthood, and probably by adolescence, the past and the future can be viewed within the common frameworks of time patterns that extend into the past and into the future. Image representations of a number of time patterns allow us equal facility in thinking backward and forward in time (Friedman, 1989). The mix of processes underlying past and future differentiation, some shared and some distinct, raise the question of how children and adults can distinguish between the past and the future. This is the subject of the following section.

III. Differentiation Between the Past and the Future A. DIFFERENCES AND SIMILARITIES BETWEEN THE PAST AND THE FUTURE

In the introduction to this chapter, I noted that in some scientific descriptions of the physical world, the past–present–future distinction is

254

William J. Friedman

unnecessary, whereas in everyday experience the distinction seems a basic aspect of reality. I suggested that the psychological separation of the past and the future is partly a social construction, and that basic cognitive processes, such as perceptual characteristics of memories (Johnson et al., 1988), probably also allow us to distinguish past events from those that have not yet occurred. Even from a psychological point of view, however, there appear to be a number of challenges to distinguishing the past–future status of events. First, thoughts about both past and future events share the status of being relatively active in memory. This was revealed by a study of adults’ answers to the question, ‘‘What day is today?’’ (Koriat & Fischhoff, 1974), which showed that the proximity of the weekend led to especially rapid responses on both Monday and Friday. If thoughts about recent and impending events are both active in memory, an event’s accessibility itself cannot tell us whether it already occurred or will occur soon. Another difficulty is that one of the mechanisms used by young children to differentiate future distances can easily produce erroneous information. If children have heard repeatedly that a particular event is coming soon, this proposition may still be active in memory after the event has occurred, leading them to retrieve the now-erroneous information that the event is imminent. Third, many important events occur in cyclic patterns and can be thought of as belonging to both the past and the future. It is easy to imagine that young children would be confused by adults’ statements, such as, ‘‘You just had a birthday, so you won’t have another one for a long time.’’ Finally, when children first develop representations of a time pattern, there may be special difficulties dealing with temporal directionality. A child might be able to access the information that another event is close to the present without clearly distinguishing recent pastness from future proximity. In the following discussion, I will present evidence that children do sometimes confuse the past and the future and consider which, if any, of these factors could be responsible. B. EVIDENCE OF PAST–FUTURE CONFUSION

1. Studies of Memory for Time or Future Differentiation In a number of the studies of past and future differentiation discussed in earlier sections, children judged impending events as being a short time ago and recent events as belonging to the near-future. When Friedman, Gardner, and Zubin (1995) tested children in the autumn, they found that most children less than 6 years of age, and many children as old as 8 years of age, judged their birthday to be more recent than Christmas when it was, in fact, longer ago than Christmas (and thus would occur again within the next few months). In another of their studies, children from preschool through


255

second grade compared the recency of their birthday and Christmas in a testing session in early December. Here the near-futurity of Christmas disrupted the usual ability to detect the greater recency of birthdays from the previous 3 months. Similar confusion is found when nonverbal scaling methods are used. Friedman and Kemp (1998) asked 3- to 6-year-olds to judge distances in the past by placing cards on a wooden ruler. Children whose birthdays were coming within the next 3 months and those tested in the weeks before Valentine’s Day underestimated the distances of these events in the past. In the studies of future differentiation of annual events (Friedman, 2000), symmetrical errors were found. Most 4-year-olds responded ‘‘yes’’ to the question, ‘‘Is Hallowe’en coming soon?’’ in the weeks after the holiday, and the picture-pointing task showed similar results: 5-year-olds assigned Valentine’s Day to the near-future when tested about a week after the holiday. When the picture-pointing task was used to measure future differentiation of daily activities (Friedman, 2002), the future distance of the most recent event, breakfast, was underestimated by children as old as first and second graders. Finally, in an unpublished study, 3- and 4-year olds tested in the morning were asked,’’ Have you had yet today?’’ for several daily activities, including breakfast and lunch.1 Children were very accurate for breakfast, but they responded at chance levels for lunch. These errors do not seem to be an artifact of a single method, because they are found with both verbal questions and spatial tasks. The errors appear on very different timescales as well: parts of the day and parts of the year. Past–future confusion seems to be a sufficiently widespread phenomenon to merit explanation, but how should it be explained? Is it the product of underlying processes contributing erroneous information, difficulty dealing with directionality in early representations, or the absence of a clear conceptual distinction between the past and the future? Is there a single explanation for all of the errors? I will return to these questions after presenting several studies specifically designed to measure past–future differentiation. 2. Studies of Past–Future Differentiation A feature of all of the studies showing past–future confusion is that children judged only one of the categories at a time. When children were asked to judge distances in the past or in the future, they had no way of indicating that they knew an event was near but in the other category. Perhaps children making these errors really knew that an event was not near 1

I am grateful to Robert Siegler for suggesting this method.

256

William J. Friedman

in the ‘‘direction’’ they were asked to judge but were more impressed by its general proximity than the fact that it was close in the wrong direction. If this is the reason for children’s errors, it might be possible to eliminate them by giving children a clear choice between the past and the future. In two studies (Friedman, 2001c), I have used a spatial representation designed to highlight the difference between the past and the future. The representation is a picture of a road stretching horizontally across the bottom part of a simple landscape. A separate cutout of a car was moved halfway across the visible length of the road to indicate the direction of travel. Children made their past–future judgments by pointing to a circle on the part of the road the car had passed or the part it would come to. During the training phase, they were shown the correct responses for the stimuli: yesterday, tomorrow, last week, and next week. In the first study, a group of mainly 4-year-old children were tested about one week before and one week after Valentine’s Day. In addition to this holiday, they judged the annual events summer and Christmas and the daily activities breakfast and lunch. The question of principal interest was whether children would be able to discriminate the past–future status of Valentine’s Day shortly before and shortly after the holiday. The results showed that before Valentine’s Day 80% of children correctly assigned the holiday to the future (accuracy that was highly significant), but after the Valentine’s Day only 60% correctly pointed to the circle representing the past (a proportion that did not differ significantly from chance). Evidently, children had a clear sense of Valentine’s Day’s futurity about a week before the event, but at about the same amount of time after the holiday many children were confused about its past–future status. These young children were aware of the pastness of one event, Christmas, but only at the first testing time. By one week after Valentine’s Day, Christmas judgments fell to chance levels, perhaps because of interference associated with Valentine’s Day’s recency. The future status of summer was recognized at both testing times by a greater proportion of children than would be expected by chance. The 4-year-olds in the Friedman (2001c) study were confused about the past–future status of Valentine’s Day about a week after the event, and the phenomenon is not just an artifact of testing children on only the past or the future direction. One explanation for the confusion in this situation is that children still had accessible in memory now-outdated propositions that Valentine’s Day is coming soon. Children must have heard such statements many times in the weeks before the holiday, or they would not have been so accurate at the first testing time. The main clues to its pastness after the holiday occurred are probably relatively active memories of what happened on Valentine’s Day. However, even these memories are probably mixed with


257

memories of preparing for the event when Valentine’s Day was still in the future. On the very different scale of the day, about three-quarters of the children who were tested in the morning (the large majority of the sample) assigned lunch to the future, but the proportion assigning breakfast to the past was very close to the null expectation of .50. It seems unlikely that the accurate judgments of the futurity of lunch were the product of recent adult statements that lunch is coming soon. Probably children either used representations of the order of daily activities or knew in some general sense that lunch is the next major activity after preschool. The difficulty that these young children had with the breakfast item echoes the findings of earlier studies of daily activities, in which the future distance of breakfast was repeatedly underestimated. As for the findings for lunch, it is unlikely that the errors for breakfast are the result of remembered propositions, in this case outdated propositions that breakfast is coming soon. Instead, the error may be related to relatively vivid memories of that morning’s breakfast or to representation-based knowledge of where breakfast occurs within the day. The second explanation requires that backward relations are especially confusing in early representations of the pattern of daily activities. In fact, when 4-year-olds attempt to extract information about the relative times of daily activities, they think more readily about relations that are forward than those that are backward in time (Friedman, 1990b). The problems that early representations may create highlight the fact that all of the events for which past–future confusion has been demonstrated occur in cyclic patterns. As was noted earlier, the fact that there are multiple referents of cyclic events, including the most recent occurrence and the next one in the future, may create difficulties. Indeed, both past and future responses were strictly correct for each of the annual events and daily activities: each had happened before, and each would happen again. In principle, judgments of events that are closer in the future could be influenced by memories of a previous occurrence of an event, and children may know that even relatively recent events, such as Valentine’s Day at the second testing time, will come again. For these reasons it is important to study children’s ability to distinguish the past–future status of noncyclic events. Consequently, in the second study parents of 4- and 6-year-olds were asked to nominate unusual or unique events from each of several past and future temporal-distance categories, ranging from earlier or later the same day to several months ago or several months in the future. During the same visit, children were trained on the road representation of the past and future and asked to point to the appropriate circle for each of the events their parents had provided. Events belonging to the past were judged as accurately as events belonging to the future, regardless of distance within

258

William J. Friedman

these categories. Overall, the 4-year-olds were correct on .64 of the past events and .67 of the future events. The corresponding means for the 6-yearolds were .88 and .84. All of these proportions differ significantly from chance levels, but the age increase in accuracy was significant. Additional analyses revealed that the age differences were unlikely to be the result of parents discussing events more frequently with older children or an artifact of them nominating more salient events. This study shows that 4-year-olds can discriminate the past–future status of autobiographical events but that their performance is relatively poor. The finding that more than one-third of the 4-year-olds err with unusual or unique events shows that difficulty discriminating pastness and futurity is not limited to cyclic events. The frequent errors also indicate that the problems are not always caused by the directional properties of early representations. Children in both age groups probably did not base their judgments on the locations of the events in time patterns, at least for events more distant than the same day. Instead, both the 4-year-olds’ difficulties and the 6-year-olds’ accuracy probably must be explained by memories of past events and memory for adults’ references to both past and future events. If children in both age groups have similar kinds of information available, why are 6-year-olds much more accurate? The reasons for the age change cannot be determined from this study, but, as a study to be discussed in Section III.C. shows, it occurs at the same time as a change in children’s understanding of causal relations between the past and future and the present, and, at least for 6-year-olds, the two abilities are correlated. These findings may reflect a common change in children’s understanding of the past and future as distinct categories. This adds another entry to the list of possible explanations of children’s difficulty discriminating pastness and futurity: the absence of a clear conceptual distinction between the two categories. 3. Conclusions Adults are usually struck by the seemingly clear psychological differences between the past and the future, including the fact that detailed, perceptual memories are only associated with past events. But the developmental evidence summarized in this section shows that memory provides both useful and confusing information about the past–future status of events, particularly in the mix of accuracy and inaccuracy shown by 4-year-olds. These children accurately recognized the pastness of Christmas a little over one month after it occurred (though not 2 weeks later, just after Valentine’s Day), and they were aware of the futurity of Valentine’s Day when it was about one week in the future and summer when it was about 4 months in the future. But when asked to compare distances in the past, children of this age


259

often judge events that are anticipated in the near-future to be recent. When they make judgments of distances in the future, recent events often intrude in the category of near-future events. The confusion is not restricted to tasks that query children about a single direction. On the road task, where the past and future are the two choices, 4-year-olds erred on a recent holiday (Valentine’s Day) and a recent daily activity (breakfast). Furthermore, the problems are not restricted to cyclic contents: children of this age also made many errors when discriminating the past–future status of unusual autobiographical events. Pastness may be clearest when one has memories for an event but outdated propositions about its futurity are not easily accessible. Children tested in early February probably correctly assigned Christmas to the past because they could recall the event (and perhaps had accessible past-referring propositions, e.g., memories for the question, ‘‘What did you get for Christmas?’’), but could no longer easily access propositions laid down in memory when the holiday was approaching. But one week after Valentine’s Day, many 4-year-olds were confused about its past–future status, probably because of readily accessible, but outdated, propositions about its approach (and perhaps the general strong impression of its nearness). Futurity may be clearest when one has readily available propositions about the imminence of an event and only very weak memories for its previous occurrence. Older children and even adults probably must contend with the same mix of memories for events and memories of references to the past and future status of events (some of which are still accurate and others outdated). But by 6–7 years of age, many of the errors shown by 4-year-olds have disappeared. Two changes seem especially promising for explaining the progress in past–future differentiation made from early childhood through middle childhood. One is the development of flexible representations of time patterns, representations that allow a clear distinction between the two temporal directions. These are acquired at different ages for different time patterns, and past–future errors on the scale of the year are sometimes found well into middle childhood. The other is a conceptual understanding of the differences between the past and the future. In the following section, I will describe my beginning attempts to learn about the development of concepts of the past and the future. C. CONCEPTS OF THE PAST AND THE FUTURE

Adults’ understanding of the past and the future goes beyond the ability to assign particular events to these two categories: we also have a general understanding of ways in which the two categories differ. One distinction is

260

William J. Friedman

captured in the view that the past can be known whereas the future is uncertain. Another is found in the belief that one can influence the future but cannot change the past. Adults also use the past and the future in selective ways to explain the present. As part of our understanding that causes precede effects, we know that a past event, but not a future event, could physically affect the present state of affairs. But our conceptions of psychological causation lead us to attribute present psychological states, such as happiness, to either past or anticipated experiences. These principles underpin important observations that adults make about human experience and adaptation. When advising someone about a personal problem, we commonly refer to the impossibility of changing the past, the benefits of putting one’s efforts instead into the future, and the fact that one cannot know what the future will hold. Do children understand these ways in which the past and the future differ and are similar? Young children understand that physical causes precede their effects (Bullock, Gelman, & Baillargeon, 1982; Shultz, 1982), but there has been no research that focuses on the causal relations between the past and future and the present. I have conducted several studies to shed light on children’s understanding of this aspect of past–future differentiation (Friedman, 2001d). In these studies children were read pairs of brief vignettes describing past and future events and were asked to draw conclusions from them. The pairs assessed knowledge of the principles mentioned in the preceding paragraph: that the past, but not the future, can be known; that one can influence the future but not the past; that past, but not future, events can influence the present; and that both past and anticipated events can influence one’s present psychological state. An example of a past item is, ‘‘Michelle had a birthday party yesterday. Can she know all the presents she got? Why (or why not)?’’ In the first study, I compared 4- and 5-year-olds with 6- through 8-year-olds and in the second study compared 4- and 6-yearolds (the same 4- and 6-year-olds who judged the past–future status of parent-nominated autobiographical events). By 4–5 years of age children accept that both past and future events can influence one’s feelings in the present. For example, a large proportion knew that a birthday party yesterday or tomorrow could make one happy today. The other principles, concerning past–future differences, seemed less well understood at this age, though for at least some stories, 4-year-olds could discriminate one’s ability to be certain about past and future events. In contrast, 6-year-olds were much more consistent in applying these differences between the past and the future. Furthermore, many 6-year-olds articulated the idea that one could draw the appropriate conclusion about whether something could or could not be caused, influenced, or known because it was past or future. For example, many pointed out that one could not change


261

something that already happened or be sure about something that has not happened yet. By this age children seem to have the appreciation that pastness and futurity are the reasons for these relations with the present. Could this sensitivity to the past and the future as distinct categories help to explain the change, seen in the previous section, from 4-year-olds’ frequent confusion of the past–future status of events to 6-year-olds’ usual accuracy in discriminating the times of past and future events? General support comes from the substantial changes in both abilities during this 2-year period. Suggestive evidence also comes from correlations between the two abilities. The correlation between children’s overall accuracy on the conceptual questions and their accuracy in assigning autobiographical events to the past or the future was significant for the 6-year-olds (r ¼ .62), but not for the 4-year-olds (r ¼ .34). This pattern might be explained by a common ability not yet present in 4-year-olds but found in some 6-yearolds. This hypothetical ability causes children to attend more closely to the past–future status of life events and to general ways in which past and future events relate to the present. The idea that there is a general basis for the sharpening of the distinction between the past and the future during these 2 years requires further support.

IV. Conclusions and Directions for Future Research A. CONCLUSIONS ABOUT CHILDREN’S SENSE OF THE PAST AND THE FUTURE

In this chapter I have reviewed developmental studies that shed light on the processes underlying humans’ sense of place in time. These studies show that our view of the past and the future depends both on basic cognitive processes and the social construction of time. The clearest example of the involvement of basic processes is the information that memory mechanisms provide about the distances of events in the past. In the case of the future, basic properties of memory, such as the influences of recency and frequency, determine which of adults’ references to future events will be accessible and which difficult to retrieve. These propositions about the times of future events are also an example of the many social contributions to a sense of the past and the future: An important part of the information is often what adults actually said about when the approaching event would occur. Other cultural contributions to children’s sense of place in time are the linguistic devices and conventional representations of time patterns that, once mastered, provide children with frameworks for structuring their experience.

262

William J. Friedman

The studies also reveal that a sense of the past and the future develops substantially from early through middle childhood. By age 4, children can use distance-based memory processes to achieve a limited sense of the past, and by age 5 they can distinguish a few future-distance categories based on their memories for adults’ statements about the times of coming events. Children of these ages have also begun to use their representations of the pattern of daily events both to distinguish their distances in the future and to reconstruct the time of day when a remembered event occurred. During middle childhood representations of time patterns play an increasing role in children’s differentiation of the past and of the future. By about 7 or 8 years, children can use their knowledge of days of the week to begin to reconstruct when within the week a remembered event was likely to have occurred and to judge the future distance of the weekend. By age 10, children use representations of the months to judge distances in the future and probably distances in the past. Not only are representations of the different time patterns acquired at different ages, but representations of particular time patterns develop over time, and initial limitations are overcome as representations become more flexible. These changes allow greater differentiation of distances within the past and within the future, and probably enable children to distinguish more clearly between the past and the future. Another development, the acquisition of concepts of the past and the future, appears to be most marked between about 4 and 6 years, and this may also contribute to the ability to differentiate between events that happened in the past and those that are expected to occur in the future. B. CONCLUSIONS ABOUT COGNITIVE DEVELOPMENT

A number of findings of the studies reviewed in this chapter may also contribute to our understanding of cognitive development. A first observation is the difficulty of predicting age changes in children’s temporal abilities from general theories of cognitive development. This is perhaps clearest when one attempts to apply the concept of egocentrism to time. Piaget and Inhelder (1967) concluded from their studies of spatial representation that children progress from a stage at which they are unable to represent perspectives other than their own to a stage at which other perspectives can be considered. In the case of temporal orientation, we see that the ability to view the times of other events from the child’s current perspective is the product of a protracted construction, with other perspectives sometimes grasped at earlier ages than is the present perspective. Many children find it easier to judge the future distances of daily activities from a different perspective, the start of the day, than from the time it is now. Another example is the difficulty of applying Case’s


263

(Case & Okamoto, 1996) central conceptual structures to the development of past and future differentiation. Case proposed a series of specific age transitions in the number of dimensions that children can process. The past and the future might each be thought of as unidimensional continua, to which a mental number line might be applied. But an analysis of numbers of dimensions bears little resemblance to the pattern of development of past and future differentiation from early to middle childhood. This pattern seems better described as young children having fragmentary information about the distances of particular events, and older children learning to use particular representations in a flexible way before either the past or the future can be grasped as integral dimensions. To understand cognitive development, we need to take account of the specific representational challenges of each domain. Furthermore, we have found substantial variation in the abilities of children of a given age even from one temporal content to another. One might have expected that the development of a sense of the past or of the future would be a matter of broad gains, so that the past in general or the future in general become differentiated at particular ages. But we have seen that this is not the case. Young children can reconstruct the part of a day when a past event took place years before they can reconstruct the time on longer timescales. Similarly, differentiation of the future distances of daily activities develops more rapidly than differentiation of events on longer timescales. Even within what may appear to be a single domain, time, we need to describe the specific representations and processes used by children of a given age for each of a number of different contents. Our scientific quest for broadly applicable principles of development must be balanced by the need to be taxonomists, especially in the early stages of inquiry in an area. Third, multiple processes underlie children’s abilities, even for a single time pattern, and they often coexist at a single age. For example, memory for the times of past events involves at least three fundamentally different kinds of processes (distance-based processes, location-based processes, and order codes), and future differentiation involves at least the retrieval of propositions about the distances of particular events and processes involving the representations of time patterns (which, in turn, probably take different forms). During middle childhood, representation-based processes become more important, but the other processes apparently are used even by adults. The observation that multiple processes coexist is broadly consistent with Siegler’s (1996) theory, in which cognitive development involves changes in the frequency with which different processes are used, rather than a succession of processes. Again, we must be willing to spend considerable effort distinguishing multiple psychological constructs within

264

William J. Friedman

what initially seem narrow phenomena if we are to understand the process of development. A final conclusion concerns the role of individual construction in cognitive development. Children learn about the times of past and future events informally from adults’ references to particular events and more formally from parents’ and teachers’ instruction about the order of conventional time markers. But it is difficult to see how either source of information would enable the child to grasp the times of a multitude of events from the perspective of a constantly changing present. Remembered propositions are often obsolete, and practice with conventional time patterns, such as recitation of the days of the week or months of the year, often leads to stereotyping in the kinds of processes that can be performed (Friedman, 1986). The ability to think flexibly about the relative times of events within a time pattern may be the product of children’s repeated attempts over many years to think about the times of past and future events. Such experience is probably responsible for the development of image representations of time patterns (Friedman, 1986, 1989, 1990a), images that allow children, adolescents, and adults to view the times of past and future events from multiple perspectives within the temporal patterns. C. FUTURE DIRECTIONS

Much remains to be discovered about the development of a sense of the past and the future. A number of significant gaps in our knowledge are closely related to the issues discussed in this chapter. First, information about the part played by propositions in a differentiated sense of the future and in past–future confusion is quite limited. We need direct evidence about how frequently and in what forms adults refer to the distances of events in the future, the conditions under which children retain this information, and how they make use of it in thinking about the times of events. Second, a number of the processes that contribute to a differentiated sense of the past remain poorly understood. Adults’ memory for the times of past events depends heavily on the ability to reconstruct when the events occurred (Friedman, 1993). At least three components are necessary for reconstructing the times of past events: episodic memories that contain temporally relevant information, general knowledge about time, and executive processes that control the search for, and integration of, these two kinds of information. Each of these, in turn, must be made up of additional components, components that may be acquired at different ages. At present, little is known about these components or how they develop. Third, our knowledge about the development of children’s conceptual understanding of the past and the future is rudimentary, as is our understanding of the


265

contribution of this knowledge to the ability to distinguish the past–future status of events. A number of promising new directions exist for research on the development of a sense of the past and the future. One is an extension of the concept of metamemory (Kreutzer, Leonard, & Flavell, 1975) to time. Research on the development of metamemory shows age increases in children’s awareness of how memory works and of the use of strategies for improving memory (Short, Schatschneider, & Friebert, 1993). Although there is a substantial literature on aspects of metamemory relating to storing and recalling content, there have been no studies on metamemory for the times of past events. Thus, researchers lack information about what children and adults know about the processes underlying memory for when past events occurred or how to improve the accuracy of judgments of the times of past events. Studies of temporal metamemory can increase researchers’ understanding of metamemory development in general and can enrich the understanding of strategic processes in temporal reconstruction. Along with new information about the development of temporal reconstruction, it can also improve our understanding of children’s testimony about the times of alleged abuse. Another potentially fruitful direction is exploring how the growth of future differentiation affects children’s ability to plan (Friedman & Scholnick, 1997) and their ability to remember to do things. The development of this latter ability, which cognitive psychologists call ‘‘prospective memory’’ (Brandimonte, Einstein, & McDaniel, 1996), has been little studied, and it may be possible to design studies that will test the role of children’s sense of place in time in prospective memory performance. Finally, much needs to be learned about family–environmental variation in the development of past and future differentiation. I have assumed that parents’ discussions with children are an important source of information about the times of events. It would be of theoretical and practical significance to know whether children growing up in unstable or deprived environments, who may have little opportunity to discuss with their parents when future events will occur, have a less differentiated sense of the future (see Nelson, 1996). Answering these and other questions should enrich our understanding of the processes that allow us to view our lives as unfolding within the frameworks of time.

ACKNOWLEDGMENTS Many of the studies reported in this chapter were supported by grants from the National Institutes of Health (HD30403-01) or the National Science Foundation (SBR 95-13881 & SBR 98-15791).

266

William J. Friedman

REFERENCES Ames, L. B. (1946). The development of the sense of time in the young child. Journal of Genetic Psychology, 68, 97–126. Bates, E., Elman, J., & Li, P. (1994). Language in, on, and about time. In M. M. Haith, J. B. Benson, R. J. Roberts, Jr., & B. F. Pennington (Eds.), The development of future-oriented processes (pp. 293–321). Chicago: University of Chicago Press. Benson, J. B. (1994). The origins of future orientation in the everyday lives of 9- to 36-monthold infants. In M. M. Haith, J. B. Benson, R. J. Roberts, Jr., & B. F. Pennington (Eds.), The development of future-oriented processes (pp. 375–407). Chicago: University of Chicago Press. Benson, J. B., Talmi, A., & Haith, M. M. (2003, in press). The social and cultural context of the development of future orientation. In C. Raeff & J. B. Benson (Eds.), Social and cognitive development in the context of individual, social, and cultural processes. London: Routledge. Brandimonte, M., Einstein, G. O., & McDaniel, M. A. (Eds.) (1996). Prospective memory: Theory and applications. Hillsdale, NJ: Erlbaum. Brown, A. L. (1973). Judgments of recency for long sequences of pictures: The absence of a developmental trend. Journal of Experimental Child Psychology, 15, 473–480. Bullock, M., Gelman, R., & Baillargeon, R. (1982). The development of causal reasoning. In W. J. Friedman (Ed.), The developmental psychology of time (pp. 209–254). New York: Academic Press. Case, R., & Okamoto, Y. (1996). The role of central conceptual structures in the development of children’s thought. Monographs of the Society for Research in Child Development, 61 (Serial No. 246). Friedman, S. L., & Scholnick, E. K. (1997). The developmental psychology of planning: Why, how, and when do we plan? Mahwah, NJ: Erlbaum. Friedman, W. J. (1977). The development of children’s knowledge of cyclic aspects of time. Child Development, 48, 1593–1599. Friedman, W. J. (1986). The development of children’s knowledge of temporal structure. Child Development, 57, 1386–1400. Friedman, W. J. (1989). The representation of temporal structure in children, adolescents and adults. In I. Levin & D. Zakay (Eds.), Time and human cognition: A life-span perspective (pp. 259–304). Amsterdam: North-Holland. Friedman, W. J. (1990a). Children’s representations of the pattern of daily activities. Child Development, 61, 1399–1412. Friedman, W. J. (1990b). About time: Inventing the fourth dimension. Cambridge, MA: MIT Press. Friedman, W. J. (1991). The development of children’s memory for the time of past events. Child Development, 62, 139–155. Friedman, W. J. (1992). Children’s time memory: The development of a differentiated past. Cognitive Development, 7, 171–187. Friedman, W. J. (1993). Memory for the time of past events. Psychological Bulletin, 113, 44–66. Friedman, W. J. (1996). Distance and location processes in memory for the times of past events. In D. L. Medin (Ed.), The psychology of learning and motivation, Vol. 35 (pp. 1–41). Orlando, FL: Academic Press. Friedman, W. J. (2000). The development of children’s knowledge of the times of future events. Child Development, 71, 913–932. Friedman, W. J. (2001a). Memory processes underlying humans’ chronological sense of the past. In C. Hoerl & T. McCormack (Eds.), Time and memory: Issues in philosophy and psychology (pp. 139–167). Oxford: Oxford University Press.


267

Friedman, W. J. (2001b). [Children’s judgments of the future distances of daily activities.] Unpublished data. Friedman, W. J. (2001c). Children’s differentiation of the past and the future. Poster presented at the meeting of the Cognitive Development Society, Virginia Beach, October 27, 2001. Friedman, W. J. (2001d). Children’s understanding of causal relations between the past and future and the present. Unpublished manuscript. Oberlin College. Friedman, W. J. (2002). Children’s knowledge of the future distances of daily activities and annual events. Journal of Cognition and Development, 3, 333–356. Friedman, W. J., Gardner, A. G., & Zubin, N. R. E. (1995). Children’s comparisons of the recency of two events from the past year. Child Development, 66, 44–66. Friedman, W. J., & Huttenlocher, J. (1997). Memory for the times of ‘‘60 Minutes’’ stories and news events. Journal of Experimental Psychology: Learning, Memory, & Cognition, 23, 560–569. Friedman, W. J., & Kemp, S. (1998). The effects of elapsed time and retrieval on young children’s judgments of the temporal distances of past events. Cognitive Development, 13, 335–367. Gell, A. (2000). Time and social anthropology. In P. J. N. Baert (Ed.), Time in contemporary intellectual thought (pp. 251–265). Amsterdam: Elsevier. Haith, M. M., Benson, J. B., Roberts, R. R., & Pennington, B. F. (Eds.) (1994). The development of future-oriented processes. Chicago: University of Chicago Press. Harner, L. (1975). Yesterday and tomorrow: Development of early understanding of the terms. Developmental Psychology, 11, 864–865. Harner, L. (1976). Children’s understanding of linguistic reference to past and future. Journal of Psycholinguistic Research, 5, 65–84. Harner, L. (1980). Comprehension of past and future reference revisited. Journal of Experimental Child Psychology, 29, 170–182. Harner, L. (1981). Children talk about the time and aspects of actions. Child Development, 52, 498–506. Harner, L. (1982a). Talking about the past and the future. In W. J. Friedman (Ed.), The developmental psychology of time (pp. 141–169). New York: Academic Press. Harner, L. (1982b). Immediacy and certainty: Factors in understanding future reference. Journal of Child Language, 9, 115–124. Hudson, J. A. (2001). The anticipated self: Mother–child talk about future events. In C. Moore & K. Lemmon (Eds.), The self in time: Developmental perspectives (pp. 53–74). Mahwah, NJ: Erlbaum. Hudson, J. A. (2002). ‘‘Do you know what we’re going to do this Summer?’’: Mothers talk to preschool children about future events. Journal of Cognition and Development, 3, 49–71. Hudson, J. A., Shapiro, L. R., & Sosa, B. S. (1995). Planning in the real world: Preschool children’s scripts and plans for familiar events. Child Development, 66, 984–998. Jackson, J. L., Michon, J. A., & Vermeeren, A. (1984). The processing of temporal information. Annals of the New York Academy of Sciences, 423, 603–604. Johnson, M. K., Foley, M. A., Suengas, A. G., & Raye, C. L. (1988). Phenomenal characteristics of memories for perceived and imagined autobiographical events. Journal of Experimental Psychology: General, 117, 371–376. Koriat, A., & Fischhoff, B. (1974). What day is today? An inquiry into the process of temporal orientation. Memory & Cognition, 2, 201–205.

268

William J. Friedman

Kreutzer, M. A., Leonard, C., & Flavell, J. H. (1975). An interview study of children’s knowledge about memory. Monographs of the Society for Research in Child Development, 40 (Serial No. 159). Kuipers, B. (1978). Modeling spatial knowledge. Cognitive Science, 2, 129–153. Lucariello, J., & Nelson, K. (1987). Remembering and planning talk between mothers and children. Discourse Processes, 10, 367–389. Mathews, M. E., & Fozard, J. L. (1970). Age differences in judgments of recency for short sequences of pictures. Developmental Psychology, 3, 208–217. McCormack, T., & Russell, J. (1997). The development of recency and frequency memory: Is there a developmental shift from reliance on trace-strength to episodic recall? Journal of Experimental Child Psychology, 66, 376–392. Moore, C., & Lemmon, K. (Eds.) (2001). The self in time: Developmental perspectives. Mahwah, NJ: Erlbaum. Nelson, K. (Ed.) (1989). Narratives from the crib. Cambridge, MA: Harvard University Press. Nelson, K. (1993). The psychological and social origins of autobiographical memory. Psychological Science, 4, 7–14. Nelson, K. (1996). Language in cognitive development: Emergence of the mediated mind. Cambridge: Cambridge University Press. Nurmi, J.-E. (1991). How do adolescents see their future? A review of the development of future orientation and planning. Developmental Review, 11, 1–59. Piaget, J., & Inhelder, B. (1967). The child’s conception of space. New York: Norton. Reese, E., & Fivush, R. (1993). Parental styles of talking about the past. Developmental Psychology, 29, 596–606. Sachs, J. (1983). Talking about the there and then: The emergence of displaced reference in parent–child discourse. In K. E. Nelson (Ed.), Children’s language (Vol. 4, pp. 3–28). Hillsdale, NJ: Erlbaum. Short, E. J., Schatschneider, C. W., & Friebert, S. E. (1993). Relationship between memory and metamemory performance: A comparison of specific and general strategy knowledge. Journal of Educational Psychology, 85, 412–423. Shultz, T. R. (1982). Rules of causal attribution. Monographs of the Society for Research in Child Development, 47 (1, Serial No. 194). Siegler, R. S. (1996). Emerging minds: The process of change in children’s thinking. New York: Oxford University Press. Silverman, J. L. (1996). The development in children of future time perspective. Poster presented at the meetings of the American Psychological Association, Toronto, August 1996. Teichmann, R. (2000). The complete description of temporal reality. In P. J. N. Baert (Ed.), Time in contemporary intellectual thought (pp. 1–16). Amsterdam: Elsevier. Thompson, C., Barresi, J., & Moore, C. (1997). The development of future-oriented prudence and altruism in preschoolers. Cognitive Development, 12, 199–212. Tzeng, O. J. L., & Cotton, B. (1980). A study-phase retrieval model of temporal coding. Journal of Experimental Psychology: Human Learning and Memory, 6, 705–716. von Wright, J. M. (1973). Judgment of relative recency: Developmental trends. Journal of Psychology, 84, 3–12. Weist, R. M. (1983). Prefix and suffix information processing in the comprehension of tense and aspect. Journal of Child Language, 10, 85–96. Weist, R. M. (1989). Time concepts in language and thought: Filling the Piagetian void between two to five years. In I. Levin & D. Zakay (Eds.), Time and human cognition: A life-span perspective (pp. 63–118). Amsterdam: North-Holland.


269

Weist, R. M., Wysocka, H., Witkowska-Stadnik, K., Buczowska, E., & Konieczna, E. (1984). The defective tense hypothesis: On the emergence of tense and aspect in child Polish. Journal of Child Language, 11, 347–374. Winograd, E., & Soloway, R. M. (1985). Reminding as a basis for temporal judgments. Journal of Experimental Psychology: Learning, Memory, and Cognition, 11, 262–271.


THE DEVELOPMENT OF COGNITIVE FLEXIBILITY AND LANGUAGE ABILITIES

Gedeon O. Dea´k DEPARTMENT OF COGNITIVE SCIENCE, UNIVERSITY OF CALIFORNIA, SAN DIEGO, 9500 GILMAN DR., LA JOLLA, CA 92093-0515

I. WHAT IS FLEXIBLE COGNITION? A. PAST AND CURRENT STUDIES OF FLEXIBLE COGNITION B. FLEXIBLE COGNITION: A DEFINITION C. THE DEVELOPMENTAL ECOLOGY OF FLEXIBLE COGNITION II. DEVELOPING TOWARD . . . ? ADULTS’ FLEXIBLE COGNITIVE PROCESSING OF MEANINGS AND MESSAGES A. DISCOURSE AND SHARED MEANING: FLEXIBLE FORMATION OF MEANING B. FORMING AND SHIFTING CONCEPTUAL MAPPINGS C. NEUROPSYCHOLOGY OF FLEXIBILITY IN ADULT LANGUAGE PROCESSING III. TOWARD A MODEL OF FLEXIBLE REPRESENTATION IN LANGUAGE PROCESSING IV. CHILDREN’S FLEXIBLE THINKING ABOUT MEANINGS AND MESSAGES A. DEVELOPMENT OF FLEXIBLE NAMING B. DEVELOPMENT OF DISCOURSE FLEXIBILITY: SHIFTING IMPLICATIONS AND INSTRUCTIONS C. FLEXIBLE USE OF VERBAL CONTEXT TO INFER WORD MEANINGS D. COMMON FACTORS IN CHILDREN’S FLEXIBLE COGNITIVE PROCESSING OF MESSAGES AND MEANINGS V. QUESTIONS AND CONCLUSIONS A. HOW DO LOGICAL AND METACOGNITIVE ABILITIES INFLUENCE FLEXIBILITY? B. HOW DOES INHIBITION INFLUENCE THE DEVELOPMENT OF FLEXIBILITY? C. IS LANGUAGE CENTRAL TO FLEXIBLE COGNITION? D. METHODOLOGICAL PROBLEM: FILLING THE GAPS E. CONCLUSIONS REFERENCES


271


272

Gedeon O. Dea´k

A hallmark of human intelligence is flexible cognition: adapting inference to unfamiliar or unexpected situations, creatively combining concepts, and modifying familiar knowledge and habits to produce novel representational syntheses or action sequences. Language enhances and permits expression of flexible cognition. It permits the encoding and making public of innovative representations of present, absent, and imagined events, entities, and relations; and of mental states, ideas, and intentions. The potential for openended innovative conceptualization in natural language is demonstrated by this excerpt from a poem by Marianne Moore (1887–1972): ‘‘I remember a swan under the willows in Oxford, with flamingo-colored, mapleleaflike feet. It reconnoitered like a battleship. Disbelief and conscious fastidiousness were ingredients in its disinclination to move.’’ (‘‘Critics and Connoisseurs,’’ 1924)

The swan is imbued with a military mode of perception. Its feet are likened to animal and plant, as perspective shifts from color to shape. Anthropomorphized mental states are likened to ingredients (a culinary metaphor); ingredient itself is a metaphor for cause. Such innovative conceptual blending, epitomized in poetry and other creative activities, reveals a species-specific cognitive ability: the activation and communication of flexibly selected, combined, and modified representations. Even those of us who are not poets can grasp an uncanny synthesis such as Moore’s (e.g., we can imagine a swan ‘‘reconnoitering like a battleship’’). This ability, to make sense of unexpected combinations, reveals our normative preparedness to adapt—and understand—innovative representations of entities and events in our shared environment. The importance of this ability in human thought cannot be overstated: it is critical for mediating social interactions and sharing perspectives, for forming representations of unseen possible worlds based on heard or read descriptions, and for building socially coordinated action plans. How does the ability to understand, evaluate, and produce innovative messages develop? To understand this critical function of natural language and its development, we must consider some fundamental representational skills that might or might not be restricted to language. For example, flexible language processing requires selecting and encoding information from a dynamically changing environment, based on contextual demands that must be periodically evaluated and updated. As MacWhinney (1987) puts it, ‘‘In order to learn a language, a child must have available a rich representational system and flexible ways of deciding between representations. The child [must]

Cognitive Flexibility and Language Abilities

273

represent [an] intricate set of roles, positional patterns, cues, and conditions . . . . [therefore] language [must] . . . utilize virtually every aspect of higher cognition’’ (pp. 249–250)

These aspects, or skills, are of course, developing concurrently with language. This makes for a very challenging problem: to learn how children’s language skill relies on, and reflects, flexible thinking. That is, how do children’s minds allow them to flexibly construct representations of others’ intended meanings, and flexibly manipulate verbal structures to express dynamically changing mental representations? Part of the challenge comes from the fact that flexible cognition and language are codependent but not monogamous, so to speak. Learning language depends not only on flexible representation but on other, elusive skills of attending to and processing linguistic and nonlinguistic cues. Moreover, it depends upon an elaborate social environment, and children’s proclivities to make sense of this environment. At the same time, flexible cognition takes clients other than language: it is deployed in action systems including tool use, social interaction, spatial navigation, planning, and creative thought. Thus, flexible cognition and language learning are partly independent, and we do not know whether, and to what extent, they have been specialized for one another. This opens some fascinating questions. For example, do cognitive processes become more flexible as language is acquired? Does language learning facilitate or limit developmental changes in cognitive flexibility? Does cognitive flexibility emerge first in language use, and get recruited by other action systems? In this chapter I attempt to lay the groundwork for answering these questions by reviewing how children learn to flexibly process utterances and produce discourse-appropriate speech acts. In Section I, I define flexible cognition in a way that takes into account ecological and functional concerns of language and thought. In Section II, I briefly describe flexible language processing in adults. This suggests a new metaphorical construct, the Multi-Aspectual Representational Medium, or MARM (described in Section III). Finally, in Section IV, I review empirical and theoretical work on children’s developing ability to flexibly comprehend and produce language based on dynamic changes in their internal representational medium.

I. What is Flexible Cognition? A. PAST AND CURRENT STUDIES OF FLEXIBLE COGNITION

Before defining flexible cognition it is useful to survey past ideas and treatments of flexibility. It is notable that no historical approach considered

274

Gedeon O. Dea´k

the development of flexible cognition, which has been studied in earnest only since the 1990s. Also, historical approaches have not considered how language reflects or facilitates flexible cognition. Though a full historical review is beyond the current scope, four influential historical traditions are summarized here, with a focus on the limitations that have motivated the framework described in Section III. Early studies in the Gestaltist tradition examined adults’ flexible inferences about object functions (Duncker, 1945; Meier, 1931), and found that adults have trouble combining and using objects in innovative ways to solve problems (see also Finke, Ward, & Smith, 1992). In some situations, for example, prior knowledge of conventional object uses impedes innovative problem solving. This work thoroughly ignored language processes (but see Glucksberg & Danks, 1968) as well as developmental questions. German and Defeyter (2000), however, found that functional fixedness in one problem increased from 5 to 7 years, ostensibly as children learned the conventional uses of stimulus objects. This finding adds to the intrigue of functional fixedness effects, but their relation to other aspects of cognitive flexibility remains unclear. This is partly because functional fixedness problems are quite elaborate and difficult, and it is partly because the old Gestalt accounts are difficult to assimilate into cognitive science conceptual frameworks. A second long-lived tradition defines flexible thinking as a component of creativity, and treats both as traits that vary across individuals (Guilford, 1967; Runco, 1993; Torrance, 1988). This camp once sought to define and measure flexibility as an independent, stable trait, but had only modest success (see Hocevar & Michael, 1979; Johnson & Fishkin, 1999). Early studies found a weak but reliable correlation between older children’s verbal abilities and creative flexibility (O’Bryan & MacArthur, 1969), but those findings were hard to interpret, and since then little empirical or theoretical progress has been made. Two other traditions hold more promise for understanding flexible cognition, its development, and its role in language learning and use. One, cognitive neuroscience, has begun to study brain bases of executive functions, some of which (e.g., selective attention; active inhibition) are relevant to flexible cognition (Roberts, Robbins, & Weiskrantz, 1998). Flexible cognition seems to rely on lateral frontal cortical structures (Damasio, 1985; Grattan & Eslinger, 1991; Pandya & Yeterian, 1998), and interactions among frontal, parietal, and temporal areas and basal ganglia (Robbins, 1998). Other data implicate right frontal and temporal areas in flexible interpretation of meaning in discourse (Beeman, 1998; Fiore & Schooler, 1998; Stemmer & Joanette, 1998). Recent evidence also links children’s flexible interpretation of changing or unexpected messages to dorsolateral prefrontal areas (Diamond, 1998). Frontal cortex probably


275

plays a critical role in the development of flexible cognition and language processing, though it is likely that a variety of brain regions and neurotransmitter systems contribute materially to flexibility. The second tradition addresses flexible thinking experimentally, using task-switching methods (Allport, Styles, & Hsieh, 1994; Meiran, Chorev, & Sapir, 2000; Monsell & Driver, 2000). Participants switch from one task to another, making different judgments about the same stimuli (e.g., reading words vs. naming colors in the Stroop task). Flexibility is measured as changes in response time (RT) across a task switch; a temporary RT increase is called a switch cost. This work considers processes of attention allocation, inhibition, forward and backwards priming, and task set (see Gilbert & Shallice, 2002; Meyer & Kieras, 1997). A few studies with children have used simplified task-switch designs, but error rate rather than RT is the measure of flexibility. The constructs used to explain adults’ task-switch costs could be generalized to children’s task-switch costs,1 though no encompassing treatment has been published. In Section III, I will propose a generalized theoretical framework that can accommodate both children’s and adults’ task-switching data. B. FLEXIBLE COGNITION: A DEFINITION

Flexible cognition entails the dynamic activation and modification of cognitive processes in response to changing task demands. As task demands and context factors (e.g., instructions) change, the cognitive system can adapt by shifting attention, selecting information to guide and select upcoming responses, forming plans, and generating new activation states to feed back into the system (e.g., goals, self-correction). If these processes result in representations and actions that are well-adapted to the altered task and context, the agent can be considered flexible. I define flexible cognition as the dynamic construction and modification of representations and responses based on information (i.e., similarities, cues, relations) selected from the linguistic and nonlinguistic environment. That is, when there is a range of plausible ways to understand and respond to a problem, flexible thinkers select patterns that limit this range. The selected information must change over time as a function of shifting task 1

For example, processes investigated in task-switching studies include switch costs, task-set inhibition, proactive interference, paradoxical order effects, and preparatory facilitation (Allport, Styles, & Hsieh, 1994; Gilbert & Shallice, 2002; Mayr & Keele, 2000; Meyer & Kieras, 1997; Monsell, Yeung, & Azuma, 2000), all of which can describe flexibility of children in linguistic and nonlinguistic tasks (most of the adult task-switching studies are only superficially linguistic).

276

Gedeon O. Dea´k

demands. That is, as new problems and circumstances are imposed by the environment, the cognitive focus should shift to new, pertinent information. Flexibility is tested when changing task demands are to some degree unpredictable or novel (so the agent cannot rely on automated responses), and the conflict between alternative responses or representations is not trivial. Because flexibility is a higher-order (i.e., derivative) property of cognition, assessment requires relatively complex paradigms and measures. Traditional cognitive psychology paradigms treat responses as independent (e.g., averaging RT across all trials). But to assess flexibility, we must consider the temporal and sequential context of past events and cognitive states. For example, Ceci and Bronfenbrenner (1985) studied changes in children’s clock-checking rate as a deadline (i.e., taking cupcakes out of the oven) approached. Most 10- and 14-year-olds check the clock (i.e., select taskrelevant information) fairly often at the start of the session (possibly to calibrate an internal ‘‘clock’’), check less often until the deadline approaches, and then again check more frequently. This function of changing clock-checking rate indicates cognitive flexibility. Flexible clockcheckers adopt a covert, dynamic action plan to govern their responses over time, as their representation of task demands (based on activation states of an internal clock) changes. This definition excludes some adaptive behavior from cognitive flexibility. Making different responses in different situations is not necessarily a sign of flexibility if each response is learned separately and elicited in such simple and different contexts that cue selection is trivial. Thus, learning several S–R pairings is not tantamount to flexibility, though it is a prerequisite (because flexibility requires selection from among a number of viable responses). However, a demand to switch or re-learn S–R associations (e.g., from red ! left to red ! right) could test flexibility. Most tests of flexibility build a response set for several trials, to generate response competition, and then change task demands and assess subsequent performance accuracy or efficiency. We also distinguish flexibility from variability of behavior over time. Children’s responses naturally vary over trials or responses (Siegler, 1996), but flexibility implies more constrained, goal-directed or task-relevant (i.e., adaptive) changes in selected patterns and responses. Thus, randomly switching responses would not count as flexibility, by my definition. Finally, a common ambiguous result involves evidence that two or more samples randomly drawn from a population produce different responses to different tasks. This does not demonstrate flexibility; it merely implies it. Cognitive flexibility is a within-subjects variable: changing responses with changing task demands. This imposes challenges for testing flexibility in young children who, unlike young adults, will not indefinitely respond to pointless,


277

repetitive questions about boring stimuli. One can therefore collect only a very limited number of data points on flexibility from a single child, by administering several age-appropriate tasks within subjects. The definition above allows us to separate two often conflated concepts: flexible cognition and explicit or self-conscious reasoning (e.g., KarmiloffSmith, 1992). It is an empirical question whether flexible cognition requires controlled, explicit, or metacognitive representation, and, conversely, whether metacognitive access or control necessarily facilitates flexible thinking. There is little evidence of either dependency. Certainly, flexible cognition is most apparent when it is reflected on and reported, but it is just as clear that adults can be myopically inflexible in spite of metacognitive access and verbal self-reflection. Also, college students can be inflexible even when attempting to respond accuracy and efficiency to changing problems (Luchins, 1942), and reflecting on their failure to solve the problems (e.g., Duncker, 1945). Even when adults are made aware of demands of a novel task and the relevant information, they do not necessarily arrive at flexible, adaptive solutions (Meier, 1931). When they do find a solution, they cannot explain how (Finke, Ward, & Smith, 1992; Meier, 1937). For these reasons we assume that conscious or metacognitive thought and flexible cognition are at least partly dissociated.

C. THE DEVELOPMENTAL ECOLOGY OF FLEXIBLE COGNITION

Not all cognition is flexible. Familiar, predictable tasks or problems (e.g., social formulas like greetings; navigating the everyday route to work) are best tackled with practiced, even automatic, cognitive processes and responses. In contrast, unexpected or unfamiliar tasks require flexible cognition: task analysis, selecting task-relevant information, forming appropriate representations, and preparing novel responses. This opposition between the need for efficient response to familiar problems and the need for flexible response to novel problems poses challenges to anyone, and poses special challenges to young children. Young children are rapidly acquiring conceptual knowledge, learning routines (perceptual, motor, cognitive, language, and social), mastering tasks in complex environments (e.g., school), and acquiring skills to activate and manipulate mental states and representations. As all these develop, children engage in increasingly varied settings (e.g., preschool; elementary school) that demand flexibility. Starting day care or preschool, for example, means learning to interact with many new people whose actions, including utterances, are unpredictable. It also means learning about many materials and events, and engaging in varied tasks that are at least somewhat novel.

278

Gedeon O. Dea´k

Thus, flexible cognition in young children lies at the intersection of settings of increasing variability, and an expanding cognitive and conceptual repertoire. Language is the primary system that mediates this interchange. When preschool children encounter novel tasks they are typically in rich social contexts (e.g., play group, day care, on errands with parents) that are mediated by language. They incorporate conversation, questioning, description, explanation, narrative, and play. Verbal acts, coordinated with novel events and tasks, serve as the medium by which these unfolding events and tasks are mapped onto shared, rich representations. This suggests a critical developmental distinction: whereas in older children generating endogenous task-shifting signals is a critical skill, in preschool children many cues about task demands are explicitly verbal (e.g., from parents). Preschoolers are expected, insofar as their burgeoning language skills allow, to respond to adults’ suggestions, statements, and instructions. Similarly, preschoolers tend to narrate (i.e., verbally externalize) their plans and intentions to shift representations (e.g., in pretend play) or responses. During preschool, then, cognitive flexibility is integrally tied to overt language (Vygotsky, 1978). Changes in flexibility, therefore, couch the development of language skills from 2 to 5 years.2 For this reason, it is noteworthy that preschool children can be strikingly inflexible across changing tasks. In one sense this is unsurprising: flexible cognition involves many cognitive processes that are developing from 2 to 5 years. Some of these have been hypothesized to underlie the development of flexible cognition. One is the developing ability to inhibit prior thoughts or responses, plus memory for alternate responses and messages (Diamond, 1998). Another hypothesis is that cognitive control gradually expands to allow conscious mediation of more and more complex contingencies, and consequent improvement in task switching (Zelazo & Frye, 1997). A third proposal is that flexibility develops with the ability to notice, analyze, and select task cues, and changes in task cues (Dea´k, 2000b). These hypotheses are evaluated in Section IV. For now, it is important to note that these hypotheses require careful evaluation because evidence of 2

Not all language processing entails flexible cognition, per the foregoing definition. For example, rote speech acts (e.g., greetings; politeness routines), and some well-practiced verbal cues and responses are excluded. Also excluded are processes that rest heavily on automated retrieval or association (e.g., accessing familiar, unambiguous root words; morphosyntactic generalizations like gender and/or case agreement; phonetic adjustments like vowel harmony) would be excluded. Language processing that relies on flexible cognition includes making sense of unfamiliar narratives, engaging in informal discourse (e.g., cocktail party banter; business negotiations; mealtime conversation), or following unpredictable instructions in novel settings.


279

developing flexibility is inherently ambiguous. Though some authors (e.g., Dempster, 1992; Houde´, 2000) ascribe development to a single mechanism, the capacity to inhibit prior representations, the underspecification of this claim is easily demonstrated. Consider the finding that children younger than 36 months, when sorting pictures of animals into ‘‘animals that fly’’ and ‘‘animals that walk,’’ tend to place several successive pictures of items from both categories into the same box (Zelazo & Reznick, 1991). Such perseveration (i.e., inappropriate repetition of a prior response) is common in 2- to 4-year-olds in certain kinds of tasks. It is commonly assumed that perseveration reflects a failure to inhibit primed responses. It could, however, instead be due to weak activation of the new association, or failure to remember the current task cue. Alternatively, it might stem from failure of control over complex response choice, or failure to notice changing task cues, or failure to notice that successive questions/tasks are different. There are other possibilities, of course; the point is that we cannot just conclude that flexibility develops with cognitive inhibition. The more general difficulty here is that inflexibility is polymorphous (Dea´k, 2000b): it is not even always manifested as perseveration. In the Ceci and Bronfenbrenner’s (1985) study, for example, some children failed to take the cupcakes out, but others never reduced clock-checking. In general, there are four possible relations between forms and causes of inflexibility: (a) one form of inflexibility with a single cause; (b) one form with multiple causes (cognitive, linguistic, or both); (c) multiple forms of inflexibility with a single cause (e.g., not understanding a task prompt might cause one child to perseverate but another to haphazardly switch responses); (d) multiple forms with multiple causes. The evidence reviewed in Section IV suggests the first two and probably the third are incorrect. In short, there seem to be multiple causes and effects of cognitive inflexibility. These might change as thought and language develop. To determine this, it is useful to begin with the normative developmental ‘‘end-point’’: How is flexible cognition manifested in typical adults, and, in particular, how is it reflected in adults’ language?

II. Developing Toward . . . ? Adults’ Flexible Cognitive Processing of Meanings and Messages To understand children’s developing flexibility in language processing and production, we need to understand the mature phenotype: adults’ flexible processing and production of messages and meanings. In this section, I briefly summarize evidence of flexible cognition in the language of neurologically intact adults.

280

Gedeon O. Dea´k

A. DISCOURSE AND SHARED MEANING: FLEXIBLE FORMATION OF MEANING

Generic kind terms (e.g., object nouns) were historically assumed to have dictionary-like meanings. It is clear, though, that word meanings are partly context-specific, and are flexibly activated by adults. Aspects of word meaning are activated according to task and context factors such as proximal words and phrases (Anderson & Ortony, 1975; Barclay et al., 1974). For instance, typicality and similarity ratings of common nouns change with adjectival context (Medin & Shoben, 1988). Adults rate blue bird and black bird as more similar than blue bird and green bird, but rate blue eyes and black eyes as less similar than blue eyes and green eyes. Typicality ratings also can shift based on overt task demands (e.g., instructions to take a foreign perspective) or goals (Barsalou, 1989). Similarly, some aspects of word meaning are primed only in certain sentence contexts (e.g., an incidental property of roof such as ‘‘can be walked on’’ is evoked only by sentences about construction or repair; Barsalou, 1982, 1983). Activation of word meanings is thus best characterized as a dynamic system: semantic knowledge is selected and modified by recent experiences, cognitive activity, and the context of received messages. Discourse provides important contextual cues for flexibly constructing meaning. Fluent speakers can choose from many descriptions of an entity or event (Brown, 1958; Cruse, 1977) that highlight different attributes (e.g., collie, dog, mutt, animal, pet, Lassie, girl). Descriptions, or locutions, often are chosen based on perspectives that are collaboratively formed and modified during discourse (e.g., Brennan & Clark, 1996; Clark, 1997; Garrod & Anderson, 1987). I assume that establishing and modifying shared meaning and reference in discourse requires dynamic updating of semantic representations from an indefinitely large space of possible mappings. Competent speakers select locutions that encode and highlight aspects of mental models shared by listeners at the current point in conversation. Although adults sometimes fail to establish a shared perspective, we do not assume that this necessarily stems from cognitive inflexibility; it can, for example, result from speakers’ emotional investment in different perspectives (Danet, 1980). Yet we do assume that young children’s conversational disjunctions (e.g., toddlers’ ‘‘parallel monologues’’) are due to some kind of cognitive inflexibility. This points to a need to know how development of discourse ability is related to changes in flexible cognition—for example, what cognitive processes contribute to social dysfluencies (i.e., inflexibility) in adult discourse (Garrod & Anderson, 1987)? Although there is no comprehensive answer to this, there is


281

neuropsychological evidence that right frontal and temporal cortical regions are critical for adults’ discourse flexibility (Brownell & Martino, 1998). These regions also are implicated in processes of inhibiting and switching attention, and in updating working memory for larger units of meaning (e.g., multiple utterances, as in discourse). Perhaps the latter processes are critical to discourse fluency and flexibility; certainly they are immature in preschool children. Nevertheless, 2- and 3-year-olds can sometimes use pragmatic and discourse information to select and shift descriptions of a referent (e.g., Clark, 1997; O’Neill, 1996). Thus, whatever component cognitive processes are necessary for discourse flexibility, they are not categorically absent in preschool children. B. FORMING AND SHIFTING CONCEPTUAL MAPPINGS

Evidence of discourse and context specificity in naming and comprehension is compatible with the cognitive semantics approach, wherein utterances are conceptualized as encodings of abstract, dynamic cognitive models, or mappings. I briefly describe this approach because it is unfamiliar to many developmentalists, despite its potential to enhance our understanding of child language and its relation to flexible cognition and conceptual development. Fauconnier (1997) describes a mapping as a ‘‘correspondence between two sets that assigns to each element in the first, a counterpart in the second’’ (p. 1). For example, one waitress might say to another, ‘‘The ham sandwich at 12 wants a soda’’ (Lakoff, 1987), wherein ham sandwich designates a particular customer by reference to his or her order, 12 is a fixed designator of a table or station, and a soda stands for a more elaborate locution (i.e., ‘‘a glass of soda’’). Far more elaborate, dynamic mappings than this can emerge in everyday discourse. We seldom notice these mappings, so fluidly do we construct, consult, and update them, but they support modifiable representation of multiple referents, links, and relations. For this reason, having conversations both depends on and exemplifies cognitive flexibility. Successive ‘‘turns’’ require updating these mappings, and the shifts cannot be solipsistic: they are designed for sharing, and rely on common ground as well as conventional abstract meaning schemas, heuristics for efficient transmission of information, and negotiation of preferred locutions (e.g., Schober, 1993). A critical point is that the real-time social context of conceptual shifts means that states of the conceptual mapping scheme cannot be prestored or selected from a look-up table; they are true products of flexible thought. Also, they are not unique products of conversation; they also emerge, for example, when we hear (or read) and comprehend a story, lecture, or argument. When reading a mystery novel, for instance, we make certain

282

Gedeon O. Dea´k

inferences (e.g., who the culprit is) and the writer exploits this by setting up likely suspects, then, much later, revealing information that forces us to update or modify our extant model. The writer thereby ‘‘plays with’’ the audience’s cognitive flexibility. This highlights a more general conclusion: narrative, written or spoken, presumes the audience’s capacity for conceptual flexibility. C. NEUROPSYCHOLOGY OF FLEXIBILITY IN ADULT LANGUAGE PROCESSING

Adults’ capacity for flexible language processing can be compromised by certain brain insults, and this evidence might shed light on limitations of flexibility in children’s language. Aphasiology offers intriguing cases of reduced flexibility in naming and reference. Deficient naming, or anomia, is common in aphasia following damage to inferior left temporal and frontal cortex. Perseverative naming errors—repeating a word inappropriately over a short time—are not uncommon (Albert & Sandson, 1986; Hirsh, 1998; Papagno & Basso, 1996). One explanation is that normal inhibition of activated words is impaired, so previously-retrieved words produce response interference (Vitkovitch & Humphreys, 1991). An alternative is that activation of appropriate lexical items is reduced, so prior lexemes compete more vigorously with current lexemes (Cohen & Dehaene, 1998). However, no successful single-process account of these errors has emerged, in part because anomic errors are polymorphous. Perseverative naming seems to be influenced by exogenous factors like stimulus type (e.g., words, nonwords, pictures) and semantic content (Tranel, Damasio, & Damasio, 1997), similarity (e.g., of stimulus features, phonology, or meaning), concurrent cognitive demands, and distracting information. They also are influenced by endogenous factors such as patient age, lesion site, age-at-lesion and recovery time, and interactive factors like familiarity or age-of-acquisition. These findings might be relevant to children’s perseverative errors, which are often attributed to inhibitory failure. Findings from aphasic adults, however, suggest that children’s perseverative naming might be due to compromised understanding of the current appropriate lexical item. The relevance of aphasic naming errors to children’s errors is indicated by evidence that speech errors in anomic adults and young children are disproportionately perseverative, over a similar time-course, whereas typical adults’ errors include many anticipatory errors (Dell, Burger, & Svec, 1997; Gershkoff-Stowe, 2002; Stemberger, 1989). Injuries to other cortical regions cause another type of language inflexibility. Pragmatic flexibility, like inferring the meanings of jokes and metaphors, relies on right hemisphere processing (Beeman, 1998; Brownell


283

et al., 1983; Brownell et al., 1990; Stemmer & Joanette, 1998). Right hemisphere patients often make rigid, over-literal, and over-simplistic interpretations of jokes, stories, and indirect messages (Dennis, 1991; Grattan & Eslinger, 1990). For example, right-lesion patients may select anomalous punch lines for a joke, but cannot tell which anomalous punch lines are funny (Brownell & Martino, 1998). Young children also fail to grasp nonliteral, idiomatic or metaphoric word usage, and often do not ‘‘get’’ jokes. This parallel is complicated by a number of population differences (e.g., young children have less elaborate conceptual knowledge, less ability to infer others’ belief states, and less working memory capacity). Nevertheless, this evidence suggests a prevalent role of right temporal and bilateral frontal cortex in flexible inference about complex, socially contextualized, and nonliteral message meanings.

III. Toward a Model of Flexible Representation in Language Processing Having defined flexible cognition and discussed its ecological role in early childhood, and sketched its manifestations in adult language, we can consider evidence of children’s flexible cognition. To render such evidence interpretable, though, it will be helpful to have a theoretical framework that can encompass children’s and adults’ flexible cognitive processes in language comprehension and production. I will therefore briefly digress to sketch such a framework. Though it cannot yet yield detailed predictions, it will hopefully aid in integrating diverse findings and accounts of what develops in children’s flexible language processing. Flexible language processing critically depends on selective activation and suppression of linguistic forms and meanings. Flexibility also depends on synthesizing language cues, task demands, contextual factors, and internal cognitive states. These claims are based on the following assumptions: i. Meaningful descriptions (e.g., locutions) in speech encode a subset of information about the world. Locutions are chosen to share attention with other people by pointing out specific (real, remembered, or imagined) aspects of the world. Over successive utterances in discourse, denoted aspects of the world, and locutions that refer to them, change in unpredictable ways. Some open-ended, responsive cognitive process is needed to update underlying conceptual models and their mappings to successive utterances and locutions. ii. Conceptual knowledge is dynamic, not static (as, e.g., fixed property lists). Conceptual knowledge is activated partly based on the context of

284

Gedeon O. Dea´k

setting, task, and recent cognitive activity. Languages use a variety of syntactic, semantic, and discourse devices to reflect and influence the activation of specific conceptual information. iii. Dynamic, context-dependent processes of meaning activation and suppression are critical for understanding and producing messages. The same processes might also be central to flexible inferences about other types of information (e.g., face representation, Schyns & Oliva, 1999; tool use, Finke, Ward, & Smith, 1992; Meier, 1931; social judgment, Smith, Fazio, & Cejka, 1996). That is, there is no empirical basis for assuming that flexible meaning processing is a domain-specialized ability. Representations are defined as dynamically emergent activation states that are accessible to processing and that constrain covert and overt responses. They are seldom prepackaged or immutable, though in the service of cognitive economy some representational content, including linguistic units (e.g., morphological and syntactic dependencies; ‘‘canned’’ lexical responses) is rigid in its form and/or conditions of activation. A metaphor for dynamic, flexible representation in language processing is an amorphous mass within a fluid medium, akin to a ‘‘Lava Lamp.’’ Representations are akin to amorphous regions of higher energy within a fluid N-dimensional activation space. As these regions shift within the space, some surface or part of the region will approach the ‘‘top’’ of the medium (or lamp). This is analogous to a threshold level of activation (or energy state) that can trigger a response (e.g., lexical access). I call the space a Multiple-Aspect Representational Medium or MARM, and the representation a Multiple-Aspect Representation or MAR. It is multi-aspectual rather than multidimensional because aspects can be continuous dimensions, discrete features, nonlinear traits, or logically complex variables (e.g., hierarchical classes, thematic associates, roles): essentially any meaningful distinction that can be drawn about possible referents. Both MARM and MAR are dynamic. The shape of the MAR changes in response to input (metaphorically conceived as thermal energy currents) in the MARM (unlike a lava lamp, however, these currents are not restricted to one region of origin). These currents expand or contract different ‘‘planes’’ (i.e., representations of stimulus aspects) of the MAR. New planes can unfold as learning or attention activates a new distinction, feature, or perspective. Planes also collapse, as distracting or uninformative aspects are suppressed or neglected. Finally, a plane can be rarefied or expanded by selective suppression of, or attention to, the aspect (see Smith & Heise, 1992). It is assumed that unfolding or collapsing a MAR plane takes time (i.e., switch costs). The nature of these processes must be established empirically—for example, a relatively stable MAR (i.e., one that


285

has not shifted for some time) might require more input energy or time to shift. Input currents in a MARM are heterogeneous: they can be generated by perceived sensory data, linguistic messages, or internal activation states (e.g., a goal). But like thermal currents, input sources are both imprecise and subject to interaction and interference. Thus, if a verbal instruction fails to change the MAR, it might be that the current was not powerful enough to travel through the MARM to expand the relevant plane of the MAR, or that the relevant plane was collapsed (e.g., by suppression) so it was inaccessible to the input current, or that other currents dissipated the input current before it could influence the MAR. These possibilities correspond, respectively, to failure to encode or comprehend a message, inability to apply the instruction to the relevant aspect of the stimulus, and interference from competing, salient cues. The main claims of this metaphor are consistent with empirical evidence reviewed above. It is therefore more defensible than traditional metaphors for conceptual knowledge (e.g., filing cabinet; dictionary; conceptual network), but (I hope) more accessible than a mathematical description. It differs slightly from other dynamic systems metaphors. In Thelen and Smith (1994), for example, representational states are described as activation points in multidimensional space. This does not aptly capture the fact that multiple aspects of a represented entity each can be simultaneously described as a multidimensional vector, and it is an array of aspect-vectors that changes dynamically. This vector array is changing within a similarly dynamic space of representational possibilities which, though large, is not unconstrained (a point captured by Smith & Heise, 1992). Also, the MARM can take many kinds of input, including linguistic information, task cues, the perceptual array, and internal, self-generated inputs (e.g., top–down construal of task demands). Internally generated input eventually greatly influences children’s cognitive and linguistic flexibility (Dea´k, 2000a; Dea´k & Bauer, 1996; Donaldson, 1978; Karmiloff-Smith, 1992; Siegal, 1991; Zelazo & Frye, 1996), though it plays little role in previous dynamic system accounts (e.g., Thelen & Smith, 1994). In sum, the Lava Lamp metaphor captures the human ability to flexibly update complex representations in response to task-specific inputs and contextual contingencies.

IV. Children’s Flexible Thinking About Meanings and Messages Young children are believed to be qualitatively less cognitively flexible than are older children and adults. In this section, I explore that claim with

286

Gedeon O. Dea´k

regard to children’s language. A guiding concern is how children’s language is flexible or inflexible. Normal language errors—retrieving the wrong word, say, or misconstruing an idiom—can reveal developmental limitations of flexibility that recur with some types of brain injuries (Brownell & Martino, 1998; Cohen & Dehaene, 1998; Grattan & Eslinger, 1990; Milner & Petrides, 1984). Young children are, however, seen as qualitatively inflexible, or globally restricted from applying processes described in Section I.B. We need to sharpen that claim. Is the difference between children and adults, particularly in linguistic flexibility, qualitative or quantitative? What factors (e.g., age, verbal knowledge) predict a developmental shift toward adult-like flexibility? How many causes of inflexibility are there? For example, is children’s naming perseveration truly like aphasiac adults’ (Vitkovitch & Humphreys, 1991)? Do children’s rule-switching errors (Zelazo, Frye, & Rapus, 1996) reflect the same processing challenges as adults’ RT switch costs (Allport, Styles, & Hsieh, 1994)? In laboratory studies of 2- to 5-year-olds, cognitive inflexibility often is manifested as perseveration. In fact, perseveration is typically the only form of inflexibility studied, and, in many tasks, the only possible manifestation. It is thus ironic that perseveration is often assumed to stem from immature inhibitory processes (Dempster, 1992; Harnishfeger, 1995; Houde´, 2000). An underlying assumption is that maturation of cognitive inhibitory mechanisms allows older children to suppress prior responses, whereas younger children are compelled to repeat them. It is questionable, though, that inhibitory failure explains children’s inflexible responses to changing verbal tasks. To evaluate this claim we must examine children’s response to a wider range of flexible language processing tasks. In Section IV.A, I review evidence of children’s ability to flexibly select and re-select locutions to describe different aspects of a referent. In Section IV.B, I review evidence of children’s ability to adapt to changing discourse messages, especially changes in verbal rules. In Section IV.C, I review data on children’s ability to flexibly select linguistic cues to infer novel word meanings. Finally, in Section IV.D, I consider whether these diverse language skills reveal general age-related changes in flexible cognitive processing. A. DEVELOPMENT OF FLEXIBLE NAMING

Piaget (1954) claimed preschool children are centrated: they think about a single dimension or aspect of reality at one time (i.e., cannot form MARs). In its weak form this claim is supported by some findings (e.g., Siegler, 1981). Other findings, however, show that young children form, modify, and maintain multifaceted representations of entities and situations. In what


287

ways, then, are young children inflexible in conceptual representation and naming? Consider children’s naming errors. Toddlers tend to perseverate; to inappropriately focus on one identity or aspect of a referent. This might show Piagetian centration, but it might also be a strategic response to any difficulty of the naming task (e.g., retrieving a word that is not too familiar; uncertainty about referent’s identity; a hard-to-articulate name). Moreover, as explained previously, toddlers’ perseverative naming could stem from either inhibitory or excitory problems. The basic problem of retrieving a word for the current referent rather than a previous one might be distinct from problems of higher-order language flexibility (e.g., interpreting an idiom in light of the speaker’s ideological slant, as suggested by prior elliptical comments). The latter require conceptual flexibility. The most basic form of this is the selection of labels to highlight specific aspects of complex referents. As children’s vocabularies grow, they can produce different locutions to describe different aspects of an event, entity, individual, or category. Maybe, though, young children cannot flexibly shift the aspects they represent, and therefore cannot flexibly produce alternate labels for a referent. For example, Siegel, Saltz, and Roskind (1967) reported that children younger than 8 years believe that a ‘‘father’’ cannot also be a ‘‘doctor.’’ Similarly, Markman and Wachtel (1988) and Merriman and Bowman (1989) suggested that children have a default ‘‘one word per object’’ assumption (i.e., each thing takes one category label), so that, for example, they prefer to map a novel word onto an unlabeled rather than a nameable referent. Similarly, the appearance– reality test poses two questions (e.g., ‘‘what does this look like?’’ & ‘‘what is it really?’’) about objects that can be classified by function or by appearance, like an apple-shaped candle. Three- and 4-year-olds tend to answer both questions with the same label (e.g., candle), suggesting a rigid focus on one aspect (Flavell, Flavell, & Green, 1983; Flavell, Green, & Flavell, 1986). Such evidence suggests that preschool children are inflexible in their conceptual representations of an entity. Other findings, however, show that children as young as 2 years can represent multiple aspects of a complex referent, and produce locutions for these. Evidence for this conclusion (Clark & Svaib, 1997; Dea´k & Maratsos, 1998; Dea´k, Yen, & Pettit, 2001; Sapp, Lee, & Muir, 1999) has been reviewed elsewhere (Clark, 1997; Dea´k, 2000b). Most pertinent are findings that 2- to 4-year-olds respond to a series of pragmatically and semantically sensible questions about a complex object by producing several different, appropriate labels. They are limited in this by the breadth of their lexicon, not by their conceptual inflexibility. Children can shift perspective, and labels, within a few seconds (depending on discourse and event context);

288

Gedeon O. Dea´k

there is no evidence that they do so much slower than adults. Also, they do not simply generate and discard a series of labels, keeping only one active at a time: when asked to verify label pairs, 3- and 4-year-olds accept appropriate pairs but reject most ‘‘foil’’ pairs (where one word is replaced by a same-category associate). Neither training, nor familiarity with the specific label pair, is necessary. The following exchange (from Dea´k & Maratsos, 1998; Experiment 1), involving a novelty pen that looks like an ear of corn, exemplifies children’s facility: Experimenter: ‘‘What do you think that is?’’ Child (3;5 female): ‘‘Pen.’’ Experimenter: ‘‘What else could that be? Anything else?’’ Child: ‘‘Corn.’’ Experimenter: ‘‘Corn. And what is corn? Is corn a kind of animal?’’ Child: [shakes head] ‘‘Food.’’ [Child then affirms each label pair, and reject several foil pairs such as ‘‘eraser and corn’’]

This capacity is not isolated to novelty items. Dea´k and Maratsos found, contrary to Siegel, Saltz, and Roskind’s (1967) claims, that 3-year-olds readily, consistently accept several appropriate labels for a character in a brief vignette (e.g., ‘‘woman,’’ ‘‘doctor,’’ ‘‘mother,’’ and ‘‘person’’). The results indicate that Siegel, Saltz, and Roskind results were attributable to procedural artifacts. Such evidence falsifies three untenable claims about semantic inflexibility in preschoolers: . .

.

Preschool children cannot simultaneously keep active more than one category representation for a referent (Flavell, Green, & Flavell, 1986). Preschool children produce or accept only one word for a complex entity (Markman, 1994)—or, more specifically, do not allow both function and appearance labels for a representational object (e.g., dog puppet; Merriman, Jarvis, & Marazita, 1995). Preschool children assume that a referent entity has only one label until they receive specific input or training is provided to demonstrate otherwise (Markman, 1994).

Several other possible limits on semantic flexibility are untested and therefore plausible: .

.

Preschool children are somewhat slower or less consistent than older children in shifting a MAR to focus on (and name) different aspects of a referent. Preschool children are somewhat slower or less consistent than older children in activating different words to describe a complex entity. That


289

is, after the representation has shifted, the selection of a new description is sluggish. Two final hypotheses about limits on representational flexibility have empirical support: .

.

Preschool children attempt to simplify the task of learning a novel word. Strategies seem to include temporarily ignoring or inhibiting known words for a referent, or ignoring the novel label, while working out respective meanings (Dea´k, Yen, & Pettit, 2001). Some results (Dea´k, 2001; Hughes-Wagner & Dea´k, 1999; Liittschwager & Markman, 1994; Rice et al., 1997) suggest this increases with ‘‘cognitive load’’ (i.e., number of to-be-learned items; memory demands). When children learn novel words or infer complex semantic relations, they tend to adopt simplified schemes for mapping the words onto aspects of candidate referents. If only a few of each child’s naming responses are assessed, this tendency will make the child appear to accept only one word per referent; this is, however, misleading. Children younger than 4 or 5 years have trouble determining which aspect of a complex referent is indicated by each of several labels. Their difficulty may lie in mapping the predicate of each question to a specific stimulus aspect, and then to a corresponding label. For example, 3-yearolds can label a dinosaur-shaped crayon ‘‘dinosaur,’’ ‘‘animal,’’ and ‘‘crayon,’’ but cannot judge whether each word names ‘‘how it looks’’ or ‘‘what you do with it’’ (Dea´k, Yen, & Pettit, 2001). More strikingly, 3and 4-year-olds who perseverate in the appearance–reality test also perseverate when answering two questions (e.g., ‘‘What is this?’’ and ‘‘What does it have?’’) about nondeceptive items (e.g., picture of an ape holding cookies; Dea´k, Ray, & Brenneman, 2003). Thus, perseverative naming as in the appearance–reality test stems from nonspecificity of predicate$word mapping, not from inability to represent dual identities (Flavell, Green, & Flavell, 1986; Gopnik & Astington, 1988).

In sum, preschool children can select and interpret alternative labels for a given referent (though of course this ability will continue to develop). The process is temporarily obscured in difficult learning tasks, and there is another factor involved: discourse knowledge. When young children are asked a series of different questions, and they do not understand what each question is ‘‘about,’’ they are willing to use a strategy that adults would not consider: repeating the same answer. Adults seem to strongly expect different questions to imply different aspects of reality, and thus require different responses; preschool children do not hold this expectation.

290

Gedeon O. Dea´k

B. DEVELOPMENT OF DISCOURSE FLEXIBILITY: SHIFTING IMPLICATIONS AND INSTRUCTIONS

We have seen that some basic skills of flexible representation and locution-selection are observable in 2- and 3-year-olds. Yet young children seem rather inflexible conversational partners. Why? Certainly there is massive growth of the lexicon, and of background knowledge that contributes to common ground. But beyond this, does representational flexibility change categorically with age, as Piaget and others suggested? What is missing in preschool children from the more sophisticated adult capacity to respond flexibly to discourse or narrative? One thing that does not change categorically from childhood to adulthood is the basic problem of discourse and narrative processing. A demand of everyday communication is flexible selection or representation of descriptive terms for topics of ongoing discourse and narrative (e.g., Brennan & Clark, 1996; Garrod & Anderson, 1987; Schober, 1993). Consider, for example, the demand to construct and modify conceptual maps from narrative. The mapping flexibility that novelists or dramatists presume of adults is ‘‘scaled down’’ for children: whereas adult readers can represent a story told in reverse (e.g., Time’s Arrow by Martin Amis) or with shifting perspectives (Mrs. Dalloway by Virginia Woolf), children’s books tend to have a linear story line and fewer (and better marked) shifts. Still, many enduring children’s stories compel their audience to represent a shifting sequence of mappings (albeit fairly simple, explicit ones). In some stories, a series of parallel mappings changes predictably. For example, in the toddler’s book Are You My Mother? by P.D. Eastman (Eastman, 1960), a baby bird searches for its mother and encounters a series of (implausible) candidate mother-objects. The young audience must represent a series of mappings, to assess the plausibility that each candidate could be the mother. Many stories for preschoolers incorporate shifting conceptual mappings that are imageable and concrete (and often supported by redundant perceptual information; i.e., pictures). There are many intriguing questions about the role of narrative in children’s flexible language skills. What succession of representations is activated when children first hear a story; when they hear it multiple times? Does hearing many stories contribute to the ability to flexibly form novel conceptual mappings from narrative? At present, little evidence addresses these questions. There is evidence that preschool children have trouble integrating unexpected information with mappings constructed from narrative. Campbell and Bowe (1983) read 3- to 5-year-olds brief stories in which the nondominant meaning of a homonym (e.g., /haˆr/ ¼ rabbit) was strongly implied by context (e.g., ‘‘The hare ran across the road’’). Afterwards, when


291

asked to draw a picture of the /haˆr/, many children depicted the dominant meaning (e.g., strands of hair) though it was completely inconsistent with the narrative. So something must develop in children’s flexible construction of mappings from narrative. Though there is little data on this, there is evidence that as children become better readers, they can integrate information and judge consistency across more remote sentences (Schmidt & Paris, 1983). There are, however, related data on children’s flexible construction of meaning in discourse. This also improves considerably across the preschool years, possibly in parallel with narrative processing. In discourse, preschoolers often fail to notice that a message is uninterpretable or ambiguous (Cosgrove & Patterson, 1977; Markman, 1979; Revelle, Wellman, & Karabenick, 1985), fail to understand jokes, metaphors, and nonliteral idioms (Chukovsky, 1968; Gombert, 1992), and produce ambiguous referential messages (Glucksberg, Krauss, & Weisberg, 1966). Apparently, 2- to 5-year-olds do not use discourse context to resolve or represent ambiguity about possible meanings of a message. Given that preschool children can produce multiple labels for an entity, the problem is not one of producing multiple mappings, but of generating multiple meanings in response to complex linguistic messages (i.e., utterances with complex meanings). This difficulty is not trivial or artifactual: in training studies (Sonnenschein & Whitehurst, 1983), children younger than 6 years do not learn to detect ambiguous messages. Nor is the difficulty one of syntactic or lexical competence, both of which are fairly advanced by 4–5 years. One possible explanation rests in the limited ability of children younger than 6–7 years to detect indeterminacy (e.g., Fabricius, Sophian, & Wellman, 1987). Dea´k, Ray, and Brenneman (2003) found that individual differences in this ability predict children’s flexible response to appearance–reality questions. In the Indeterminacy Detection task, children are shown situations with indeterminate outcomes (e.g., what color chip would be pulled next from a box containing chips of many colors) or with determinate outcomes (e.g., all chips of one color). Children are asked to judge whether they could know the outcome ‘‘for sure’’ or whether they ‘‘have to guess.’’ Children who can judge whether a situation has a definite outcome or is unpredictable also select different labels for different aspects (i.e., ‘‘looks like. . .’’ and ‘‘is really. . .’’) of deceptive objects. Although such correlational data are ambiguous, one interpretation is that children’s growing awareness of possibility—the potential for alternative mental models to be consistent with given information about a question or problem—facilitates the activation and selection of multiple aspects (i.e., MARs) that cover multiple possible mappings. Nonverbal tasks show an age-related increase, between 4 and 7 years, in ability to notice multiple possible clues and answers to a problem (e.g., Vurpillot, 1968). In discourse or narrative, however, premature resolution of

292

Gedeon O. Dea´k

a message’s possible meanings would make children relatively quick to judgment and rigid in their mappings (i.e., interpretations). This account, which suggests that a higher-order logical skill (i.e., noticing indeterminancy) facilitates the development of linguistic flexibility is, however, speculative and in need of converging evidence. 1. Flexibility in Responding to Rules and Instructions The ability to respond flexibly to changing and unpredictable messages, such as instructions or rules, improves substantially with age. Luria (1959) outlined a progression of skill in responding flexibly to verbal messages. When asked to retrieve a specific familiar object, 1-year-olds fail if a different, salient object is closer. Also, after making a response several times, they persist when a different response is mandated by the task (Diamond, 1998). When given a simple rule ‘‘If [X occurs], do [action1],’’ 2-year-olds tend to produce the action before X occurs. When given a salient but rule-invalid response signal, as in ‘‘Simon Says’’ (Reed, Pien, & Rothbart, 1984), 2- and 3-year-olds often impulsively produce the response. Also, in simple switch tasks involving bi-conditional rules (e.g., ‘‘When X happens, do [action1]; when Y happens, do [action2]’’), 2-year-olds often fail to switch (Zelazo & Reznick, 1991). Switching between conditional rules for alternate responses (see below), or switching from familiar to novel responses (e.g., ‘‘say ‘night’ when you see a picture of the sun’’; Gerstadt, Hong, & Diamond, 1994), is difficult until 4 or 5 years of age. These findings seem to show a clear progression in flexible messageprocessing skill, but the forces behind this progression remain to be specified. Empirical work has focused on children’s flexible response to changing conditional rules. The Dimensional Change Card-Sort test (DCCS), refined by Zelazo and colleagues (e.g., Zelazo, Reznick, & Pinõn, 1995), distills the rule-change problem. In this task children hear an unambiguous rule for choosing one of several responses (e.g., ‘‘blue things go in this box, and red things go in that box’’). After several trials a new rule is given, demanding a response switch (e.g., ‘‘cars go in this box, and flowers go in that box’’).3 Usually each rule is easy; the question of 3 The DCCS has been likened to a simplified Wisconsin Card-Sorting Test (Milner & Petrides, 1984), but there are significant differences: in the DCCS only 1–2 rules are used (and only two stimuli), yielding a much smaller problem space. Each rule is explicitly stated, making it a deductive, not inductive, test. Therefore, although both the DCCS and WCST test response switching, only the WCST assesses response-set learning rate or changes in learning rate (Lezak, 1995, Chapter 15), dependent variables that might differentiate adult frontal patients with different syndromes (Daigneault, Braun, & Whitaker, 1992; Taylor, Saint-Cyr, & Lang, 1986). Also, perseveration in the DCCS is less informative than perseveration in the WCST because any post-switch error in the former is, by default, coded as perseverative.


293

interest is whether the rule change carries a switch cost. The DCCS is schematized in Figure 1, using lighter and darker-colored moons and stars as stimuli, for purposes of illustration. Table I summarizes results from several studies of preschoolers’ DCCS performance. Between 3 and 4 years, the probability a child will switch responses according to the post-switch rule, rather than staying with the preswitch rule, increases significantly. Because children who perseverate seem to grasp both rules, Zelazo and Frye (1996) describe this as a dissociation of knowledge and action selection. Several other findings are notable. First, the age difference holds up when pre-testing is used to eliminate children who do not understand either the task or the shape and color words. This eliminates some mundane explanations of children’s errors. Second, the rules are almost always ‘‘sort by shape’’ and ‘‘sort by color,’’ applied to two simple, familiar shapes and colors combined in two low-dimensional stimuli (e.g., simple drawings of a red car and a blue flower, or, as in Figure 1, of a light-colored

Fig. 1. Schematic illustration of the DCCS test: the pre-switch task includes an instruction and a sorting phase; the latter typically lasts from 1 to 6 trials. The post-switch task, which immediately follows, includes a re-instruction phase and a post-switch sorting phase. See text for additional information.

Gedeon O. Dea´k

294

TABLE I Summary of Reported Studies of the DCCS: Percentages of Children Who Perseverated on the Pre-switch Rule During the Post-switch Phase, by Age (3- vs. 4-year-olds) Percentage of perseverative children Study

3-year-olds

4-year-olds

Frye et al. (1995) Exp. 1 (n ¼ 40) Frye et al. (1995) Exp. 2 (n ¼ 40)1 Zelazo et al. (1996) Exp. 1 (n ¼ 30)

87% 67% 60%

68% 27% 10%

Mean

71%

35%

Note: Summarizes only those published experiments that report numbers of 3- and 4-year-olds who made 75–100% post-switch responses by a given rule (pre-switch: perseverative; post-switch: flexible). 1 Mean of two DCCS tests (given within-subjects).

moon and a darker star). These are shown repeatedly across trials. In this task perseveration might stem from: (a) interference from a specific value (e.g., child cannot stop attending to red when it occurs), (b) a value–location association (e.g., once red things are associated with the right box, the child avoids putting them in the left one), (c) a value–location–motor response contingency (e.g., after placing red things on the right, it is hard to put them on the left), (d) selective attention to one dimension (e.g., after attending to color, it is hard to attend to shape), or (e) persistence of the first abstract rule (e.g., after adopting a color-sorting rule, it is hard to adopt a shape-sorting rule). The source of 3-year-olds’ errors is therefore ambiguous. Recent studies have reduced this ambiguity. Jacques et al. (1999) showed that children who perseverate on the DCCS judge that a perseverating puppet is correct but a rule-switching puppet is incorrect; conversely, children who flexibly switch rules make the opposite judgment (i.e., perseverating puppet is wrong). Also, Zelazo, Frye, and Rapus (1996, Experiment 3) showed that 3-year-olds perseverate just as much when making a verbal (nonsorting) response. Thus, motor responses are not critical. Perner and Lang (2002) found 3-year-olds can make a withindimension rule switch that requires reversing a value$location association, and Towse et al. (2000) found that using new boxes in the post-switch trials did not improve flexibility. Thus, perseveration is not based on interference from prior motor responses or locations (eliminating (b) and (c)). Also, 3-year-olds can, in some tasks, switch responses (eliminating (a)). It seems instead that their difficulty lies in updating the current rule, or switching attention to a new, relevant dimension.


295

Other evidence sheds light on the nature of children’s switching errors, and the dissociation between knowledge and sorting response (Zelazo & Frye, 1996). For example, 2.5-year-olds show a similar dissociation in a simpler test of sorting cards into two categories (e.g., animals that fly vs. animals that walk). Sorting yields more errors than verbally verifying the relevant category each item (e.g., answering ‘‘yes’’ or ‘‘no’’ when asked ‘‘Does this one walk?’’ Zelazo & Reznick, 1991; Zelazo, Reznick, & Pinõn, 1995). Stimuli in this paradigm are more diverse than DCCS stimuli, suggesting that repetition of specific items is not necessary for perseveration, though it should be noted that two-thirds of children’s errors were perseverative, which is exactly what we would expect if children were responding by chance. Are 2.5-year-olds’ errors in Zelazo and Reznick’s (1991) classification task due to the same factors as 3-year-olds’ DCCS errors? In an important regard the classification task is different: it requires a simpler binary distinction, not a contingent choice between bi-conditional rules. That is, children’s failure in the DCCS is in switching from one category-based response to another, or in updating the aspect implicated by the latest rule. Zelazo and Frye (1996) suggest that this added complexity accounts for the higher age of mastery. However, the age difference might be due to an added discourse demand. The DCCS requires knowing that social rules often are mutually exclusive, so a new rule replaces older ones. That knowledge is likely acquired in semiformal social contexts like preschool; thus, older preschoolers are more likely than younger ones to have this discourse knowledge and use it in a rule-switching situation. In the following subsections (IV.B.1.a–c), I will elaborate on some alternative accounts. a. Evaluating Rule-following Flexibility: Symbolic and Lexical Mapping. Why do children make perseverative errors if, in both the DCCS and the classification task (Zelazo & Reznick, 1991), they know enough to solve each problem? One possibility is an unrecognized mapping demand: the ‘‘right’’ response, in both sorting tasks, is arbitrary in a way that is rather odd to children. Nothing in the stimulus array or prior knowledge stipulates, for instance, putting blue things in a box on the left, or putting ‘‘animals that fly’’ into a box on the right. This arbitrary mapping is akin to algebra or predicate logic (e.g., ‘‘Let all red things be X’’). Children might respond to such seemingly arbitrary mappings without understanding that they are merely conventional, and can be invalidated or switched by agreement or by a verbal signal. By contrast, mappings with which preschool children are more familiar—namely, word-referent mappings—also are abstract and arbitrary, but do not change spontaneously.

296

Gedeon O. Dea´k

This is, perhaps, why knowledge tests in the DCCS and the classification test (e.g., ‘‘Is this a flying thing?’’) are easier: they only require retrieval of a known (static, not arbitrarily changing) fact or designator of the referent. Another idea concerns rule meaning rather than mappings. Maybe children perseverate if they cannot select the aspect designated by the predicate of the current rule (e.g., ‘‘In the shape game. . .’’). That is, some children might have trouble selecting blue-ness or red-ness (rather than dogness) based on the aspect labeled in the rule (i.e., ‘‘shape game’’). There is very little data on children’s understanding of dimension words like ‘‘shape’’ and ‘‘color’’: Shatz and Backscheider (2001) reported that some toddlers map dimension words onto appropriate value words (e.g., ‘‘red’’), but perhaps many 3-year-olds still have a tenuous grasp of the dimension labels. Munakata and Yerys (2001) suggested that this is why some 3-year-olds do not differentiate successive rules. They modified the DCCS knowledge question (e.g., ‘‘Where does the _____ go in the shape game?’’) so items were described by both aspects (e.g., ‘‘red car’’) of a picture. As a result, 3-yearolds did poorly answering knowledge questions. It seems they cannot use the aspect named in the rule to choose the correct word from a complex locution (e.g., ‘‘red car’’). Other indirect evidence comes from a variant of the DCCS with three rules (Narasimham & Dea´k, 2001): pilot testing revealed it was indeed harder for preschool children to use the abstract dimension labels ‘‘shape,’’ ‘‘size,’’ and ‘‘color’’ to classify complex items (e.g., a small red bird), even when the value labels (e.g., ‘‘small’’) were familiar. This seemed to reflect uncertainty about the meaning of the rules. This could also explain Towse et al.’s (2000) finding that 3-year-olds’ perseveration diminished when the post-switch rule was demonstrated (like the pre-switch rule), and Bohlmann’s (2001) finding that 1–2 trials of feedback in the post-switch phase eliminated 3-year-olds’ perseveration. b. Evaluating Rule-following Flexibility: Complexity. The dominant account of children’s rule-following flexibility (Zelazo & Frye, 1996) is based on the degree of rule complexity a child can represent and use. Two-yearolds do not flexibly shift responses using a binary (or 1 level) rule contingency (see Figure 2). Three-year-olds do not flexibly choose a 2 level rule to select subordinate contingent responses (e.g., DCCS). Thus, from 2 to 4 years, children acquire controlled flexibility to make progressively more complex rule-based responses. Challenges to the Cognitive Complexity and Control, or CCC, theory (Zelazo & Frye, 1996) have focused on whether contingency complexity is indeed the critical limiting factor in children’s flexibility, and whether


297

Fig. 2. Illustration of levels of complexity of the contingencies (rules) in the DCCS, after Zelazo and Frye (1996). I have added an overarching principle/rule concerning expectations for adapting successive responses to verbal instructions.

analyses of rule complexity are adequate (e.g., Perner, Stummer, & Lang, 1999). It is possible that Zelazo and Frye underestimate the complexity of the DCCS because an additional, implicit pragmatic rule or principle is needed to respond flexibly or consistently in different verbal or questioning situations: on every trial, use the last rule to select a response, until a different rule is given. Figure 2 adds this principle to the apex of the hierarchy. This should guide responses to each item, regardless of prior responses and the number of rule switch or time since a rule switch. The principle is presumed (but not stated) in, besides the DCCS, many formal problem-solving activities (e.g., school work), though not all social situations impose it; for example, free play in preschool is guided by the principle choose from allowable activities and switch at will until instructed to stop. Adults do not typically explicate these principles, so children may choose the wrong one. If, in the DCCS, children use a principle like make the most familiar response until corrected, which works perfectly well in many social situations, they will perseverate. This suggests children’s perseveration might stem from incorrect interpretation of the pragmatics of rule-following tasks, instead of (or in addition to) limits on controllable rule complexity. Though there is limited evidence for this hypothesis, Perner and Lang (2002) gave 3- and 4-year-old children a standard DCCS, as well as three easier switching tasks. Three-year-olds produced the usual pattern of perseveration only in the standard DCCS, and only when it was given first. In contrast, they flexibly

298

Gedeon O. Dea´k

switched rules for giving cards to cartoon characters, based on stated preferences. For instance, changing the rules from Mickey Mouse likes red things and Donald Duck likes black things to Mickey likes circles and Donald likes squares elicited near-ceiling performance. This task is as complex as the DCCS, so complexity was not governing performance. More strikingly, when 3-year-olds did any such easier task first, they then generalized the rule-switching strategy to the DCCS. This suggests that perseveration follows adoption of the wrong implicit pragmatic principle, and a ‘‘task set’’ for adopting the correct principle will yield nonperseverative performance. Another problem for the CCC is that complexity limits do not seem to generalize across linguistic reasoning tasks. The syntax and morphology of a language, for instance, form complex contingency systems wherein multiple variables must be considered to produce or interpret utterances (Maratsos, 1998). Two- and 3-year-olds accurately sift through multiple, ambiguous variables (e.g., definiteness, case, noun class, number, tense, aspect, etc.) and learn, without explicit help, to assign ambiguous constituents to roles in a sentence, to inflect constituents correctly, etc. (e.g., MacWhinney, 1978). Such variables are denoted, across languages, by diverse, hard-to-analyze, unpredictable markers. Yet for most markers in most languages, 3-year-olds can track and adapt to several unpredictably changing variables at once, and in real time (i.e., quickly). For instance, MacWhinney, Ple´h, and Bates (1985) found that Hungarian 2- to 3-year-olds use case and animacy cues to assign nouns in transitive sentences to subject or object slots. Each child did this accurately over many sentences that varied unpredictably (and repeated constituents, thus requiring flexibility). Other studies have shown that English-speaking 2-year-olds take into account syntactic, semantic, and conceptual contingencies to interpret sentences; for instance, they use definiteness and animacy to infer the referent of a novel noun (Katz, Baker, & Macnamara, 1974). Also, 2-year-olds consider object location and listeners’ knowledge when deciding how much descriptive and deictic information is needed to request an object from an adult (O’Neill, 1996). Detailed analysis of such examples reveals the difficulty of objectively formalizing the complexity of different linguistic contingencies, but it is clear that 2- and 3-year-olds, in everyday discourse, interpret messages with respect to combinations of linguistic cues that exceed complexity limits proposed by Zelazo and Frye (1996). Another question concerns branching complexity (i.e., rule breadth, not depth) and flexibility across changing rules. In the DCCS, stimuli, questions, and responses are simple and repetitive. Dea´k (2000b) speculated that this reduces 3-year-olds’ flexibility, because they are ‘‘lulled,’’ during pre-switch trials, into believing that they have deduced the right response for each card, and thereafter stop analyzing verbal input (i.e., instructions). When the new


299

rule is stated, children’s certainty in ‘‘known’’ responses (plus the extensive similarity of post- and pre-switch trials) outweighs the new instruction and corresponding opposite responses. For older children, in contrast, the most recent rule principle allows the new instruction to prevail. Perhaps making the test more difficult and less predictable would push 3-year-olds from comfortable, practiced, confident responses toward reliance on a verbal rule. That is, children might rely on verbal cues (e.g., rules) when other salient cues become less predictable. This leads to a counterintuitive prediction: perseveration should decrease in more complex and variable tasks. Narasimham and Dea´k (2001) tested this in a modified card-sorting test. The 3DCCS varies three aspects of the stimuli, so cards are sorted by three different rules. This allows two rule shifts, rather than one, and increases the number of 1 level contingencies from four (in the DCCS) to nine (due to increased branching of the contingency tree at the 1 and 2 levels). Thus, though the 3DCCS is unaltered in complexity as defined by Zelazo and Frye (i.e., branching depth), it has more variable and diverse stimuli as well as greater rule cardinality. In the 3DCCS, children sort cards showing an animal (dog, bird, or fish) in one of three colors (red, blue, or yellow) and sizes (‘‘little,’’ ‘‘middle,’’ ‘‘big’’). Six cards are randomly chosen so that each value is depicted twice, in different combinations. These are sorted into one of four boxes, each defined by a unique sample card (e.g., small yellow dog, medium red bird, large blue fish, and long black snake [distractor]). Children sort each test card three times using three different rules: shape, color, and size. The procedure is illustrated in Figure 3. Results from the 3DCCS are compatible with data from the DCCS. As shown in Figure 4, 3-year-olds make fewer correct responses overall than 4-year-olds, and 3-year-olds’ rule compliance declines more sharply after the first block. Four-year-olds are more likely to follow both post-switch rules, demonstrating an age-related increase in flexibility using a more variable test with the same contingency depth as the DCCS. However, children produced a wider range of response patterns in the 3DCCS than in the DCCS. Some children (25%) followed one of the two post-switch rules; others (10%) changed their responses rather indiscriminately. Also, whereas in the DCCS most children follow one rule (pre- or post-switch) on all post-switch trials, in the 3DCCS they were not so consistent: 20% changed their selected aspect at least once within a block. Thus, a more variable test revealed that the dichotomy of flexible versus perseverative children (defined as up to 17% errors or at least 83%) is not so sharp as it appears in the DCCS (see Table II). So increasing number and diversity of rules and stimuli does not eliminate age differences, but it reveals a general problem with our focus on perseveration: in many popular tasks such as the DCCS, the only possible error is perseveration.

300

Gedeon O. Dea´k

Fig. 3. Schematic illustration of the Three-Dimensions Change Card-Sort test (3DCCS). The pre-switch task is followed by two post-switch tasks, each with a different rule, and six diverse sorting stimuli per rule. On each trial children must choose between four boxes, one of which shows a shape match, one with a color match, one with a size match, and one with a distractor. See text for additional information.

In sum, much evidence is consistent with the CCC theory, but a few findings (e.g., Perner & Lang, 2002) are not, and there are serious concerns that the theory cannot predict children’s competency in language processing tasks that do not involve rules, but nevertheless require knowledge of complex contingencies among syntactic, morphological, and semantic cues. c. Rule-following Complexity: Inhibitory Accounts. The most prominent alternative to the CCC focuses on general inhibitory capacity (Dempster, 1992; Houde´, 2000). Immaturity of prefrontal cortex (which persists until adulthood) is believed to prevent efficient inhibition of prior responses or representations, thus increasing perseveration. This account is appealing and parsimonious. If correct, it reduces preschoolers’ inflexibility to a single cognitive process. Unfortunately, the account is so vague that it is unfalsifiable without much elaboration. Also, perseveration is not

301


Fig. 4. Mean number of correct responses (out of 6) to the first, second, and third rules in the 3DCCS, by age (n ¼ 61). Bars show standard errors.

TABLE II Number of Flexible, Partly Flexible, and Perseverative 3-year-olds (n ¼ 16) and 4-year-olds (n ¼ 26) in the 3DCCS Response patterns Age

Flexible

Partly flexible

Perseverative

3-year-olds 4-year-olds

2 (12%) 14 (54%)

6 (38%) 6 (23%)

8 (50%) 6 (23%)

necessarily caused by inhibitory failure. We must specify exactly what perseverative children fail to inhibit, what neural processes are involved, and what tests could falsify a claim that every case of perseveration is due to inhibitory failure. Consider the claim for a prefrontal cortical mechanism. No study has showed an increased rate, or biological milestone, in maturation of prefrontal cortex between 3 and 4 years. This would be important to demonstrate to support the claim that improvement in rule-switching flexibility is a result of maturation of frontal cortex-controlled inhibitory

302

Gedeon O. Dea´k

processes. Also, no study supports a general inhibitory trait that predicts age and individual differences in children’s flexibility. A study by Carlson and Moses (2001) found little coherence across tests of inhibition in preschool children. Across 10 tasks, all hypothesized to require cognitive inhibition of responses to rules or requests, the mean simple correlation was r ¼ .28. The mean, with age, gender, and verbal IQ partialled out, was r ¼ .16. Thus, no unitary underlying trait was evident, and more homogeneous subsets of tasks, characterized by conflict between competing responses to instructions (e.g., DCCS), were not much more cohesive: their mean partial correlation averaged only r ¼ .22. The idea that a general inhibitory trait can account for age and individual differences in flexibility thus remains unconfirmed. Perhaps, however, a better specified inhibitory account is viable. Diamond (1998) argued that holding several relations in mind (e.g., lower-order response contingencies), while inhibiting highly activated responses, is difficult for children younger than 6 years, due to immaturity of a region of prefrontal cortex. That region is implicated in nonhuman primates’ performance on tests of working memory-plus-inhibition (e.g., delayed nonmatch to sample; Diamond & Taylor, 1996). These demands are at levels in the DCCS that challenge typically developing 3-year-olds; the same demands are presumed greater in another test, the Stroop Day/Night (Gerstadt, Hong, & Diamond, 1994), that challenges older children. In that test, children must inhibit a learned verbal association: they are instructed to say ‘‘day’’ when shown a picture of the moon, and ‘‘night’’ when shown a picture of the sun. Note, however, that the Stroop task has not been shown to have objectively greater inhibition-plus-memory demands. Regardless, Diamond et al. (1997) found that children with mild, treated phenylketonuria (PKU), which impairs the prefrontal system, perform worse than same-age controls on these tests. For example, children with PKU as old as 5 years showed increased error rates in the DCCS. Diamond’s theory provides much-needed sharpening of the general inhibitory account, but it cannot explain all available data. At issue is the claim that memory for rules or verbal input is a second demand that makes the DCCS, delayed nonmatch to sample, and Stroop tests difficult for preschool children. However, Dea´k, Ray, and Brenneman (2003) found that several verbal capacities, but not verbal working memory span, predicted preschoolers’ perseverative appearance–reality errors. Also, Zelazo, Reznick, and Pinon˜ (1995) found little or no effect of working memory load on children’s rule-following flexibility. Finally, Perner and Lang’s (2002) finding that 3-year-olds fail the DCCS but succeed at other rule-switching tasks with similar inhibition and memory demands cannot easily be explained by Diamond’s theory.


303

Thus, the role of working memory in flexible rule-following remains hypothetical. Diamond is probably correct that some process of encoding and retrieving representations of rules or instructions is critical to flexibility. Inability to remember rules would surely impair flexible rule switching. Also, switching from one rule to another sometimes will require suppression of familiar responses. In some sense, then, Diamond’s theory is probably correct, but there is little evidence that individual and age differences in working memory and inhibitory capacity can account for the difference between children who perseverate and those who are flexible. d. Rule-following Complexity: Summary. These accounts do not exhaust the range of possible accounts of age and individual differences in flexible response to rules and instructions. For example, Dea´k (2000b) emphasizes the role of growing understanding of pragmatic cues to task demands and changes in the development of flexibility. But this account is not necessarily incompatible with either the CCC or an inhibition-plusmemory account. For example, a pragmatic cues account shares with CCC an emphasis on correctly representing task demands, though it differs from CCC by incorporating the child’s knowledge of discourse conventions. Each account also can be accommodated by the MARM metaphor. For example, complexity can be coded as the number of aspects that remain active in the MAR. Working memory can be coded as the number of aspects that remain sufficiently active to be available for response or description. Pragmatic cues draw children’s attention to the input currents intended by the adult. Rule-switching paradigms have yielded important data on the development of children’s linguistic flexibility. Yet they represent a limited range of language tasks. Though caregivers sometimes give young children explicit instructions or rules, they probably avoid giving series of changing instructions. Also, caregivers might tolerate a fairly high rate of noncompliance, and repeat rules or give assistance (e.g., feedback) as needed. Thus, it is unclear how results from rule-switching tests extend to everyday language processing. This limitation must be addressed with naturalistic studies of adult–children communication. For now, however, we can compare results from tests of rule-switching to results from other tests of linguistic flexibility. One of these, inferring word meanings, is considered next. C. FLEXIBLE USE OF VERBAL CONTEXT TO INFER WORD MEANINGS

1. Age-related Changes in Flexible Induction of Word Meanings In a few short years, children learn the meanings of thousands of words and locutions. The extensive literature on children’s word learning and

304

Gedeon O. Dea´k

vocabulary development (Bloom, 2000; Dea´k, 2000a) has only begun to address how children can flexibly respond to changing cues to the meanings of the unfamiliar words they hear. That is, though many studies have tested children’s use of various cues to infer a word’s possible meaning, few have tested how children adapt their representation of possible meanings as they hear new, unpredictable cues and messages. Yet flexible cue selection is critical because everyday discourse carries many kinds of cues to meaning, and these change from utterance to utterance. Even the same utterance can carry opposite meanings in two contexts (e.g., ‘‘Well, that was a great movie!’’). As in this case, prosody may be the critical cue; often, though, it will not. The next utterance’s meaning might rest on a different combination of syntactic, lexical, semantic, discourse, and paralinguistic cues. Even this does not exhaust the complexity of the problem: listeners must also consider physical, social, emotional, and cognitive contextual variables (e.g., nearby objects; speaker’s interests; recent notable events). In short, the meanings of messages hinge on a changing, unpredictable series of diverse, shifting linguistic, paralinguistic, and nonlinguistic cues. So far I have described the general problem, but preschool children face a particularly concentrated version: to build a lexicon from the unfamiliar words that liberally pepper the utterances they hear. For this they have available a wide spectrum of cues to words’ meanings; yet their grasp of these cues is profoundly limited. Therefore, they are faced with the need to infer more meaning from less useable information. The dual challenge to preschoolers, then, is to use an unpredictable array of cues to interpret the meanings of changing messages with unknown words, and to learn the words as well. This challenge could be exacerbated by cognitive inflexibility, as in rule-switching tasks (Section IV.B). However, the outcome of the challenge is near-complete fluency, and a sizeable lexicon, by the fourth birthday. This is a paradox: preschoolers are inflexible in rule-switching tasks, but they learn very many words from a shifting, uncertain platform of linguistic cues. Perhaps young children are more flexible when inferring meanings from probabilistic, unpredictable cues than when following simple, deterministic rules. How can we make sense of this apparent paradox? I have studied how preschool children meet the practical demand to infer meanings and learn words by flexibly miming an unstable cue-lode. A starting assumption is that the most useful information about a novel word meaning is in its predicate—the phrases and words that surround it and form a coherent message meaning. Novel words are hard to interpret from nonverbal or syntactic cues alone (Dea´k, 2000b), but the predicate context of a novel word typically carries enough semantic, syntactic, and morphological information to powerfully constrain its likely meaning (Dea´k, 2000a,b; Goodman,


305

McDonough, & Brown, 1998). It is therefore critical that children pull information about word meaning from the predicate context. Predicates are, however, changeable and unpredictable. Especially when children hear several unfamiliar words within a situation or conversation, they must adapt to each word’s specific predicate context. This might be particularly important in settings with low-frequency referents (e.g., zoos; museums), or during special activities (e.g., preschool field trips). But more centrally, children might hear multiple unfamiliar words in conversation (see Beals & Tabors, 1995, on word learning at mealtimes), when hearing a new story, or when accompanying parents on errands. The Flexible Induction of Meaning (FIM) test requires children to infer the meanings of novel words by using changing predicate cues to flexibly shift attention among aspects of the referent. In the FIM-Ob children infer meanings of words for object properties (Dea´k, 2000b). In a newer version, designated FIM-An (for animates; Narasimham & Dea´k, 2001), children infer meanings of novel words for properties of strange creatures. The logic of the test is as follows: .

Sets of items are presented several times. Each set includes a standard and four comparison items that share different properties with the standard. In FIM-Ob, comparison objects have novel (i.e., not readily nameable) body shapes, materials, and affixed parts that differ from set to set. In each set one comparison object has the same shape as the standard, one is made of the same material, one has same affixed part, and a fourth is a dissimilar distractor. An example is shown in Figure 5.

Fig. 5. Example of set from the FIM-Ob test (Dea´k, 2000b), including standard object and four comparison objects (same-shape, same-material, same-part, and distractor).

306

.

.

.

Gedeon O. Dea´k

Each time a given set is presented the standard is described, somewhat ambiguously (like real word-learning situations), by a novel word modified by a predicate cue. The predicate implies a specific referent meaning. In the FIM-Ob, each word follows one of three predicates: ‘‘looks like a. . .’’ (or ‘‘is a’’), ‘‘is made of,’’ or ‘‘has a.’’ These imply the novel shape, material, or part, respectively. After hearing a predicate-word declarative (e.g., ‘‘This one is made of plexar’’), the child is asked to generalize the word to one of the comparison items (e.g., ‘‘Find another one that is made of [novel word]’’). Inductive responses are classified as predicate-appropriate (e.g., judging the same-material object to be ‘‘[also] made of plexar’’) or predicate-inappropriate. Each set is presented several times with a different predicate and novel word. For example, the first time a set is shown the child might hear that the standard ‘‘is made of stylar,’’ and, on subsequent trials, that it ‘‘has a graggle,’’ and ‘‘looks like an introm.’’ Flexibility is related to predicate-appropriate responses, particularly in later trials (i.e., generalizing the second and third words for a set to the properties implied by those predicates). A useful dependent measure is the number of predicate-appropriate switches: number of post-switch (i.e., second or third trial) choices of previously unselected, predicate-appropriate objects as referents.

Because children choose from several comparison items over several trials, for each of several sets, this paradigm can reveal more varied flexible and inflexible response patterns than can other tests of flexibility (e.g., DCCS). This allows testing (described below) of simple generalizations like ‘‘3-year-olds perseverate; 4-year-olds don’t.’’ Also, unlike the DCCS, FIM tests inductive flexibility: cues are not deterministic rules but predicate cues with probabilistic implications relative to some physical array. Nevertheless, 5- and 6-year-olds make mostly appropriate responses in the FIM-Ob (Dea´k, 2000b). Thus, the predicate cues are sufficiently informative. The FIM also permits control of temporal and sequential parameters relevant to response set, interference, and flexibility. For example, predicate order might be relevant because even if children grasp each predicate meaning, some are easier than others. Predicates that specifically imply a single available aspect or property have high implicature specificity and permit easy mapping. Predicates that are weakly associated with a single aspect are, at worst, equally associated with two or more aspects, thus have low implicature specificity, and are harder to map. Baseline task difficulty is a critical factor to consider because it is ecologically important (i.e., when we switch from one task to another, it is rare that the tasks are equally


307

interesting, easy, and motivating to us), because the difficulty of any given task is likely to change with age, and because difficulty might interact with order. For instance, switching to an easier versus a harder task impacts adults’ task-switching costs (Monsell, Yeung, & Azuma, 2000). In the FIM, flexibility might be greater when switching to an easy versus a hard predicate. When children’s first inference about a set is based on an easy predicate, they make more perseverative errors on subsequent trials than when the first predicate is difficult. In the FIM-Ob, ‘‘is made of’’ specifically implies material kind, whereas the less-specific predicate ‘‘looks like a’’ (or ‘‘is a’’) is ambiguous to preschool children. When the first word follows ‘‘is made of,’’ 3- to 6-year-olds usually generalize it to a same-material object. In postswitch trials, though, when generalizing words following ‘‘looks like a,’’ 3- and 4-year-olds often perseverate by selecting the same-material object again (Dea´k, 2000b). Figure 6 shows the mean number of predicateappropriate responses in the first, second, and third trials, contingent on the initial predicate. The decline of appropriate responses in later blocks, especially in 3-year-olds, reflects perseveration on initial responses. This, in turn, depends partly on whether the first inference was supported by highly specific predicate–aspect implication. Selecting a strongly cue-implied aspect interferes with 3-year-olds’ later responses to the same stimulus array. Predicate order does not, however, fully explain the development of flexible induction of word meaning. Increasing sensitivity to the implications of various predicate cues, and awareness that successive word meanings should be independently inferred, also contribute. To better show this, children’s appropriate switches (Dea´k, 2000b, Experiments 1–2) were weighted by a predicate order difficulty coefficient.4 Weighted means are shown in Figure 7. Appropriate switches increased with age, from 2.9 (out of 12) at 3 years to 5.9 at 4 years. Four-year-olds, but not 3-yearolds, made more predicate-appropriate switches than expected by chance. It is critical to note, however, that 3-year-olds make more appropriate responses than expected in the first block, suggesting that they can draw the implications of these predicates, but cannot reliably do so in the face of conflicting prior responses. The shift from inflexible responding at 3 years 4 The specific weighting procedure was to take, for each group, the ratio of mean first-block appropriate responses to a given predicate by the mean for all three predicates. Only first block responses are used because they are not complicated by switching. These ratios deviate from 1.0 to the extent that the predicates differ in specificity (e.g., an easy predicate receives a weight above 1.0). The reciprocal of the weight is multiplied by the number of correct switches produced by a child in response to that predicate in a later block. In this way, correct switches to an easier predicate receive less ‘‘credit’’ than correct switches to a harder predicate.

308

Gedeon O. Dea´k

Fig. 6. Mean appropriate inferences about the first, second, and third words for an object, based on first predicate and child’s age (from Dea´k, 2000b, Experiments 1–2 aggregated). Bars show standard errors.

Fig. 7. Mean predicate-appropriate response switches (weighted to reflect cue order difficulty) in blocks 2 and 3 (from Dea´k, 2000b, Experiments 1–2). Bars show standard errors.

to partly flexible responding at 4 years and fully flexible responding at 5 or 6 years is only moderated, but not dependent, on predicate order. Nevertheless, predicate specificity is a critical factor in word learning. The FIM-Ob was designed in part to pit this claim against the shape bias


309

hypothesis—the idea that children learn to map object count nouns onto shape rather than other properties like color or size (see Smith, 1999, for review). Dea´k (2000b) found that predicate implications, and children’s adaptation to predicates, overrides simple associations between aspects (shape) and a syntactic class (count noun). Three- and 4-year-olds did not selectively extend nouns predicated by ‘‘looks like a’’ or ‘‘is a’’ to sameshape objects, and 4-year-olds selectively generalized count nouns predicated by ‘‘has a’’ to small parts rather than objects with the same body shape. It was not that shapes were subtle or uninteresting—in a pre-test, children judged the same-shape objects to be more similar than the other comparison objects to the standards.5 Nor was it that a same-shape interpretation of words predicated by ‘‘looks like a. . .’’ was conceptually implausible, because 5- and 6-year-olds overwhelmingly made this interpretation. Finally, there is no syntactic ambiguity—‘‘looks like a’’ and ‘‘is a’’ must modify count nouns (if the noun phrase is a single lexeme). The best interpretation, then, is that because the predicate ‘‘looks like a. . .’’ is semantically ambiguous (i.e., has nonspecific implications), children with less semantic knowledge either cannot or will not use this cue to guide their response—in fact, its ambiguity seems to dissuade them from choosing the most perceptually salient match! Dea´k (2000b) concluded that a shape bias might emerge when children generalize words for simplified drawings or objects that emphasize shape (Dea´k & Bauer, 1996), but in general, young children will flexibly choose from a variety of aspects as meanings of novel words, based on the specific meaning context of a word, not on rigid associations between properties and syntactic categories. The data from the FIM, however, show a restriction on younger children’s flexibility in inferring multiple word meanings within a situation. Though early word learning is often described as precocious, even 4-yearolds had trouble using predicate cues when the inferred referents of previously learned words were present. This was true even if the later predicate cues were interpretable, and if children were given preliminary practice with those predicate cues. Children’s errors show a blind spot in their lexical problem solving: they do not require word meanings to be consistent with the semantic implications of the immediate linguistic context. The other meaningful elements within an utterance that modify a word should be the final arbiters of its meaning—not, for example, what one happens to be thinking about when the word is uttered. Yet 3- and 4-year-olds in the FIM sometimes map a new word onto the same referent of a previous word. This shows a baffling ‘‘leakage’’ of implication across utterances, and thus 5

The fact that standard and same-shape objects differed by one part, incidentally, did not seem to reduce shape-based choices; see Dea´k (2000b) for details.

310

Gedeon O. Dea´k

ignorance of the relative importance of different kinds of cues (i.e., local predicates vs. past utterances) for determining word meaning, as well as ignorance that different words for a referent typically refer to different aspects (excepting the rare true synonym). Thus, in situations that test these conflicts, some preschool children do not demonstrate practical knowledge of these basic characteristics of word meanings. 2. Individual Differences in Children’s Flexible Induction of Meaning In addition to age differences, the FIM reveals individual differences in preschool children’s flexibility. Even among children who make more predicate-appropriate switches than expected by chance, some children shift their responses to all three predicates; others shift for only one of the two predicates. If the former is defined by at least 80% appropriate responses to each predicate (25% is expected by chance), no 3-year-old, 28% of 4-yearolds, and 73% of 6-year-olds meet this higher criterion (i.e., fully flexible). Partial flexibility—defined by at least 80% appropriate responses to only two predicates—shows a notable pattern: children are most likely to perseverate in post-switch responses to ‘‘looks like a’’ words, and never perseverate in post-switch responses to ‘‘is made of’’ words. Thus, children on the cusp of flexible word learning are heavily dependent on predicate specificity. Among inflexible children, some (about 25–30%) consistently perseverate, seldom if ever switching responses after the first. Half of these children perseverate on single aspect; for example always choosing the same-material objects. The chosen aspect is usually that implied by the first predicate (indicating that even these children pay some attention to predicate context), unless the first predicate is ‘‘looks like a.’’ The remaining children persistently focus on a specific item from each set, with no apparent pattern across sets. Perhaps these children do not notice the predicate cue, or do not know whether it should override salient perceptual similarities. Other inflexible 3- to 6-year-olds (about 16% of sample), mostly 3-year-olds, are indiscriminate: they switch some responses over trials, but not based on predicate cues. Perhaps these children notice a change between successive questions (e.g., a different novel word), and expect different words to have different meanings, but fail to notice or draw the implication of each predicate cue. Perhaps they then switch responses in hopes of receiving feedback from the adult (Speer, 1984). I have described these results to show that perseveration is not the inevitable alternative to flexibility. Also, perseveration is probabilistic (i.e., the same child might perseverate from task A to B, but not C) and mediated by factors like task order (see also Perner & Lang, 2002). This is revealed only by more complex tests than the DCCS or the Stroop. Perseveration


311

also stems from different kinds of interference (e.g., attributes or specific items). Conversely, flexibility also is graded and context dependent. Understanding patterns of meaning (e.g., in the lexicon), and various meaning cues, is important for adaptation to changing contexts of meaning. Children who do not notice changing meaning cues (e.g., predicates), or do not know their implications, will be inflexible. They might perseverate (at least in some inferences), or select meanings haphazardly. How children construe meaning cues, and the importance of changes in these cues across utterances, determines their flexibility in inferring meaning. Other, poorly understood factors determine how inflexibility is manifested. In the FIM at least three variables shift from trial to trial: object array, words, and predicate cues. If children do not expect different words to have different meanings, and different predicates to imply different stimulus aspects, there is no reason why they should not perseverate. It does not mean that they are incapable of inhibiting prior responses; it is at least as plausible that they do not recognize the demand to suppress those responses. In contrast, if children realize that the question is changing across trials, but fail to focus on the relevant information (i.e., predicate cue), they might respond indiscriminately. Even if these claims are true, they do not presuppose that individual differences in ability to flexibly induce meaning are stable within a child. That is, are some 3- and 4-year-olds consistently more flexible in inferring meaning? To address this, Narasimham and Dea´k (2001) administered both the FIM-Ob and the FIM-Al to a group of 3- and 4-year-olds. The partial correlation between appropriate switches in the two tests was r ¼ .53, indicating that they tap the same skills (though the FIM-Al was slightly harder than the FIM-Ob). 3. Age and Individual Differences in Flexibility: Relation to Inhibition Perhaps all these data can be explained more simply: maybe some preschool children lack the inhibitory capacity to de-select prior responses, and this explains perseveration in the FIM. This is inconsistent, however, with a control test (Dea´k, 2000b, Experiment 3) that used stimuli analogous to FIM-Ob sets, except critical attributes and words were familiar (e.g., a square made of paper with an affixed button). Children were asked, for instance, to find another object ‘‘made of paper’’ or (on another trial) one that ‘‘has a button.’’ Here the demand to use predicate cues is reduced (because property labels are familiar), but the demand to inhibit prior responses is held constant. Three- and 4-year-olds performed very well in this test, indicating that they can inhibit prior responses when redundant cues are available. This further suggests that a central inhibitory capacity cannot account for age and individual differences in children’s flexibility.

312

Gedeon O. Dea´k

Young children’s perseveration in the FIM still must be explained. Recent work (Dea´k & Narasimham, under review) tested the relation among perseveration, inhibition, and flexibility in the FIM-Ob. To test whether perseveration stems from inability to inhibit prior responses, we varied response set activation strength by manipulating the delay between responses or between blocks, and by changing the number of successive questions with a single predicate cue. In one study, a group of 3-year-olds were given the FIM-Ob with a 90-sec delay between blocks (trials were blocked by predicate). During the delay, children were primed for the next predicate (e.g., if ‘‘has a’’ was next, object parts were pointed out and described using that phrase). If inhibitory capacity is important, the delay and priming should reduce proactive interference and thereby reduce perseveration. Yet perseveration was no less common in this group than in the original, no-delay sample (Dea´k, 2000b). In a second study, one group of 3- and 4-year-olds received a random mix of predicates in each block of trials, and another group had trials blocked by set (i.e., all three questions about a set were given in succession). If inhibitory ability determines flexibility, mixing predicates should reduce perseveration because there is no chance to build a response set. Conversely, blocking by sets should increase perseveration because there is no chance for release from proactive interference between responses. Alternately, rapid predicate switching in both conditions might increase indiscriminate responding. Yet neither condition influenced children’s response patterns: the previously described age difference in flexibility was replicated, but none of the groups differed from each other, or from the original sample, in mean number of appropriate switches. Thus, switching rate and interval between successive problems does not seem to affect flexibility, at least within the parameters studied. In a third study, 3- and 4-year-olds assigned to high- or low-interference groups did six familiar-attribute trials (as in Dea´k, 2000b, Experiment 3, described above) before the FIM-Ob. In the high-interference group, all six trials used the same predicate as the first FIM-Ob test block, so children completed 12 (six easy; six hard) trials with one predicate before switching. The low-interference group’s familiar-attribute trials used all three predicates, so they switched predicates several times before starting the first test block. If response interference causes perseveration, repetition of one type of response should increase perseveration. However, there was no difference between the two groups. Also, there was no correlation between any measure of flexibility in the FIM and any measure of flexibility in the Stroop Day/Night test—another strike against the idea that differences in flexibility depend on a general inhibitory trait.


313

These findings suggest perseveration does not change with the number of pre-switch trials (see also Zelazo, Frye, & Rapus, 1996), number of intervening trials, or number of task switches. Perhaps, then, perseveration in 3- to 5-year-olds is not sensitive to interference from prior responses, as mediated by repetition of one type of response, or by delay since the last response. This narrows the possible roles of cognitive inhibition in flexible induction of meaning. 4. Age and Individual Differences in Flexibility: Verbal Knowledge and Memory Perhaps children’s flexibility in inferring word meanings follows their comprehension of predicate cues. Several findings make this an unlikely explanation for all FIM findings. First, 3-year-olds make more predicateappropriate responses than expected by chance in the first block, but not in later blocks (Dea´k, 2000b); this difference disappears with age. Second, in an unpublished study (Dea´k, 1995), 3-year-olds produced words for object shapes, materials, and parts, in response to questions that used the FIM-Ob predicates. Children were asked ‘‘What does this look like?’’ and ‘‘What is this made of?’’ about a wooden star, paper rectangle, metal disk, glass cube, cloth letter A, sponge heart, Play-Doh ball, and a plastic triangle. They were asked ‘‘What does this look like?’’ and ‘‘What does this have?’’ about a toy teacup, dinosaur, fire truck, raccoon, telephone, rabbit, biplane, and snail. Predicate-appropriate answers were shape or object kind labels for ‘‘looks like’’ questions (mean ¼ 14.7 of 16 labels), material kind terms for ‘‘made of’’ (mean ¼ 3.5 of 8 responses) and part labels for ‘‘has a’’ (mean ¼ 7.3 of 8 responses). This confirms that even 3-year-olds have reasonably accurate knowledge of these predicate meanings. Third, Yen (1997) showed children objects such as a large star covered with smaller stars and with a mediumsized wooden star in the center, and asked, on different trials, which other object ‘‘looks like a star,’’ ‘‘is made of star,’’ and ‘‘has a star.’’ The comparison objects were, for example, a large star covered with circles, a square covered with small stars, or a triangle with a medium-sized star attached. Thus, the word is familiar but totally ambiguous, so the predicate is the only useful cue. Three-year-olds made mostly predicate-appropriate responses, suggesting that they can use the predicate in ambiguous situations. Though most preschoolers at least marginally comprehend the predicates, some added processing demands might reduce children’s application of this knowledge. Perhaps, as Bishop (1997) suggests, working memory demands (e.g., novel words; complex stimulus array) impair performance especially when sentences are complex and delivered quickly. In the original FIM protocol, however, sentences were spoken slowly, enunciated clearly, and

314

Gedeon O. Dea´k

repeated. Also, they were syntactically simple. Thus, I suspect working memory is not a major factor in age and individual differences, but this has not been tested directly. Rather, FIM-Ob test flexibility seems to depend on children’s tendency to consistently attend to predicate cues and notice when they switch, and their understanding that the local predicate should ‘‘trump’’ other response cues. 5. Summary: Children’s Flexible Induction of Word Meanings Children’s ability to flexibly use predicate information to constrain inferences of word meanings undergoes substantial development from 3 to 6 years. Many obvious possible causes of development can be ruled out; see Table III for a summary of findings. A fascinating question is whether this development is reiterated when language skills are transferred to the modality of written language. Making inferences from text is a critical reading skill by which many new words are learned. Individual differences in flexible induction of (spoken) word meanings might predict later differences in children’s ability to infer meaning from text (see Yuill & Oakhill, 1991). On the one hand, variance in preschool children’s inductive flexibility might reflect age-specific attainments, for example, awareness of the ‘‘operating principles’’ of language. That is, flexible children assume that different novel words have distinct meanings. They notice when speakers signal a change to a new predicate—akin to a clear topic change. This sets the stage for flexible selection of cues to meaning. Because these principles are already familiar when children begin reading, different processes might account for flexible induction in reading. On the other hand, Olson (1977) suggests that ability to decontextualize messages, and use text meaning per se to draw inferences, is crucial for acquiring written language. This ability also can be seen as essential in using the predicate, rather than distal context (e.g., prior responses), to infer meaning.

D. COMMON FACTORS IN CHILDREN’S FLEXIBLE COGNITIVE PROCESSING OF MESSAGES AND MEANINGS

The empirical results reviewed here suggest a complex developmental pattern. The DCCS tests rule-use flexibility; the FIM tests word learning flexibility. Each uses unique stimuli, cues, and procedures. Nevertheless, they show roughly parallel results: significant improvement from 3 to 4 years, many perseverative errors in younger preschoolers, and no impact of factors like number of pre-switch trials and number of rule switches. Before concluding that we have found a general developmental phenomenon in flexible language processing, however, recall that 3-year-olds perform


315

TABLE III Summary of Questions and Findings about Preschool Children’s Flexible Induction of Word Meanings Question

Result

Can children use predicate cues (e.g., ‘‘has a. . .’’; ‘‘is made of. . .’’) to flexibly infer novel word meanings?

About one-third of 3-year-olds are above chance; most perseverate or switch responses indiscriminately.a Two-thirds of 4-year-olds are above change in post-switch responses; a minority still perseverate.a

Is there independent evidence of this age difference?

Yes, in children’s judgments about the breakability or location of unusual ‘‘hybrid’’ objects, based on different predicates.b

Is the FIM test too confusing or tedious for children?

No: In a control task using objects with familiar attributes and labels, performance is near ceiling.c

Do children perseverate because they cannot ignore previously chosen objects?

No: See previous entry.c

Do 3-year-old children understand the predicate cues?

Most are above chance in mapping the FIM-Ob predicates to the appropriate aspects.d

Is training on the task necessary or important?

Brief training on the predicate meanings has no effect on performance.a

Effects of between question delay, number of successive same-predicate trials, or number of predicate switches?

There is no evidence that any of these factors significantly affect performance.e

Is a child’s performance stable or predictable across wordlearning tasks?

Preschoolers’ performance on two versions of the FIM task has a partial correlation of r ¼ .53. Performance on the Stroop Day/Night task does not correlate with the FIM.e

Note: aDea´k (2000b, Experiments 1–2); bKalish & Gelman (1992); cDea´k (2000b, Experiment 3); Unpublished studies (see text); eDea´k & Narasimham (under review).

d

flexibly in some analogous tasks (Dea´k, 2000b, Experiment 3; Perner & Lang, 2002). The picture is not clear and simple. Perhaps a broader survey of cognitive changes during the preschool years will provide some clues. In general, preschool children are rapidly getting better at solving problems by selecting task-relevant aspects of complex stimuli. In many cases, knowing what is ‘‘task-relevant’’ depends on semantic and pragmatic sensitivity to task content. Children by 2 and 3 years can produce MARs, but cannot reliably use explicit instructions as input to shift the active aspect of the MAR. The problem seems to lie in choosing the aspect implied by a specific request, by virtue of the semantic content per se. Thus,

316

Gedeon O. Dea´k

though 3-year-olds can produce several words for an object, they cannot judge whether these words specify ‘‘what it looks like’’ or ‘‘what it does’’ (Dea´k, Yen, & Pettit, 2001). Similarly, 4-year-olds, but not 3-year-olds, flexibly select aspects (e.g., ‘‘fur’’ or ‘‘bowl’’) from novel compounds (e.g., ‘‘fur bowl’’) according to specific aspects implied by the question (e.g, ‘‘Will it break?’’ vs. ‘‘Does it go in the kitchen?’’) (Kalish & Gelman, 1992). This skill—to use specific semantics of questions to choose an answer— cannot be deficient because it is ecologically irrelevant. In preschoolers’ everyday settings (e.g., home, school, laboratory), tasks and events are described by command, instruction, and description. Older children are expected to organize their actions in accordance with the ‘‘text’’ of instructions. The importance of message interpretation skills suggests that 2- and 3-year-olds’ inflexibility is due to some rather pervasive cognitive limitation. One possibility is that message interpretation and responsiveness requires an ‘‘uncertainty stance’’: tolerance for (and expectation of) indeterminacy of upcoming messages, and assumption that some cues, which change unpredictably, can resolve this ambiguity. The most informative cues to the current utterance’s meaning are not homogeneously distributed over time. All else being equal, cues to an unfamiliar term’s meaning are concentrated in the same utterance as the term. The fact that, for instance, we talked about an object’s material a few minutes ago is no guarantee that the next unfamiliar word also will refer to material. Yet many 3-year-olds do not know the scope or ‘‘sphere of influence’’ of different linguistic cues—for example, a novel word’s predicate context ‘‘trumps’’ prior responses. In flexible rule use, children’s problem might be inferring the operative principle for rule selection. For instance, ‘‘Where does this go in the shape game?’’ presents a conflict between the current predicate cue’s (i.e., ‘‘shape game’’) implication (e.g., dogs go in the left box), and a previously practiced implication and response. In both cases, selecting the current message meaning is critical. Thus, in quite different tests, flexibility requires knowing which linguistic cues should govern inferences about meaning.

V. Questions and Conclusions Available data on children’s flexible language processing contradict outdated and simplistic views of its development. These data also highlight difficult questions. Outlining the most pressing of these is important for guiding future empirical efforts. These are summarized overleaf.


317

A. HOW DO LOGICAL AND METACOGNITIVE ABILITIES INFLUENCE FLEXIBILITY?

I have hypothesized that flexibility requires sensitivity to the independent indeterminacy of the meaning of each question or utterance in a series. Existing empirical evidence for this dependency is, however, only correlational. Experimental tests that manipulate or train children’s awareness of indeterminacy would be informative. B. HOW DOES INHIBITION INFLUENCE THE DEVELOPMENT OF FLEXIBILITY?

Inhibitory processes probably play an important role in children’s rapidly developing ability to respond flexibly to changing messages and meanings. Available data and theory do not, however, provide a coherent or falsifiable account that adds teeth to this vague supposition. I suspect our concepts of cognitive inhibition are too primitive to advance much further, and a radical reconceptualization of the construct must precede any substantive progress. The best approach using the available construct of cognitive inhibition is to specify the kind of information that might interfere with children’s response switching. This can reveal developmental changes; for instance, children 3 years and older are seldom confounded by changing stimulus locations, whereas 1- and 2-year-olds are susceptible to location-based interference (Zelazo, Reznick, & Spinazzola, 1998). C. IS LANGUAGE CENTRAL TO FLEXIBLE COGNITION?

The MARM metaphor implies a general, multimodal representational process that is reflected in flexible language processing, and in nonlinguistic perception and action systems. For example, from 3 to 6 years there is improvement in the flexible deployment of encoding and recall strategies (Ceci & Howe, 1978; Miller et al., 1986), spatial inference (e.g., Fabricius, Sophian, & Wellman, 1987; Hermer-Vazquez, 1997), graphical representation (Picard & Vinter, 1999), and mental state inferences (Wellman, 1990). This parallel development might result from domain-general representational changes, for example in metarepresentation (Karmiloff-Smith, 1992), but it is also possible that language plays a unique role in cognitive flexibility. Language reflects and facilitates our most pervasive, open-ended manifestations of cognitive flexibility. The basic function of language is fast, flexible production and reconstruction of a practically unlimited range of selectively sculpted mental representations. No other behavior system in nature matches this potential for flexible representation. Though laboratory

318

Gedeon O. Dea´k

tests of flexible cognition (e.g., task-switching) typically ignore the role of symbolic knowledge, virtually all of these tasks use instructions to orient participants to the task, and abstract symbols to cue task switches. Perhaps, then, basic symbol mapping knowledge is needed for cognitive flexibility (Deacon, 1997). One candidate is the expectation that messages are unpredictable in meaning, and that some linguistic cues determine a message’s meaning, even if the listener does not endorse or fully understand it. This is critical knowledge shared by no cognitive system, to our knowledge, outside of humans above the age of 2 years. This knowledge might facilitate flexibility in processing problems that are not primarily linguistic. For example, children’s ability to produce complex descriptions of specific locations predicts their ability to flexibly use spatial cues to infer object location (Hermer-Vazquez, 1997). Even adults, in functional fixedness paradigms, often need detailed cues labels or verbal ‘‘hints’’ to eventually solve the problems (Glucksberg & Danks, 1968; Meier, 1937). Though such evidence does not resolve ‘‘chicken-or-egg’’ questions about the evolution of flexible cognition and language, it suggests that development of ability to produce and understand complex, specific locutions is linked to the development of ability to choose between conceptual distinctions in complex, ambiguous situations or arrays. Perhaps knowing or inducing labels for different aspects of a stimulus provides a conceptual toehold for switching attention among aspects of reality. D. METHODOLOGICAL PROBLEM: FILLING THE GAPS

Flexibility in any system, linguistic or otherwise, is difficult to study in a controlled manner. Meaningful series of responses must be elicited, and critical dependent measures concern change over time. Thus, some of the challenges of microgenetic methods (Siegler & Crowley, 1991), such as fairly dense observations, also characterize rich tests of flexibility. A practical challenge is that child participants, unlike most adults, do not cheerfully tolerate long series of boring test questions or problems. This limits the number of responses and task switches obtainable from children within a session. Another challenge of comparing flexibility across a wide age range is that preferred dependent measures change with age. One possibility is to use RTs to measure switch costs in young children, but this introduces new challenges. For instance, young children’s response times are quite variable, so if each child contributes only a few switch cost data points, large samples must be tested to detect reliable group differences. A related problem is that the distribution of preschoolers’ response times has a very long tail, yet most of the long responses seem valid: children may stay attentive, oriented, and ‘‘on-task,’’ and take as long as 30 sec to respond to an FIM trial! Because long


319

responses cannot be discarded as anomalous, they present a sticky analytic problem. More generally, changes with age in the nature of switch costs reveal a critical aspect of the development of flexibility. In simple tasks, children’s switch costs are manifested as a high rate of perseverative errors, whereas adults show a modest, transitory increase in RT. What does this indicate about changes in inhibition, executive control, and strategic responses to imputed principles of a task? For example, why are adults’ task switch costs limited to 1–2 post-switch trials (Monsell, Yeung, & Azuma, 2000), whereas preschool children’s perseveration persists for many post-switch trials (Zelazo, Frye, & Rapus, 1996)? No data are available to answer this question. E. CONCLUSIONS

Cognitive flexibility allows humans to adapt successive inferences or responses to changing task demands, by selecting task-relevant information that may change unpredictably. It is a higher-order cognitive ability, because it concerns controlled changes in cognitive activity over time, problems, or tasks. Cognitive flexibility is required for everyday language processing, because most of us do not inhabit fully predictable and familiar linguistic environments, or use only learned scripts and sequences to produce and understand words, utterances, and discourse. A generalized description of cognitive flexibility is the metaphorical Multi-Aspectual Representational Medium, wherein activation of different aspects of representations (e.g., of a physical array, sequence of events, or mental state) dynamically shifts in response to varied, changing task-relevant input forces. This input is often in the medium of natural language. In investigating the development of cognitive flexibility as reflected in, and influenced by, language use, a persistent challenge is that age-related changes in flexibility coincide with multiple changes in brain, cognitive, and language development. Nevertheless, some interim conclusions can be drawn. From 2 to 6 years, and even from 3 to 5 years, there is substantial increase in ability to adapt descriptive locutions to changing (linguistic) task cues, and ability to adapt to changing meanings of successive verbal messages. This applies to a variety of speech acts (e.g., declarative utterances in narrative, instructions or rules, descriptive sentences, questions), and to practical problems of following adults’ instructions, and inferring the meanings of ambiguous words or the referents of complex locutions. There is no evidence that this developmental change is the direct result of maturation of a central, executive capacity to inhibit active responses, whereby activation grows over repeated responses and decays over time or intervening activity. Young children are not mechanically unable to inhibit prior responses (Dea´k, 2000b, Experiment 3; Perner & Lang, 2001), nor are

320

Gedeon O. Dea´k

they unable to switch labels, on short order, for a given referent (e.g., Dea´k & Maratsos, 1998). Of course, inhibitory ability might contribute to the development of flexibility, but this hypothesis has little support. In contrast, there is recent evidence that children’s awareness of changing verbal information (e.g., predicates), and ability to selectively map alternative locutions to different predicates or questions, is a factor in the development of flexible language use. A more speculative hypothesis is that sensitivity to indeterminacy of messages (e.g., questions; unfamiliar words) contributes to flexibility in language processing. These speculations are rich grounds for future research efforts. Finally, it is worth noting that available evidence explores flexibility in three age groups: 2- to 6-year-olds, young adults, and elderly adults. Although there is some evidence pertaining to cognitive flexibility in older children (Ceci & Howe, 1978; Cepeda, Kramer & Gonzalez de Sather, 2001), such evidence is so scant that few inferences can be drawn about development in the vast gulf between kindergarten and college. We are therefore far from being able to describe and explain typical and atypical life-span changes in flexible linguistic and nonlinguistic cognitive processing. Investigating this will, in part, depend on methods and tests that can compare cognitive flexibility across wide age ranges.

ACKNOWLEDGMENT Preparation of this chapter was supported by NSF award BCS-0092027. Address correspondences to the author at [email protected].

REFERENCES Albert, M. L., & Sandson, J. (1986). Perseveration in aphasia. Cortex, 22, 103–115. Allport, A., Styles, E. A., & Hsieh, S. (1994). Shifting intentional set: Exploring the dynamic control of tasks. In C. Umilta & M. Moscovitch (Eds.), Attention and performance XV (pp. 421–452). Cambridge: MIT Press. Anderson, R. C., & Ortony, A. (1975). On putting apples into bottles—A problem of polysemy. Cognitive Psychology, 7, 167–180. Barclay, J., Bransford, J. D., Franks, J., McCarrell, N., & Nitsch, K. (1974). Comprehension and semantic flexibility. Journal of Verbal Learning and Verbal Behavior, 13, 471–481. Barsalou, L. W. (1982). Context-independent and context-dependent information in concepts. Memory & Cognition, 10, 82–93. Barsalou, L. W. (1983). Ad hoc categories. Memory & Cognition, 11, 211–227. Barsalou, L. W. (1989). Intraconcept similarity and its implications for interconcept similarity. In S. Vosniadou & A. Ortony (Eds.), Similarity and analogical reasoning (pp. 101–140). Cambridge: C.U.P.


321

Beals, D. E., & Tabors, P. O. (1995). Arboretum, bureaucratic and carbohydrates: Preschoolers’ exposure to rare vocabulary at home. First Language, 15, 57–76. Beeman, M. (1998). Coarse semantic coding and discourse comprehension. In M. Beeman & C. Chiarello (Eds.), Right hemisphere language comprehension (pp. 255–284). Mahwah, NJ: Lawrence Erlbaum. Bishop, D. V. (1997). Uncommon understanding: Development and disorders of language comprehension in children. Hove, UK: Taylor & Francis. Bloom, P. (2000). How children learn the meaning of words. Cambridge, MA: MIT Press. Bohlmann, N. (2001). The effects of feedback on perseverative errors in problem solving with 3- to 5-year-old children. Unpublished MA thesis, San Diego State University, San Diego, CA. Brennan, S. E., & Clark, H. H. (1996). Conceptual pacts and lexical choice in conversation. Journal of Experimental Psychology: Learning, Memory, & Cognition, 22, 1482–1493. Brown, R. (1958). How shall a thing be called? Psychological Review, 65, 14–21. Brownell, H., & Martino, G. (1998). Deficits in inference and social cognition: The effects of right hemisphere brain damage on discourse. In M. Beeman & C. Chiarello (Eds.), Right hemisphere language comprehension (pp. 309–328). Mahwah, NJ: Lawrence Erlbaum. Brownell, H., Michelow, D., Powelson, J., & Gardner, H. (1983). Surprise but not coherence: Sensitivity to verbal humor in right hemisphere patients. Brain and Language, 18, 20–27. Brownell, H. H., Simpson, T. L., Bihrle, A. M., Potter, H. H., & Gardner, H. (1990). Appreciation of metaphoric alternative word meaning by left and right brain-damaged patients. Neuropsychologia, 28, 375–383. Campbell, R. N., & Bowe, T. B. (1983). Text and context in early language comprehension. In M. Donaldson, R. Grieve, & C. Pratt (Eds.), Early childhood development and education (pp. 115–126). Oxford: Basil Blackwell. Carlson, S. M., & Moses, L. J. (2001). Individual differences in inhibitory control and children’s theory of mind. Child Development, 72, 1032–1053. Ceci, S. J., & Bronfenbrenner, U. (1985). ‘‘Don’t forget to take the cupcakes out of the oven’’: Prospective memory, strategic time-monitoring, and context. Child Development, 56, 152–164. Ceci, S. J., & Howe, M. J. A. (1978). Age-related differences in free recall as a function of retrieval flexibility. Journal of Experimental Child Psychology, 26, 432–442. Cepeda, N. J., Kramer, A. F., & Gonzalez de Sather, J. C. (2001). Changes in executive control across the life span: Examination of task-switching performance. Developmental Psychology, 37, 715–730. Chukovsky, K. (1968). From two to five (Rev. ed., M. Morton, Trans.). Berkeley: University of California Press. Clark, E. V. (1997). Conceptual perspective and lexical choice in acquisition. Cognition (Vol. 64, pp. 1–37). Clark, E. V., & Svaib, T. A. (1997). Speaker perspective and reference in young children. First Language (Vol. 17, pp. 57–74). Cohen, L., & Dehaene, S. (1998). Competition between past and present: Assessment and interpretation of verbal perseverations. Brain, 121, 1641–1659. Cosgrove, J. M., & Patterson, C. J. (1977). Plans and the development of listener skills. Developmental Psychology, 13, 557–564. Cruse, D. (1977). The pragmatics of lexical specificity. Journal of Linguistics, 13, 153–164. Daigneault, S., Braun, C., & Whitaker, H. (1992). Early effects of normal aging on perseverative and non-perseverative prefrontal measures. Developmental Neuropsychology, 8, 99–114. Damasio, A. R. (1985). The frontal lobes. In K. Heilman & E. Valenstein (Eds.), Clinical neuropsychology (pp. 339–376). New York: Oxford University Press.

322

Gedeon O. Dea´k

Danet, B. (1980). ‘Baby’ or ‘fetus’? Language and the construction of reality in a manslaughter trial. Semiotica, 32, 187–219. Deacon, T. W. (1997). The symbolic species: The co-evolution of language and the brain. New York: Norton. Dea´k, G. O. (1995). Unpublished doctoral dissertation, University of Minnesota, Twin Cities. Dea´k, G. O. (2000a). Chasing the fox of word learning: Why ‘‘constraints’’ fail to capture it. Developmental Review, 20, 29–80. Dea´k, G. O. (2000b). The growth of flexible problem solving: Preschool children use changing verbal cues to infer multiple word meanings. Journal of Cognition and Development, 1, 157–192. Dea´k, G., & Bauer, P. J. (1996). The dynamics of preschoolers’ categorization choices. Child Development, 67, 740–767. Dea´k, G. O., & Maratsos, M. (1998). On having complex representations of things: Preschoolers use multiple words for objects and people. Developmental Psychology, 34, 224–240. Dea´k, G. O., & Narasimham, G. (under review). Is perseveration caused by inhibition failure? Evidence from preschool children’s flexible inferences about word meanings. Dea´k, G. O., Ray, S. D., & Brenneman, K. (2003). Children’s perseverative appearance – reality errors are related to emerging language skills. Child Development. Dea´k, G. O., Ray, S. D., & Pick, A. D. (2003). Flexibility in preschoolers’ object classification: Effects of age, task difficulty, and rule reminders. Unpublished manuscript. Dea´k, G. O., Yen, L., & Pettit, J. (2001). By any other name: When will preschoolers produce multiple labels for a referent? Journal of Child Language, 28, 787–804. Dell, G. S., Burger, L. K., & Svec, W. R. (1997). Language production and serial order: A functional analysis and a model. Psychological Review, 104, 123–147. Dempster, F. N. (1992). The rise and fall of the inhibitory mechanism: Toward a unified theory of cognitive development and aging. Developmental Review, 12, 45–75. Dennis, M. (1991). Frontal lobe function in childhood and adolescence: A heuristic for assessing attention regulation, executive control, and the intentional states important for social discourse. Developmental Neuropsychology, 7, 327–358. Diamond, A. (1998). Evidence of the importance of dopamine for prefrontal cortex functions early in life. In A. Roberts, T. Robbins, & L. Weiskrantz (Eds.), The prefrontal cortex (pp. 144–164). Oxford: University Press. Diamond, A., & Taylor, C. (1996). Development of an aspect of executive control: Development of the abilities to remember what I said and to ‘‘Do as I say, not as I do.’’ Developmental Psychobiology, 29, 315–334. Diamond, A., Prevor, M. B., Callender, G., & Druin, D. P. (1997). Prefrontal cortex cognitive deficits in children treated early and continuously for PKU. Monographs of the Society for Research in Child Development, 62 (4, Serial No. 252). Donaldson, M. (1978). Children’s minds. Glasgow: Fontana. Duncker, K. (1945). On problem solving. Psychological Monographs, 58 (5, Whole No. 270). Eastman, P. D. (1960). Are you my mother? New York: Random House. Fabricius, W. V., Sophian, C., & Wellman, H. M. (1987). Young children’s sensitivity to logical necessity in their inferential search behavior. Child Development, 58, 409–423. Fauconnier, G. (1997). Mappings in thought and language. Cambridge: Cambridge University Press. Finke, R. A., Ward, T. B., & Smith, S. M. (1992). Creative cognition: Theory, research, and applications. Cambridge, MA: MIT Press. Fiore, S. M., & Schooler, J. W. (1998). Right hemisphere contributions to creative problem solving: Converging evidence for divergent thinking. In M. Beeman & C. Chiarello (Eds.), Right hemisphere language comprehension (pp. 349–371). Mahwah, NJ: Lawrence Erlbaum.


323

Flavell, J. H., Flavell, E. R., & Green, F. L. (1983). Development of the appearance-reality distinction. Cognitive Psychology, 15, 95–120. Flavell, J. H., Green, F. L., & Flavell, E. R. (1986). Development of knowledge about the appearance-reality distinction. Monographs of the Society for Research on Child Development, 51 (1, serial no. 212). Frye, D., Zelazo, P. D., & Palfai, T. (1995). Theory of mind and rule-based reasoning. Cognitive Development, 10, 483–527. Garrod, S., & Anderson, A. (1987). Saying what you mean in dialogue: A study in conceptual and semantic co-ordination. Cognition, 27, 181–218. German, T. P., & Defeyter, M. (2000). Immunity to ‘functional fixedness’ in young children. Psychonomic Bulletin and Review, 7, 707–712. Gershkoff-Stowe, L. (2002). Object naming, vocabulary growth, and the development of word retrieval abilities. Journal of Memory and Language, 46, 665–687. Gerstadt, C. L., Hong, Y., & Diamond, A. (1994). The relationship between cognition and action: Performance of children 3½–7 years old on a Stroop-like day-night task. Cognition, 53, 129–153. Gilbert, S. J., & Shallice, T. (2000). Task switching: A PDP model. Cognition, 44, 297–337. Glucksberg, S., & Danks, J. H. (1968). Effects of discriminative labels and of nonsense labels upon availability of novel function. Journal of Verbal Learning & Verbal Behavior, 7, 72–76. Glucksberg, S., Krauss, R. M., & Weisberg, R. (1966). Referential communication in nursery school children: Method and some preliminary findings. Journal of Experimental Child Psychology, 3, 333–342. Gombert, J. E. (1992). Metalinguistic development. Chicago: University of Chicago. Goodman, J. C., McDonough, L., & Brown, N. B. (1998). The role of semantic context and memory in the acquisition of novel nouns. Child Development, 69, 1330–1344. Gopnik, A., & Astington, J. W. (1988). Children’s understanding of representational change and its relation to the understanding of false belief and the appearance-reality distinction. Child Development, 59, 26–37. Grattan, L. M., & Eslinger, P. J. (1990). Higher cognition and social behavior: Cognitive flexibility and empathy after brain injury. Neuropsychology, 3, 175–185. Grattan, L. M., & Eslinger, P. J. (1991). Frontal lobe damage in children: A comparative review. Developmental Neuropsychology, 7, 283–326. Guilford, J. P. (1967). The nature of human intelligence. New York: McGraw-Hill. Harnishfeger, K. K. (1995). The development of cognitive inhibition: Theories, definitions, and research evidence. In F. N. Dempster & C. J. Brainerd (Eds.), Interference and inhibition in cognition (pp. 175–204). San Diego: Academic Press. Hermer-Vazquez, L. (1997). Cognitive flexibility as it emerges over evolution and human development: The case of spatial reorientation. Unpublished doctoral thesis, Cornell University. Hirsh, K. W. (1998). Perseveration and activation in aphasic speech production. Cognitive Neuropsychology, 15, 377–388. Hocevar, D., & Michael, W. (1979). The effects of scoring formulas on the discriminant validity of tests of divergent thinking. Educational & Psychological Measurement, 39, 917–921. Houde´, O. (2000). Inhibition and cognitive development: Object, number, categorization, and reasoning. Cognitive Development, 15, 63–73. Hughes-Wagner, J., & Dea´k, G. (1999, April). Learning semantic relations between novel words. Presented at the Society for Research on Child Development meeting, Albuquerque, NM. Jacques, S., Zelazo, P. D., Kirkham, N., & Semecesen, T. (1999). Rule selection versus rule execution in preschoolers: An error-detection approach. Developmental Psychology, 35, 770–780.

324

Gedeon O. Dea´k

Johnson, A. S., & Fishkin, A. S. (1999). Assessment of cognitive and affective behaviors related to creativity. In A. Fishkin, B. Cramond, & P. Olszewski-Kubilius (Eds.), Investigating creativity in youth (pp. 265–306). Cresskill, NJ: Hampton. Kalish, C. W., & Gelman, S. A. (1992). On wooden pillows: Multiple classification and children’s category-based inductions. Child Development, 63, 1536–1557. Karmiloff-Smith, A. (1992). Beyond modularity: A developmental perspective on cognitive science. Cambridge, MA: MIT Press. Katz, N., Baker, E., & Macnamara, J. (1974). What’s in a name? A study of how children learn common and proper names. Child Development, 45, 469–473. Lakoff, G. (1987). Women, fire, and dangerous things. Chicago, IL: University of Chicago. Lezak, M.D. (1995). Neuropsychological assessment (3rd ed.). London: Oxford University Press. Liittschwager, J., & Markman, E. M. (1994). 16- and 24-month-olds’ use of mutual exclusivity as a default assumption in second-label learning. Developmental Psychology, 6, 955–968. Luchins, A. S. (1942). Mechanization in problem solving: The effect of Einstellung. Psychological Monographs, 54 (6, Whole No. 248). Luria, A. R. (1959). The directive function of speech in development and dissolution. Word, 341–452. MacWhinney, B. (1978). The acquisition of morphophonology. Monographs of the Society for Research on Child Development, 43 (1–2, serial no. 174). MacWhinney, B. (1987). The competition model. In B. MacWhinney (Ed.), Mechanisms of language acquisition (pp. 249–308). Hillsdale, NJ: Lawrence Erlbaum. MacWhinney, B., Ple´h, C., & Bates, E. (1985). The development of sentence interpretation in Hungarian. Cognitive Psychology, 17, 178–209. Maratsos, M. (1998). The acquisition of grammar. In D. Kuhn & R. Siegler (Volume Eds.), W. Damon (Series Ed.), Handbook of child psychology (5th ed.): Volume 2: Cognition, perception, and language (pp. 421–466). New York: Wiley. Markman, E. M. (1979). Realizing that you don’t understand: Elementary school children’s awareness of inconsistencies. Child Development, 50, 643–655. Markman, E. M. (1994). Constraints on word meaning in early language acquisition. Lingua, 92, 199–227. Markman, E. M., & Wachtel, G. F. (1988). Children’s use of mutual exclusivity to constrain the meanings of words. Cognitive Psychology, 20, 121–157. Mayr, U., & Keele, S. W. (2000). Changing internal constraints on action: The role of backward inhibition. Journal of Experimental Psychology: General, 129, 4–26. Medin, D. L., & Shoben, E. J. (1988). Context and structure in conceptual combination. Cognitive Psychology, 20, 158–190. Meier, N. R. (1931). Reasoning in humans. Journal of Comparative Psychology, 12, 181–194. Meier, N. R. (1937). Reasoning in rats and human beings. Psychological Review, 44, 365–378. Meiran, N., Chorev, Z., & Sapir, A. (2000). Component processes in task switching. Cognitive Psychology, 41, 211–253. Merriman, W. E., & Bowman, L. L. (1989). The mutual exclusivity bias in children’s word learning. Monographs of the Society for Research on Child Development, 54 (3–4, serial no. 220). Merriman, W. E., Jarvis, L. H., & Marazita, J. M. (1995). How shall a deceptive thing be called? Journal of Child Language, 22, 129–149. Meyer, D. E., & Kieras, D. (1997). A computational theory of executive cognitive processes and multiple-task performance: Part I, Basic mechanisms. Psychological Review, 104, 3–65. Miller, P. H., Haynes, V., DeMarie-Dreblow, D., & Woody-Ramsey, J. (1986). Children’s strategies for gathering information in three tasks. Child Development, 57, 1429–1439.


325

Milner, B., & Petrides, M. (1984). Behavioural effects of frontal-lobe lesions in man. Trends in Neuroscience, 403–407. Monsell, S., & Driver, J. (Eds.) (2000). Control of cognitive processes: Attention and performance XVIII. Cambridge, MA: MIT Press. Monsell, S., Yeung, N., & Azuma, R. (2000). Reconfiguration of task-set: Is it easier to switch to the weaker task? Psychological Research, 63, 250–264. Munakata, Y., & Yerys, B. (2001, October). Perseveration in card sorting: Do children know more or less than we think? Paper presented at the second Cognitive Development Society meeting, Virginia Beach, VA. Narasimham, G., & Dea´k, G. O. (2001, October). Inductive and deductive flexibility in 3- and 4-year-old children: Effects of task and complexity. Poster presented at the Cognitive Development Society meeting, Virginia Beach, VA. O’Bryan, K. G., & MacArthur, R. S. (1969). Reversibility, intelligence, and creativity in nine-year-old boys. Child Development, 40, 33–45. Olson, D. R. (1977). From utterance to text: The bias of language in speech and writing. Harvard Educational Review, 47, 257–281. O’Neill, D. K. (1996). Two-year-old children’s sensitivity to a parent’s knowledge state when making requests. Child Development, 67, 659–677. Pandya, D. N., & Yeterian, E. H. (1998). Comparison of prefrontal architecture and connections. In A. Roberts, T. Robbins, & L. Weiskrantz (Eds.), The prefrontal cortex (pp. 51–66). Oxford: Oxford University Press. Papagno, C., & Basso, A. (1996). Perseveration in two aphasic patients. Cortex, 32, 67–82. Perner, J., & Lang, B. (2002). What causes 3-year-olds’ difficulty on the Dimensional Change Card Sorting task? Infant and Child Development, 11, 93–105. Perner, J., Stummer, S., & Lang, B. (1999). Executive functions and theory of mind: Cognitive complexity or functional dependence? In P. D. Zelazo, J. Astington, & D. Olson (Eds.), Developing theories of intention: Social understanding and self-control (pp. 133–152). Mahwah, NJ: Lawrence Erlbaum. Piaget, J. (1954). The construction of reality in the child. New York: Ballantine. Picard, D., & Vinter, A. (1999). Representational flexibility in children’s drawings: Effects of age and verbal instructions. British Journal of Developmental Psychology, 17, 605–622. Reed, M. A., Pien, D. L., & Rothbart, M. K. (1984). Inhibitory self-control in preschool children. Merrill-Palmer Quarterly, 30, 131–147. Revelle, G. L., Wellman, H. M., & Karabenick, J. D. (1985). Comprehension monitoring in preschool children. Child Development, 56, 654–663. Rice, C., Koinis, D., Sullivan, K., Tager-Flusberg, H., & Winner, E. (1997). When 3-year-olds pass the appearance-reality test. Developmental Psychology, 33, 54–61. Robbins, T. W. (1998). Dissociating executive functions of the prefrontal cortex. In A. Roberts, T. Robbins, & L. Weiskrantz (Eds.), The prefrontal cortex (pp. 117–130). Oxford: Oxford University Press. Roberts, A., Robbins, T., & Weiskrantz, L. (Eds.) (1998). The prefrontal cortex. Oxford: Oxford University Press. Runco, M. (1993). Cognitive and psychometric issues in creativity research. In S. Isaksen, M. Murdock, R. Firestein, & D. Treffinger (Eds.), Understanding and recognizing creativity: The emergence of a discipline (pp. 331–368). Norwood, NJ: Ablex. Sapp, F., Lee, K., & Muir, D. (2000). Three-year-olds’ difficulty with the appearance-reality distinction: Is it real or is it apparent? Developmental Psychology, 36, 547–560. Schmidt, C. R., & Paris, S. G. (1983). Children’s use of successive clues to generate and monitor inferences. Child Development, 54, 742–759. Schober, M. F. (1993). Spatial perspective-taking in conversation. Cognition, 47, 1–24.

326

Gedeon O. Dea´k

Schyns, P. G., & Oliva, A. (1999). Dr. Angry and Mr. Smile: When categorization flexibly modifies the perception of faces in rapid visual presentations. Cognition, 69, 243–265. Shatz, M., & Backscheider, A. (2001, December). The development of non-object categories in 2-year-olds. Presented at the Early Language Acquisition Conference, Lyon. Siegal, M. (1991). Knowing children: Experiments in conversation and cognition. Hillsdale, NJ: Erlbaum. Siegel, I. E., Saltz, E., & Roskind, W. (1967). Variables determining concept conservation in children. Journal of Experimental Psychology, 74, 471–475. Siegler, R. S. (1981). Developmental sequences within and between concepts. Monographs of the Society for Research in Child Development, 46 (2, Serial No. 189). Siegler, R. S. (1996). Emerging minds. Oxford: University Press. Siegler, R. S., & Crowley, K. (1991). The microgenetic method: A direct means for studying cognitive development. American Psychologist, 46, 606–620. Smith, E. R., Fazio, R. H., & Cejka, M. (1996). Accessible attitudes influence categorization of multiply categorizable objects. Journal of Personality and Social Psychology, 71, 888–898. Smith, L. B. (1999). Children’s noun learning: How general learning processes make specialized learning mechanisms. In B. MacWhinney (Ed.), The emergence of language (pp. 207–303). Mahwah, NJ: Erlbaum. Smith, L. B., & Heise, D. (1992). Perceptual similarity and conceptual structure. In B. Burns, (Ed.), Advances in Psychology—percepts, concepts, and categories: The representation and processing of information (Vol. 93, pp. 233–272). Amsterdam: Elsevier. Sonnenschein, S., & Whitehurst, G. J. (1983). Training referential communication skills: The limits of success. Journal of Experimental Child Psychology, 35, 426–436. Speer, J. R. (1984). Two practical strategies young children use to interpret vague instructions. Child Development, 55, 1811–1819. Stemberger, J. P. (1989). Speech errors in early child language production. Journal of Memory and Language, 28, 164–188. Stemmer, B., & Joanette, Y. (1998). The interpretation of narrative discourse of brain-damaged individuals within the framework of a multilevel discourse model. In M. Beeman & C. Chiarello (Eds.), Right hemisphere language comprehension (pp. 329–348). Mahwah, NJ: Erlbaum. Taylor, A. E., Saint-Cyr, J. A., & Lang, A. E. (1986). Frontal lobe dysfunction in Parkinson’s disease. Brain, 109, 845–883. Thelen, E., & Smith, L. B. (1994). A dynamic systems approach to the development of cognition and action. Cambridge, MA: MIT Press. Torrance, E. P. (1988). The nature of creativity as manifest in its testing. In R. Sternberg (Ed.), The nature of creativity: Contemporary psychological perspectives (pp. 43–75). Cambridge University Press. Towse, J. N., Redbond, J., Houston-Price, C., & Cook, S. (2000). Understanding the dimensional change card sort: Perspectives from task success and failure. Cognitive Development, 15, 347–365. Tranel, D., Damasio, H., & Damasio, A. R. (1997). On the neurology of naming. In H. Goodglass & A. Wingfield (Eds.), Anomia: Neuroanatomica and cognitive correlates (pp. 65–90). San Diego: Academic Press. Vitkovitch, M., & Humphreys, G. (1991). Perseverant responding in speeded picture naming: It’s in the links. Journal of Experimental Psychology: Learning, Memory, & Cognition, 17, 664–680. Vurpillot, E. (1968). The development of scanning strategies and their relation to visual differentiation. Journal of Experimental Child Psychology, 6, 622–650.


327

Vygotsky, L. (1978). Mind in society. Cambridge, MA: Harvard University Press. Wellman, H. M. (1990). Children’s theories of mind. Cambridge, MA: MIT Press. Yen, L. (1997). Unpublished study, Vanderbilt University. Yuill, N., & Oakhill, J. (1991). Children’s problems in text comprehension. Cambridge: Cambridge University Press. Zelazo, P. D., & Frye, D. (1996). Cognitive complexity and control: A theory of the development of deliberate reasoning and intentional control. In M. Stamenov (Ed.), Language structure, discourse, and the access to consciousness. (pp. 113–153) Amsterdam: John Benjamins. Zelazo, P. D., Frye, D., & Rapus, T. (1996). An age-related dissociation between knowing rules and using them. Cognitive Development, 11, 37–63. Zelazo, P. D., & Reznick, S. J. (1991). Age-related asynchrony of knowledge and action. Child Development, 62, 719–735. Zelazo, P. D., Reznick, J. S., & Pinõn, D. E. (1995). Response control and the execution of verbal rules. Developmental Psychology, 31, 508–517. Zelazo, P. D., Reznick, J. S., & Spinazzola, J. (1998). Representational flexibility and response control in a multistep multilocation task. Developmental Psychology, 34, 203–214.


A BIO-SOCIAL-COGNITIVE APPROACH TO UNDERSTANDING AND PROMOTING THE OUTCOMES OF CHILDREN WITH MEDICAL AND PHYSICAL DISORDERS

Daphne Blunt Bugental and David A. Beaulieu DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF CALIFORNIA SANTA BARBARA, SANTA BARBARA, CALIFORNIA, CA 93106

I. INTRODUCTION II. THEORETICAL MODELS EMPLOYED TO UNDERSTAND THE EXPERIENCES OF CHILDREN WITH MEDICAL AND PHYSICAL DISORDERS A. MEDICAL MODEL B. SOCIAL MODEL C. AN INTERACTIVE MODEL: PARENTING RESPONSES AS A FUNCTION OF PARENTAL COGNITIONS AND CHILDREN’S CHARACTERISTICS D. AN INTERACTIVE MODEL: PARENTAL INVESTMENT AS A FUNCTION OF THEIR RESOURCES AND A CHILD’S REPRODUCTIVE POTENTIAL E. RESILIENCE AND THRIVING MODELS F. COMPARISON AND INTEGRATION OF MODELS III. OPTIMIZING THE OUTCOMES OF CHILDREN WITH MEDICAL AND PHYSICAL DISORDERS A. WHO WERE THE FAMILIES? B. WHAT WAS THE NATURE OF THE EMPOWERMENT INDUCTION? C. DID PARENTAL EMPOWERMENT SERVE TO ENHANCE THE OUTCOMES OF CHILDREN WITH MPDs? D. SUMMARY IV. INTEGRATION REFERENCES

I. Introduction Children born with medical or physical disorders (MPDs) have traditionally been viewed from the framework of a deficit model. Problems are seen as resulting from the child’s disorder—which in turn—is


329


330

Daphne B. Bugental and David A. Beaulieu

viewed as predictive of a wide range of social and cognitive problems. One limitation within this approach is that it fails to consider the fact that many disorders carry no necessary implications for cognitive or social problems (e.g., as is the case for orthopedic, hearing or visual disorders). A second limitation is that it fails to consider that many of the later problems experienced by MPD children follow from the reactions of others. Some social reactions (e.g., avoidance) may constitute a general response to all children who are ‘‘different.’’ Other social reactions may be specific to the nature of the stimulus the child poses for others. For instance, a child with an orthopedic may appear to be dependent. Thus, many of the problems that may be experienced by children follow from the perceptions of others rather than the specific etiology of the child’s condition. In this chapter, we consider the theoretical approaches that have been applied—or might be applied—to the life outcomes experienced by MPD children. We begin by considering the traditional medical model, and move on to consider the social model that emerged to challenge this position. We then consider developmental models that focus on the interactive effects that result as a function of child and parent characteristics and resources. Across this discussion, we draw from the fields of developmental psychology, social psychology, developmental neuroscience, and evolutionary psychology. We also summarize research that shows some of the ways that problems may be prevented and positive outcomes fostered for MPD children as a result of an integrative bio-social-cognitive approach to intervention.

II. Theoretical Models Employed to Understand the Experiences of Children with Medical and Physical Disorders Theoretical models employed to understand the experiences of children with MPDs have reflected the basic conceptions and content of a number of relevant fields. Health care professionals, rehabilitation professionals, and educators—consistent with their remedial approach—initially focused on MPDs as conditions that resided primarily within the child. As social scientists took interest in the topic, attention turned to the ways in which others respond to those with MPDs. When developmental psychologists entered the scene, efforts were made to determine the ways in which the characteristics of the MPD child interact with the social responses of others to influence their outcomes. In particular, attention was given to the ways in which parents (and to a lesser extent, teachers and peers) respond to children with MPDs. Special attention was given to the ways in which

Children with Medical and Physical Disorders

331

parents think about such children, and how the interpretations they make influence their emotions and behaviors. As a relatively new entry to this field of inquiry, evolutionary psychology has directed interest to the ways in which parents invest or fail to invest in children with MPDs based upon the availability of resources. Finally, the field of developmental neuroscience offers potential insights into the processes that may foster child resilience (or even thriving) in the face of the early challenges posed by medical and physical disorders. A. MEDICAL MODEL

Historically, health care scientists and educators have utilized a medical model in conceptualizing the nature and effects of children’s medical and physical disorders. From this framework, children’s MPDs are understood as conditions that reside within an individual (Hedlund, 2000; Llewellyn & Hogan, 2000; Marks, 1997; Olkin, 2002). The goal of this approach has been to find medical procedures that would prevent or remediate such disorders, or to find ways to best prepare such children to optimize their life experiences through rehabilitation and educational programs tailored to the specific disorder. Although the focus on the specific etiology of various MPDs is useful in the design of relevant medical or educational programs, it does not provide an understanding of the social experiences of such children. In doing so, it has failed to give full consideration to factors that influence the child’s social and emotional well-being, an outcome that may be more influenced by the responses of others than by the child’s specific disorder (Lavigne & Faier-Routman, 1993). The medical model is still heavily represented among professionals within traditional, clinical, rehabilitation, and educational disciplines. Although disability is increasingly recognized as a social policy issue, there continues to be a strong emphasis on the specific etiology of children with MPDs (e.g., Glueckauf, 2000). As a result, the preponderance of work in this field focuses on disorder-specific treatment and remediation programs (as may be seen in Frank & Elliott’s Handbook of Rehabilitation Psychology, 2000 and Wang, Reynold, & Walberg’s Handbook of Special Education, 1991). In the future, it is likely that those who focus on medical or educational remediations will come to include a more complete consideration of the social as well as the medical origins of many of the problems experienced by MPD children. In doing so, it can be expected that greater consideration will be given to the implications of medical, rehabilitation, and educational procedures on the quality of the MPD child’s social life.

332


B. SOCIAL MODEL

1. Social Stigma In the late 1940s, considerable public concern was addressed to the plight of those World War II veterans who came home with disabling injuries. These concerns triggered efforts to know more about the psychological and social effects (as well as direct medical effects) of MPDs. The Social Science Research Council, in response to this need, asked Roger Barker (a prominent social psychologist) to review the literature on the adjustment issues that confront those with physical handicaps. Responding to this invitation, Barker arranged for a set of papers written by social psychologists to be published in the Journal of Social Issues. Within these papers, a new perspective emerged on the topic of disability. The argument was introduced that those with MPDs were subject to discrimination, and may be best understood as a minority group. That is, such individuals were subject to stereotyping and discrimination in many of the same ways as were those from racial or ethnic minorities (e.g., limitations were placed upon their access to public facilities; educational opportunities). This position was a break with past views of disability as reflecting a ‘‘flaw’’ within the individual and with past recommendations that disabled individuals best accommodate if they ‘‘accept their fate’’ (as pointed out by Meyerson, 1988). Although the topic of stigma soon assumed importance as an area of social psychology, attention turned more to the stigmatizing effects of race and gender than those of disabilities. Nonetheless, a number of social psychologists continued to argue for disabilities as a social rather than simply a physical problem (e.g., Hebl & Kleck, 2000). With this interest came attention to the shared ways that others respond to individuals with MPDs. For example, Jones et al. (1984) argued that the response shown to those who are disabled (or have other potentially stigmatizing conditions) varies with others’ beliefs about their condition (e.g., how controllable it is or how it was acquired). For example, Juvonen (1991) argued that since one’s weight is seen as controllable, overweight children may be deemed ‘‘blameworthy’’ and thus receive particularly negative reactions from others. Her findings provided evidence for this prediction and the cognitive approach taken by Jones and others pointed out that to a large extent, the ways others react to those with medical or physical disorders depends upon their beliefs. Although the original message regarding the social implications of disabilities failed to stimulate extensive research, it did stimulate social action. As its most important outcome, the American Disabilities Act (ADA) was passed in 1990. With the removal of barriers (physical and


333

social) to the world of the able-bodied, those with disabilities were increasingly able to become full participants in community life. The ADA, consistent with the papers first appearing in the Journal of Social Issues in 1948 (and with the later writings of others, e.g., Pape & Tarvydas, 1993), located the central problems of the disabled as existing within the social– political environment, not intrinsic flaws within the individual. In sum, as those with physical and medical disabilities came to be understood as a minority group, the problems they experienced were increasingly conceptualized as following from the biased ways in which they were dealt with within the broader social community; and their limits were seen as primarily imposed upon them by their social–political environment. Comments offered by disability activists echo this point, for example, ‘‘Viewed from this perspective, it is clear that it is the barriers, both physical and attitudinal, that need to be changed, not the impairments or the bodies with which we live. I have asked many disabled persons what causes them more difficulty, the disability or the discriminatory barriers put in their way. The answer is overwhelmingly the latter.’’ (Fries, 1997, pp. 7–8)

2. Moving Beyond Stigma Although the increasing focus on stigmatization processes has added to our understanding of the experiences of those with MPDs, the story is still incomplete. The focus on such individuals as ‘‘flawed’’ is only one part of the interpretive biases that may be invoked by others. To the extent children with MPDs are visibly different (e.g., they may use a wheelchair, their faces may appear unattractive, their limbs may appear distorted), they will be at risk to elicit fear or apprehension from others—as is found in reactions to adults who are visibly different (Blascovich et al., 2001). We propose here that the social problems experienced by children with MPDs are more completely understood when consideration is given to the stimulus value that such children have for others. Medical or physical conditions that share little or nothing in terms of their etiology may be similar in the responses they elicit from others. Thus, we need to consider the social stimulus features of MPDs. For example, some medical or physical disorders interfere with the child’s capacity to negotiate the physical environment—a pattern that easily leads them to be seen as dependent or immature. Other MPDs interfere with the child’s ability to become socially engaged with the world—a pattern that may lead others to think of their apparent unresponsiveness as reflecting noncompliance or rejection. However, focusing on different groupings we must be mindful of the possibility that a child’s particular disorder may have multiple social

334


consequences. For example, a child with a hearing impairment may be seen as unresponsive but may also experience social stigma if he or she wears a visible hearing aid. Children with severe visual impairments may not only experience limitations in their ability to negotiate the physical environment (and thus appear dependent), they may also be interpreted by others as socially disengaged or inattentive. Thus, for many types of MPDs, children will be faced with multiple social challenges as a result of the different features of their disorder. We need to think in terms of the social aspects of different MPDs rather than discrete groupings of MPDs. In this paper, we focus on three aspects of MPDs that will elicit atypical responses from others. These include children’s apparent dependence, apparent unresponsiveness, and appearance impairment. Of all the child characteristics that one might identify, one may ask, ‘‘Why these particular characteristics?’’ Each of these child characteristics have been found to pose a challenge to the caregiving system, and to set the child at risk for maltreatment. However, the different groupings of MPDs differ somewhat in the type of maltreatment risk. MPDs that are associated with continuing levels of high dependence (e.g., orthopedic disorders) set the child at risk for parental neglect (Bugental, in press; Sullivan & Knutson, 2000). MPDs that are associated with apparent child unresponsiveness (e.g., hearing disorders) set the child at risk for physical abuse (Bugental, in press; Sullivan & Knutson, 2000). MPDs that are associated with appearance impairments set the child at risk for abandonment or even infanticide (Bugental, in press; Weiss, 1998). Parents expect their offspring to become capable of taking care of themselves, to be receptive to caregivers’ influence attempts, and ultimately to get married and have children. When children’s characteristics pose perceived or actual threats to these expectations, negative caregiving outcomes are more probable. a. The Social Impact of Children Who Appear to be Dependent. Children with disorders that limit their ability to negotiate the physical environment share some common experiences as a result of these constraints. For example, children with severe visual or orthopedic disorders experience barriers to their mobility. As a result, both groupings of children may view the world in distinctive ways. In addition, their mobility constraints may lead them to be seen by others in distinctive ways. For example, they may easily be seen as needy, child-like, or dependent. Children who experience long-term limitations in their mobility may require high levels of protective care from their parents. As a result, parents of such children experience exceptionally high levels of distress (e.g., Sloper & Turner, 1993). Distress is typically higher among mothers than fathers. The heightened distress experienced by mothers appears to be an indirect


335

consequence of the child’s disorder; that is, children with mobility constraints place an exceptional demand on her time (Waddington & BuschRossnagel, 1992). As a result, mothers may have difficulty in meeting both their own needs as well as those of other family members. Not surprising, then, is the observation that those mothers who have greater ‘‘resources’’ (e.g., access to transportation and employment) experience lower levels of distress (Sloper & Turner, 1993). Furthermore, children with mobility constraints have been found to suffer elevated levels of parental neglect; they do not, however, experience elevated levels of physical abuse by parents (Sullivan & Knutson, 2000). In addition, parents of mobility-constrained children demonstrate greater conflict than do other parents. Such conflict, in turn, predicts differential divorce rates; that is, the parents of mobility-impaired children (e.g., visual and orthopedic impairments) show higher divorce rates than do the parents of children without MPDs or the parents of children who have other types of MPDs (Hodapp & Krasner, 1995; Joesch & Smith, 1997). It is important to note that this elevation in divorce rates remain significant even when the effects of other factors (e.g., socio-economic status) are controlled statistically. Women regularly report that the apparent inability of the father to accept the child contributed to the divorce (Sloper & Turner, 1993). This shows another source of distress for these mothers. That is, while mothers may invest in their mobility-impaired children, their overall access to resources is undermined by fathers who may be less willing to invest in these children. b. The Social Impact of Children Who Appear to be Unresponsive. As a second aspect of MPDs, children with a variety of disorders may have difficulty responding to or maintaining social contact with others. As a result, they may easily be seen by others as disrespectful, rejecting or noncompliant. Children who are hearing impaired or who have attentional problems represent examples of such MPDs. For example, children who have difficulty maintaining their attention on a task have been found to be interpreted by parents as defiant or uncontrollable (Johnston & Freeman, 1997). Children who appear to be inattentive are at risk for triggering negative emotional responses from others. Their interactions with parents more easily become conflictual and parents show a harsher style of interaction with them than with other children. Indeed, it has been suggested (Zirpoli & Bell, 1987) that a child’s apparent unresponsiveness may have stronger effects on parental reactions than the actual MPD itself. Ultimately, such children are at elevated risk for physical abuse (as shown in the greatly elevated levels of abuse among hearing impaired or learning disabled

336


children, Sullivan & Knutson, 2000). In addition, children who appear inattentive are at elevated levels of risk for neglect. However, this aspect of children’s MPDs is not associated with interparental conflict. That is, the presence of childhood disorders that are accompanied by apparent unresponsiveness have not been found to be related to parental divorce rates (Hodapp & Krasner, 1995; Joesch & Smith, 1997). c. The Social Impact of Children with Appearance Impairments. A third aspect of many MPDs involves appearance anomalies associated with the child’s disorder. Children who violate parental expectations in the ways they look often come to share many of the same subjective and social experiences. Although such variations in appearance take many different forms (e.g., the use of a hearing aid, loss of a limb), they overlap in the extent to which they may trigger feelings of fear and disgust in others— feelings that in turn may foster social avoidance. Children with appearance anomalies at birth face a particular risk in the first few months of life (Daly & Wilson, 1988; Marks, 2001). They are more likely to be abandoned than are children who have equivalent or even more severe health risks—but who look the same as other children (Weiss, 1998). Such appearance anomalies are also associated with elevated rates of infanticide—even within countries that have laws and ethical principles that act against this practice (Daly & Wilson, 1984, 1988). At a lower level of negativity, parents initially are unresponsive to their appearance-impaired infants (Wasserman et al., 1987). The negative reactions of parents to appearance anomalies in newborns largely disappear across the course of infancy, and are often replaced with exceptionally high levels of attachment. For example, mothers who judge their infants (with craniofacial anomalies) to be most unattractive early on were found to be most attached to their infants at one year (Coy, Speltz, & Jones, 2002; Spelz et al., 1997). Once mothers get attached to such children, they are unusually responsive to them (Wasserman et al., 1987) At the same time, many parents seek to surgically ‘‘normalize’’ the appearance of their appearance-impaired children. Fathers, somewhat more than mothers, are motivated to have their children’s appearance (and behavior) be socially accepted (Lamb & Meyer, 1991). However, many mothers share this same motivation and when they do, they tend to respond positively to their children’s surgical ‘‘normalization.’’ For example, mothers whose children with cleft palate had reconstructive surgery were found to respond more positively to them following this event (Lindenfelser, 1997). Whereas parents of children with MPDs are typically at risk for elevated levels of conflict and divorce, this is not the case for the parents of children


337

with appearance impairments. Although such parents show elevated levels of depression and feelings of hopelessness (Benson et al., 1991), this does not appear to influence the marital relationship. Instead, such parents are at lower risk for divorce than are parents of children who do not have MPDs or parents of children with other types of MPDs (Joesch & Smith, 1997). Although empirical evidence is lacking on the reasons for this pattern of reduced risk, we speculate that the frequently observed withdrawal of support by friends and extended family in response to the birth of an appearance-impaired infant (Benson et al., 1991) fosters increased closeness between parents. In summary, a social model of the experiences of children with MPDs focuses on the general interpretive biases that occur in the perceptions, and ultimately the actions of others. When children are the recipients of such biased responses, the possibilities are increased for self-fulfilling expectations. The work of Steele (e.g., Steele, 1998) has demonstrated the ways in which those who are subject to negative biases (e.g., minorities, women, those from low SES backgrounds) may come to experience ‘‘stereotype threat’’ in achievement domains for which they are seen as less able. When the stereotypes applied to them are made salient, their performance on relevant tasks declines. In the same way, children who are stereotyped as disabled, as having special needs, or as handicapped can be expected to experience performance-limiting anxiety when their MPDs are made salient. In short, settings and experiences that categorize such children as ‘‘different’’ (and implicitly deficient) limit their opportunities to function in the same ways as do other children. If the characteristics of their MPD also make them appear to be dependent, unresponsive, or flawed, they will experience the added difficulty of being interpreted on the basis of their assumed rather than their actual characteristics. C. AN INTERACTIVE MODEL: PARENTING RESPONSES AS A FUNCTION OF PARENTAL COGNITIONS AND CHILDREN’S CHARACTERISTICS

Although considerable attention has been given to the ways in which parents respond to their MPD children, little attention has been given to individual differences in their responses. For the most part, researchers in this area have focused on differences between mothers and fathers. Little systematic attention has been given to other kinds of parental variations in their responses to MPD children. One kind of variation that might predict responses to such children involves parental cognitions concerning caregiving relationships. For example, if the caregiving relationship becomes difficult, is this someone’s fault or is it interpreted in more benign ways that do not imply blame? This leads to concerns with variations in parental

338


beliefs or attributions—a topic that has received considerable theoretical and empirical attention but not as specifically applied to MPD children. As an exception, Affleck and his colleagues (Affleck et al., 1985) explored the extent to which maternal cognitions predicted the later outcomes of their at-risk newborns. Parents of MPD children frequently ask many ‘‘why’’ questions, for example, ‘‘Why did this happen to me?’’ ‘‘Why did this happen to my child?’’ (Shapp et al., 1992). Although such questions are usual early on, they typically decline when and if parents come to accommodate the unexpected event. However, the way these early questions are answered influences both the later well-being of the parents and the wellbeing of the child. Children have been found to be more likely to experience social, cognitive, and emotional problems at later ages when parents engage in other-blame (Affleck et al., 1985). If the blaming process also extends to the child, parents’ affective and behavioral responses are influenced by this attributional bias. Although we have some general ideas about the damaging effects of parental blame (as summarized in Bugental & Happaney, 2002, and Bugental & Johnston, 2000), there is relatively little evidence with respect to the possibility that parents will blame children differentially, based upon the social implications of their disorder. As an exception, Bugental and her colleagues have shown that parents who see themselves as ‘‘power disadvantaged,’’ that is, who see the child as having greater power than themselves, are particularly reactive to children who appear to be unresponsive (e.g., Bugental et al., 2002; Bugental et al., 2000). So, for example, a mother may be more likely to respond to a child’s incessant cries with anger if she believes that the child is being deliberately annoying and intentionally unresponsive to her calming efforts. That is, when parents come to think of themselves as the ‘‘victims’’ of the child, they are more likely to engage in harsh or neglectful parenting practices (Bugental, 2001); however, this pattern is only found when there is a combination of power-biased parental cognitions and a child who appears to be socially unresponsive. In a series of investigations, Bugental and her colleagues have explored the kinds of ‘‘power defensive’’ responses shown by adults with a low perceived balance of power when they are faced with a child who might be interpreted as ‘‘threatening’’ (e.g., a child who is unresponsive to their influence attempts). This response pattern begins with an activation of their ‘‘emergency’’ physiological response systems; for example, their levels of cortisol production (Lin et al., 2002) and their heart rate (Bugental et al., 2000) both increase. They often begin their interactions with such children in a deferential manner (e.g., using soft voices or ‘‘false smiles,’’ that is, they show the kinds of smiles that are not associated with happy affect, Bugental et al., 1997). They eventually shift to a harsher style; for example, they


339

increasingly use a harsh tone of voice and derogatory style of speech (Bugental & Happaney, 2000; Bugental & Lewis, 1998, 1999). As a laboratory analogue of physical abuse, we also found that these parents—when faced with ambiguity in their ability to control an unrelated child trainee’s behavior—used excess physical force in operating computer controls that were believed to display negative feedback to the child (Bugental et al., 2000). No equivalent response patterns have been found for such parents during their interaction with children who appear to be fearful or dependent (Bugental & Shennum, 1984). Children who appear to be unresponsive (e.g., those with attentional or hearing disorders) may well pose a particular threat to those adults who see themselves as lacking power in the caregiving relationship. In contrast, parents who see themselves as having control in the relationship are more likely to respond positively to these same children. From a bio-social-cognitive perspective (e.g., Tomaka et al., 1997), it could be predicted that those parents who see an ‘‘unresponsive’’ child as ‘‘manageable’’ would be more likely to see such a child as a (positive) challenge rather than a ‘‘threat.’’ In summary, a social cognitive approach to childhood MPDs tracks the ways in which caregivers come to think of children with MPDs as a threat. In doing so, they also are likely to respond with physiological responses that mobilize physiological reactions that are consistent with threat. Thus, the biased perceptions of caregivers predict harsh caregiving, as mediated by mobilization of the body’s emergency response systems. This approach targets parental cognitions as a contributor to later problems of children with MPDs, and suggests the utility of introducing cognitively based intervention and prevention programs. D. AN INTERACTIVE MODEL: PARENTAL INVESTMENT AS A FUNCTION OF THEIR RESOURCES AND A CHILD’S REPRODUCTIVE POTENTIAL

All children with MPDs in some way violate parental expectations. Possibly as a reflection of these violations, all groupings of MPD children are at elevated risk for maltreatment of some type (Sullivan & Knutson, 2000). However, many MPD children receive exceptional levels of parental care and flourish in the process. Insights to this apparent contradiction are provided by parental investment theory, an approach derived from evolutionary psychology. Unlike other theoretical approaches, parental investment theory proposes a process by which MPD children may receive polarized parental responses based upon contextual variables. Although many theories offer explanations concerning the negative reactions of parents to offspring with MPDs, evolutionary psychology also specifies the

340


circumstances under which parents (both human and nonhuman) provide enhanced care for their high-risk offspring. 1. Parental Investment Theory Within evolutionary theory, MPD offspring are conceptualized as having low reproductive potential (e.g. Daly & Wilson, 1984). This is most clearly illustrated if one realizes that in our evolutionary past, compared with other children, those born with MPDs would have faced not only much greater risks to survival but would have faced greater obstacles to successful mating. In other words, those born with MPDs would have been less likely than other children to survive to adulthood and thus produce progeny. Within the earlier writings of evolutionary psychology, this feature of MPD children was brought to bear in accounting for the high associated levels of infanticide (among both human and nonhuman parents). The basic mechanism invoked was that parents are more likely to invest in offspring who appear healthy (and thus appear to have high reproductive potential) and to abandon or neglect offspring who appear unhealthy (and appear to have low reproductive potential) (Daly & Wilson, 1984; Trivers, 1974). That is, as a result of variations in offspring’s reproductive potential, there should have been selective pressure for an implicit cost–benefit analysis of parental investment. Parents should utilize this implicit cost–benefit analysis in deciding the relative costs and benefits they can expect to obtain as a result of investment in one offspring as opposed to other offspring (that they currently have or may have in the future). Offspring with MPDs regularly involve higher costs than do other offspring. In some cases, such costs are financial. In other cases, they pose the cost of time and energy. In still other cases, they involve social and personal costs, for example, the parents of socially stigmatized child may themselves be stigmatized (referred to by Goffman, 1963, as a courtesy stigma). On the benefit side of the equation, even though the MPD child may be of low reproductive potential, to an older mother nearing menopause such a child may provide a final opportunity to further her reproductive success (as pointed out by Daly & Wilson, 1984). We want to emphasize that this implicit cost–benefit analysis is implicit or automatic rather than deliberate. For example, the disgust response to an appearance-impaired child is no more deliberate than the experience of sexual affinity to members of the opposite sex. That is, just as individuals who ‘‘decided’’ to engage in sex with members of the opposite sex left more progeny than those who did not, individuals who ‘‘decided’’ to invest in their offspring based in part on the child’s reproductive potential would have had more progeny than those who invested indiscriminately.


341

However, the story was still incomplete. Why do some MPD offspring— under some circumstances—receive more than ‘‘usual’’ levels of care? What are the circumstances under which parents would seem to act against their own reproductive interests by extra investment in a child that represents a low reproductive potential? Before the implicit costs and benefits can be tallied in this situation, consideration needs to be given to the resources available to the parents. As originally suggested by Trivers (1974) and as more fully developed by others (e.g., Mann, 1992), parents invest differentially in their costly offspring on the basis of the availability of resources. Parental investment theory—in its most complete form—predicts that parents with low resources should invest relatively more in their high reproductive potential offspring. Conversely, parents with high resources should invest relatively more in their low reproductive potential offspring. The underlying logic of these predictions is best illuminated by focusing on the diminishing returns aspect of parental investment, and this aspect of parental investment is best illustrated with a simple thought experiment: 1. Imagine three sets of parents each with a set of twins. 2. Each set of twins consists of an unhealthy twin with a low reproductive potential and a healthy twin with a high reproductive potential. 3. Both twins are hungry and on average the healthy child needs two units of food in order to meet their nutritional needs and the unhealthy child (due to higher risks of survival) needs on average four units. 4. Assume that parents respond to ‘‘diminishing returns.’’ Regardless of how the parents decide to distribute their food units a child gains one unit of survival if his or her nutritional needs are to be met. And for each incremental food unit beyond what is necessary, the child’s survival units increase by half the gains of the previous unit of food. For example, if the parents have accumulated four units of food and have decided to give all the units to the healthy twin, the twin will gain one unit of survival for the initial two units of food but will only gain half a unit of survival for the third, and quarter of a unit of survival for the fourth (resulting in 13⁄4 units of survival). 5. Finally, assume that the child’s units of survival can be reconceptualized by the parents as units of the child’s reproductive potential. Now imagine that the first set of parents has two units of food, the second has four and the third has eight. For the first set of parents the most profitable distribution of food units is to give both units to the healthy child. For the second set of parents, due to diminishing returns (i.e., statement 4), there exists two equally profitable distributions. That is, the parents could give all four food units to the healthy child and receive a 13⁄4 unit increase in their offspring’s reproductive potential, or they could give three food units to the healthy child and one food unit to the unhealthy child and receive the

342


same payoff. Again due to diminishing returns, for the last and most resource-rich set of parents, the most profitable allocation of food units is to give the healthy child three food units and the unhealthy child the remaining five food units, resulting in a three-unit increase in their offspring’s reproductive potential. The bottom line is that the diminishing returns aspect of parental investment assumes that at some point the cost of parental investment in one offspring (in this case the healthy one) becomes disproportionately high compared with the benefits of investing in another offspring (in this case the unhealthy one). Furthermore, as illustrated by our thought experiment, because an unhealthy child often initially needs more investment than a healthy one, the costly diminishing returns associated with investing in a healthier child may never arise unless parents have an abundance of resources. In other words, if parents have high resources, they can possibly afford to invest in a child with a low reproductive potential because it will not meaningfully deplete their ability to simultaneously meet the needs of a child that may have a higher reproductive potential. As a result of this enhanced investment, a high-risk child’s initial low reproductive potential increases (i.e., the child is more likely to survive and have progeny of their own) and other children do not suffer in the process. 2. Evidence from the Nonhuman World Evidence from nonhuman models is particularly valuable in that investment in offspring based upon cost and resources cannot be explained on the basis of higher-level cognitive mechanisms. Among humans, cognitive mechanisms are particularly likely to be invoked when parents show high investment in an MPD child. Sometimes such explanations are negative; for example, such parents may be thought of as reflecting guilt or denial processes. Other explanations are more positive; for example, such parents may be thought of as engaging in positive reappraisal, a mechanism used to explain the ways in which people may infuse positive meaning into stressful life events (e.g., Folkman, 1997). However, the decisions made by non-human parents regarding care of the young do not involve conscious appraisal processes or foresight. Nonetheless, nonhuman parents appear to be sensitive to both the relative cost of offspring, as well as the presence of resources in the environment in making their decisions (Maestripieri & Carroll, 1998). Nonhumans (as is true for humans) show particular risk-aversion to investing in unrelated offspring. In species in which females mate with multiple partners, males often engage in selective litter destruction based upon the duration of time since the male’s last mating episode, and thus the probability that the young were sired by them (vom Saal & Howard, 1982).


343

In this case we see the operation of a clock-like computational mechanism as a regulator of responses to the young. Although nonhuman females are more likely to be involved in protecting the young from male predation than in infanticide, they display selective investment rules—rules that are organized on the basis of brood quality and resource availability. Thus, they may abandon a brood that is small and thus a poor investment. Selective investment of this type not only occurs naturally (e.g., Mock & Parker, 1986), it may also be induced by experimental brood reduction (Armstrong & Robinson, 1988). Although instances of abandonment of individual offspring is less common, occasional instances are found. For example, royal penguins are more likely to eject the first-laid egg (St. Clair et al., 1995). The adaptiveness of this response is shown by the fact that first-laid eggs are more likely to be flawed, as evidenced by the fact that they are less likely to hatch than are later-laid eggs. However, the predominant means of protecting parental investment across species involves shifting care practices in response to the availability of resources. Investment in the young is often terminated when resources are low, for example when there are food shortages (Mock, 1984). In sum, there is substantial evidence to support the conjecture that nonhuman parents show reduced investment in their offspring when the potential pay-off is low. The next question then becomes, Is there any reason to believe that nonhuman parents would—under some circumstances—benefit by the provision of extra care to their high-cost offspring? To answer this question, Davis and Todd (1999) conducted a computer simulation study to test whether birds would increase their reproductive success by following the strategies predicted by parental investment theory. Within the model, they tested the ultimate outcomes of different feeding strategies. In assessing the success of different feeding strategies (e.g., feeding the largest chick first, feeding the smallest chick first), Todd and David varied the availability of food. The level of reproductive success was defined on the basis of the combined weight of all the chicks when they reached the age that they would be expected to leave the nest. As predicted, the most successful feeding strategy depended on the availability of food. Thus, when food was relatively easily available, the winning strategy was to feed the smallest chick first. Alternatively, when food was less plentiful, the most successful strategy was to feed the largest chick first. Consistent with the outcomes of this simulation of parental investment theory are findings derived from naturalistic observations. For example, pied flycatcher dams (mothers) have been observed to selectively vary the feeding strategies they employ based upon the availability of food (Gottlander, 1987). As predicted, they are more likely to feed their largest

344


chick (the one most likely to survive) first when food is scarce. In contrast, when food is plentiful, they feed the smallest chick first. In the absence of training or higher-level cognitive processes, birds appear to follow the rule associated with the diminishing return aspect of parental investment; that is, when resources are plentiful, one benefits most from investing relatively more in the offspring with low reproductive potential. At this point, any consideration of the neural substrates of relevant computational processes is conjectural. Although there is a long history of nonhuman research showing sensitivity to quality and quantity of food, less is known about sensitivity to the reproductive fitness of the young. In the examples offered here, fitness decisions of nonhumans were based upon simple computations of numerosity (brood size) and birth sequence (firstvs. later-laid eggs). As one idea, it has been speculated that vocalizations of the young simultaneously convey their high levels of need (a cost), but also their vigor (and potential for reproductive success) (Lumaa et al., 1998). Support for this notion has come from primatologists who have studied parental investment in malformed infants on Awajishima Island in Japan (Nakamichi, 1986). For unknown reasons, there are a high number of such infants in this setting—thus allowing extensive observation. These infants were found to vocalize almost three times as often as by nondeformed infants; mothers, in turn, responded even more frequently to the calls of disabled infants than they did to the calls of able-bodied infants. 3. Evidence from Humans There is also suggestive evidence that the investment patterns of human parents may mirror the investment patterns of nonhuman parents (Mann, 1992). On the negative side of the story, children’s medical and physical disorders set the young child at risk for both infanticide (Daly & Wilson, 1984) and maltreatment (Brown et al., 1999). Infanticide continues to exist in countries in which there are clearly enforced laws against such practices. The question then becomes, to what extent are such practices more common among parents with low resources? And is there any evidence to support the alternative outcome of enhanced care of offspring with low reproductive potential in the presence of high resources? There has been extensive evidence supporting the notion that parents with low resources (e.g., low socio-economic status, low access to family members who can provide resources) are at elevated risk both for infanticide and child maltreatment. Little evidence is available, however, with regard to the selective manifestation of these negative outcomes among MPD children. There is, of course, considerable comorbidity of child risk and parental risk (i.e., lack of resources). Thus, teenage mothers are more likely


345

to lack resources, to have children born with medical or physical complications, and to be at elevated risk for child abandonment or maltreatment (Daly & Wilson, 1984). However, this does not constitute evidence that lack of parental investment follows from the combined circumstances of child risk and low maternal resources. Better evidence is provided by the care provided to offspring at differential risk within the same family. For example, Mann (1992) reported that mothers invest differentially in twin pairs, based upon the risk status of the individual twin and their access to resources. Mothers were often observed to show greater investment in the twin who manifested the stronger cues to health (Mann, 1992). At the same time, there were exceptions. Mothers were more likely to withdraw investment from an at-risk child if they lacked resources, and provide enhanced investment if they had adequate resources. There are also some indications that at-risk children experience quite different outcomes in countries (or areas) that vary in the general availability of resources. As an example, severely disadvantaged mothers in the shantytowns of Brazil often allow their high-risk infants to die of benign neglect (Scheper-Hughes, 1985). In the US (in particular, within advantaged middle-class families), mothers often show exceptionally high level of investment in their high-risk (e.g., premature) infants (Field, 1982). In short, MPD children may experience polarized outcomes, contingent upon the resources that are available to parents as individuals or that are available to community members as a whole. Thus, there appears to be suggestive support for the central predictions made by parental investment theory. Finally, we note that parental investment theory has focused primarily on life essentials in the consideration given to parental resources, that is, the ability of parents to provide food to the young (‘‘provisioning’’). When we apply these formulations to human families, it is appropriate to extend the time frame and scope of provisioning to include the presence of skills necessary to acquire such resources. This includes such skills as knowing how to gain access to family-relevant resources, for example, community services, supportive others, and marketable skills. From an evolutionary standpoint, the successful management of these parenting tasks ultimately involves the maximization of the parent’s own reproductive success. Parents would then be expected to show a particularly high investment in their MPD children if they possess the skills and means to provide access to resources needed for the child’s long-term welfare. In the absence of such affordances, they would be expected to show a particularly low investment in their MPD children. Although parents’ overall investment in their children as a function of their access to resources can be predicted from many different theoretical

346


perspectives, parental investment theory provides the most parsimonious account of their selective investment (or failure to invest) in high-risk children. In contrast with the other models we have considered, a parental investment model provides a rationale for the heightened benefits that MPD children may receive as well as the circumstances that lead to abandonment, abuse, and neglect of these same kinds of children. In doing so, it offers ideas as to the ways in which child maltreatment may be prevented and child outcomes enhanced through community programs that foster the resources of parents. E. RESILIENCE AND THRIVING MODELS

An approach that is applicable to our topic but that has been relatively untested empirically is a child resilience model. That is, some children, as a result of their temperament or early experiences may be resistant to some of the more negative consequences they may confront as a result of an MPD. For example, children’s dispositional features may serve to ‘‘protect’’ them against more negative experiences (e.g., Rutter, 2002). Just as children who are characterized by more positive emotionality, sociability, and communication skills are better able to effectively cope with change and adverse life experiences (Fletcher, 1996; Wermer, 1995), they may be expected to more effectively cope with MPDs. At the same time, of course, other childhood temperament features may create a heightened vulnerability to the negative effects of MPDs. Thus, children whose MPDs are accompanied by social anxiety or low sociability (variables that are genetically influenced, e.g., Daniels & Plomin, 1985; Fyer, 1993; Schmitz & Reif, 1994) may show a particular vulnerability to the negative effects of their disorders. Resilience models do not focus exclusively on the buffering effects of the child’s genetic endowment. Children who experience highly stressful MPDs (including associated medical treatment) may under some circumstances acquire enhanced resilience in the face of future stress. The literature on child resilience has included an interest in the kinds of parenting experiences that allow children to cope with early adversity. For example, Masten and her colleagues (e.g., Masten & Coatsworth, 1998; Masten et al., 1999) have been concerned with the buffering effects of parenting style on children’s resilience in the face of severe adversity. When parents are authoritative (warm but firm) and have high expectations, child resilience is enhanced. The developmental neuroscience model has moved one step beyond the concept of resilience and has considered the potential benefits of early adversity. This possibility is consistent with the emerging interest in


347

the concept of ‘‘thriving’’ (e.g., Affleck & Tennen, 1996; Bugental, in press; Carver, 1998; Epel, McEwen, & Ickovics, 1998; Luthar, Cicchetti, & Becker, 2000). Across fields, the notion of thriving suggests the possibility that adverse experience—under some circumstances—may produce later benefits. Parents, as managers of their children’s environment, may provide the kind of experiences that foster opportunities for the child to acquire increased resilience as a function of early experience. For example, parents who provide opportunities for their mobility-impaired children to master the physical environment may also foster resilience. Developmental neuroscientists have suggested some of the ways in which stress immunization may be mediated at the level of the brain. As an example, drawn from early experiences of nonhumans, the combination of early stress (e.g., handling of rat pups) followed by high maternal comfort (e.g., extended licking and grooming) fosters resilience and easy habituation to future stress (e.g., Levine, 1957; Meaney et al., 1985). At a more general level, McEwen and his colleagues (McEwen, 1998, 2000) have suggested that repeated activation of the body’s emergency response systems— followed by recovery—may foster thriving (‘‘allostasis’’). In summary, a resilience model allows consideration of the ways in which children may resist the adverse consequences sometimes associated with MPDs—either as a function of their own temperament or the buffering experiences they have early in life. A thriving model takes us one step further to consider the possibility that very challenging early experiences—if they are followed by recovery—may actually empower the child to become stronger and better able to manage future stress than are children who have not had such experiences.

F. COMPARISON AND INTEGRATION OF MODELS

With the first model we considered (a medical model), the focus was on the kinds of risk that MPD children would face—how they would suffer socially, emotionally, and cognitively as a result of their condition. This model stimulated efforts to produce educational and medical remediational programs for such children. As a cost, it categorized children in a way that fostered stereotyping and stigma. As we moved through other models, we considered ways in which the risk associated with childhood MPDs could be prevented—or at least, alleviated—as a result of the actions of others. If others did not stigmatize or otherwise categorize such children on the basis of their presumed characteristics—and if they did not pose external limits on their access to opportunities—risk for later problems would be much reduced or even

348


prevented. When consideration was given to individual differences in caregiver tendencies to blame others for difficult caregiving experiences, the possibility emerged that interpretive biases might be reduced or prevented—and thus prevent some of the more negative experiences of MPD children. With consideration of a parental investment model, alternative explanations were offered for negative reactions that caregivers may have to MPD children. This model (as noted by Maestripieri & Carroll, 1998) is particularly applicable to parental neglect. Thus, children who appear to be dependent may be seen as posing an exceptionally high caregiving cost. Those with appearance anomalies may be seen as flawed, and thus unlikely to be a good investment. In contrast, children who appear to be unresponsive are less likely than other children to be neglected but are subject to risk for physical abuse—a reaction that is more driven by anger than investment threat. With the application of parental investment theory, a new possibility emerged. Children with MPDs, under the right circumstances, might even receive heightened care from caregivers. Finally, a resilience model opened up consideration of the ways in which children may be buffered against later problems as a function of their early experiences—particularly, when considered in combination with positive temperament features. When the parental investment model is paired with a resilience model, the possibility of ‘‘thriving’’ emerges as a result of MPDs. That is, heightened parental investment combined with early mastery experiences may lead to stress immunization. That is, children with MPDs— under the right set of circumstances—may thrive and show exceptional social-emotional strengths and resistance to future stress—a combination that may foster exceptionally positive life outcomes. Models that focus on parental cognitions and investment decisions provide insights into the social contexts that foster positive versus negative outcomes for children with MPDs. Developmental neuroscience and behavioral genetics add key insights with respect to the variables that mediate and moderate these outcomes. Across all approaches that consider the (negative and positive) contributions that are made by others in the early lives of MPD children, advantages can be seen for shifting caregivers away from a focus on the problems, costs and blame assigned to MPD children—and toward acquisition of positive problem-solving skills and benefits within the caregiving relationship. This leads us to consider specific programmatic ways to produce such changes. In doing so, we will draw primarily from the last three models (models concerned with the interaction of child characteristics and parental responses).


349

III. Optimizing the Outcomes of Children with Medical and Physical Disorders Given past evidence regarding the factors that influence the experiences of children with medical and physical disorders, it is logical to ask how this knowledge may be utilized to optimize the outcomes of such children. As a step in this direction, we conducted a community-based home visitation program to foster more positive family experiences early in the child’s life. The program had two general goals: (1) preventing physical abuse of at-risk (MPD) children during infancy, and (2) fostering the health of at-risk children (MPD) during infancy. In designing the program to achieve the first goal, we borrowed heavily from our own past work on the kinds of parental cognitions that increase the likelihood of harsh or abusive parenting (Bugental et al., 2002). The program made use of an empowerment manipulation in which parents acquired skills in optimizing their outcomes as caregivers. The concept of empowerment (as used here) refers both to: (1) parents’ guided practice in problem-solving activities directed to reaching desired ends, and (2) parents’ perceived ability to reach those ends. In this case, the ‘‘desired end’’ is their effectiveness in facilitating the long-term well-being of their children. On a more exploratory basis, we also tested the extent to which child health was promoted by the empowerment manipulation (Bugental & Ellerson, 2002). In doing so, we reasoned that the parental empowerment manipulation could also be conceptualized as a parental ‘‘resource-holding potential’’ (RHP) manipulation. Whereas the concepts of parental power or resources refer to the parents’ current access to family-relevant resources, the concepts of empowerment and RHP refer to the parents’ ability to gain access to future resources and optimize the long-term welfare of the young. From this framework, we expected that parental investment (and thus, children’s health outcomes) would be influenced by the child’s level of risk (in this case, the child’s medical risk or cost), as moderated by variations in parents’ resource-holding potential. If empowered parents are conceptualized as having increased RHP, they would be expected to invest more in (high-risk) MPD rather than (low-risk) non-MPD children; their MPD children, in turn, would be expected to show increased health benefits. In contrast, parents not participating in the program were expected to show greater investment in (low-risk) non-MPD than in (high-risk) MPD children; MPD children, in turn, were expected to show increased health problems. Due to the diminishing returns associated with increased investment in non-MPD children, no differences were expected to be shown by children without MPDs as a result of their parents’ participation in different conditions.

350


A. WHO WERE THE FAMILIES?

Participants were 96 families recruited during the third trimester of the mother’s pregnancy. Follow-up measures were completed at the end of one year on 72 families. Parents were eligible if they were at moderate risk to become abusive. Risk was based on scores on the Family Stress Checklist (Murphy, Orkow, & Nicola, 1985), a measure that has been validated against future abuse. The average education of mothers was low (7.8 years), and single parenthood (48% of the families) was common. Most (97%) of the mothers were relatively recent immigrants from Mexico. Their average age was 25.5 years. Children were identified as at medical risk based upon records obtained at the hospital where they were born. Thirty-eight per cent of newborn were categorized as at mild or moderate medical risk (for future health problems). Risk status was based upon the presence of mild or moderate prematurity (3 weeks or greater) and/or relatively low Apgar scores (80 ug/dl) have been well documented and can include ataxia, coma, convulsions, severe mental retardation, and sensory impairment (Goyer, 1993; Lockitch, 1993). This level of exposure rarely occurs in childhood. The large majority of children exposed to significant amounts of environmental lead have body lead levels between 10 and 20 ug/dl (Berney, 1996; Lin-Fu, 1992; Schaffer, Szilagyi, & Weitzman, 1994). According to survey data, over 2 million children in the United States may be at risk for lead exposure at or greater than a level of 10 ug/dl (Goyer, 1993; Lockitch, 1993). World wide this figure may only be the tip of the iceberg given that over 85% of the world’s children are found in developing countries, where there is typically less regulation of toxic substances such as lead (Wolf, Jimenez, & Lozoff, 1994). The impact upon children’s development of having a body lead burden in the 10–20 ug/dl range is a critical question. There are a number of reasons why young children should be at particular risk for developmental consequences associated with chronic low level lead exposure resulting in a body lead burden between 10 and 20 ug/dl. Among these are the rapid development of the central nervous system in the first year of life, which means a heightened risk of interference by environmental toxins upon developing as opposed to a developed brain system (Ernhart et al., 1989; Lockitch, 1993). In addition, the immature central nervous system appears to be less able to clear lead that passes through the blood–brain barrier (Anderson, Pueschel, & Linakis, 1996; Castellino & Castellino, 1995). Furthermore, the higher metabolism rate of young children leads to a greater absorption of lead (Goyer, 1993; Winneke, Lilienthal, & Kramer, 1996). In addition to developmental factors, continuing exposure of children to environmental lead remains a problem due to the long half-life of lead in the environment and in the child’s skeletal structure (Bogden, Oleske, & Louria, 1997), as well as the lack of significant reduction of lead exposure following family-based intervention to control residential lead levels (Lanphear, Eberly, & Howard, 2000). In addition, infrahuman research indicates that the earlier the exposure of the organism to lead, the greater the long-term retention of lead (Han et al., 1997). Finally, characteristic behaviors of young children such as mouthing or playing on the ground are more likely to result in a greater intake of ambient lead sources, such as flaking lead-based paint or lead in soil (Ernhart et al., 1989).

378

Theodore D. Wachs

What does the research evidence tell us about the impact of chronic lowlevel lead exposure upon children’s development? A number of reviewers have summarized evidence indicating statistically significant but small negative relations between indices of cognitive function and body lead levels in the 10–25 ug/dl range (Bellinger, 1995; Dietrich, 1995; Goyer, 1993; Needleman, 1993; Needleman & Gatsonis, 1990). For example, Winneke, Lilienthal, and Kramer (1996) conclude that an increase in body lead burden from 10 to 20 ug/dl is linked to a loss of between 1 and 3 IQ points. At the same time, many of these same reviewers also report notable inconsistencies across studies, with not all studies finding significant relations between body lead burden and cognitive outcomes (Goyer, 1993). Furthermore, there remains the question of whether relations between lead and developmental outcomes are statistically meaningful after controlling for covariates such as family social class (Bellinger, 1995; Dietrich, 1995; Phelps, 1999). Fewer studies have focused on the relation between lead exposure in childhood and noncognitive behavioral development. Again, the results are not totally consistent. Infants and children with higher body lead burdens have been reported as being more distractible (Burnette et al., 1999; Leviton et al., 1993; Needleman et al., 1979), more irritable (Miller, Massaro, & Massaro, 1990), more withdrawn or more aggressive (Wasserman et al., 1998, 2001), less socially responsive (Bellinger et al., 1984), and less active (Padich, Dietrich, & Pearson, 1985). These differences, although statistically significant, again involve relatively small effect sizes (1–4% unique variance). Furthermore, other studies have reported no relation between lead exposure and children’s mood (Bellinger et al., 1994; Dietrich et al., 1987) or psychiatric diagnoses (Gillberg et al., 1982). The overall pattern of results indicates that although body lead levels in the 10–25 ug/dl range are sometimes related to poorer cognitive and behavioral performance in childhood, the evidence is not always consistent, and when significant effects are found they tend to be relatively modest. What does this pattern of findings tell us about the implications of exposure to a common bio-ecological environmental toxin for children’s development? The traditional explanatory framework used to explain the impact of lead upon development is based on a main effect model, wherein it was assumed that lead burden ! CNS deficit ! impaired development (Miller, Massaro, & Massaro, 1990; Pearson & Dietrich, 1985). Within this framework a critical assumption is that all individuals should react similarly to the same level of lead exposure (Bellinger, 1995), and that a single dose–response relation should link lead exposure to indices of

The Bio-ecological Environment and Development

379

development (Pearson & Dietrich, 1985). However, the validity of this assumption of equivalent reactivity has been questioned by a number of reviewers, given evidence from infrahuman studies on the context-dependent nature of consequences of lead exposure plus the number of potential moderators of lead exposure effects in human populations (Bellinger, 1995, 2001; Ruff, 1999; Wasserman, 1995; Winneke, 1995). Such questioning of the traditional view is quite consistent with the conceptual framework of this chapter, namely that the impact of bio-ecological influences can only be understood when such influences are viewed as operating within a complex system of linked multiple influences upon development. This means looking at what other bio-ecological and psychosocial influences covary and interact with lead exposure.4 A number of bio-ecological and psychosocial factors have been considered as possible moderators of the impact of lead exposure upon development. One possible moderator is gender. Although some investigators report a greater impact of lead exposure on males (Thompson et al., 1989), others report either no gender differences or greater reactivity for females (McMichael et al., 1992), or differential effects for males versus females depending upon the outcome measure utilized (Leviton et al., 1993). A second potential moderator is age. Whereas some studies have suggested that different functions will be affected, depending upon the age at which the child is exposed to lead (Shaheen, 1984; Wasserman et al., 1994), the overall pattern of results does not offer strong evidence for a sensitive period for postnatal lead exposure effects (Dietrich, 1995). A third potential moderator is nutritional status. Infrahuman and human data indicate that deficits in intake of calcium (Fullmer, 1992; Johnson, 2001; Mahaffey et al., 1982; Miller, Massaro, & Massaro, 1990), iron (Hammad, Sexton, & Langenberg, 1996; Lynch, 1997; Mahaffey, 1990; Miller, Massaro, & Massaro, 1990; Yip, 1989) and carbohydrates act to increase the absorption of lead (Lucas, Sexton, & Langenberg, 1996). This pattern of results would suggest that lead-exposed children with nutritionally deficient diets should be most at risk for developmental deficits. Unfortunately, the limited evidence on this hypothesis is not totally consistent. Although three studies have reported that adequacy of children’s dietary intake can act to moderate the cognitive consequences of lead exposure (Gardner et al., 1998; Lester, Horst, & Thatcher, 1988; 4 Inconsistencies in the lead-behavior literature may also reflect the operation of specificity processes, with results varying depending on the outcome measure used. For example, Wasserman et al. (2000) have reported that although lead-exposed children show deficits in measures of fine motor skills and visual motor performance, their gross motor performance appears to be unaffected.

380

Theodore D. Wachs

Ruff et al., 1996), additive but not interactive relations between lead, iron, and children’s cognitive performance have been reported in two other studies (Wasserman et al., 1992, 1994), with neither additive nor interactive effects being found in a sixth study (Wolf, Jimenez, & Lozoff, 1994). In contrast to the inconsistent findings concerning gender, age, or nutritional status as moderators of lead exposure, there is a consistent body of evidence in regard to both developmentally important covariates of lead exposure and at least one moderating mechanism. Children who are exposed to lead are also more likely to be exposed to other toxic heavy metals like cobalt and mercury (Lewis et al., 1992). Also a consistent body of evidence documents that children with higher levels of lead exposure are more likely to have greater exposure to other psychosocial risk factors including child abuse, minority group, or poverty status, are more likely to have mothers who are of lower intelligence or are less involved with their children, and are more likely to have pervasive developmental disorder (Bithoney, Vandeven, & Ryan, 1993; Dietrich et al., 1987; Milar et al., 1980; Schroeder & Hawk, 1987; Shannon & Graef, 1996; Vorhees & Mollnow, 1987). Linking together both covariates and moderators, Bellinger (1995) hypothesized that children with a history of, or greater exposure to, biological or psychosocial risk conditions will be more sensitive to low level lead exposure than nonrisk children. Supporting Bellinger’s (1995) hypothesis studies have shown accentuation of the effects of low level lead exposure in children with a history of jaundice during the neonatal period (Damm et al., 1993; Lyngbye, Hansen, & Grandjean, 1991), children from low SES families (Bellinger et al., 1991) or children whose parents have less than a high school education (Rabinowitz, Wang, & Soong, 1991). This pattern of results is congruent with the second thesis of this chapter in indicating that the impact of chronic lead exposure upon children’s development is best understood with reference to processes involving organism–environment covariance (lead exposure in childhood covaries with exposure to other biosocial risk conditions) and organism–environment interaction (children with greater exposure to bio-social risk conditions will be more sensitive to lowlevel lead exposure than children with less exposure to such conditions). Within this conceptual framework the impact of lead exposure on young children will systematically vary, depending on the extent of covariance between lead exposure and exposure to other bio-ecological or psychosocial risk factors. G. NUTRITIONAL STATUS

Malnutrition, chronic undernutrition, and micronutrient (trace mineral and vitamin) deficiencies represent some of the most common risk factors to


381

which young children are exposed (UNICEF, 1998). Approximately 30% of the world’s children are moderately to severely malnourished (defined as weight for height 2 standard deviations below the mean—Grantham-McGregor, Fernald, & Sethuraman, 1999a), and approximately 60% of the world’s children are iron deficient (Allen & Casterline-Sabel, 2001). These figures do not take into account the even higher numbers of children who, although not clinically malnourished, are chronically undernourished, with their intake of energy or specific micronutrients consistently below recommended allowances (Simeon & Grantham-McGregor, 1990; Stoltzfus, 2001). Structurally, characteristics of the bio-ecological dimension of nutrition fit within the overall environmental taxonomy described by Bronfenbrenner. Quality or quantity of nutritional intake can be viewed as a proximal microsystem influence, given that not all children living in a given geographic area have equivalent levels of nutrition, even after controlling for availability of nutrients or equating for family economic resources (Wachs, 2002). Nutrition also can be viewed as a mesosystem influence, given that links between inadequate nutrition and susceptibility to gastrointestinal illness can interact in a synergistic negative cycle (Keusch, 1990). In addition, nutrition may function at the exosystem level as well, when poorly nourished parents are less able to provide appropriate caregiving for their children (Grantham-McGregor, 1984). A representative summary of findings on the relations between different types of nutritional deficiencies and children’s cognitive performance is shown in Table I. A parallel table relating nutritional deficits to children’s social–emotional outcomes is shown in Table II. The overwhelming majority of findings summarized in Tables I and II are based on studies on children in developing countries, but even in developed countries inadequate nutrition (e.g., iron deficiency anemia) has been related to an increased risk of mental retardation (Hurtado, Claussen, & Scott, 1999) and to deficits in school achievement (Palti, Meijer, & Adler, 1985). 1. Prenatal Malnutrition Intrauterine growth retardation (IUGR) has long-term biomedical consequences (Barker et al., 1993; James, 1997) and has also been related to postnatal physical growth rates, physical performance (Martorell et al., 1998) and to early central nervous system development (Rao & Georgieff, 2000). However, as shown in Table I, evidence is equivocal on the cognitive consequences of prenatal malnutrition. There are several possible reasons why the evidence is equivocal. First, there could be a moderating influence of postnatal nutrition and psychosocial influences on the initial consequences of inadequate prenatal nutrition (Grantham-McGregor, Fernald, & Sethuraman, 1999a). Second, intrauterine growth retardation can be caused

382

Theodore D. Wachs TABLE I

Summary of Findings Relating Nutritional Deficiencies to Child Cognitive Performance Nutrient

Summary of findings

References

Prenatal period Intrauterine growth retardation Iodine deficiency

Findings inconsistent

Grantham-McGregor, Fernald, and Sethuraman (1999a), Stein et al. (1975) Increased risk of cognitive Grantham-McGregor, Fernald, deficits including mental retardation and Sethuraman (1999b) Lower levels of both general and specific cognitive performance as well as an increased risk of poor school achievement

Grantham-McGregor (1995) Grantham-McGregor, Fernald, and Sethuraman (1999a)

Iron deficiency anemia Zinc deficiency

Lower levels of both general and specific cognitive performance Results for cognitive function inconsistent

Lozoff (1998), Watkins and Pollitt (1998) Black (1998)

Chronic undernutrition

Increased risk of less adequate cognitive and academic performance

Gorman (1995), Sigman, Espinosa, and Whaley (1998), Wachs (1995)

Increased intake of essential fatty acids in infancy

Evidence inconsistent with regard to Willatts et al. (1998), Wroble et al. the beneficial effects of fatty acid (2002) supplementation for enhancing visual acuity and early cognitive performance

Chronic mild micronutrient deficits

Greater intake of animal source iron related to better cognitive performance in undernourished infants and school age children Lower cognitive performance in iron-deficient, nonanemic children Slower activity in the EEG power spectrum of iron-deficient, nonanemic children Intake of B vitamins related to higher levels of infant alertness and toddler symbolic play

Moderate–severe postnatal malnutrition Severe micronutrient deficiencies

Sigman Espinosa, and Whaley (1998), Wachs (1995)

Bruner et al. (1996) Halterman et al. (2001) Otero et al. (1999)

Wachs (1995)

by a variety of influences in addition to malnutrition (Kleinman, 1992). Third, there is the possibility of ‘‘sleeper effects’’ given some evidence suggesting that cognitive consequences associated with intrauterine growth retardation are more likely to appear in the middle childhood period than


383

TABLE II Summary of Findings Relating Nutritional Deficiencies to Children’s Social–Emotional Function Nutrient

Summary of findings

References

Prenatal nutrition

Prenatal zinc supplementation Merialdi et al. (1998) related to higher fetal activity level Wachs et al. (2003) More negative affect and lower attention in neonates with iron deficiency anemia or a deficiency in serum iron levels at birth

Moderate–severe postnatal malnutrition

Behavior characterized by apathy, irritability and lower activity level Reduced attention to the environment; fewer interactions with objects in their environment; reductions in the variety of exploratory behaviors utilized as well as increased levels of negative affect and wariness Lower attention and lower exploration in the school years by children who were moderately to severely malnourished in infancy

Geber and Dean (1956), Grantham-McGregor (1984) Chavez and Martinez (1984), Lester (1975), Meeks-Gardner et al. (1999)

Reduced activity level, lower reactivity, higher inhibition, wariness, and negative affect Zinc supplementation related to higher activity levels in zinc-deficient children Reductions in the amount, duration, and variability of play behavior in infancy Lower activity, reactivity, and less adequate emotional reactivity in infancy and lower activity and sociability in school-age children

Lozoff and Wachs (2001)

Barrett and Radke-Yarrow (1985), Richardson et al. (1972)

Severe micronutrient deficits Iron deficiency anemia Zinc deficiency

Chronic undernutrition

Black (1998)

Sigman et al. (1989a), Wachs et al. (1993), Walka et al. (2000) Pollitt et al. (2000), Sigman et al. (1989b)

Chronic low level micronutrient deficiencies B vitamin deficiency

Wachs (2000) Reduced reactivity, poorer state control, and reduced activity level in infancy and reduced activity in school-age children

384

Theodore D. Wachs

earlier in development (Grantham-McGregor, Fernald, & Sethuraman, 1999a). Finally, specific prenatal micronutrient deficits rather than general prenatal protein energy deficits may well be more critical for later behavioral development. As seen in Table I, long-term cognitive deficits can be found for individuals with prenatal iodine deficiency (Grantham-McGregor, Fernald, & Sethuraman, 1999b). As shown in Table II, prenatal activity level has been linked to prenatal zinc supplementation (Merialdi et al., 1998) and temperament in newborns has been related to indices of perinatal iron deficiency (Wachs et al., 2003). 2. Moderate–Severe Postnatal Malnutrition As shown in Table I, one consequence of moderate–severe postnatal malnutrition is an increased risk of cognitive impairment. Cognitive deficits linked to moderate–severe malnutrition are shown both for measures of general intelligence (Grantham-McGregor, Fernald, & Sethuraman, 1999a) and for more specific information processing measures (Rose, 1994). Although the cognitive deficits associated with early moderate–severe malnutrition can persist into adulthood (Galler & Barrett, 2001), these deficits are diminished when nutritional interventions occur following early malnutrition (Grantham-McGregor, Fernald, & Sethuraman, 1999a). For example, a long-term study carried out in Guatemala showed small but significant differences in adult cognitive performance favoring malnourished children who were nutritionally supplemented during the first several years of life, as compared with malnourished children who received a nutritional placebo (Pollitt et al., 1993). There also are characteristic behavioral patterns that result from moderate–severe malnutrition (see Table II). Many of these symptoms vanish as the child is nutritionally rehabilitated (Grantham-McGregor, Stuart, & Powell, 1991). However, long-term follow-up studies of nutritionally rehabilitated malnourished children indicate greater distractibility and lower reactivity, emotional control and activity level than is seen in matched nonmalnourished controls (Grantham-McGregor, 1995; Simeon & Grantham-McGregor, 1990). 3. Severe Mineral and Vitamin Deficits As also seen in Tables I and II, iron deficiency anemia, the most severe form of nutritional iron deficiency, is linked to lower cognitive function and specific patterns of behavioral deficits. The cognitive and behavioral consequences of iron deficiency anemia early in life may not be completely remediable, even when iron supplementation is given (GranthamMcGregor, Fernald, & Sethuraman, 1999b; Lozoff & Wachs, 2001). Outside of iron deficiency anemia, far less evidence is available at the human level on the consequences of severe postnatal deficits in other


385

micro-nutrients. Limited evidence has suggested that low maternal vitamin B6 levels may affect neonatal reactivity (McCullough et al., 1990). Though results have been inconsistent with regard to relations between zinc deficiency and cognitive performance, as shown in Table II, some studies have related zinc deficiency to reduced activity level (Black, 1998). 4. Chronic Undernutrition There has been a shift toward defining chronic undernutrition based on indices of dietary quality, rather than on indices of dietary quantity (e.g., energy intake, Wachs, 1995). A poor quality diet is low in animal source foods, which means a diet that is lower in critical micronutrients or one in which critical micronutrients are less bio-available to the individual (Allen, 1993). As shown in Tables I and II, poor quality diets have been linked to cognitive, academic, and behavioral deficits in childhood. In terms of the consequences of poor dietary quality, one re-emerging issue is whether iron deficiency that has not yet reached the stage of iron deficiency anemia can influence children’s cognitive performance. For many years the generally accepted conclusion was that iron deficiency without anemia was not associated with cognitive deficit (Watkins & Pollitt, 1998). However, there are a number of reasons which suggest the need to re-evaluate this conclusion. First, the evidence is not totally consistent. Non-anemic rat pups born to anemic mothers showed poorer early neurodevelopmental performance than pups born to nonanemic mothers (Felt & Lozoff, 1996). At the human level some investigators have reported deficits in cognitive and neural functioning in children who are iron deficient but not anemic (see Table I). In addition, studies on the cognitive impact of iron deficiency without anemia have generally used global measures of cognitive functioning. Such measures are less likely to be sensitive to effects of iron deficiency that are specific to particular cognitive processes (Wauben & Wainright, 1999). Supporting this hypothesis, Stoltzfus et al. (2001) found that the effects of iron supplementation upon language is found across the full range of iron deficiency, whereas the impact of such supplementation for motor development occurs only for infants with iron deficiency anemia. 5. Going Beyond Nutrition and the Brain The level and quality of dietary intake influences the structural development of the central nervous system (Levitsky & Strupp, 1995; Rao & Georgieff, 2000; Tacconi, Calzi, & Salmona, 1997) as well as the efficiency of neurotransmitter metabolism (Christensen, 1996; Guilarte, 1993). For example, early iron deficiency anemia may have relatively permanent effects on brain myelination and on the number of D2 dopamine receptors in the prefrontal cortex (Lozoff, 1998), and there is increasing evidence delineating

386

Theodore D. Wachs

the role played by essential fatty acids for both CNS structural development (e.g., the occipital cortex) and neurotransmitter function (e.g., number of dopamine receptors) (Wauben & Wainright, 1999; Yehuda et al., 1998). What is described by this evidence is a process of nutrition ! CNS structure and function ! behavior and development: However, inspection of the research literature also indicates that the nature and impact of nutritional deficiencies upon behavioral–developmental variability are also subject to the same underlying processes which we have previously documented with regard to other bio-ecological influences. Consistent with what has been shown for other dimensions of the bioecological environment, nutritional influences also operate across time, covary and interact with other psychosocial and bio-ecological influences. Studies supporting this conclusion are shown in Table III. As shown in

TABLE III Process Mechanisms Illustrating How Specific Nutritional Influences are Embedded in a Network of Multiple Influences Process mechanism

Linkage pattern

Organism–Environment covariance Passive covariance

Reactive covariance

Malnourished or chronically undernourished children are more likely to be exposed to inadequate housing, developmentally inhibiting parental rearing styles, inadequate schooling, and biological risk factors such as low birth weight, and a compromised immune system (Grantham-McGregor & Ani, 2001; Keusch, 1990; Pollitt, 1988; Ricciuti, 1993). A variety of disturbances in parent–child relations in undernourished populations have been reported, including lower parental sensitivity (Valenzuela, 1997), reduced positive affect by parents (Lozoff et al., 1998), greater attempts to retain close physical contact when their infants moved away from them (Lozoff, Klein, & Prabucki, 1986), and lower quality vocalization patterns by parents toward their infants (Meeks-Gardner et al., 1999). There are systematic differences in patterns of parent–child interaction between nutritionally supplemented and non-nutritionally supplemented infants (Chavez, Martinez, & Yaschine, 1971). Evidence from both developed (Hagekull, Bohlin, & Rydell, 1997; Vandiver, 1997) and developing countries (DeVries, 1984; ScheperHughes, 1987) indicates systematic differences in parent feeding patterns as a function of variability in infant temperament. continued


387

TABLE III Continued Process mechanism Psychosocial environment bio-ecological environment covariance

Linkage pattern Demands of maternal work environment can impact upon who in the family will be the primary child caregiver; primary caregiver characteristics in turn relate to the adequacy of child nutrition (Engle, 1991).

Temporal influences Long-term sensitization

Children with a history of chronic undernutrition show poorer cognitive performance after an experimentally induced brief fasting period (e.g., missing breakfast) than do children with more adequate nutritional histories (Grantham-McGregor, Chang, & Walker, 1998; Pollitt, Cueto, & Jacoby, 1998).

Interaction Psychosocial moderators of the impact of nutritional deficiencies

Individual child moderators

The detrimental influences of early malnutrition can be attenuated when previously malnourished children are subsequently reared in more adequate psychosocial environments (Colombo, de la Parra, & Lopez, 1992; Paine et al., 1992; Winick, Meyer, & Harris, 1975). In orderly school environments breakfast feeding facilitated children’s on-task behavior in the classroom, whereas in chaotic school environments supplementary breakfast feeding led to reduced on-task classroom behavior (Grantham-McGregor, Chang, & Walker, 1998). Nutritionally supplemented children showed either higher pro-social or anti-social behavior patterns depending upon the harshness of ecological conditions in their villages (Barrett, Radke-Yarrow, & Klein, 1982). Reactions by teachers to children with differing degree of undernutrition varied as a function of contextually based expectancies about the types of behavioral patterns that are appropriate for male versus female school children (Wachs et al., 1995). Evidence for gender differences in reaction to nutritional supplementation is found in some studies (Sazawal et al., 1996; Walka et al., 2000).

Table III, early nutritional deficits may have temporal consequences through sensitizing the individual to later nutritional deficits (GranthamMcGregor, Chang, & Walker, 1998; Pollitt, Cueto, & Jacoby, 1998). As also shown in Table III, there is ample evidence for covariance between nutritional deficiencies and the child’s encountering other bioecological or psychosocial risk conditions. Particularly interesting for developmental researchers is the covariance between deficiencies in child nutritional intake and patterns of transactions with their parents (e.g., Lozoff et al., 1998; Valenzuela, 1997). This covariance may be passive

388

Theodore D. Wachs

in nature, reflecting the possibility that inadequately nourished children are more likely to have inadequately nourished parents who have less energy available to allow them to stimulate their children appropriately (Grantham-McGregor, 1984). In addition, mothers with poor vitamin B6 status are less responsive to their newborn infants’ vocalizations and are less effective at soothing when their infant is distressed (McCullough et al., 1990). However, there is also the possibility of reactive covariance, such that the signals displayed by undernourished or micronutrient deficient infants (e.g., increased proximity seeking or fewer displays of positive affect) may influence the nature of caregiver–child transactions (Pollitt, 2000). The research depicted in Table III reveals a variety of differences in the interaction patterns shown by parents of adequately versus inadequately nourished infants. Such differences may be the result of inadequate parental nutrition, but in the few studies that have contrasted the relative salience of parents’ and children’s diets, the results suggest the operation of reactive rather than passive infant–environment covariance. Specifically, in a crosscountry study of inadequately nourished infants and their mothers in Egypt and Kenya, relations between the quality of child nutritional intake and parenting behaviors remained significant even after statistically controlling for maternal nutrition (Wachs et al., 1992). Similarly, Lozoff has reported significant differences in maternal behavior patterns toward anemic versus nonanemic infants, even though virtually all of the mothers studied (97%) were nonanemic (Lozoff et al., 1998). Finally, the impact of inadequate nutritional intake can be accentuated or attenuated by the occurrence of other developmental influences (see Table III). Much of the available research has involved the moderating influence of the child’s psychosocial context (Colombo, de la Parra, & Lopez, 1992; Paine et al., 1992; Winick, Meyer, & Harris, 1975). Going beyond just accentuation or attenuation, contextual characteristics can dramatically alter the patterns of behavior exhibited by nutritionally supplemented children (Barrett, Radke-Yarrow, & Klein, 1982; Grantham-McGregor, Chang, & Walker, 1998) or the reactions of others to the behavior patterns of undernourished children (Wachs et al., 1995). Better nutrition will apparently energize children’s behavior, but the nature of this energizing depends on the specific characteristics of the child’s psychosocial context.

IV. Bringing it All Together: A Converging Structural and Conceptual Framework Evidence I have presented in this chapter documents the relevance of the bio-ecological environment for children’s development and shows the


389

similarity between structural characteristics of the psychosocial environment (Bronfenbrenner & Crouter, 1983) and the bio-ecological environment. In addition, I have illustrated how the same process mechanisms that link the psychosocial environment to development also operate within the bio-ecological environment. One major implication of these findings is that understanding individual variability in children’s behavioral development will require assessment and integration of influences from both the psychosocial and the bio-ecological environments. Such assessment and integration is more likely to occur in the context of collaborative research between biologically and psychosocially oriented developmental scientists. However, collaborative research efforts have been hampered by the lack of a common conceptual framework linking together bio-ecological and psychosocial influences. In the final section of this chapter I first present a structural framework encompassing the dimensions of both the bioecological and the psychosocial environments. I then present a processoriented approach that integrates the contributions of dimensions of both the bio-ecological and psychosocial environments within a single integrated framework. A. AN INTEGRATED STRUCTURAL FRAMEWORK

The structural framework shown in Figure 1 is derived from Bronfenbrenner’s environmental model, but expands his conceptualization to include dimensions from the bio-ecological environment as well as potential links between the bio-ecological and psychosocial environments. As can be seen in Figure 1, the psychosocial macrosystem is nested under the bio-ecological macrosystem. The rationale for this is based on anthropological evidence noted earlier, referring to how physical ecological features like climate or altitude can structure the nature of cultures living in specific ecologies (Witkin & Berry, 1975). Thereafter, the model shown in Figure 1 illustrates parallel dimensions between the bio-ecological and psychosocial environments, with bidirectional links up and down the hierarchy and across parallel dimensions.5 The darker directional arrows shown in Figure 1 reflect the asymmetrical directionality inherent in hierarchical structures, with top down influences likely to be stronger than 5

There is a third, parallel environmental universe that could be added to this figure, namely an individual’s subjective perceptions of the characteristics of his or her environment. Analysis of the individual’s subjective environment reveals a hierarchical structure similar to that seen for the objective psychosocial environment (Wachs, 1999). For simplicity’s sake I am not including the subjective environmental structure as part of Figure 1, but this aspect of the environment clearly is important when considering the nature of environmental influences on development.

390

Theodore D. Wachs

Fig. 1. Integrated model of the bio-ecological and psychosocial environments. From: T.D. Wachs Nutritional Deficiences as a biological context for development. Chapter in W. Hartup & R. Silbereisen (Eds.): Growing Points in Developmental Science (pp. 74). Copyright 2002 by Psychology Press. Reprinted by permission.

bottom up influences (Wachs, 2000). Parallels between the structural dimensions of the psychosocial environment (microsystem, mesosystem, exosystem, and macrosystem) and the structural dimensions of the bioecological environment have been delineated earlier in this chapter. Empirical support for the links between the different levels of the psychosocial environment have been documented in earlier reviews (Wachs, 1992, 2000). Links between the different levels of the bio-ecological environment have been described previously in this chapter (e.g., accumulation of proximal bio-ecological risk conditions at high altitudes, the nesting of food availability or viral/bacterial exposure under climate). Relations between the different dimensions of the bio-ecological and psychosocial environments have been documented in multiple places in this chapter as well (e.g., cultural structuring or poverty and children’s exposure to environmental hazards; impact of maternal parasitic infection upon maternal caregiving—including providing adequate nutrition for her


391

child; findings concerning the influence of child nutrition upon patterns of parent–child relations). These findings underscore the importance of links between the bio-ecological and psychosocial environments. The implications for developmental scientists of the expanded environmental structure proposed here, integrating the bio-ecological and psychosocial environments, are similar to the implications drawn by Bronfenbrenner when he proposed his expanded psychosocial environmental framework. Within the framework of Bronfenbrenner’s environmental model one can only understand the contributions of any level of the psychosocial environment structure by integrating the characteristics and contributions of all other levels (Bronfenbrenner & Crouter, 1983). In this sense the psychosocial environment functions as an integrated system. In the expanded view of the environment proposed here, we can only understand the contributions of any level of the bio-ecological environment by understanding the characteristics and contributions of other levels of the bio-ecological environment. Even more critically, within the expanded environmental framework I have proposed, we can only understand the contributions of the psychosocial environment to children’s development by also understanding the characteristics and contributions of the bioecological environment within which the child develops. The reverse also holds; we can only understand the contributions of the bio-ecological environment to developmental variability by also understanding the characteristics of the psychosocial environment within which children develop. Viewed in this way, the bio-ecological and psychosocial environments function together as an integrated system. B. AN INTEGRATED PROCESS FRAMEWORK: FUNCTIONAL ISOLATION

Figure 1 describes the structure of the environment, integrating both the bio-ecological and psychosocial domains. I now present a specific process example, functional isolation, illustrating how the integrated contributions from both domains translate into variability in children’s development. The concept of functional isolation arose from observational studies reporting higher levels of apathy and lower levels of activity in severely malnourished young children (Geber & Dean, 1956; Grantham-McGregor, 1984), as well as from infrahuman research documenting changes in exploration and learning patterns in malnourished animals (Levitsky & Strupp, 1984; Strupp & Levitsky, 1995). Based on these findings, Levitsky and Barnes (1972) hypothesized that the developmental impact of malnutrition involves not only direct influences of malnutrition upon the developing central nervous system, but also the indirect influences of the individual’s reduced

392

Theodore D. Wachs

Fig. 2. An expanded conceptual framework for functional isolation processes.

exploration and involvement with the environment. Levitsky and Barnes described these indirect influences upon development using the term functional isolation. As originally conceived, the functional isolation hypothesis offered a means of going beyond purely biological explanations of nutritionally driven developmental deficits in young children. In an expanded conceptualization I have added two additional pathways to Levitsky and Barnes’ original functional isolation model (Wachs, 2002). As shown in Figure 2, the first replaces the original unidirectional pathway between parent–child transactions and children’s nutrition with a bidirectional pathway. The second adds a new bidirectional path between children’s involvement with the environment and central nervous system development and function.6 6 The functional isolation framework shown in Figure 2 could be further expanded by adding an intervening child characteristics term in the path between nutritional deficits and parent– child transactions, with an arrow from CNS to this additional term. However, given the sketchy state of our knowledge on the role played by nutrition upon individual variability in child characteristics such as temperament that have the potential to influence parent–child transactions, I have not formalized this additional term in the current functional isolation framework (Wachs, 2000).


393

Although originally formulated to explain the developmental consequences of moderate–severe protein calorie malnutrition, converging evidence has led to an elaboration of the range of functional isolation to also include the developmental impact of trace mineral deficits, such as iron deficiency anemia (Lozoff, 1998), and zinc deficiency (Black, 1998). The expanded functional isolation framework presented here may also be useful as a means of conceptualizing the impact of other dimensions of the bioecological environment including exposure to infectious agents, parasites and environmental toxins. Within the functional isolation framework shown in Figure 2 developmental deficits are considered to be the result of both bio-ecological and psychosocial mechanisms. In this sense the functional isolation framework is a process that results from the integrated bio-ecological and psychosocial environmental structure described in Figure 1. Biologically, nutritional deficiencies are hypothesized to lead to an altered CNS, which in turn is associated with deficits in specific aspects of development, depending upon which CNS functions are most affected. Psychosocially, nutritional deficiencies are hypothesized to be associated with reduced environmental involvement by the child, as well as with developmentally inappropriate caregiver–child transactions (e.g., treating children as if they were younger than they are). Reduced environmental involvement by the child and developmentally inappropriate caregiver–child transactions feed back onto child nutrition and CNS development and function in a negative synergistic loop. Reduced environmental involvement and developmentally inappropriate parent–child transactions are also hypothesized to produce reduced developmental competence as well. Empirical support exists for each pathway shown in Figure 2. Evidence relating nutritional deficits to altered CNS structure and function, to children’s involvement with the environment, and to behavior of undernourished children and subsequent child–caregiver transactions has been summarized previously, as has evidence on the neural consequences of differences in environmental rearing conditions. Evidence also is available illustrating how level of CNS maturation and neurotransmitter action influence child– environment transactions (Diamond, 1990; Lozoff, 1998; Nelson, 1995). Indirect evidence supporting the validity of the pathway from parenting ! child nutrition comes from studies relating adequacy of children’s nutritional intake to parent or caregiver characteristics such as age (Engle, 1991) and parental educational or intellectual level (Sandiford et al., 1997; Wachs & McCabe, 2001). More direct evidence comes from cross-cultural (Engle et al., 1996) and anthropological studies (Scheper-Hughes, 1987) documenting how culturally driven differences in parental value systems about feeding practices and desirable child characteristics can act to influence

394

Theodore D. Wachs

the child’s level of nutritional intake. Studies of young children with nonorganic failure to thrive implicate low maternal sensitivity, responsivity, and nurturance as etiological factors related to this nutritionally based developmental disorder (Black & Krishnakumar, 1999; Hagekull, Bohlin, & Rydell, 1997; Wolke, Skuse, & Mathiesen, 1990). Finally, a number of investigators have related the use of passive versus active feeding patterns by caregivers to differential feeding patterns by young children, leading to either childhood undernutrition or obesity (Bentley et al., 2000; Engle & Lhotska, 1999). Ultimately, each of the final pathways in the functional isolation framework seen in Figure 2 has been shown to be a necessary influence for variability in children’s behavior and development (Wachs, 2000). Although all the individual paths of the functional isolation framework have been validated, the validity of the overall framework has not been adequately tested. A satisfactory test of the overall framework will require collaborative multidisciplinary studies involving nutrition researchers, environmentally oriented developmental psychologists and neuroscientists. Borrowing a theme from the movie Field of Dreams—‘‘Build it and they will come’’—the existence of an integrated framework for multiple influences may serve to promote such sorely needed collaborative efforts.

V. Conclusions As noted at the outset, two primary conclusions can be drawn from the present chapter. The first is the importance of the bio-ecological environment for understanding variability in children’s behavior and development. The second is that similar environmental structure and process mechanisms mediate the influence of the psychosocial and bioecological environments on children’s behavior and development. These conclusions have both practical applications, as well as important implications for future developmental research and developmental theory. In regard to practical applications, one implication is the importance of multidomain interventions with at-risk children. This statement is supported by the results from intervention studies, showing how both stronger and longer lasting developmental gains occur when nutritional supplementation is combined with psychosocial stimulation in populations of malnourished children (Powell, Walker, & Grantham-McGregor, 1998), as well as evidence indicating stronger recovery of cognitive function when treatment for parasitic infection is combined with treatment for iron deficiency anemia (Boivin & Giordani, 1993). In regard to research implications, the fact that individual biological, psychological, and cultural influences taken in isolation have proven to be


395

‘‘necessary but not sufficient’’ explanations of variability in individual development emphasizes the necessity for collaborative multidisciplinary research between biological, psychological, and anthropological scientists. Such collaboration, although often an ideal, is difficult to achieve in practice due to different conceptual frameworks and different research models used by developmental scientists from different disciplines. By documenting common structures and processes underlying the operation of bio-ecological and psychosocial influences, and by providing a conceptual model built around multidomain contributions to developmental variability, I have attempted to offer a common conceptual and research framework for interdisciplinary collaboration. In regard to theory, the vast majority of developmental theories are founded upon studies of middle-class children living in western developed countries (Schoepflin & Muller-Brettel, 1990; Super & Harkness, 1999). However, the vast majority of the world’s children live in non-western developing countries. Although many of the bio-ecological environmental influences discussed here do operate in developed countries, they are far more likely to be encountered by children living in non-western contexts. To what extent will our developmental theories, which rarely consider the influence of the bio-ecological environment, generalize to the majority of the world’s children for whom bio-ecological risk factors are a constant presence? As stated by Kagitcibasi (1996, p. 7): ‘‘Any psychological theory claiming universality, as they all do, must be demonstrated to hold crossculturally.’’ Based upon what has been presented here, I would take Kagitcibasi’s challenge one step further and argue that any psychological theory attempting to explain individual developmental variability must not only be tested cross-culturally, but must also encompass the full range of the child’s context, which includes both the psychosocial and the bio-ecological environments.

REFERENCES Allen, L. (1993). The nutrition CRSP: What is marginal malnutrition and does it affect human function. Nutrition Reviews, 51, 255–267. Allen, L., & Casterline-Sabel, J. (2001). Prevalence and causes of nutritional anemias. In U. Ramakrishnan (Ed.), Nutritional anemias (pp. 7–22). Boca Raton: CRC Press. Anderson, A., Pueschel, S., & Linakis, J. (1996). Pathophysiology of lead poisoning. In S. Pueschel, J. Linakis, & A. Anderson (Eds.), Lead poisoning in childhood (pp. 75–96). Baltimore: Brookes. Anderson, V., Bond, L., Catroppa, C., Grimwood, K., Keir, E., & Nolan, T. (1997). Child bacterial meningitis. Journal of the International Neuropsychological Society, 3, 147–158.

396

Theodore D. Wachs

Atlas, J. (1989). Birth seasonality in developmentally disabled children. Psychological Reports, 64, 1213–1214. Barker, D., Gluckman, P., Godfrey, K., Harding, J., Owens, J., & Robinson, J. (1993). Fetal nutrition and cardiovascular disease in adult life. The Lancet, 341, 938–941. Barr, R., Boyce, W., & Zeltzer, L. (1994). The stress-illness association in children. In R. Haggerty, L. Sherrod, N. Garmezy, & M. Rutter (Eds.), Stress, risk and resilience in children and adolescents (pp. 182–224). Cambridge: Cambridge University Press. Barrett, D., & Radke-Yarrow, M. (1985). Effect of nutritional supplementation on children’s responses to novel frustrating and competitive situations. American Journal of Clinical Nutrition, 42, 102–120. Barrett, D., Radke-Yarrow, M., & Klein, R. (1982). Chronic malnutrition and child behavior. Developmental Psychology, 18, 541–556. Bates, J. (2001). Adjustment style in childhood as a product of parenting and temperament. In T. D. Wachs & G. Kohnstamm (Eds.), Temperament in context (pp. 173–200). Mahwah, NJ: Erlbaum. Behrman, J. (1988). Nutrition, health, birth order and seasonality: Intrahousehold allocation among children in rural India. Journal of Development Economics, 28, 43–62. Bellinger, D. (1995). Interpreting the literature on lead and child development. Neurotoxicology and Teratology, 17, 201–212. Bellinger, D. (2001). Future directions for neurobehavioral studies of environmental neurotoxicants. Neurotoxicology, 22, 645–656. Bellinger, D., Needleman, H., Leviton, A., Waternaux, C., Rabinowitz, M., & Nichols, M. (1984). Early sensory-motor development and prenatal exposure to lead. Neurobehavioral Toxicology and Teratology, 6, 387–402. Bellinger, D., Sloman, J., Leviton, A., Rabinowitz, M., Needleman, H., & Waternaux, C. (1991). Low level lead exposure and children’s cognitive function in the preschool years. Pediatrics, 87, 219–227. Bellinger, D., Titelbaum, L., Hu, H., & Needleman, H. (1994). Attentional correlates of dentin and bone lead levels in adolescents. Archives Environmental Health, 49, 98–105. Bender, B. (1999). Learning disorders associated with asthma and allergies. School Psychology Review, 28, 204–214. Bengtsson, T., & Lindstrom, M. (2000). Childhood misery and disease in later life: The effects of mortality in old age of hazards experienced in early life, southern Sweden, 1760–1894. Population Studies, 54, 263–277. Bentley, M., Engle, P., Penny, M., Creed-Kanashiro, H., Marin, R., Bartolini, R., & Caulfield, L. (2000). Care and nutrition: differences in feeding style and infant nutrition status among low-income urban and rural Peru caregivers. FASEB Journal, Abstract #356.2. Berger, J., Aguayo, V., Miguel, J., Lujan, C., Tellez, W., & Traissac, P. (1997). Definition and prevalence of anemia in Bolivian women of childbearing age living at high altitudes. Nutrition Reviews, 55, 247–256. Berkman, D., Lescano, A., Gilman, R., Lopez, S., & Black, M. (2002). Effects of stunting, diarrhoeal disease, and parasitic infection during infancy on cognition in late childhood: A follow-up study. Lancet, 359, 564–571. Berney, B. (1996). Epidemiology of childhood lead poisoning. In S. Pueschel, J. Linakis, & A. Anderson (Eds.), Lead poisoning in childhood (pp. 15–36). Baltimore: Brookes. Bithoney, W., Vandeven, A., & Ryan, A. (1993). Elevated lead levels in reportedly abused children. Journal of Pediatrics, 122, 719–720. Bjorklund, D., & Pellegrini, A. (2002). The origins of human nature. Washington, DC: American Psychological Association.


397

Black, M. (1998). Zinc deficiency and child development. American Journal of Clinical Nutrition Supplement, 68, 464S–469S. Black, M., & Krishnakumar, A. (1999). Predicting longitudinal growth curves of height and weight using ecological factors for children with and without early growth deficiency. Journal of Nutrition, 129, 539S–543S. Bogden, J. D., Oleske, J. M., & Louria, D. B. (1997). Lead poisoning—one approach to a problem that won’t go away. Environmental Health Perspectives, 105, 1284–1287. Boivin, M., & Giordani, B. (1993). Improvements in cognitive performance for school children in Zaire, Africa, following an iron supplement and treatment for intestinal parasites. Journal of Pediatric Psychology, 18, 249–264. Boivin, M., Giordani, B., Ndanga, K., & Maky, M. (1993). Effects of treatment for intestinal parasites and malaria on the cognitive abilities of school children in Zaire, Africa. Health Psychology, 12, 220–226. Bornstein, M. (1995). Handbook of parenting. Mahwah, NJ: Erlbaum. Boyce, W., & Jemerin, J. (1990). Psychobiological differences in childhood stress response. Developmental and Behavioral Pediatrics, 11, 86–94. Bradbury, T., & Miller, G. (1985). Season of birth in schizophrenia: A review of evidence, methodology, and etiology. Psychological Bulletin, 98, 569–594. Bradley, R. (1999). The home environment. In S. Friedman & T. D. Wachs (Eds.), Conceptualization and measurement across the life span (pp. 31–58). Washington, DC: American Psychological Association. Bradley, R., & Corwyn, R. (1999). Parenting. In L. Balter & C. Tamis-LeMonda (Eds.), Child psychology: A handbook of contemporary issues (pp. 339–362). Philadelphia: Psychology Press. Bradley, R., Whiteside, L., Mundfrom, D., Casey, P., Kelleher, K., & Pope, S. (1994). Early indications of resilience and their relation to experiences in the home environments of low birth weight premature children living in poverty. Child Development, 65, 346–360. Branca, F., Pastore, G., Demissie, T., & Ferro-Luzzi, A. (1993). The nutritional impact of seasonality in children and adults of rural Ethiopia. European Journal of Clinical Nutrition, 47, 840–850. Bronfenbrenner, U. (1989). Ecological systems theory. Annals of Child Development, 6, 187–249. Bronfenbrenner, U. (1999). Environments and developmental perspective. In S. Friedman & T. D. Wachs (Eds.), Measuring the environment across the life span (pp. 3–30). Washington, DC: American Psychological Association. Bronfenbrenner, U., & Ceci, S. (1994). Nature-nurture, reconceptionalized in developmental perspective. Psychological Review, 101, 568–586. Bronfenbrenner, U., & Crouter, A. (1983). The evolution of environmental models in developmental research. In R. Lerner (Ed.), Handbook of child psychology, Vol. 1: History, theory and methods (Vol. 4, pp. 357–414). New York: Wiley. Brown, B. (1999). Measuring the peer environment of American adolescents. In S. Friedman & T. D. Wachs (Eds.), Conceptualization and measurement across the life span (pp. 59–90). Washington, DC: American Psychological Association. Brown, J., & Sherman, L. (1995). Policy implications of new scientific knowledge. Journal of Nutrition—Supplement, 125, 2281S–2284S. Bruner, A. B., Joffe, A., Duggan, A. K., Casella, J. F., & Brandt, J. (1996). Randomised study of cognitive effects of iron supplementation in non-anaemic iron-deficient adolescent girls. The Lancet, 348, 992–996. Buka, S., Tsuang, M., Torrey, E. F., Klebanoff, M., & Bernstein, D. (2001). Maternal infections and subsequent psychosis among offspring. Archives of General Psychiatry, 58, 1032–1037.

398

Theodore D. Wachs

Burnette, M., Ebbert, R., Edwards-Hawver, C., Soma, S., & Canfield, R. (1999). Pediatric lead exposure is associated with poor impulse control during free play. Paper presented to the American Psychological Society, Denver, Colorado. Castellino, N., & Castellino, P. (1995). Lead metabolism. In N. Castellino, P. Castellino, & N. Sannolo (Eds.), Inorganic lead exposure: Metabolism and intoxication (pp. 115–202). Boca Raton: Lewis. Cataldo, M., Finney, J., Richman, G., Riley, A. et al. (1992). Behavior of injured and uninjured children and their parents in simulated hazardous setting. Journal of Pediatric Psychology, 17, 73–80. Chase-Landsdale, P., Lindsay, C., Andrew, J., & Kiernen, K. (1995). The long term effects of parental divorce on the mental health of young adults. Child Development, 66, 1614–1634. Chavez, A., & Martinez, C. (1984). Behavioral measurement of activity in children and their relation to food intake in a poor community. In E. Pollitt & P. Amante (Eds.), Energy intake and activity (pp. 303–321). New York: Liss. Chavez, A., Martinez, C., & Yaschine, J. (1971). Nutrition, behavioral development and mother–child interaction in young rural children. Federation Proceedings, 34, 1574–1582. Christensen, L. (1996). Diet behavior relationships: Focus on depression. Washington, DC: American Psychological Association. Collins, W., Maccoby, E., Steinberg, L., Hetherington, E., & Bornstein, M. (2001). Toward nature with nurture. American Psychologist, 56, 171–173. Colombo, M., de la Parra, A., & Lopez, I. (1992). Intellectual and physical outcome of children undernourished in early life is influenced by later environmental conditions. Developmental Medicine and Child Neurology, 34, 611–622. Committee on Environmental Health. (1997). Environmental tobacco smoke: A hazard to children. Pediatrics, 99, 639–642. Connolly, K. (1998). Mental and behavioral effects of parasitic infection. Nutrition, health and child development. Pan-American Health Organization Scientific Monograph #566 (pp. 138–161). Washington, DC: PAHO. Connolly, K., & Kvalsig, J. (1993). Infection, nutrition and cognitive performance in children. Parasitology, 107, S187–S200. Crouter, A., & McHale, S., (1993). The long arm of the job. In T. Luster & L. Okagaki (Eds.), Parenting: An Ecological Perspective (pp. 179–202). Hillsdale NJ: Erlbaum. Crow, T. (1989). Viruses and schizophrenia: The virogene hypothesis. Biologist, 36, 10–14. Damm, D., Grandjean, P., Lyngbye, T., Trillingsgaard, A., & Hansen, O. (1993). Early lead exposure and neonatal jaundice: Relation to neurobehavioral performance at 15 years of age. Neurotoxicology & Teratology, 15, 173–181. Davidson, C., & Rabinowitz, M. (1992). Lead in the environment. In H. Needleman (Ed.), Human lead exposure (pp. 66–84). Boca Raton: CRC Press. DeFries, J., Plomin, R., & Fulker, D. (1994). Nature and nurture during middle childhood. Cambridge, MA: Blackwell. DeVries, M. (1984). Temperament and infant mortality among the Masai of East Africa. American Journal of Psychiatry, 141, 1189–1194. Diamond, A. (1990). The development and neural basis of memory function as indexed by the AB and delayed response tasks in human infants and human monkeys. In A. Diamond (Ed.), The development and neural basis of higher cognitive functions. Annals of the New York Academy of Sciences, 608, 267–309. Dietrich, K. (1995). Environmental neurotoxicants and psychological development. In K. Yeates, M. Ris et al. (Eds.), The science and practice of neuropsychology (pp. 206–234). New York: The Guilford Press.


399

Dietrich, K., Kraft, K., Shukla, R., Bornschein, R., & Succop, P. (1987). The neurobehavioral effects of early lead exposure. In S. Schroeder (Ed.), Toxic substances and mental retardation. Monographs of the American Association on Mental Deficiency (Vol. 8, pp. 71–96). Washington, DC: AAMD. Dumaret, A. (1985). IQ, scholastic performance and behavior of sibs raised in contrasting environments. Journal of Child Psychology and Psychiatry, 26, 553–580. Elder, G. (1998). The life course and human development. In R. Lerner (Ed.), Handbook of child psychology (Vol. 1, pp. 939–991). New York: Wiley. Engle, P. (1991). Maternal work and child care strategies in Peri-Urban, Guatemala: Nutritional effects. Child Development, 62, 954–965. Engle, P., & Lhotska, L. (1999). The role of care in programmatic actions for nutrition. Food and Nutrition Bulletin, 20, 121–135. Engle, P., Zeitlin, M., Medrano, Y., & Garcia, L. (1996). Growth consequences of low income Nicaraguan mothers’ theories about feeding one year olds. In S. Harkness & C. Super (Eds.), Parents’ cultural belief systems (pp. 428–446). New York: Guilford. Ernhart, C., Morrow-Tlucak, Wolf, A., Super, D., & Drotar, D. (1989). Low level lead exposure in the prenatal and early preschool periods: Intelligence prior to school entry. Neurotoxology and Teratology, 11, 161–170. Evans, G. (1999). Measurement of the physical environment as stressor. In S. Friedman & T. D. Wachs (Eds.), Measuring environment across the life span (pp. 249–278). Washington, DC: American Psychological Association. Farmer, J., & Peterson, L. (1995). Injury risk factors in children with attention deficit hyperactivity disorder. Health Psychology, 14, 325–332. Felt, B. T., & Lozoff, B. (1996). Brain iron and behavior of rats are not normalized by treatment of iron deficiency anemia during early development. American Institute of Nutrition, 693–701. Frank, D., Roos, N., Meyers, A., Napoleone, M., Peterson, K., Cather, A., & Cupples, L. (1996). Seasonal variation in weight-for-age in a pediatric emergency room. Public Health Reports, 111, 366–371. Friedman, S., & Amato, J. (1999). The child-care environment: Conceptualizations, assessments and issues. In S. Friedman & T. D. Wachs (Eds.), Measuring environment across the life span (pp. 127–166). Washington, DC: American Psychological Association. Fullmer, C. (1992). Intestinal interactions of lead and calcium. Neurotoxology, 13, 799–807. Galler, J., & Barrett, L. R. (2001). Children and famine: long-term impact on development. Ambulatory Child Health, 7, 85–95. Gardner, J., Walker, S., Chang, S., Vutchkov, M., & Lalor, G. (1998). Undernutrition and elevated blood levels: Effects on psychomotor development among Jamaican children. Public Health Nutrition, 39, 177–179. Garralda, E., & Rangel, L. (2002). Chronic fatigue syndrome in children and adolescents. Journal of Child Psychology & Psychiatry & Allied Disciplines, 43, 169–176. Geber, M., & Dean, R. (1956). The psychological changes accompanying Kwashiorkor. Courrier, 6, 3–15. Gillberg, C., Noren, J., Wahlstrom, J., & Rasmussen, P. (1982). Heavy metals and neuropsychiatric disorders in six-year-old children. Acta Paedopsychiatrica, 48, 253–263. Gonez, C., Donayre, M., Villena, A., & Gonzales, G. (1993). Hematocrit levels in children at sea level and at high altitude: Effect of adrenal androgens. Human Biology, 65, 49–55. Gonzales, G., & Villena, A. (1996). Body mass index and age at menarche in Peruvian children living at high altitude and at sea level. Human Biology, 68, 265–270.

400

Theodore D. Wachs

Gorman, K. (1995). Malnutrition and cognitive development. Journal of Nutrition—Supplement, 125, S2239–S2244. Gottleib, G. (1998). Normally occurring environmental and behavioral influences on gene activity: From central dogma to probabilistic epigenesis. Psychological Bulletin, 105, 792–802. Goyer, R. (1993). Lead toxicity: Current concerns. Environmental Health Perspectives, 100, 177–187. Grantham-McGregor, S. (1984). The social background of childhood malnutrition. In B. Schurch & J. Brozek (Eds.), Malnutrition and behavior (pp. 358–374). Lausanne: IDECG. Grantham-McGregor, S. (1995). A review of studies of the effect of severe malnutrition on mental development. Journal of Nutrition—Supplement, 125, 2233S–2238S. Grantham-McGregor, S. (2002). Linear growth retardation and cognition. Lancet, 359, 542. Grantham-McGregor, S., & Ani, C. (2001). A review of studies on the effect of iron deficiency on cognitive development in children. Journal of Nutrition—Supplement, 131, 649–668. Grantham-McGregor, S., Chang, S., & Walker, S. (1998). Evaluation of school feeding programs: Some Jamaican examples. American Journal of Clinical Nutrition, Supplement, 67, 785S–789S. Grantham-McGregor, S., Fernald, L., & Sethuraman, K. (1999a). The effects of health and nutrition on cognitive and behavioral development in children in the first three years of life. Part 1: Low birth weight, breast feeding and protein energy malnutrition. Food and Nutrition Bulletin, 20, 53–75. Grantham-McGregor, S., Fernald, L., & Sethuraman, K. (1999b). Effects of health and nutrition on cognitive and behavioral development in children in the first three years of life. Part 2: Infections and micronutrient deficiencies: Iodine, iron and zinc. Food and Nutrition Bulletin, 20, 76–99. Grantham-McGregor, S., Lira, P., Ashworth, A., Morris, S., & Assuncao, A. (1998). The development of low birth weight term infants and the effects of the environment in Northeast Brazil. Journal of Pediatrics, 132, 661–666. Grantham-McGregor, S., Stuart, M., & Powell, C. (1991). Behavior of severely malnourished children in a Jamaican hospital. Developmental Medicine and Child Neurology, 33, 706–714. Gratz, R. (1992). Unintentional injury in early childhood group settings. Children’s Health Care, 21, 239–249. Greenough, W., & Black, J. (1992). Induction of brain structure by experience. In M. Gunnar & C. Nelson (Eds.), Developmental behavioral neuroscience (pp. 153–232). Hillsdale, NJ: Erlbaum. Grundman, M. (1996). Historical context of father absence. International Journal of Behavioral Development, 19, 415–431. Guilarte, T. (1993). Vitamin B6 and cognitive development. Nutritional Review, 51, 193–198. Gunnar, M. (2000). Early adversity and the development of stress reactivity and regulation. In C. Nelson (Ed.), The effects of early adversity on neurobehavioral development (pp. 163–200). Mahwah, NJ: Erlbaum. Guntheroth, W., Lohmann, R., & Spiers, P. (1992). A seasonal association between SIDS deaths and kindergarten absences. Public Health Reports, 107, 319–323. Hagekull, B., Bohlin, G., & Rydell, A. (1997). Maternal sensitivity, infant temperament and the development of early feeding problems. Infant Mental Health Journal, 18, 92–106. Halterman, J., Kaczorowski, J., Aligne, A., Auinger, P., & Szilagyi, P. (2001). Iron deficiency and cognitive achievement among school-age children and adolescents in the United States. Pediatrics, 107, 1381–1386. Hammad, T., Sexton, M., & Langenberg, P. (1996). Relationship between blood lead and dietary iron intake in preschool children. Annals of Epidemiology, 6, 30–33.


401

Han, S., Qiao, X., Kemp, F., & Bogden, J. (1997). Lead exposure at an early age substantially increases lead retention in the rat. Environmental Health Perspectives, 105, 412–417. Harper, L., & Huie, K. (1998). Free play use of space by preschoolers from diverse backgrounds: Factors influencing activity choices. Merrill-Palmer Quarterly, 44, 423–446. Harris, J. (1995). Where is the child’s environment? A group socialization theory of development. Psychological Review, 102, 458–489. Harris, N., Crawford, P., Yangzom, Y., Pinzo, L., Gyaltsen, P., & Hudes, M. (2001). Nutritional and health status of Tibetan children living in high altitudes. The New England Journal of Medicine, 344, 341–347. Hurtado, E., Claussen, A., & Scott, K. (1999). Early childhood anemia and mild or moderate mental retardation. American Journal of Clinical Nutrition, 69, 115–119. Hutchinson, M. K., & Sandall, S. (1995). Congenital TORCH infections in infants and young children: Neurodevelopment sequelae and implications for intervention. Topics in Early Childhood Special Education, 15, 65–82. Jacobson, S., & Jacobson, J. (2000). Teratogenic insult and neurobehavioral function in infancy and childhood. In C. Nelson (Ed.), The effects of early adversity on neurobehavioral development (pp. 61–112). Mahwah, NJ: Erlbaum. James, W. (1997). Long-term fetal programming of body composition and longevity. Nutrition Reviews—Supplement, 55, S31–S40. Johnson, M. (2001). Functional brain development during infancy. In G. Bremner & A. Fogel (Eds.), Blackwell handbook of infant development (pp. 169–190). Oxford: Blackwell. Jordan, K., Ayers, P., Jahn, S., Taylor, K., Huang, C. F., Richman, J., & Jason, L. (2000). Prevalence of fatigue and chronic fatigue syndrome-like illness in children and adolescents. Journal of the Chronic Fatigue Syndrome, 6, 3–21. Kagitcibasi, C. (1996). Family and human development across cultures. Mahwah, NJ: Erlbaum. Kaplan, H., & Dove, H. (1987). Infant development among the Ache of eastern Paraguay. Developmental Psychology, 23, 190–198. Karp, R. (1993). Malnourished children in the United States. New York: Springer. Keusch, G. (1990). Malnutrition, infection and immune function. In R. Suskind & L. LewinterSuskind (Eds.), The malnourished child (pp. 37–55). New York: Raven. Kinney, D., Jacobsen, B., Jansson, L., Faber, B., Tramer, S., & Suozzo, M. (2000). Winter birth and biological family history in adopted schizophrenics. Schizophrenia Research, 44, 95–103. Kleinman, J. (1992). The epidemiology of low birthweight. In S. Friedman & M. Sigman (Eds.), The psychological development of low birthweight children (pp. 21–36). Ablex: Norwood. Kochanska, G. (1997). Multiple pathways to conscience for children with different temperaments. Developmental Psychology, 33, 228–240. Konstantareas, M., Hauser, P., Lennox, C., & Homatidis, S. (1986). Season of birth in infantile autism. Child Psychiatry and Human Development, 17, 53–65. Kovalinko, P., Hoven, C., Wicks, J., Moore, R., Mandell, D., & Liu, H. (2000). Seasonal variations in internalizing, externalizing, and substance use disorders in youth. Psychiatry Research, 94, 103–119. Kvalsvig, J., & Becker, P. (1988). Selective exposure of active and sociable children to schistosomiasis. Annals of Tropical Medicine and Parasitology, 82, 471–474. Lanphear, B., Eberly, S., & Howard, C. (2000). Long-term effect of dust control on blood lead concentrations. Pediatrics, 106, 820. Leftwich, M., & Collins, F. (1994). Parental smoking, depression and child development. Journal of Pediatric Psychology, 19, 557–569. Lester, B. (1975). Cardiac habituation of the orienting response to an auditory signal in infants of varying nutritional status. Development Psychology, 11, 432–442.

402

Theodore D. Wachs

Lester, M., Horst, R., & Thatcher, R. (1988). Protective effects of zinc and calcium against heavy metal impairment of children’s cognitive function. Nutrition and Behavior, 3, 145–161. LeVine, R., & LeVine, S. (1988). Parental strategies among the Gusii of Kenya. In R. LeVine, P. Miller, & M. West (Eds.), Parental behaviors in diverse societies (pp. 27–36). San Francisco: Jossey-Bass. Leviton, A., Bellinger, D., Allred, E., Rabinowitz, M., Needleman, H., & Schoenbaum, S. (1993). Pre- and post-natal low level lead exposure and children’s dysfunction and school. Environmental Research, 60, 30–43. Levitsky, D., & Barnes, R. (1972). Nutritional and environmental interactions in behavioral development of the rat. Science, 176, 68–72. Levitsky, D., & Strupp, B. (1984). Functional isolation in rats. In J. Brozek & B. Schurch (Eds.), Malnutrition and behavior: Critical assessment of key issues (pp. 411–420). Lausanne: Nestle Foundation. Levitsky, D., & Strupp, B. (1995). Malnutrition and the brain. Journal of Nutrition—Supplement, 125, 2212S–2220S. Lewis, M., Worobey, J., Ramsay, D., & McCormack, M. (1992). Prenatal exposure to heavy metals: Effect on childhood cognitive skills and health status. Pediatrics, 89, 1010–1015. Lin-Fu, J. (1992). Modern history of lead poisoning. In H. Needleman (Ed.), Human lead exposure (pp. 24–85). Boca Raton: CRC Press. Lockitch, G. (1993). Perspectives on lead toxicity. Clinical Biochemistry, 26, 371–381. Lozoff, B. (1998). Exploratory mechanism for poorer development in iron-deficient anemic infants. In Nutrition, health and child development. Pan American Health Organization Scientific Monograph Number 566 (pp. 162–178). Washington, DC: PAHO. Lozoff, B., Klein, N., & Prabucki, K. (1986). Iron deficient anemic infants at play. Journal of Developmental and Behavioral Pediatrics, 7, 152–158. Lozoff, B., Klein, N., Nelson, E., McClish, D., Manuel, M., & Chacon, M. (1998). Behavior of infants with iron deficiency anemia. Child Development, 69, 24–36. Lozoff, B., & Wachs, T. D. (2001). Functional correlates of nutritional anemias in infancy and early childhood: Child development and behavior. In U. Ramakrishnan (Ed.), Nutritional anemias (pp. 69–88). Boca Raton: CRC Press. Lucas, S., Sexton, M., & Langenberg, P. (1996). Relationship between blood lead and nutritional factors in preschool children. Pediatrics, 97, 74–78. Lynch, S. (1997). Interaction of iron with other nutrients. Nutrition Reviews, 55, 102–110. Lyngbye, T., Hansen, O., & Grandjean, P. (1991). Low level lead absorption and neonatal jaundice. Neurotoxicology, 12, 810. Magnusson, D. (1995). Individual development: A holistic integrated model. In P. Moen, G. Elder & K. Luscher (Eds.), Examining lives in context: Perspectives on the ecology of human development (pp. 19–60). Washington, DC: American Psychological Association. Mahaffey, K. (1990). Environmental lead toxicity: Nutrition as a component of intervention. Environmental Health Perspectives, 89, 75–78. Mahaffey, K., Rosen, J., Chesney, R., Peeler, J., Smith, C., & DeLuca, H. (1982). Association between age, blood level concentration and serum 1, 25-dihydroxycholecalciferol levels in children. American Journal of Clinical Nutrition, 35, 1327–1331. Marlowe, M., Moon, C., Errera, J., Jacobs, J. et al. (1986). Low mercury levels and childhood intelligence. Journal of Orthomolecular Medicine, 1, 43–49. Martorell, R., Ramakrishnana, U., Schroeder, D., Melgar, P., & Neufeld, L. (1998). Intrauterine growth retardation, body size, body composition and physical performance in adolescence. European Journal of Clinical Nutrition Supplement, 52, S43–S53.


403

Marx, J. (1997). Iron deficiency in developed countries: Prevalence, influence of lifestyle factors and hazards of prevention. European Journal of Clinical Nutrition, 51, 491–494. Masten, A., Best, K., & Garmezy, N. (1990). Resilience in development. Development and Psychopathology, 2, 425–444. Matheny, A. (1991). Children’s unintentional injuries and gender: Differentiation by environmental and psychosocial aspects. Children’s Environmental Quarterly, 8, 51–61. McCarton, C., Brooks-Gunn, J., & Tonascia, J. (1994). The cognitive, behavioral, and health, status of mainland Puerto Rican children in the Infant Health and Development Program. In G. Lamberty & C. G. Coll (Eds.), Puerto Rican women and children: Issues in health, growth, and development. Topics in social psychiatry (pp. 161–189). New York: Plenum Press. McCullough, A., Kirksey, A., Wachs, T. D., McCabe, G., Bassily, N., Bishry, Z., Galal, O., Harrison, G., & Jerome, N. (1990). Vitamin B-6 status of Egyptian mothers: Relation to infant behavior and maternal infant interactions. American Journal of Clinical Nutrition, 51, 1067–1074. McMichael, A., Baghurst, P., Vimpani, G., Robertson, E., Wigg, N., & Tong, S. (1992). Sociodemographic factors modifying the effect of environmental lead on neuropsychological development in early childhood. Neurology and Teratology, 14, 321–327. McSwain, R. (1981). Care and conflict in infant development. Infant Behavior and Development, 4, 225–246. Mednick, S., Machon, R., Huttunen, M., & Bonett, D. (1988). Adult schizophrenia following prenatal exposure to an influenza epidemic. Archives of General Psychiatry, 45, 189–192. Meeks-Gardner, J., Grantham-McGregor, S., Himes, J., & Chang, S. (1999). Behaviour and development of stunted and nonstunted Jamaican children. Journal of Child Psychology & Psychiatry & Allied Disciplines, 5, 819–827. Merialdi, M., Caulfield, L., Zavaleta, N., Figueroa, A., & DiPietro, J. (1998). Adding zinc to prenatal iron and folate tablets improves fetal neurobehavioral development. FASEB Journal, 12, A346. Milar, C., Schroeder, S., Mushak, A., Dolcourt, J., & Grant, L. (1980). Contribution of the caregiving environment to increased lead levels of children. American Journal of Mental Deficiency, 84, 339–344. Miller, G., Massaro, T., & Massaro, E. (1990). Interactions between lead and essential elements. Neurotoxicology, 11, 99–120. Morrongiello, B., & Dawber, T. (1998). Toddlers’ and mothers’ behaviors in an injury-risk situation: Implications for sex differences in childhood injuries. Journal of Applied Developmental Psychology, 19, 625–639. Mouridsen, S., Nielsen, S., Rich, B., & Isager, T. (1994). Season of birth in infantile autism and other types of childhood psychoses. Child Psychiatry & Human Development, 25, 31–43. Narita, K., Sasaki, T., Akaho, R., Okazaki, Y., Kusumi, I., Kato, T., Hashimoto, O., Fukuda, R., Koyama, T., Matsuo, K., Okabe, Y., Nando, S., Hohjoh, H., & Tokunaga, K. (2000). Human leukocyte antigen and season of birth in Japanese patients with schizophrenia. American Journal of Psychiatry, 157, 1173–1175. National Research Council (2000). From neurons to neighborhoods. Washington, DC: National Academy Press. Needleman, H. (1993). The current status of childhood low-level lead toxicity. Neurotoxicology, 14, 161–166. Needleman, H., & Gatsonis, C. (1990). Low-level lead exposure and the IQ of children. Journal of the American Medical Association, 263, 673–678.

404

Theodore D. Wachs

Needleman, H., Gunnoe, C., Leviton, A., Reed, R., Peresie, H., Meher, C., & Barrett, P. (1979). Deficits in psychologic and classroom performance of children with elevated dentine lead levels. New England Journal of Medicine, 300, 689–695. Nelson, C. (1995). The ontogeny of human memory. Developmental Psychology, 31, 723–738. Nelson, C. (2000). The neurobiological bases of early intervention. In S. Meisels & J. Shonkoff (Eds.), Handbook of early childhood intervention (2nd ed. pp. 204–227). Cambridge, England: Cambridge University Press. Novotny, T., Hopkins, R., Shillam, P., & Janoff, E. (1990). Prevalence of giardia lamblia and risk factors for infection among children attending day-care facilities in Denver. Public Health Reports, 105, 72–75. Nsamenang, A. (1992). Human development in cultural context. Newbury Park, CA: Sage. Otero, G. A., Aguirre, D. M., Porcayo, R., & Fernandez, T. (1999). Psychological and electroencephalographic study in school children with iron deficiency. International Journal of Neuroscience, 99, 113–121. Padich, R., Dietrich, K., & Pearson, D. (1985). Attention activity level on lead exposure at 18 months. Environmental Research, 38, 137–143. Paine, P., Dorea, J., Pasquali, J., & Monteior, A. (1992). Growth and cognition in Brazilian school children. International Journal of Behavioral Development, 15, 160–183. Palti, H., Meijer, A., & Adler, B. (1985). Learning achievement and behavior at school of anemic and non-anemic infants. Early Human Development, 10, 217–223. Parmelee, A. (1986). Children’s illness: Their beneficial effects on behavioral development. Child Development, 57, 1–10. Patterson, G., & Bank, L. (1989). Some amplifying mechanisms for pathologic processes in families. In M. Gunnar & E. Thelan (Eds.), Systems and development: The Minnesota Symposium on Child Psychology (Vol. 22, pp. 167–209). Hillsdale, NJ: Erlbaum. Pawson, I., & Jest, C. (1978). The high altitude areas of the world and their cultures. In P. Baker (Ed.), The biology of high altitude people (pp. 17–46). Cambridge: Cambridge University Press. Pearson, D., & Dietrich, K. (1985). The behavioral toxicology and teratology of childhood. Neurotoxicology, 6, 165–182. Phelps, L. (1999). Low-level lead exposure: Implications for research and practice. School Psychology Review, 28, 477–492. Pollitt, E. (1988). A critical view of three decades of research on the effects of chronic energy malnutrition on behavioral development. In B. Schurch & M. Scrimshaw (Eds.), Chronic energy deficiency (pp. 77–93). Lausanne: IDECG. Pollitt, E. (2000). A developmental view of the undernourished child. European Journal of Clinical Nutrition—Supplement, 54, S2–S10. Pollitt, E., & Saco-Pollitt, C. (1996). On the role of the physical environment in the development of intelligence. In D. Detterman (Ed.), Current topics in human intelligence (Vol. 5, pp. 163–172). Norwood, NJ: Ablex. Pollitt, E., Cueto, S., & Jacoby, E. (1998). Fasting and cognition in well and undernourished school children. American Journal of Clinical Nutrition Supplement, 67, 779S–785S. Pollitt, E., Gorman, K., Engle, P., Martorell, R., & Rivera, J. (1993). Early supplementary feeding and cognition: Effects over two decades. Monograph of the Society for Research in Child Development, 58, 1–99. Pollitt, E., Saco-Pollitt, C., Jahari, A., Husaini, M., & Huang, J. (2000). Effects of an energy and micronutrient supplement on mental development and behavior under natural conditions in undernourished children in Indonesia. European Journal of Clinical Nutrition—Supplement, 54, S80–S90.


405

Potasova, A. (1998). Cognitive functions of preschool children affected by environmental toxins. Studia Psychologica, 40, 335–338. Powell, C., Walker, S., & Grantham-McGregor, S. (1998). Early childhood stimulation programs in Jamaica. In Nutrition health and child development (pp. 198–208). Pan American Health Organization Scientific Monograph Number 566. Washington, DC: PAHO. Rabinowitz, M., Wang, J., & Soong, W. (1991). Dentine lead and child intelligence in Taiwan. Archives of Environmental Health, 46, 351–360. Ramey, C., & Ramey, S. (1998). Early intervention. American Psychologist, 53, 210–225. Rao, R., & Georgieff, M. (2000). Early nutrition and brain development. In C. Nelson (Ed.), The effects of early adversity on neurobehavioral development (pp. 1–30). Mahwah, NJ: Erlbaum. Ricciuti, H. (1993). Nutrition and mental development. Current Directions in Psychological Science, 2, 43–46. Richardson, S., Birch, H., Grabie, E., & Yoder, K. (1972). The behavior of children in school who were severely malnourished in the first two years of life. Journal of Health and Social Behavior, 13, 276–284. Roberts, J., Burchinal, M., & Clarke-Klein, S. (1995). Otitis media in early childhood and cognitive, academic and behavior outcomes at 12 years of age. Journal of Pediatric Psychology, 20, 645–660. Rose, S. (1994). Relation between physical growth and information processing in infants born in India. Child Development, 65, 889–902. Rowe, D. (1994). The limits of family influence. New York: Guilford. Ruff, H. (1999). Population-based data and the development of individual children: The case of low to moderate lead levels and intelligence. Journal of Developmental and Behavioral Pediatrics, 20, 42–49. Ruff, H., Markowitz, M., Bijur, P., & Rosen, J. (1996). Relationships among blood lead levels iron deficiency and cognitive development in two year old children. Environmental Health Perspectives, 104, 180–185. Rutter, M. (1993). Developmental psychopathology as a research perspective. In D. Magnusson & P. Casaer (Eds.), Longitudinal research on individual development (pp. 127–152). Cambridge: Cambridge University Press. Rutter, M. (1996). Transitions and turning points in developmental psychopathology. International Journal of Behavioral Development, 19, 603–626. Rvachew, S., Slawinski, E., Williams, M., & Green, C. (1996). The impact of early onset otitis media on prelinguistic speech development. Journal of Speech-Language Pathology & Audiology, 20, 247–255. Rvachew, S., Slawinski, E., Williams, M., & Green, C. (1999). The impact of early onset otitis media on babbling and early language development. Journal of the Acoustical Society of America, 105, 467–475. Saco-Pollitt, C. (1989). Ecocultural context and developmental risk. In J. Nugent, B. Lester, & T. Brazelton (Eds.), The cultural context of infancy (Vol. 1, pp. 3–25). Norwood, NJ: Ablex. Sameroff, A. (1995). General systems theory and developmental psychopathology. In D. Cicchetti & D. Cohen (Eds.), Developmental psychopathology, Vol. 1: Theory and method (pp. 659–695). New York: Wiley. Sampson, R., & Laub, J. (1994). Urban poverty and the family context of delinquency. Child Development, 65, 523–540. Sandiford, P., Cassel, J., Sanchez, G., & Coldham, C. (1997). Does intelligence account for the link between maternal literacy and child survival? Social Science and Medicine, 45, 1231–1239.

406

Theodore D. Wachs

Satz, P., Zaucha, K., McCleary, C., Light, R., Asarnow, R., & Becker, D. (1997). Mild head injury in children and adolescents. Psychological Bulletin, 122, 107–131. Sazawal, S., Bentley, M., Black, R., Dhingra, P., George, S., & Bhan, M. (1996). Effect of zinc supplementation on observed activity in low socioeconomic Indian preschool children. Pediatrics, 98, 1132–1137. Scarr, S. (1992). Developmental theories of the 1990s. Child Development, 63, 1–19. Schaffer, S., Szilagyi, P., & Weitzman, M. (1994). Lead poisoning risk determination in an urban population through use of a standardized questionnaire. Pediatrics, 93, 159–163. Scheper-Hughes, N. (1987). A basic strangeness: Maternal estrangement and infant death. In C. Super (Ed.), The role of culture in developmental disorder (pp. 131–153). San Diego: Academic Press. Schoepflin, U., & Muller-Brettel, M. (1990). International journal of behavioral development: Scope and trends. International Journal of Behavioral Development, 13, 393–406. Schore, A. (1994). Affect regulation and the origin of the self. Hillsdale, NJ: Erlbaum. Schroeder, S., & Hawk, B. (1987). Psycho-social factors, lead exposure and IQ. In S. Schroeder (Ed.), Toxic substances and mental retardation. Monographs of the American Association on Mental Deficiency (Vol. 8, pp. 97–138). Washington, DC: AAMD. Scola, P. (1991). Infections. In J. Matson & J. Mulick (Eds.), Handbook of mental retardation (2nd ed., pp. 151–157). New York: Pergamon. Segall, M., Dasen, P., Berry, J., & Poortinga, Y. (1990). Human behavior in global perspective. New York: Pergamon. Shaheen, S. (1984). Neuromaturation and behavior development. Developmental Psychology, 20, 542–550. Shannon, M., & Graef, J. (1996). Lead intoxication in children with pervasive developmental disorders. Journal of Toxicology—Clinical Toxicology, 34, 177–181. Shonkoff, J. (1994). Health surveillance and the development of children. In S. Friedman & C. Haywood (Eds.), Developmental follow-up (pp. 113–128). San Diego: Academic Press. Sigman, M., Espinosa, M., & Whaley, S. (1998). Mild malnutrition and the cognitive development of Kenyan school children. In Nutrition Health and Child Development. Pan American Health Organization Scientific Monograph Number 566 (pp. 91–103). PAHO: Washington, DC. Sigman, M., Neumann, C., Baksh, M., Bwibo, N., & McDonald, M. (1989a). Relation between nutrition and development in Kenyan toddlers. Journal of Pediatrics, 115, 357–364. Sigman, M., Neumann, C., Zansen, A., & Bwido, N. (1989b). Cognitive abilities of Kenyan children in relation to nutrition, family characteristics and education. Child Development, 60, 1463–1474. Simeon, D., & Grantham-McGregor, S. (1990). Nutritional deficiencies and children’s behavior and mental development. Nutrition Research Review, 3, 1–24. Singer, L., Arendt, R., Farkas, K., Minnes, S., Huang, J., & Yamashita, T. (1997). Relationship of prenatal cocaine exposure and maternal postpartum psychological distress to child developmental outcome. Development and Psychopathology, 9, 473–489. Sourander, A., Koskelainen, M., & Helenius, H. (1999). Mood, latitude, and seasonality among adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 38, 1271–1276. Stein, Z., Susser, M., Saenger, G., & Marolla, F. (1975). Famine and human development. London: Oxford University Press. Steinberg, L., Darling, N. E., & Fletcher, A. C. (1995). Authoritative parenting and adolescent adjustment: An ecological journey. In P. Moen, G. Elder, & K. Luscher (Eds.), Examining lives in context: Perspectives on the ecology of human development (pp. 423–466). Washington, DC: American Psychological Association.


407

Steinberg, L., Dornbusch, S., & Brown, B. (1992). Ethnic differences in adolescent achievement. American Psychologist, 47, 723–729. Stoltzfus, R. (2001). Defining iron deficiency in public health terms. Journal of Nutrition— Supplement, 131, 565S–567S. Stoltzfus, R., Dreyfuss, M., Chwaya, H., & Albonico, M. (1997). Hookworm control as a strategy to prevent iron deficiency. Nutrition Reviews, 55, 223–232. Stoltzfus, R., Kvalsvig, J., Chwaya, H., Montresor, A., Albonico, M., Tielsch, J., Savioli, L., & Pollitt, E. (2001). Effects of iron supplementation and anthelmintic treatment on motor and language development of preschool children in Zanzibar: Double blind, placebo controlled study. British Medical Journal, 323, 1389–1393. Strupp, B., & Levitsky, D. (1995). Enduring cognitive effects of early malnutrition. Journal of Nutrition—Supplement, 125, 2221S–2232S. Super, C., & Harkness, S. (1999). The environment in cultural and developmental research. In S. Friedman & T. D. Wachs (Eds.), Measuring environment across the lifespan (pp. 279–336). Washington, DC: American Psychological Association. Tacconi, M., Calzi, F., & Salmona, M. (1997). Grain lipids and diet. In M. Hilbrand & R. Spitz (Eds.), Lipids, health and behavior (pp. 197–226). Washington, DC: American Psychological Association. Talbert, J., & McLaughlin, M. (1999). Assessing the school environment. In S. Friedman & T. D. Wachs (Eds.), Measuring environment across the life-span (pp. 197–228). Washington, DC: American Psychological Association. Tamis-LeMonda, C., & Bornstein, M. (1994). Specificity in mother-toddler play reactions across the second year. Developmental Psychology, 30, 283–292. Thompson, G., Raab, G., Hepburn, W., Hunter, R., Fulton, M., & Laxen, D. (1989). Blood lead levels and children’s behavior. Journal of Child Psychology and Psychiatry, 30, 515–528. Tronick, E., Thomas, R., & Daltabuit, M. (1994). The Quechua Manta pouch: A caregiving practice for buffering the Peruvian infant against the multiple stressors of high altitude. Child Development, 5, 1005–1013. UNICEF. (1998). The state of the world’s children 1998: Malnutrition causes, consequences and solutions. Nutrition Reviews, 56, 115–123. Valenzuela, M. (1997). Maternal sensitivity in a developing society: The context of urban poverty and infant chronic undernutrition. Developmental Psychology, 33, 845–855. Vandell, D., & Posner, J. (1999). Conceptualization and measurement of children’s after school environment. In S. Friedman & T. D. Wachs (Eds.), Measuring the environment across the life span (pp. 167–196). Washington, DC: American Psychological Association. Vandiver, T. (1997). Relationship of mother’s perceptions and behaviors to the duration of breastfeeding. Psychological Reports, 80, 1375–1384. Vorhees, C., & Mollnow, E. (1987). Behavioral teratogenesis: Long term influences on behavior from early exposure to environmental agents. In J. Osofsky (Ed.), Handbook of infant development (2nd ed., pp. 913–971). New York: Wiley. Wachs, T. D. (1992). The nature of nurture. Newbury Park, CA: Sage. Wachs, T. D. (1995). Relation of mild to moderate malnutrition to human development: Correlational studies. Journal of Nutrition Supplement, 125, 2245S–2254S. Wachs, T. D. (1999). Celebrating complexity: Conceptualization and assessment of the environment. In S. Friedman & T. D. Wachs (Eds.), Measuring the environment across the life span (pp. 357–392). Washington, DC: American Psychological Association. Wachs, T. D. (2000). Necessary but not sufficient: The respective roles of single and multiple influences on human development. Washington, DC: American Psychological Association.

408

Theodore D. Wachs

Wachs, T. D. (2002). Nutritional deficiencies as a biological context for development. In W. Hartup & R. Silbereisen (Eds.), Growing points in developmental science (pp. 64–84). New York: Psychology Press. Wachs, T. D., & Gruen, G. (1982). Early experience and human development. New York: Plenum. Wachs, T. D., & McCabe, G. (2001). Relation of maternal intelligence and schooling to offspring nutritional intake. International Journal of Behavioral Development, 25, 444–449. Wachs, T. D., & Plomin, R. (1991). Conceptualization and measurement of organism environment interaction. Washington, DC: American Psychological Association. Wachs, T. D., Bishry, Z., Yunis, F., McCabe, G., Sweifi, E., Harrison, G., Shaheen, F., Galal, O., Jerome, N., & Kirksey, A. (1995). Nutritional intake and context as predictors of cognition and adaptive behavior of Egyptian school age children. International Journal of Behavioral Development, 18, 425–450. Wachs, T. D., Moussa, W., Bishry, Z., Yunis, F., Sobhy, A., McCabe, G., Jerome, N., Galal, O., Harrison, G., & Kirksey, A. (1993). Relations between nutrition and cognitive performance in Egyptian toddlers. Intelligence, 17, 151–172. Wachs, T. D., Pollitt, E., Cueto, S., Jacoby, E., & Kanashiro, H. (2003). Structure and nutritional correlates of individual variability in neonatal temperament. Paper presented to the Society for Research in Child Development, Tampa. Wachs, T. D., Sigman, M., Bishry, Z., Moussa, W., Jerome, N., Neumann, C., Bwibo, N., & McDonald, M. (1992). Caregiver child interaction patterns in two cultures in relation to nutrition. International Journal of Behavioral Development, 15, 1–18. Walka, H., Triana, N., Jahari, A., Husaini, M., & Pollitt, E. (2000). Effects of an energy and micronutrient supplement on play behavior in undernourished children in Indonesia. European Journal of Clinical Nutrition—Supplement, 54, S91–S106. Wasserman, G. (1995). Effects of early lead exposure: Time to integrate and broaden our efforts. Neurotoxicology and Teratology, 17, 243–244. Wasserman, G., Graziano, J., Factor-Litvak, D., Morna, N., Muni-Redjepi, E., Hadzialjevic, S., Slavkovich, V., Kline, J., Shrout, P., & Stein, Z. (1992). Independent effects of lead exposure and iron deficiency anemia on developmental outcome at age 2 years. Journal of Pediatrics, 121, 695–703. Wasserman, G., Graziano, J., Factor-Litvak, P., Popovac, D., Morina, N., Musabegovic, A., Vrenezi, N., Capuni-Paracka, S., Lekic, V., & Preteni-Redjepi, E. (1994). Consequences of lead exposure and iron supplementation on childhood development at age 4 years. Neurotoxicology and Teratology, 16, 233–240. Wasserman, G. A., Liu, X., Pine, D. S., & Graziano, J. H. (2001). Contribution of maternal smoking during pregnancy and lead exposure to early child behavior problems. Neurotoxicology & Teratology, 23, 13–21. Wasserman, G., Musabegovic, A., Liu, X., Kline, J., Factor-Litvak, P., & Graziano, J. (2000). Lead exposure and motor functioning in 4 ½-year-old children: the Yugoslavia prospective study. Journal of Pediatrics, 137, 555–561. Wasserman, G., Staghezza-Jaramillo, B., Shrout, P., Popovac, D., & Graziano, J. (1998). The effect of lead exposure on behavior problems in preschool children. American Journal of Public Health, 88, 481–486. Watkins, W., & Pollitt, E. (1996). Effect of removing Ascaris on the growth of Guatemalan school children. Pediatrics, 97, 871–876. Watkins, W., & Pollitt, E. (1997). Stupidity or worms: Do intestinal worms impair mental performance? Psychological Bulletin, 121, 171–191. Watkins, W., & Pollitt, E. (1998). Iron deficiency and cognition among school aged children. In Nutrition, health and child development. Pan American Health Organization


409

Scientific Monograph No. 566 (pp. 179–197). Washington, DC: Pan American Health Organization. Watts, S., Brieger, W., & Yacoob, M. (1989). Guinea worm: An in-depth study of what happens to mothers, families and committees. Social Science & Medicine, 29, 1043–1049. Wauben, P. M., & Wainright, P. E. (1999). The influence of neonatal nutrition on behavioral development: A critical appraisal. Nutrition Reviews, 57, 35–44. Webster, A., Bamford, J., Thyer, N., & Ayles, R. (1989). The psychological, educational and auditory sequelae of early persistent secretory otitis media. Journal of Child Psychology and Psychiatry, 30, 529–546. Werner, E. (1990). Protective factors in individual resilience. In S. Meisels & J. Shonkoff (Eds.), Handbook of early childhood intervention (pp. 97–116). Cambridge: Cambridge University Press. Willatts, P., Forsyth, J., DiModugno, M., Varma, S., & Colvin, M. (1998). Effect of long-chain polyunsaturated fatty acids in infant formula on problem solving at 10 months of age. The Lancet, 352, 688–691. Winick, M., Meyer, K., & Harris, R. (1975). Malnutrition and environmental enrichment by early adoption. Science, 190, 1173–1175. Winneke, G. (1995). Lead in child development: Uncertainties, possibilities and explanations. Neurotoxicology and Teratology, 17, 245–247. Winneke, G., Lilienthal, H., & Kramer, U. (1996). The neurobehavioral toxicology and teratology of lead. Archives of Toxicology Supplement, 18, 57–70. Witkin, H., & Berry, J. (1975). Psychological differentiation in cross-cultural perspectives. Journal of Cross-Cultural Psychology, 6, 4–87. Wolf, A., Jimenez, E., & Lozoff, B. (1994). No evidence of developmental ill effects of low level lead exposure in a developing country. Journal of Developmental and Behavioral Pediatrics, 15, 224–231. Wolke, D., Skuse, D., & Mathiesen, B. (1990). Behavioral style in failure to thrive infants. Journal of Pediatric Psychology, 15, 237–243. Wroble, M., Mash, C., Williams, L., & McCall, R. (2002). Should long chain polyunsaturated fatty acids be added to infant formula to promote development? Applied Developmental Psychology, 23, 99–112. Yehuda, S., Rabinovitz, S., Casasso, R., & Mosotofsky, D. (1998). Fatty acids and brain peptides. Peptides, 19, 416–419. Yip, R. (1989). The interaction of lead and iron. (pp. 179–182). In L. Filer (Ed.), Dietary iron: Birth to two years. New York: Raven.


PATHWAYS TO EARLY LITERACY: THE COMPLEX INTERPLAY OF CHILD, FAMILY, AND SOCIOCULTURAL FACTORS

Megan M. McClelland DEPARTMENT OF HUMAN DEVELOPMENT AND FAMILY SCIENCES, OREGON STATE UNIVERSITY CORVALLIS, OREGON 97330

Maureen Kessenich DEPARTMENT OF PEDIATRICS PERINATAL CENTER/NEONATAL DEVELOPMENTAL FOLLOW-UP CLINIC, LOYOLA UNIVERSITY MEDICAL CENTER MAYWOOD, ILLINOIS 60153

Frederick J. Morrison DEPARTMENT OF PSYCHOLOGY UNIVERSITY OF MICHIGAN ANN ARBOR, MICHIGAN 48109

I. INTRODUCTION A. VIEWING DEVELOPMENT FROM A DYNAMIC SYSTEMS PERSPECTIVE B. THE NATURE OF EARLY LITERACY DEVELOPMENT C. THE EARLY EMERGENCE OF VARIATION IN CHILDREN’S LITERACY SKILLS II. CHILD FACTORS AND EARLY LITERACY DEVELOPMENT A. IQ B. LANGUAGE AND PHONOLOGICAL SKILLS C. SOCIAL SKILLS D. LEARNING-RELATED SOCIAL SKILLS AND ACADEMIC ACHIEVEMENT E. TEMPERAMENT III. PARENTING AND EARLY LITERACY DEVELOPMENT A. FAMILY LEARNING ENVIRONMENT B. COGNITIVE STIMULATION C. PARENTING STYLE D. THE IMPACT OF PARENTING ON EARLY LITERACY SKILLS: A COMPREHENSIVE MODEL IV. SOCIOCULTURAL FACTORS AND EARLY LITERACY DEVELOPMENT A. SOCIOECONOMIC STATUS


411


412

Megan M. McClelland et al. B. RACE/ETHNICITY C. CHILD CARE

V. SCHOOLING INFLUENCES AND CHILDREN’S EARLY LITERACY DEVELOPMENT VI. DYNAMIC RELATIONS BETWEEN CHILD, FAMILY, AND SOCIOCULTURAL FACTORS, SCHOOLING INFLUENCES AND EARLY LITERACY SKILLS A. A COMPLEX MODEL OF CHILD LITERACY ACQUISITION B. INTERACTIONS BETWEEN CHILD, FAMILY, AND SOCIOCULTURAL FACTORS, AND EARLY LITERACY SKILLS C. INTERVENING RELATIONS BETWEEN CHILD, FAMILY, AND SOCIOCULTURAL FACTORS AND EARLY LITERACY SKILLS D. THE INTERPLAY BETWEEN CHILD TEMPERAMENT, PARENTING, AND LEARNING-RELATED SOCIAL SKILLS E. CHILD–SCHOOLING INTERACTIONS IN CHILDREN’S EARLY LITERACY GROWTH F. COMBINING VARIABLE-BASED AND PERSON-ORIENTED ANALYTIC METHODS IN EARLY LITERACY DEVELOPMENT VII. CONCLUSION REFERENCES

I. Introduction Modern conceptualizations in developmental science suggest that our understanding of children’s early growth and learning will be enhanced by viewing development from a dynamic, multilevel, and interactive framework (Cairns, Elder, & Costello, 1996; Gottlieb, Wahlsten, & Lickliter, 1998; Thelen & Smith, 1998). As a consequence, research has begun to move away from a focus on primarily microanalytic, laboratorybased methodologies to incorporate a more applied and dynamic ecological model of development. This paradigm shift has stemmed, in part, from a growing appreciation of the complex relations that exist between many levels of influence that shape children’s development. As the need to see development from a multilevel, interactive framework has grown, researchers have sought ways to bring this perspective to life. This has led to advances in methodologies and analytic techniques, which have allowed researchers to examine complex relations among different factors affecting children’s development. In the present chapter we use the ecological and dynamic system perspectives as a framework for describing the nature and sources of children’s early literacy development. We discuss important child, family, sociocultural, and schooling factors influencing children’s literacy acquisition

Pathways to Early Literacy

413

such as children’s social skills and temperament, the family learning environment and aspects of parenting, socioeconomic status, and race/ ethnicity, and the effect that classroom instruction has on children’s literacy skills. We outline the multiple pathways to literacy development as well as the dynamic relations among these factors. Finally, we provide examples of how the pathways to literacy can be described using a combination of variable-based and person-oriented analytic procedures. A. VIEWING DEVELOPMENT FROM A DYNAMIC SYSTEMS PERSPECTIVE

At the end of the 20th century, the field of developmental psychology underwent a paradigm shift, moving away from a static view of human development and toward a dynamic, interactive, and multilevel framework (Bronfenbrenner, 1989; Gottlieb, Wahlsten, & Lickliter, 1998; Thelen & Smith, 1998). This change in conceptualization arose partly from limitations in seeing development as a set of universal and unchanging properties and from a growing appreciation of context and environmental influences shaping development (Cairns, Elder, & Costello, 1996; Morrison & Ornstein, 1996). For example, research focusing on aspects of children’s cognitive skills such as memory, literacy, and academic skills indicate that culture and context play a large role in the trajectory of children’s learning and that growth in these areas depends in part on being in a formalized educational setting (Ceci & Roazzi, 1994; Rogoff, 1998). Moreover, environmental influences on children’s development are well documented. Researchers have demonstrated the importance for children’s early literacy development of family factors such as parenting and the family learning environment (Griffin & Morrison, 1997; Hart & Risley, 1995; McClelland, Morrison, & Holmes, 2000; Morrison & Cooney, 2002), sociocultural factors such as socioeconomic status (SES) and race/ ethnicity (Bachman, Morrison, & Bryant, 2002; Jencks & Phillips, 1998), and schooling influences such as amount and type of instruction (Freese et al., 2002; McDonald Connor, Morrison, & Katch, 2002). As theoretical perspectives have embraced dynamic and ecological views of development, researchers have focused on variables that interact on multiple levels to determine development. Appreciation of the complex relations among variables has challenged researchers to find ways to adequately capture the complexity of development that incorporates these multilevel relations. In the remainder of the chapter we examine the nature and sources of early literacy acquisition as a way to illustrate the multiple child, family, and sociocultural factors influencing development. In addition, the focus on

414

Megan M. McClelland et al.

literacy will illuminate some of the complex dynamic pathways shaping development that emerge from the application of recent variable-based and person-oriented strategies. B. THE NATURE OF EARLY LITERACY DEVELOPMENT

It is clear that significant numbers of children and adults do not acquire the literacy and numeracy skills needed for success in school and in the workplace (Mullis & Jenkins, 1990; Rayner et al., 2001; Steinberg, 1996; US Department of Education, 1991). Moreover, substantial variability in children’s early literacy skills emerges even before formal schooling begins (Alexander & Entwisle, 1988; Hart & Risley, 1995; Morrison, Griffith, & Williamson, 1993; Morrison et al., 1995; Shonkoff & Phillips, 2000; Stevenson et al., 1976; Weinert & Helmke, 1998). As a consequence, a solution to America’s literacy problems must address the mosaic of interrelated forces in the child, family, school, and sociocultural environment that shape early literacy acquisition well before children enter kindergarten. Accumulating evidence points to a number of characteristics of children that predict later academic achievements such as IQ (Morrison, Griffith, & Williamson, 1993; Seigel, 1981; Smith et al., 1972), language and phonological skills (Rayner et al., 2001), learning-related social skills (Agostin & Bain, 1997; Bronson, Tivnan, & Seppanen, 1995; Frosch et al., 1998; Green & Francis, 1988; McClelland et al., 2000) and temperament (Kagan, 1998; Rothbart & Bates, 1998). Furthermore, parenting factors such as cognitive stimulation and the family learning environment, parental warmth, sensitivity, and responsivity, as well as control and discipline are related to children’s early cognitive and language growth and academic performance at school entry (Bradley & Caldwell, 1984; Coates & Lewis, 1984; Estrada et al., 1987; Gottfried, 1984; Hess et al., 1984; Kessenich & Morrison, 2002; Morrison & Cooney, 2002; Olson, Bates, & Bayles, 1984; Roberts, Burchinal, & Durham, 1999). Sociocultural factors such as socioeconomic status, race/ethnicity, as well as amount and quality of child care have also been linked to children’s preschool cognitive and language skills (Bachman, Morrison, & Bryant, 2002; NICHD Early Child Care Research Network, 2000; Walker et al., 1994). Moreover, researchers have begun to demonstrate relations between the quality of early schooling and amount of direct instruction, and children’s literacy outcomes (Freese et al., 2002). Emerging research indicates that it is the complex interactions among these sources of variability that combine to influence children’s early literacy development (Molfese, DiLalla, & Lovelace, 1996; Morrison et al., 2002). The impact of child, family, schooling, and sociocultural factors on children’s literacy skills is best understood within the context of each of the other


415

imposing factors. Once we identify the complex array of mediational and moderational relations among these various influences, we will have a more comprehensive appreciation of the origins of children’s literacy development. C. THE EARLY EMERGENCE OF VARIATION IN CHILDREN’S LITERACY SKILLS

Early elementary school children vary considerably in their academic competence (Alexander & Entwisle, 1988; Morrison, Griffith, & Williamson, 1993; Shonkoff & Phillips, 2000; Stevenson, Chen, & Lee, 1993). By the time children complete first grade, they are already demonstrating a broad range of skills in subjects such as vocabulary, reading, general knowledge, and mathematics. Morrison, Griffith, and Williamson (1993) discovered that individual differences in children’s early literacy skills either remain unchanged or are magnified as children progress through early elementary school. Yet few researchers have focused on the degree of variability in early literacy skills prior to school entry. Morrison, Griffith, and Williamson (1993) also detected substantial individual differences in vocabulary comprehension at kindergarten entry, with age equivalencies on the Peabody Picture Vocabulary Test ranging from 2 to 11 years of age. Using a large national sample of children from the NICHD Study of Early Child Care and Youth Development, Raviv, Kessenich, and Morrison (2002) found substantial individual differences in cognitive and language skills as early as 2 and 3 years of age. For example, at age 3, developmental age equivalencies on the Reynell Expressive Language and Vocabulary Comprehension subscales ranged from 1 to 5 years of age. Moreover, scores on the Bracken Basic Concepts Scale, which measures recognition of letters, numbers, shapes, and colors, also demonstrated a wide range of skill levels. Also notable was the stability in cognitive and language skills found between 2 and 3 years of age. Twenty-four-month Bayley and 36-month Bracken scores, which both measure cognitive skills, were strongly correlated (r ¼ .53, p < .0001), as were 24-month MacArthur (a language measure) and the 36-month Reynell Developmental Language subscales (r ¼ .32 for Expressive Language, and r ¼ .41 for Receptive Language, p < .0001). In summary, sizeable individual differences in children’s early literacy skills emerge before school entry and are reasonably stable from 2 to 3 years (Shonkoff & Phillips, 2000). Clearly, the next important question centers on the sources of this early emerging variability. In fact, a growing body of research indicates a variety of characteristics in the child, family, school, and sociocultural environment that combine to shape children’s early literacy outcomes (Morrison et al., 1995; Shonkoff & Phillips, 2000). It is important to explore both the individual and interactive influences that

416


these various factors have on children’s academic skills in order to successfully address the nation’s literacy problem.

II. Child Factors and Early Literacy Development A child’s individual characteristics exhibit a large influence on his or her literacy development and help determine whether that child will make a successful transition to kindergarten. Historically, the discussion of child factors affecting early literacy development has focused on cognitive characteristics such as intelligence and IQ, as well as other factors such as language and phonological skills (Adams, 1990; Rayner et al., 2001). However, other factors such as temperament and social skills also contribute to literacy development (Kessenich & Morrison, 2002; McClelland & Morrison, 2003; Morrison et al., 2002). A. IQ

A large body of evidence has documented the relation between child IQ and cognitive, literacy, and academic skills (Morrison, Griffith, & Williamson, 1993; Seigel, 1981; Smith et al., 1972). For example, Morrison, Griffith, and Williamson (1993) found that children’s IQ exhibited a strong influence on reading, vocabulary, general knowledge, and mathematics between kindergarten and second grade. In addition, IQ at 24 months (measured with the Bayley Scales of Infant Development—II) was significantly predictive of cognitive and language skills at 36 months (measured with the Bracken Basic Concept Scale and the Reynell Developmental Language Scales; Kessenich & Morrison, 2002). B. LANGUAGE AND PHONOLOGICAL SKILLS

There is a substantial literature documenting the influence of children’s language and phonological skills in children’s reading and literacy acquisition (Adams, 1990; Hart & Risley, 1995; Rayner et al., 2001; Snow, Burns, & Griffin, 1998). As we have already noted, there are large individual differences in children’s language and vocabulary skills prior to school entry (Hart & Risley, 1995; Stipek & Ryan, 1997). This variability in language skills has important implications for children’s phonological awareness and, in turn, for learning to read: phonological awareness is the most important predictor of early reading skills (Adams, 1990; Rayner et al., 2001). Phonological awareness is defined as the extent to which a child recognizes the internal structure of words and can perform tasks such as


417

identifying the beginning or ending sounds of words. Children who have strong phonological skills have an easier time learning to read than children with weak phonological skills (Rayner et al., 2001). The relation between a child’s phonological skill and reading ability is bidirectional. Children’s phonological awareness enables them to learn to read more easily and as they are exposed to instructions in spelling and sound, they continue to refine their understanding of phonology. Moreover, reading programs that provide direct instruction in phonological skills can improve children’s reading skills (Blachman, 1989; Wise, Ring, & Olson, 1999). C. SOCIAL SKILLS

A growing body of research has indicated the importance of children’s early social behavior on school adaptation and achievement (DeRosier, Kupersmidt, & Patterson, 1994; Dishion, 1990; Ladd, 1990; Ladd & Price, 1987). Children entering school with poor social behavior often have a plethora of problems including peer rejection, behavior problems, and low levels of academic achievement (Alexander, Entwisle, & Dauber, 1993; Cooper & Farran, 1988; McClelland, Morrison & Holmes, 2000). In addition, teachers report that children come into school with differing levels of social skills and that these skills are critical for early school success (Foulks & Morrow, 1989). For example, in one study some teachers reported at least 50% of children entering kindergarten did not have the basic social competencies needed to do well in school, such as following directions, working independently, and having adequate academic skills (RimmKaufman, Pianta, & Cox, 2000). Most research focusing on children’s early social behavior and school achievement has concentrated on social behavior in general, without specifying the aspects of social behavior that are especially important in school performance. However, increasing evidence suggests that aspects of children’s learning-related social skills, which tap the domains of independence, self-regulation, responsibility, and cooperation, are particularly important for early school performance and the transition to school (Bachman & Morrison, 2002a; Cooper & Farran, 1988; Cooper & Speece, 1988; McClelland, Morrison & Holmes, 2000). D. LEARNING-RELATED SOCIAL SKILLS AND ACADEMIC ACHIEVEMENT

There has been an increased interest in how to define learning-related social skills with researchers from a number of theoretical perspectives labeling

418


these skills differently. For example, some of the terms used include executive functioning skills (Bronson, 2000; Karmiloff-Smith, 1993), self-regulation (Bronson, 2000; Shonkoff & Phillips, 2000), mastery skills (Bronson, 1994; Bronson, Tivnan, & Seppanen, 1995), and social competence (Rose-Krasnor, 1997; Wentzel, 1991, 1993). Although these terms come from a variety of perspectives, they reflect a similar constellation of skills and encompass a number of behaviors relating to attention, self-regulation, independence, organization, and cooperation. For simplicity, in the present chapter we use the term learning-related social skills to describe behaviors such as listening and following directions, participating appropriately in groups (such as taking turns), staying on task, and organizing work materials (Cooper & Farran, 1991; McClelland, Morrison & Holmes, 2000). Existing research has pointed to the importance of children’s learningrelated social skills for early school success and school adjustment. For example, Ladd, Birch, and Buhs (1999) found that children’s classroom participation and their ability to be cooperative and independent in kindergarten was an important predictor of early school achievement. In addition, Bronson, Tivnan, and Seppanen (1995) found that prekindergarten children who spent more time uninvolved in the classroom and had difficulty with rules or the teacher scored lower on a standardized cognitive achievement measure. These children also exhibited more risk indicators such as family problems, lower parental education, and behavioral or emotional problems. Once children make the transition to school, learning-related social skills continue to be linked to a child’s academic success. These early skills can be said to ‘‘set the stage’’ for later social behavior and academic performance by providing the foundation for positive classroom behavior. In a study examining the relation between classroom behavior and school performance, Alexander, Entwisle, and Dauber (1993) found that children who were interested in classroom activities and were able to focus and pay attention performed significantly better on academic outcomes in the first grade and fourth grade. In addition, McClelland, Morrison, and Holmes (2000) studied the unique contribution of learning-related social skills to children’s academic achievement at the beginning of kindergarten and at the end of second grade. They found that learning-related skills uniquely predicted literacy and academic outcomes at both time points after controlling for the effects of children’s IQ, age at school entrance, amount of preschool experience, ethnicity, parents’ education, and family learning environment. These investigators also examined characteristics of those children with poor learning-related skills, and the relation of poor learning-related skills


419

to academic achievement at school entry and at the end of second grade. Children with poor learning-related skills were found to differ from the overall sample on a number of child, family, and sociocultural variables including: significantly lower IQs, more behavior difficulties, poorer family learning environments, and more medical problems such as hearing and language problems. Finally, children with low learning-related skills scored lower on academic outcomes at the beginning of kindergarten and at the end of second grade, and learned at significantly slower rates than their peers between school entry and second grade on measures of reading recognition and mathematics (McClelland, Morrison & Holmes, 2000). They continued to perform poorly on reading and mathematics at the end of sixth grade and fell increasingly more behind their peers in reading and mathematics between kindergarten and sixth grade (McClelland & Hansen, 2001). Children with poor learning-related skills evidently start formal schooling behind their peers on many literacy indices and continue to perform at lower levels in reading and mathematics between kindergarten and sixth grade. These results also support the importance of learning-related social skills at the beginning of school and continuing to sixth grade (McClelland, Morrison & Holmes, 2000; McClelland & Hansen, 2001). Given evidence pointing to the importance of children’s learning-related social skills for early literacy and academic achievement, it is important to look at factors that influence the development of early social behavior. One factor that has emerged as being particularly salient for children’s social skills is child temperament. E. TEMPERAMENT

Rothbart and Bates (1998) have defined temperament as a subset of personality that describes individual differences in self-regulation, emotionality, motor activity, attention, and reactivity, which are relatively stable over time. Most of the research in temperament has looked at the relations between children’s temperament and social behavior. The literature has not found much evidence for direct links between children’s temperament and early literacy skills, but instead has focused on support for indirect and interacting relations between temperament, social behavior, and literacy or academic achievement (e.g., Rothbart & Bates, 1998). To better understand how children’s temperament and social behavior may indirectly or interactively influence literacy skills, it is useful to examine the relation between temperament and social behavior. Temperament is linked consistently to social behavior and adjustment (Kagan, 1998; Rothbart & Bates, 1998). In one study, Rothbart, Ahadi, and

420


Hershey (1994) examined relations between measures of temperament and social behaviors defined by empathy, guilt/shame, aggression, help-seeking, and negativity for 80 6- to 7-year-olds. Children scoring high on temperamental traits such as irritability, anger, and discomfort exhibited more antisocial behavior in elementary school (Rothbart, Ahadi, & Hershey, 1994). The results of this study and other research (see Rothbart & Bates, 1998) have linked temperament to children’s social adjustment and have also demonstrated how aspects of temperament can be linked to children’s social skills as they develop over the preschool and early school years. One temperamental component, emotion regulation, is specifically related to the development of social skills (Kopp, 1989; Shonkoff & Phillips, 2000). Emotion regulation, defined as the ability to cope with high levels of positive and negative emotions (Kopp, 1989), is related to social adjustment (Rubin et al., 1995). In addition, researchers have argued that a child’s self-regulation is an important aspect of social adjustment (Bronson, 2000; Kopp, 1982, 1989, 1991; Shonkoff & Phillips, 2000). Selfregulation has been identified as occurring when a child ‘‘goes along with the caregiver expectations in the absence of external monitors’’ (Kopp, 1989, p. 350). The link between temperament and literacy skills is probably complex, and likely works through a child’s social skills. For example, McClelland (2002) found that the effortful control temperament dimension (characterized by behaviors relating to inhibitory control, attention, and perceptual sensitivity) was related to early learning-related social skills in 3- to 5-yearolds, with the relation strengthening over time. Taken together, the contribution of children’s temperament and social skills to early literacy development is complex and most likely involves interacting and/or intervening relations. Although there are direct links between cognitive development and literacy skills, and between social skills and literacy skills, examining interacting and intervening relations as well as other factors such as temperament provides for a more complex, multilevel, and interactive framework describing children’s literacy development.

III. Parenting and Early Literacy Development Parenting comprises a constellation of factors such as parental style, warmth/sensitivity, control/discipline, cognitive stimulation, and the family learning environment that combine to shape academic growth through a variety of direct and indirect pathways (Collins et al., 2000; Morrison & Cooney, 2002). To better understand children’s literacy development, it is necessary to explore the effects these parenting factors have on children’s emerging skills.


421

A. FAMILY LEARNING ENVIRONMENT

The most obvious aspect of parenting to influence early literacy outcomes is the family learning environment. The family learning environment was originally defined in the literature by the frequency of parent–child book reading (Bus, van Ijzendoorn, & Pellegrini, 1995), but it has been extended to include measures such as the number of reading materials at home (e.g., newspapers, child and adult magazines and books), frequency of library visits, parents’ independent reading, duration of parent–child shared book reading, and frequency of nonliteracy related activities such as TV viewing (Griffin & Morrison, 1997). The family learning environment is predictive of children’s early literacy skills in preschool and early elementary school. For example, research by Morrison and colleagues found that the literacy environment (e.g., number of reading materials at home, frequency of library visits, parents’ independent reading and parent–child shared book reading, and frequency of nonliteracy related activities) predicts children’s reading and vocabulary skills, but not mathematics skills, at age 5 (Griffin & Morrison, 1997; Morrison & Cooney, 2002). In addition, Teale (1986) has documented an association between the family learning environment and children’s literacy development; and Payne, Whitehurst, and Angell (1994) found that the family learning environment explained 12–19% of the variance in children’s language skills. Many researchers have documented the relation between one component of the family learning environment, namely parent–child book reading, and later literacy skills (Bus, van Ijzendoorn, & Pellegrini, 1995; Haden, Reese, & Fivush, 1996; Havlik & Haden, 2002; Reznick, 1997; Whitehurst et al., 1994). During book reading, parents have the opportunity to engage in various behaviors that facilitate learning, such as labeling, open-ended questioning, and elaboration. Parents who engage in such activities foster better literacy development in their children (Haden, Reese, & Fivush, 1996; Havlik & Haden, 2002; Whitehurst et al., 1994). For example, in a study by Whitehurst et al. (1994), Head Start children whose parents participated in a book-reading intervention performed better in tests of emergent literacy skills. In a study of 3- to 5-year-olds, Haden, Reese, and Fivush (1996) identified three distinct maternal styles of reading—Describers, Comprehenders, and Collaborators—and these styles were related differentially to children’s later literacy skills at age 6 for unfamiliar books. ‘‘Describer mothers’’ emphasized descriptions of objects and characters in the story. ‘‘Comprehender mothers’’ embellished and expanded on indirectly specified information, linking the text to real world knowledge and experiences. ‘‘Collaborator mothers’’ combined frequent higher-level comments relating to inferences, predictions, and print knowledge with

422


some lower-level descriptive comments. Children with Comprehender mothers scored higher on a measure of story comprehension as compared to children of mothers who were Describers or Collaborators. Children with Collaborator mothers scored higher on the WRAT, which assesses letter and word recognition and pronunciation. And children with Describer mothers scored lower on measures of receptive vocabulary, word recognition, and story comprehension. Thus, different maternal styles during mother–child book reading apparently are associated with distinct literacy outcomes in young children. B. COGNITIVE STIMULATION

Maternal cognitive stimulation is associated directly with children’s early cognitive, academic, and language abilities (Bornstein, 1985; Bradley & Caldwell, 1984; Coates & Lewis, 1984; Elardo, Bradley, & Caldwell, 1975; Hess et al., 1984; Kessenich & Morrison, 2002; Landry et al., 1997; Olson, Bates, & Bayles, 1984; Siegel, 1981; Tamis-LeMonda & Bornstein, 1989). Cognitive stimulation refers to activities such as labeling, scaffolding, and elaboration, and has been measured using tools such as the HOME Inventory checklist (Caldwell & Bradley, 1984) and observations of parent– child structured interactions. Using the HOME Inventory, which measures the general cognitive environment in the home using a standard checklist, Bradley and Caldwell (1984) found that Total HOME scores at 6, 12, and 24 months correlated with IQ scores at age 3 (rs ¼ .50, .58, and .71; p < .05) and age 4½ years (rs ¼ .44, .53, and .57; p < .05). The association between the HOME Inventory and children’s cognitive and language competencies during the preschool years has been documented in numerous other studies. Furthermore, the HOME continues to predict developmental outcomes from first grade through age 10 (Bee et al., 1982; Bradley, Caldwell, & Rock, 1988; Elardo, Bradley, & Caldwell, 1975). Whereas the HOME Inventory assesses components of the family learning environment other than direct parental cognitive stimulation (e.g., number of books and educational toys at home), a new structured interaction observation measure was used in the context of the NICHD Study of Early Child Care and Youth Development in order to evaluate cognitive stimulation independent of these other aspects of the family learning environment. Using data from the NICHD study, Kessenich and Morrison (2002) found that this observation measure of cognitive stimulation, which was based on a structured play interaction between mother and child at age 2, significantly predicted cognitive and language outcomes at age 3. Thus, measures of the home environment that assess behaviors such as parental


423

involvement and stimulation demonstrate strong relations to both cognitive and language skills, even at 2 and 3 years of age. In attempting to understand the relation between parenting practices and early literacy, researchers have focused almost exclusively on the family learning environment and cognitive stimulation as predictors of literacy outcomes. These are important influences, but children’s academic outcomes are also affected by other parental factors that, on the surface, might not have such obvious links to literacy, such as parenting style. C. PARENTING STYLE

The concept of parenting style was originally developed by Baumrind (1971) and comprised of three distinct types: authoritarian, authoritative, and permissive parenting. These dimensions were identified on the basis of varying degrees of parental warmth and control. Maccoby and Martin (1983) subsequently expanded this configuration by distinguishing between permissive parenting that was indulgent versus neglectful. Most studies with Caucasian-American samples and ethnically diverse samples as a whole have demonstrated a positive relation between authoritative parenting and higher levels of academic achievement (Dornbusch et al., 1987; Glasgow et al., 1997; Lamborn et al., 1991; Steinberg et al., 1994). In contrast, in studies of Asian, Asian-American, and Black populations in isolation better school performance has been associated positively with more authoritarian parenting (Chao, 1994; Darling & Steinberg, 1993; Dornbusch et al., 1987). Researchers have proposed several hypotheses regarding the source of these ethnic differences. One view is that more authoritarian parenting practices may be adopted by socioeconomically disadvantaged Black families living in unsafe neighborhoods in order for parents to better ensure the safety of their children by insisting on strict obedience and adherence to rules. With regard to Asian and Asian-American parenting, Chao (1994) pointed to the emphasis on authoritarian-like principles such as chiao shun and guan in the Chinese culture. The concept of chiao shun refers to ‘‘training’’ children in appropriate or expected behaviors, and the term guan means ‘‘to govern’’ as well as ‘‘to love.’’ Thus, parental control and authority appear to be synonymous with expressions of love and concern in Asian cultures (Chao, 1994). Furthermore, Darling and Steinberg (1993) hypothesized that differences in parenting style may be linked to cultural variations in the goals parents have toward socializing their children. There is much research to be done in order to fully understand the differences in effective parenting styles across cultures, as well as across various ages.

424


Little research has attempted to look at the effect of parenting style on early literacy prior to school entry. Such research might lead to interesting findings regarding the learning patterns that develop in early childhood as a result of various parenting styles and levels of control and responsivity. The impact of parenting style on literacy can also be understood by examining the two components—warmth and control—separately. 1. Parental Warmth/Sensitivity/Responsivity Parenting style is, in part, defined by various levels of warmth, sensitivity, and emotional responsivity, as measured by positive, affectionate, responsive, and nonintrusive behaviors of mothers toward their children. Such behaviors have emerged as salient influences on children’s cognitive and language outcomes. Parenting behaviors have been reliably examined using structured measures such as the HOME Inventory checklist (Caldwell & Bradley, 1984) as well as through observed interactions between parents and children. Interaction patterns between mothers and children aged 12–48 months have been identified as significant predictors of concurrent cognitive and language skills as well as subsequent academic achievement (Estrada et al., 1987; Hess et al., 1984; Kessenich & Morrison, 2002; Murray & Hornbaker, 1997; Olson, Bates, & Kaskie, 1992). Coates and Lewis (1984) found that a mother’s responsivity and sensitivity to the distress of her 3-month-old infant accounted for more than 25% of the variance in the child’s verbal IQ score at 6 years of age. In addition, Kessenich and Morrison (2002) determined that an observational measure of maternal warmth and sensitivity predicted equally significant amounts of variance in Bracken Basic Concept scores and Reynell Developmental Language scores at 3 years of age as compared to a measure of maternal cognitive stimulation. Both cognitive stimulation and warmth/sensitivity were assessed using qualitative ratings of explicit maternal behaviors (1 ¼ Not at all characteristic, 4 ¼ Highly characteristic) observed during a structured play interaction between mother and child. Cognitive stimulation was defined as behaviors such as describing, labeling, or asking questions about toys, objects, attributes of objects, or experiences; sensitivity was characterized in terms of positive emotional regard, lack of intrusive behavior, and awareness of the child’s affect, interests, or response to stimulation. In general, research suggests that parenting qualities such as warmth, sensitivity, and responsivity during the first few years of life are related to children’s later cognitive and language development. In order to establish the mechanisms through which maternal warmth and sensitivity influence children’s cognitive and language outcomes, it is necessary to investigate the


425

role of mediating factors. For example, warm, sensitive parenting may give young children a sense of security, stability, and self-assurance which in turn enables them to comfortably explore their world—providing them with a ‘‘secure base’’ from which to interact and learn from their environment (Ainsworth et al., 1978; Bowlby, 1988; Main & Solomon, 1986). 2. Parental Control and Discipline Parental control is the second of two components incorporated within Baumrind’s representation of parenting style. A large part of parenting has to do with the methods used to manage and discipline a child. Setting and maintaining consistent rules and limits helps to provide a supportive, structured environment in which children can develop. For example, Morrison and Cooney (2002) demonstrated an indirect relation between children’s literacy skills and parental control/discipline by way of a child’s learning-related social skills (e.g., listening and following directions, cooperation, independence, self-regulation). Thus, a child’s ability to listen and follow directions, mediated the relation between parental control/ discipline and a child’s acquisition of literacy skills. It is posited that higher levels of consistent, authoritative parental control promote more cooperation, compliance, and independence in young children. These learningrelated social behaviors, in turn, enable children to acquire the important literacy skills they need to succeed in the classroom (McClelland, 2002; McClelland, Morrison & Holmes, 2000). D. THE IMPACT OF PARENTING ON EARLY LITERACY SKILLS: A COMPREHENSIVE MODEL

Several parenting factors have demonstrated significant direct or indirect associations with children’s literacy skills. Yet evidence suggests that specific dimensions of parenting may be differentially related to various literacy skills. Morrison and Cooney (2002) used an a priori structural equation model to examine different aspects of parenting and their influence on children’s literacy and social skills. Four aspects of parenting were assessed: Parental Warmth and Responsiveness, the Strength of Parental Beliefs about Child Qualities, Parental Control, and the Quality of the Family Learning Environment. Parenting factors were significantly associated with children’s academic and social skills at the beginning of kindergarten. For example, the Quality of the Family Learning Environment was related to children’s vocabulary, general knowledge, reading, and mathematics skills; Parental Warmth and Responsiveness was associated with children’s vocabulary and general knowledge skills; and the Strength of Parental

426


Beliefs about Child Qualities was related to children’s learning-related social skills at the beginning of kindergarten. These results suggest that multiple pathways simultaneously influence children’s literacy growth. For example, Parental Warmth and Responsiveness was related to children’s vocabulary and general knowledge skills, skills which are not directly taught in the classroom like reading and mathematics, but which are indirectly taught by parents. In addition, the Quality of the Family Learning Environment, which assesses characteristics of the home such as number of books owned, and hours of TV watched by the child, was related to all of the academic outcomes. Thus, a global conceptualization of parenting is useful for specifying how aspects of parenting affect literacy skills differentially, making it important to examine the dynamic and complex paths through which parenting impacts children’s early literacy development.

IV. Sociocultural Factors and Early Literacy Development A. SOCIOECONOMIC STATUS

Socioeconomic status (SES) has long been considered as an important background variable when conducting research on human development. Traditional indicants of SES, including income, education level, and occupational status, have continually demonstrated associations with children’s cognitive, language, and literacy development (Brooks-Gunn, Duncan, & Britto, 1999; Dodge, Petit, & Bates, 1994; Duncan, BrooksGunn, & Klebanov, 1994; Lempers, Clark-Lempers, & Simons, 1989; Smith & Dixon, 1995). For example, in a meta-analysis of over 100 studies, the average correlation between SES indices and children’s IQ scores was .40, and the average correlation between SES and verbal achievement was .31 (White, 1982). When the focus is shifted to literacy skills per se, SES remains influential. In a study by Walker et al. (1994), SES indices such as maternal education, family income, and occupational status were significantly associated with expressive language at 36 months, and receptive language through third grade. Furthermore, a review of numerous longitudinal studies determined that income level predicted children’s verbal and intelligence test scores from age 2 to 5, even after controlling for other family characteristics such as parental education level and family structure (Brooks-Gunn, Duncan, & Britto, 1999). In addition, Duncan, Brooks-Gunn, and Klebanov (1994) found that income-to-needs ratio and maternal education level were significant predictors of children’s IQ at age 5. Hart and Risley (1995) discovered that the size and richness of children’s vocabularies at age 3


427

varied substantially as a function of their socioeconomic status. Finally, Stipek and Ryan (1997) found that economically advantaged preschoolers outperformed economically disadvantaged kindergartners on a number of cognitive tests. Although the association between SES and developmental outcomes is well documented, the particular mechanism by which SES exerts its influence is less clear. A distal factor such as socioeconomic status may shape literacy outcomes by influencing more proximal factors at home or to the child, such as parental behaviors and children’s learning-related social skills. Yet, research rarely integrates proximal child and parenting factors and distal socioeconomic variables into a comprehensive conceptual model. Such a model would better clarify the relations among child, parenting, and socioeconomic characteristics that combine to influence children’s literacy development. B. RACE/ETHNICITY

Significant racial differences in academic skills such as vocabulary, general knowledge, mathematics, and reading are evident among Black and White children at the elementary and high school levels (Applebee, Langer, & Mullis, 1989; Bachman, Morrison, & Bryant, 2002; Jencks & Phillips, 1998; Phillips, Crouse, & Ralph, 1998; Stevenson, Chen, & Uttal, 1990). Even at kindergarten entry, Black children demonstrate poorer literacy skills than White children, and this discrepancy is either maintained or magnified over the school year (Cooney, 1999). Yet after controlling for variables such as parental education, children’s IQ, and the family learning environment, the unique effect of race on kindergarten reading, mathematics, and general knowledge scores disappears, which is consistent with research on school-aged children (Alexander & Entwisle, 1988; Stevenson, Chen, & Uttal, 1990). Thus, it is possible that socioeconomic factors and the family learning environment may mediate the effect of race on literacy outcomes. However, Bachman, Morrison, and Bryant (2002) found that the family learning environment significantly mediated the effect of parental education on children’s literacy skills for White families, but not for Black families. In Black families, the mediational relation broke down in terms of the weaker link between parental education level and the family learning environment. Whereas 79% of White parents with higher levels of education (more than 12 years) reported average or above average family learning environment scores, only 21% of Black parents with higher education levels scored average or above average on the family learning environment questionnaire.

428


Research comparing Asian and American schoolchildren has also demonstrated differences in academic outcomes, with Asian and AsianAmerican children outperforming their White and Black peers (Stevenson, Chen, & Lee, 1993; Stevenson, Lee, & Stigler, 1986). Again, this discrepancy is thought to be due, in part, to differences in parental behaviors and the home environment. For example, Asian and Asian-American parents exhibit greater control and involvement over their children’s educational development than White or Black parents (Chao, 1994; Dornbusch et al., 1987; Jose et al., 1995; Steinberg, Dornbusch, & Brown, 1992). In summary, significant differences in children’s literacy skills clearly exist across various racial and ethnic groups (Cooney, 1999; Jencks & Phillips, 1998). What is not yet clear is whether these differences are the result of variations in socioeconomic factors, parenting behaviors, cultural beliefs, or a combination of these influences. Further research is needed in order to decipher the myriad of possible mediational and moderational relations among this complex array of factors.

C. CHILD CARE

The number of young children in child care settings increased dramatically at the end of the 20th century, with almost 75% of children involved in full or part-time care. In studies examining the effects of day care on children’s outcomes, research has often focused on the amount of time spent in child care. However, quality of child care, as indicated by caregiver–child ratios, group size, resources, environment, and caregiver training, is likely to be more revealing. Several studies have investigated the impact of early child care on later literacy outcomes. Although some researchers have found no significant associations between quantity or quality of child care and later cognitive and academic outcomes (Chin-Quee & Scarr, 1994; Howes, 1988; Larsen, Hite, & Hart, 1983), others have found that involvement in child care can have a compensatory effect on the outcomes of at-risk children from disadvantaged families (Burchinal, Lee, & Ramey, 1989; Christian, Morrison, & Bryant, 1998; Desai, Chase-Lansdale, & Michael, 1989; Golden et al., 1978; McCartney, 1984; NICHD Early Child Care Research Network, 2000). For example, O’Brien Caughey, DiPietro, and Strobini (1994) found that children from impoverished environments who were involved in child care had higher mathematics and reading scores than impoverished children who were not involved in child care. After distinguishing between various types of child care (e.g., center-based, home-based, family care), O’Brien Caughey, DiPietro, and Strobini (1994) discovered that this compensatory effect was specific to


429

center-based care for children from impoverished homes. However, O’Brien Caughey, DiPietro, and Strobini (1994) also discovered that children from middle-class backgrounds who were enrolled in child care during infancy exhibited poorer reading skills in kindergarten than their homereared peers. This discrepancy in the compensatory effects of child care is believed to be due to variations in the amount of cognitive stimulation and learning opportunities that children receive in child care, with impoverished children experiencing more than they would at home, and children from middle-class families receiving less than they would at home (Desai, Chase-Lansdale, & Michael, 1989). Similar results were obtained by Christian, Morrison, and Bryant (1998): the amount of time spent in child care was positively associated with kindergarten mathematics skills, but only for children from low literacy environments with less educated mothers. Thus, the effect of child care on children’s mathematics skills was moderated by the level of maternal education and the quality of the family learning environment. Unfortunately, information regarding the quality of the child care was not gathered for this study. In the NICHD Study of Early Child Care and Youth Development (2000), at-risk children from poor home environments who were enrolled in day care performed better on cognitive and language measures than children raised at home. In addition to quantity of child care, quality of child care—based on characteristics such as caregiver–child ratios, group size, resources, environment, and caregiver training—has been shown to relate to children’s later cognitive, language, and social skills. McCartney (1984) found that children enrolled in child care programs with poorer environments and resources and fewer language learning opportunities demonstrated inferior language skills as compared with children from better child care environments. Ruopp et al. (1979) discovered that children from child care environments with better trained teachers were more cooperative and more persistent on assigned activities, and obtained higher scores on a standardized measure of preschool skills. Furthermore, Ruopp et al. (1979) ascertained that children from child care settings with smaller group sizes performed better on tests of kindergarten and first-grade readiness. Finally, Burchinal et al. (2000) found that higher quality child care was related to higher scores on measures of cognitive and language development, even after adjusting for important child and family factors. Specifically, child care settings with professionally recommended caregiver–child ratios had children with better language skills, and those that met recommendations regarding teacher education had girls with better cognitive and receptive language skills. In summary, both the amount and the quality of child care influence children’s cognitive, literacy, and language outcomes. Amount of child care

430


is associated with positive influences on later cognitive, academic, and literacy outcomes for children from disadvantaged homes, yet this positive influence is not as clear for children from more economically advantaged homes. Quality of child care, as measured by teacher training, group size, resources and environment, has also demonstrated significant relations with children’s literacy-related outcomes.

V. Schooling Influences and Children’s Early Literacy Development Thus far, our discussion of pathways to early literacy acquisition has centered on relations between child, family, and sociocultural influences and children’s literacy skills. However, given the large variability in children’s literacy skills prior to formal schooling, it is important to document what happens to children’s skills once they enter kindergarten. The influence that schooling and classroom instruction have on children’s literacy development is another level of influence that must be examined within the ecological paradigm of early literacy development. For example, schooling may reduce, maintain, or increase the variability in children’s literacy skills evident before children enter kindergarten. If schooling and classroom instruction exerts a positive influence on children’s literacy development then it may be possible to decrease the amount of variability in children’s skills and ensure that more children succeed academically. Obviously, schooling exerts a strong effect on children’s literacy skills. For example, classrooms differ substantially in the amount and type of time spent on instructional activities, which directly affects children’s early literacy development (Pianta et al., 2002; Pressley et al., 1998). In addition, the type of instruction combined with the individual characteristics of a child exerts a strong impact on the development of children’s literacy skills (McDonald Connor, Morrison, & Katch, 2002). This variability across classrooms is evident in study by Freese et al. (2002) of 58 kindergarten and first-grade classrooms within a single school district. Classrooms—particularly those in kindergarten—differed substantially in the time spent on noninstructional activities. One kindergarten classroom was engaged in noninstructional activities for an average of 83 minutes per day, whereas another kindergarten classroom spent an average of 167 minutes in noninstructional activities. Differences in time spent on noninstructional activities were smaller in first grade, although still apparent. In addition, substantial variability was found for time spent on language arts and mathematical instruction across kindergarten and first grade. For


431

example, one kindergarten teacher spent an average of 20 minutes per day teaching language arts while another kindergarten teacher spent an average of 90 minutes per day on language arts instruction. In the first grade, one teacher spent an average of 57 minutes per day teaching language arts while another teacher spent an average of 134 minutes teaching language arts (Freese et al., 2002). These findings suggest that kindergarten and first grade students in the same school district were receiving quite different educational experiences. While some children were spending more time engaged in instructional activities, particularly language arts activities, other children were spending considerably more time in noninstructional activities such as transition time and behavior management. These results demonstrate that in addition to other important child, family, and sociocultural factors, children’s developmental trajectories in literacy skills may differ based on the amount and type of instruction given in kindergarten and first grade.

VI. Dynamic Relations Between Child, Family, and Sociocultural Factors, Schooling Influences and Early Literacy Skills In general, research has primarily focused on separate links between child, family, and sociocultural factors and children’s literacy acquisition (Kessenich & Morrison, 2002; Morrison et al., 2002). However, to describe and explain pathways in children’s early literacy development, researchers must examine complex relations between child, family, and sociocultural factors and children’s literacy skills. A growing body of evidence indicates that children’s literacy acquisition is the result of dynamic direct, indirect, and interacting relations between these variables. Moreover, advances in analytic tools such as structural equation modeling (SEM) and hierarchical linear modeling (HLM) allow researchers to better examine multiple factors influencing literacy skills and the complex relations among them. Thus, research that incorporates a dynamic systems perspective with new analytic tools furthers the understanding of literacy development and the processes underlying children’s academic trajectories. A. A COMPLEX MODEL OF CHILD LITERACY ACQUISITION

Bachman and Morrison (2002a) identified important child, family, and sociocultural factors in order to develop a complex model of child literacy acquisition using a sample of 382 kindergarten children (see Figure 1). They

432


Fig. 1. Model of child, family, and sociocultural factors related to child literacy outcomes at kindergarten. Direct effects of race, parental education, and family learning environment on the four child outcomes are not included in the figure for clarity of presentation.

found that seven core child, family, and sociocultural variables accounted for a large portion of the variance across four literacy outcomes of reading recognition, receptive vocabulary, general knowledge, and mathematics. These variables included three child factors (IQ, school entrance age, and learning-related social skills), one family factor (family learning environment), and three sociocultural factors (race/ethnicity, amount of preschool experience and parents’ education). Specifically, at the beginning of kindergarten, these seven variables accounted for 64% of the variance in children’s receptive vocabulary skills, 55% of the variance in general knowledge skills, 34% of the variance in reading recognition skills, and 49% of the variance in mathematics skills. Using these seven variables, Bachman and Morrison (2002a) used SEM to test a model predicting children’s literacy skills at the beginning of kindergarten (see Figure 1). SEM is a family of statistical techniques that define and estimate general models of variables including charting cause and effect relations (Klem, 2000). When examining direct effects between predictors and outcomes, using SEM is analogous to conducting a series of simultaneous multiple regressions and allows researchers to directly estimate measurement error and increase the power of the analysis compared with more basic regression techniques (Kline, 1998).


433

In addition, SEM allows for examination of indirect (mediator) and interactive (moderator) relations among variables in the model. Results of the Bachman and Morrison (2002a) study revealed that the seven variables had both direct and indirect influences on children’s literacy skills at kindergarten (see Figure 1). For example, direct relations were found between parents’ education, children’s IQ, learning-related social skills, entrance age, and the four literacy outcomes of receptive vocabulary, general knowledge, reading, and mathematics. Direct links were also found between race/ethnicity and receptive vocabulary and general knowledge; between the family learning environment and all literacy skills except mathematics; and between the amount of preschool experience and mathematics skills. All relations were positive with the exception of race/ ethnicity, where being African-American was related to lower receptive vocabulary and general knowledge scores. In addition to direct effects, race/ethnicity and parents’ education influenced child literacy skills indirectly through other variables (see Figure 1). For example, race/ethnicity operated indirectly through the family learning environment, child’s IQ, and children’s learning-related social skills to influence children’s literacy skills. In addition, parents’ education operated through children’s IQ and the family learning environment to influence literacy skills. Together, the results from this study demonstrate that the pathways to child literacy skills are complex and involve multiple variables (Bachman & Morrison, 2002a). B. INTERACTIONS BETWEEN CHILD, FAMILY, AND SOCIOCULTURAL FACTORS, AND EARLY LITERACY SKILLS

Researchers have also found evidence for complex interacting relations between child, family, and sociocultural factors and children’s early literacy skills. For example, McClelland, Morrison, and Bryant (2000) found that parents’ education interacted with children’s learning-related skills to increase children’s vocabulary and general knowledge skills at kindergarten. In addition, a child factor, IQ, interacted with children’s learning-related skills to influence reading skills at the fall of kindergarten. These interactions were found to have an augmenting effect in which high learning-related skills were associated with stronger academic performance when combined with high levels of IQ and parents’ education. Further analysis of the interactions between parents’ education, learning-related skills, and literacy outcomes suggested that high learning-related skills helped children in academic areas such as vocabulary and general knowledge, which are not directly focused on in the school environment,

434


but are learned in the home environment and are affected by parents’ education. Less firm interpretations could be made about the interaction between IQ and learning-related skills and reading because there was little actual difference in the reading scores of children with high or low learning-related skills and a high or low IQ. In addition, children’s age equivalents between the groups for reading scores were similar. Having high learning-related skills and a high IQ may help children listen and sit still when learning about reading more than just having a high IQ. However, the interaction also showed that children with high learning-related skills and a low IQ did similarly to children with low learning-related skills, and a high IQ. Having a high IQ may help children even if they have low learning-related social skills, but if children had a lower IQ, having strong learning-related skills helped them perform better on reading in kindergarten. This study found that high learning-related skills and parents’ education were associated with stronger vocabulary and general knowledge skills in kindergarten. The interaction between learning-related social skills and children’s IQ on reading skills in kindergarten indicated that having either a high IQ or strong learning-related social skills helped children perform better on reading. Overall, results from this study add to our understanding of variables influencing literacy skills and help to characterize the relations between child and family factors and early literacy acquisition. C. INTERVENING RELATIONS BETWEEN CHILD, FAMILY, AND SOCIOCULTURAL FACTORS AND EARLY LITERACY SKILLS

Although some research has documented interactions, or moderational relations, among child, family, and sociocultural variables, other studies have demonstrated mediational, or intervening, relations between such variables and children’s early literacy-related skills. Raviv, Kessenich, and Morrison (2002) found that the influence of socioeconomic factors such as income-to-needs ratio and maternal education on children’s expressive and receptive language skills at 3 years of age was partially mediated by parenting behaviors such as cognitive stimulation and sensitivity. Furthermore, cognitive skills at age 3 partly mediated the relation between parental behaviors (i.e., cognitive stimulation and sensitivity) and 3-year-old expressive and receptive language skills. This study provides additional evidence of the complex and indirect relations linking parenting, sociocultural factors and children’s early literacy development.


435

D. THE INTERPLAY BETWEEN CHILD TEMPERAMENT, PARENTING, AND LEARNING-RELATED SOCIAL SKILLS

In addition to studying the complex factors contributing to literacy development, research has found evidence for dynamic relations between child and family factors and children’s social skills. As discussed previously, children’s learning-related social skills are significant predictors of children’s academic and literacy outcomes (McClelland, Morrison & Holmes, 2000). For example, McClelland (2002) examined the emergence of children’s early learning-related social skills in preschool and found that the relation between aspects of parenting, child temperament, and child social skills is complex and changes over time. In the study, the effortful control dimension of temperament, which comprises of behaviors relating to inhibitory control, attention, and perceptual sensitivity, moderated, or interacted with, parental warmth to influence children’s learning-related skills. When parental warmth was high, a child’s temperament was not related to his or her learning-related skills, but when parental warmth was low, effortful control became an important indicator of learning-related skills at 3–4 years of age. The longitudinal nature of the study highlighted how the relation between child temperament, parenting, and children’s learning-related social skills changed with age. The moderating interaction between child temperament and parenting found when children were 3–4 years old changed to a mediated or indirect effect when children and their families were studied again one year later, when children were 4–5 years of age. A child’s effortful control significantly mediated the relation between the parental warmth and learning-related skills at 4–5 years of age after controlling for background characteristics such as child age, parents’ education, preschool experience, and ethnicity. Specifically, at ages 4 and 5, higher levels of parental warmth were related to higher levels of effortful control in children, which were then related to higher learning-related skills in children (McClelland, 2002). These results demonstrate how characteristics of children and parenting are part of dynamic and complex pathways of development. Although a moderational interaction between child temperament and parental warmth on children’s learning-related skills was found when children were 3–4 years old, this relation changed to be mediational when children and their families were studied one year later. Taken together, the complex relations between child and family factors suggest that the pathways to children’s early social and literacy development are dynamic and change over time.

436


E. CHILD–SCHOOLING INTERACTIONS IN CHILDREN’S EARLY LITERACY GROWTH

In addition to links between child, family, and sociocultural factors and literacy skills, different instructional practices benefit children in differing ways depending on their entering language and reading skills. For example, in a study of growth in word-decoding in first grade, McDonald Connor, Morrison, and Katch (2002) examined the impact of three instructional dimensions: teacher-managed versus child-managed, explicit versus implicit, and changes in instruction over the school year. Different instructional variables were most effective for children with specific combinations of skills upon entering first grade. Specifically, for children with low vocabulary and low word-decoding skills, the more teacher-managed explicit instruction they received in first grade, the higher their spring scores. Conversely, the more child-managed explicit instruction they received, the worse they performed. Finally, teachers who started with low amounts of child-managed implicit instruction in the fall but increased steadily in winter and spring produced higher performance in their low vocabulary, low word-decoding children. A contrasting pattern emerged for children who started first grade with high vocabulary and high word-decoding skills. For these children, increasing amounts of teacher-managed explicit instruction had no discernible effect, but increasing amounts of child-managed implicit instruction yielded higher spring first grade scores. Finally, steady amounts of child-managed implicit instruction produced greater spring worddecoding scores. Thus, children’s growth in word-decoding skills in first grade apparently depends on the match between the child’s level of beginning skill and the type of instruction presented by their teacher. One interpretation is that the teacher’s change in instruction is in response to the child’s progress, but the reverse may also be true and children are responding to changes in teacher’s instruction. Regardless of the direction of the association, this finding is consistent with the view that children’s pathways to early literacy acquisition may depend on a ‘‘goodness-of-fit’’ match of children’s characteristics with the type and amount of instruction provided by the teacher (Foorman et al., 1998; Juel & Minden-Cupp, 2000). Although the concept of goodness-of-fit has historically been used to describe the fit between a child’s temperament or characteristics and a parent’s characteristics (e.g., Rothbart & Bates, 1998), it is also relevant in describing children’s literacy skills trajectories. For example, children who started first grade with low vocabulary and word-decoding skills benefited the most over the year when paired with teachers who used


437

more teacher-managed explicit instruction. In contrast, children who came to first grade with strong vocabulary and word-decoding skills showed more growth in word-decoding when allowed more child-managed implicit instruction. Thus, classroom instruction that can better match the needs of children and that takes into account children’s entering vocabulary and word-decoding skill level can optimize children’s learning over the course of the year. F. COMBINING VARIABLE-BASED AND PERSON-ORIENTED ANALYTIC METHODS IN EARLY LITERACY DEVELOPMENT

Advances in statistics have allowed researchers to increasingly rely on analytic strategies that combine variable-based and person-oriented procedures to better examine the dynamic relations between child, family, and sociocultural factors and how they influence early literacy skills. Variable-based procedures are correlational methods that attribute significant associations between variables to a general linear model for all individuals in a sample (Bachman & Morrison, 2002b; Magnusson & Bergman, 1988). Examples of variable-based analytic procedures are regression analyses, discriminant analysis, SEM, and growth curve analyses using hierarchical linear modeling. A limit of variable-based procedures is that significant relations may be due to extreme individual scores rather than reflecting a general pattern for the sample as a whole. Consequently, many researchers advocate the use of person-oriented approaches in addition to variable-based procedures (e.g., Bachman & Morrison, 2002b; Magnusson & Bergman, 1988; Roesner, Eccles, & Sameroff, 1998). Person-oriented procedures focus on locating meaningful subgroups of individuals and examining relations among individuals. Examples of person-oriented procedures include clustering methods, latent class analysis, and the homogeneous grouping strategy. A particularly useful strategy is to combine variable-based procedures with person-oriented methods. For example, Morrison et al. (2002) used the homogeneous grouping strategy, a person-oriented procedure, and HLM, a variable-based method, to examine the unique and interactive influences of child and family variables on children’s literacy development. The homogeneous grouping strategy is a person-oriented procedure that creates subgroups or clusters of individuals prior to data analysis based on theoretically driven questions. In contrast, HLM is a variablebased method that incorporates multiple levels of analysis, modeling individual growth curves at the first level, and variations in individual’s growth curves using other predictor variables at the second level

438


(Bryk & Raudenbush, 1992; Weinfurt, 2000). When combining the homogeneous grouping strategy with HLM, individuals are divided into similar subgroups on the basis of theoretically driven questions and the growth curves of those subgroups on outcome variables are then charted (Morrison et al., 2002). In Morrison et al.’s (2002) study, the influences of children’s IQ and family learning environment scores were examined to determine their effect on children’s early literacy skills from kindergarten through the end of second grade. Using the homogeneous grouping strategy, two similar groups of children with higher and lower IQ scores were created and then divided again on the basis of high versus low family learning environment scores. This strategy was then combined with growth curve analysis using HLM to examine the independent and combined influence of children’s IQ and family learning environment on the growth of reading recognition, mathematics, receptive vocabulary, and general knowledge skills between kindergarten and second grade. Morrison et al. (2002) found that children’s literacy skills followed specific developmental pathways. For example, although children’s IQ and family learning environment exerted strong influences on child literacy skills, the pattern of influence was not the same. There was an additive effect of child’s IQ and family learning environment on children’s vocabulary and reading, but an interactive effect between children’s IQ and family learning environment on children’s general knowledge skills at kindergarten. In this interaction, the effect of a higher family learning environment on children’s general knowledge skills was stronger for higher IQ children than for lower IQ children at kindergarten. In addition, different relations were found for the influence of children’s IQ and family learning environment on children’s literacy growth over the first 3 years of schooling. For general knowledge, only a child’s family learning environment affected growth rates, with children in higher family learning environments growing more rapidly in general knowledge than children in lower family learning environments. In contrast, a distinct pattern emerged for children’s growth in vocabulary where only children’s IQ affected growth rates. Lower IQ children grew faster in receptive vocabulary between kindergarten and second grade compared with higher IQ children, whereas the influence of the family learning environment was maintained over time but did not predict growth rates. As a result, vocabulary in the higher and lower IQ children had converged slightly by the end of second grade. A separate pattern of results also emerged for growth in children’s reading skills. Both IQ and family learning environment affected growth rates: children with higher IQs grew more in reading than children with lower IQs,


439

and children from higher family learning environments experienced more growth in reading than children from lower family learning environments. Finally, for mathematics, only IQ affected growth rates during the first 3 years of school, where children with higher IQs grew at a faster rate in mathematics than children with lower IQs. The results of the Morrison et al.’s (2002) study point to the specific and complex influence of child and family factors on the growth of children’s early literacy skills. Both children’s IQ and the family learning environment strongly influenced children’s literacy skills, but the pattern of the influence on children’s growth differed across the literacy skills. For reading, both IQ and family learning environment influenced growth, whereas for general knowledge, growth rates were due mostly to the family learning environment. In addition, for mathematics, only child IQ affected growth rates between kindergarten and second grade. These findings point to the high degree of specificity present in the developmental trajectories of children’s early literacy skills, which were influenced by children’s IQ and the family learning environment. This pattern of results would not have been uncovered if the data were analyzed from just a person-oriented procedure such as the homogeneous grouping strategy, or solely from a variable-based method such as growth curve analyses.

VII. Conclusion Children’s early literacy development is not a static or one-dimensional process but instead involves complex relations between multiple levels of influence. At one level, child, family, sociocultural factors influence literacy skills directly as well as more complex ways involving interacting and intervening variables. At another level, aspects of instructional influences such as the amount and type of instruction influence literacy development. Specifically, pathways to early literacy acquisition differ based on a child’s characteristics such as IQ, language, social skills, and temperament; family characteristics such as the quality of the family learning environment and parenting; sociocultural factors such as parents’ education and ethnicity; and instructional dimensions such as teacher-managed versus childmanaged, explicit versus implicit, and changes in instruction over the school year. Furthermore, the optimal literacy development may involve a match between child, family, and sociocultural characteristics, and the type of instruction provided by the teacher. Research must utilize advances in analytic techniques that reflect variablebased approaches such as SEM and growth curve analyses (HLM), as

440


well as person-oriented procedures such as the homogeneous grouping strategy. Moreover, the analytic techniques that show the most promise are those that combine variable-based and person-oriented procedures such as the homogeneous grouping strategy combined with HLM. Together, these methodological and analytic techniques are important tools that can help disentangle the multiple layers of influence in literacy development.

ACKNOWLEDGMENT The research presented in this chapter was completed with funding from grants NICHD R01-HD27176 and NSF BCS-0111754 to Frederick J. Morrison.

REFERENCES Adams, M. J. (1990). Beginning to read: Thinking and learning about print. Cambridge, MA: MIT Press. Agostin, T. M., & Bain, S. K. (1997). Predicting early school success with developmental and social skill screeners. Psychology in the Schools, 34, 219–228. Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, NJ: Erlbaum. Alexander, K. L., & Entwisle, D. R. (1988). Achievement in the first 2 years of school: Patterns and processes. Monographs of the Society for Research in Child Development, 53 (2, Serial No. 218). Alexander, K. L., Entwisle, D. R., & Dauber, S. L. (1993). First-grade classroom behavior: Its short- and long-term consequences for school performance. Child Development, 64, 801–814. Applebee, A. N., Langer, J. A., & Mullis, I. N. S. (1989). Crossroads in American education: A summary of findings. Princeton: NJ. Educational Testing Service. Bachman, H. J., & Morrison, F. J. (2002a). Early literacy: Toward a comprehensive model. Unpublished manuscript. Bachman, H. J., & Morrison, F. J. (2002b, April). Comparing variable-based and person-oriented procedures directly: SEM vs. clustering. Paper presented at the biennial Conference on Human Development, Charlotte, NC. Bachman, H. J., Morrison, F. J., & Bryant, F. B. (2002). Beyond social class: Home literacy promotion as a proximal source of Black-White differences in academic skills at kindergarten entry. Unpublished manuscript. Baumrind, D. (1971). Current patterns of parental authority. Developmental Psychology Monograph, 4, 1–103. Bee, H. L., Barnard, K. E., Eyres, S. J., Gray, C. A., Hammond, M. A., Spietz, A. L., Snyder, C., & Clark, B. (1982). Prediction of IQ and language skill from perinatal status, child performance, family characteristics, and mother–infant interaction. Child Development, 53, 1134–1156. Blachman, B. A. (1989). Phonological awareness and word recognition: Assessment and intervention. In A. G. Kamhi & H. W. Catts (Eds.), Reading disabilities: A developmental language perspective (pp. 133–158). Boston: Little, Brown.


441

Bornstein, M. H. (1985). How infant and mother jointly contribute to developing cognitive competence in the child. Proceedings of the National Academy of Sciences, 82, 7470–7473. Bowlby, J. (1988). A secure base: Parent–child attachment and healthy human development. New York: Basic Books. Bradley, R. H., & Caldwell, B. M. (1984). 174 children: A study of the relationship between home environment and cognitive development during the first 5 years. In A. W. Gottfried (Ed.), Home environment and early cognitive development: Longitudinal research (pp. 329–342). Orlando, FL: Academic Press. Bradley, R. H., Caldwell, B. M., & Rock, S. L. (1988). Home environment and school performance: A ten-year follow-up and examination of three models of environmental action. Child Development, 59, 852–867. Bronfenbrenner, U. (1989). Ecological systems theory. In R. Vasta (Ed.), Six theories of child development: Revised formulations and current issues (pp. 185–246). London, UK: Jessica Kingsley Publishers. Bronson, M. B. (1994). The usefulness of an observational measure of young children’s social and mastery behaviors in early childhood classrooms. Early Childhood Research Quarterly, 9, 19–43. Bronson, M. B. (2000). Self-regulation in early childhood: Nature and nurture. New York, NY: The Guilford Press. Bronson, M. B., Tivnan, T., & Seppanen, P. S. (1995). Relations between teacher and classroom activity variables and the classroom behaviors of prekindergarten children in Chapter 1 funded programs. Journal of Applied Developmental Psychology, 16, 253–282. Brooks-Gunn, J., Duncan, G. J., & Britto, P. R. (1999). Are socioeconomic gradients for children similar to those for adults? Achievement and health of children in the United States. In D. P. Keating & C. Hertzman (Eds.), Developmental health and the wealth of nations: Social, biological, and educational dynamics (pp. 94–124). New York: The Guilford Press. Bryk, A. S., & Raudenbush, S. W. (1992). Hierarchical linear models: Applications and data analysis methods. Newbury Park, CA: Sage Publications. Burchinal, M., Lee, M., & Ramey, C. (1989). Styles of day-care and preschool intellectual development in disadvantaged children. Child Development, 60, 128–137. Burchinal, M. R., Roberts, J. E., Riggins, R., Zeisel, S. A., Neebe, E., & Bryant, D. (2000). Relating quality of center-based child care to early cognitive and language development longitudinally. Child Development, 71, 338–357. Bus, A. G, van Ijzendoorn, M. H., & Pellegrini, A. D. (1995). Joint book reading makes for success in learning to read: A meta-analysis on intergenerational transmission of literacy. Review of Educational Research, 65, 1–21. Cairns, R. B., Elder, G. H., Jr., & Costello, E. J. (1996). Developmental science. New York: Cambridge University Press. Caldwell, B. M., & Bradley, R. H. (1984). Home observation for measurement of the environment. Little Rock: University of Arkansas Press. Ceci, S. J., & Roazzi, A. (1994). The effects of context on cognition: Postcards from Brazil. In R. J. Sternberg (Ed.), Mind in context (pp. 74 –101). New York: Cambridge University Press. Chao, R. K. (1994). Beyond parental control and authoritarian parenting style: Understanding Chinese parenting through the cultural notion of training. Child Development, 65, 1111–1119. Chin-Quee, D. S., & Scarr, S. (1994). Lack of early child care effects on school-age children’s social competence and academic achievement. Early Development and Parenting, 3, 103–112.

442


Christian, M. K., Morrison, F. J., & Bryant, F. B. (1998). Predicting kindergarten academic skills: Interactions among child care, maternal education, and family literacy environments. Early Childhood Research Quarterly, 13, 485–505. Coates, D. L., & Lewis, M. (1984). Early mother–infant interaction and infant cognitive status as predictors of school performance and cognitive behavior in six-year-olds. Child Development, 55, 1219–1230. Collins, W. A., Maccoby, E. E., Steinberg, L., Hertherington, E. M., & Bornstein, M. H. (2000). Contemporary research on parenting: The case for nature and nurture. American Psychologist, 52, 218–232. Cooney, R. R. (1999, April). How early are racial gaps evident in academic achievement? In F. J. Morrison (Chair), Racial differences in academic achievement: When and why? Symposium paper presented at the biennial meeting of the Society for Research in Child Development meeting, Albuquerque, NM. Cooper, D. H., & Farran, D. C. (1988). Behavioral risk factors in kindergarten. Early Childhood Research Quarterly, 3, 1–19. Cooper, D. H., & Farran, D. C. (1991). The Cooper-Farran Behavioral Rating Scales. Brandon, VT: Clinical Psychology Publishing Co., Inc. Cooper, D. H., & Speece, D. L. (1988). A novel methodology for the study of children at risk for school failure. The Journal of Special Education, 22, 186–198. Darling, N., & Steinberg, L. (1993). Parenting style as context: An integrative model. Psychological Bulletin, 113, 487–496. DeRosier, M. E., Kupersmidt, J. B., & Patterson, C. J. (1994). Children’s academic and behavioral adjustment as a function of the chronicity and proximity of peer rejection. Child Development, 65, 1799–1813. Desai, S., Chase-Lansdale, P. L., & Michael, R. T. (1989). Mother or market? Effects of maternal employment on the intellectual ability of 4-year-old children. Demography, 26, 545–561. Dishion, T. J. (1990). The family ecology of boys’ peer relations in middle childhood. Child Development, 61, 874–892. Dodge, K. A., Petit, G. S., & Bates, J. E. (1994). Socialization mediators of the relation between socioeconomic status and child conduct problems. Child Development, 65, 649–665. Dornbusch, S. M., Ritter, P., Leiderman, P., Roberts, D., & Fraleigh, M. (1987). The relation of parenting style to adolescent school performance. Child Development, 58, 1244–1257. Duncan, G., Brooks-Gunn, J., & Klebanov, P. K. (1994). Economic deprivation and early childhood development. Child Development, 65, 296–318. Elardo, R., Bradley, R., & Caldwell, B. M. (1975). The relation of infants’ home environments to mental test performance from six to thirty-six months: A longitudinal analysis. Child Development, 46, 71–76. Estrada, P., Arsenio, W. F., Hess, R. D., & Holloway, S. D. (1987). Affective quality of the mother–child relationship: Longitudinal consequences for children’s school-relevant cognitive functioning. Developmental Psychology, 23, 210–215. Foorman, B. R., Francis, D. J., Fletcher, J. M., Schatschneider, C., & Mehta, P. (1998). The role of instruction in learning to read: Preventing reading failure in at-risk children. Journal of Educational Psychology, 90, 37–55. Foulks, B., & Morrow, R. D. (1989). Academic survival skills for the young child at risk for school failure. Journal of Educational Research, 82, 158–165. Freese, M. K., Morrison, F. J., Griffin, E. A., & Williams, S. (2002). Classroom instruction in kindergarten and first grade: What happened in school today? Unpublished manuscript.


443

Frosch, C. A., Caskie, G. I. L., Cox, M. J., Morrison, F. J., & Goldman, B. D. (1998). Early aggression and the development of verbal skills in pre-kindergarten children. Unpublished manuscript, Center for Developmental Science, University of North Carolina at Chapel Hill. Glasgow, K. L., Dornbusch, S. M., Troyer, L., Steinberg, L., & Ritter, P. L. (1997). Parenting styles, adolescents’ attributions, and educational outcomes in nice heterogeneous high schools. Child Development, 68, 507–523. Golden, M., Rosenbluth, L., Grossi, M., Policare, H., Freeman, H., & Brownlee, E. (1978). The New York City infant day care study. New York: Medical and Health Research Association of New York City. Gottfried, A. W. (1984). Home environment and early cognitive development: Longitudinal research. Orlando: Academic Press. Gottlieb, G., Wahlsten, D., & Lickliter, R. (1998). The significance of biology for human development: A developmental psychobiological systems view. In W. Damon (Series Ed.) & R. M. Lerner (Vol. Ed.), Handbook on child psychology: Vol. 1. Theoretical models of human development (5th ed., pp. 233–273). New York: Wiley. Green, L., & Francis, J. (1988). Children’s learning skills at the infant and junior stages: A follow-on study. British Journal of Educational Psychology, 58, 120–126. Griffin, E. A., & Morrison, F. J. (1997). The unique contribution of home literacy environment to differences in early literacy skills. Early Child Development and Care, 127–128, 233–243. Haden, C. A., Reese, E., & Fivush, R. (1996). Mothers’ extratextual comments during storybook reading: Stylistic differences over time and across texts. Discourse Processes, 2, 135–169. Hart, B., & Risley, T. R. (1995). Meaningful differences in the everyday experience of young American children. Baltimore, MD: Paul H. Brookes Publishing Co. Havlik, B. S., & Haden, C. A. (2002, April). Reading between the lines: Linkages between mother–child bookreading and preschoolers’ language and literacy skills. Paper presented at the biennial Conference on Human Development, Charlotte, NC. Hess, R. D., Holloway, S. D., Dickson, W. P., & Price, G. G. (1984). Maternal variables as predictors of children’s school readiness and later achievement in vocabulary and mathematics in sixth grade. Child Development, 55, 1902–1912. Howes, C. (1988). Relations between early child care and schooling. Developmental Psychology, 2, 53–57. Jencks, C., & Phillips, M. (1998). The Black-White test score gap. Washington, DC: Brookings. Jose, P. E., Huntsinger, C. S., Huntsinger, P. R., & Liaw, F. R. (1995, April). Parental values and practices relevant to young children’s social development in Taiwan and the U.S. Paper presented at the biennial Conference of the Society for Research in Child Development. Juel, C., & Minden-Cupp, C. (2000). Learning to read words: Linguistic units and instructional strategies. Reading Research Quarterly, 35, 458–492. Kagan, J. (1998). Biology and the child. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp. 177–235). New York: Wiley. Karmiloff-Smith, A. (1993). Self-organization and cognitive change. In M. H. Johnson (Ed.), Brain development and cognition (pp. 592–618). Oxford, UK: Blackwell. Kessenich, M. A., & Morrison, F. J. (2002). The ABCs of early literacy development: A model of preschool cognitive and language skills. Unpublished manuscript.

444


Klem, L. (2000). Structural equation modeling. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding more multivariate statistics (pp. 227–259). Washington, DC: American Psychological Association. Kline, R. B. (1998). Principles and practice of structural equation modeling. New York, NY: Guilford Press. Kopp, C. B. (1982). Antecedents of self-regulation: A developmental perspective. Developmental Psychology, 18, 199–214. Kopp, C. B. (1989). Regulation of distress and negative emotions: A developmental view. Developmental Psychology, 25, 343–354. Kopp, C. B. (1991). Young children’s progression to self-regulation. In M. Bullock (Ed.), The development of intentional action: Vol. 22. Cognitive, motivational, and interactive processes (pp. 38–54). Basel: Karger. Ladd, G. W. (1990). Having friends, keeping friends, making friends, and being liked by peers in the classroom: Predictors of children’s early school adjustment? Child Development, 61, 1081–1100. Ladd, G. W., & Price, J. M. (1987). Predicting children’s social and school adjustment following the transition from preschool to kindergarten. Child Development, 58, 1168–1189. Ladd, G. W., Birch, S. H., & Buhs, E. S. (1999). Children’s social and scholastic lives in kindergarten: Related spheres of influence? Child Development, 70, 1373–1400. Lamborn, S. D., Mounts, N. S., Steinberg, L., & Dornbusch, S. M. (1991). Patterns of competence and adjustment among adolescents from authoritative, authoritarian, indulgent, and neglectful families. Child Development, 62, 1049–1065. Landry, S. H., Smith, K. E., Miller-Loncar, C. L., & Swank, P. R. (1997). Predicting cognitivelanguage and social growth curves from early maternal behaviors in children at varying degrees of biological risk. Developmental Psychology, 33, 1040–1053. Larsen, J. M., Hite, S. J., & Hart, C. H. (1983). The effects of preschool on educationally advantaged children: First phases of a longitudinal study. Intelligence, 7, 345–352. Lempers, J. D., Clark-Lempers, D., & Simons, R. L. (1989). Economic hardship, parenting, and distress in adolescence. Child Development, 60, 25–39. Maccoby, E. E., & Martin, J. (1983). Socialization in the context of the family: Parent–child interaction. In P. H. Mussen (Series Ed.) & E. M. Hetherington (Vol. Ed.), Handbook of child psychology: Vol. 4. Socialization, personality, and social development (4th ed., pp. 1–101). New York: Wiley. Magnusson, D., & Bergman, L. R. (1988). Individual and variable-based approaches to longitudinal research on early risk factors. In M. Rutter (Ed.), Studies of psychosocial risk: The power of longitudinal data (pp. 45–61). Cambridge: Cambridge University Press. Main, M., & Solomon, J. (1986). Discovery of an insecure-disorganized/disoriented attachment pattern. In T. B. Brazelton, M. W. Yogman et al. (Eds.), Affective development in infancy (pp. 95–124). Norwood, NJ: Ablex. McCartney, K. (1984). Effects of quality of day care environment on children’s language development. Developmental Psychology, 20, 244–260. McClelland, M. M. (2002). The emergence of work-related skills in preschool children. Unpublished doctoral dissertation, Loyola University Chicago. McClelland, M. M., & Hansen, E. E. (2001, June). A follow-up of children with poor work-related skills: Do problems persist at the end of elementary school? Poster presented at the annual meeting of the Society for the Scientific Study of Reading, Boulder, CO. McClelland, M. M., & Morrison, F. J. (2003). The emergence of learning-related social skills in preschool children. Early Childhood Research Quarterly, 18, 206–224.


445

McClelland, M. M., Morrison, F. J., & Bryant, F. B. (2000, April). Multiple pathways to early academic performance: The role of work-related skills. Poster presented at the biennial Conference on Human Development, Memphis, TN. McClelland, M. M., Morrison, F. J., & Holmes, D. H. (2000). Children at-risk for early academic problems: The role of learning-related social skills. Early Childhood Research Quarterly, 15, 307–329. McDonald Connor, C., Morrison, F. J., & Katch, L. (2002). Beyond the reading wars: Exploring the effect of child-instruction interactions on growth in early reading. Unpublished manuscript. Molfese, V. J., DiLalla, L. F., & Lovelace, L. (1996). Perinatal, home environment, and infant measures as successful predictors of preschool cognitive and verbal abilities. International Journal of Behavioral Development, 19, 101–119. Morrison, F. J., & Cooney, R. R. (2002). Parenting and academic achievement: Multiple pathways to early literacy. In J. Borkowski, S. Ramey, & M. Bristol-Power (Eds.), Parenting and the child’s world: Influences on academic, intellectual, and social-emotional development (pp. 141–160). Mahwah, NJ: Erlbaum. Morrison, F. J., & Ornstein, P. A. (1996). Cognitive development. In R. B. Cairns, G. H. Elder, Jr., & J. Costello (Eds.), Developmental science (pp. 121–134). New York: Cambridge University Press. Morrison, F. J., Griffith, E. M., & Williamson, G. (1993, November). Two strikes from the start: Individual differences in early literacy skills. Paper presented at the annual Head Start Research Conference, Washington, DC. Morrison, F. J., Griffith, E. M., Williamson, G., & Hardway, C. L. (1995, April). The nature and sources of early literacy. Paper presented at the biennial Conference of the Society for Research in Child Development, Indianapolis, IN. Morrison, F. J., Ornstein, P. A., Hardway, C. L., & Pelphrey, K. A. (2002). Charting developmental pathways: A methodological strategy. Unpublished manuscript. Mullis, I. V. S., & Jenkins, L. B. (1990). The reading report card, 1971–1988: Trends from the nation’s report card. Princeton, NJ: Educational Testing Service. Murray, A. D., & Hornbaker, A. V. (1997). Maternal directive and facilitative interaction styles: Associations with language and cognitive development of low risk and high risk toddlers. Development and Psychopathology, 9, 507–516. NICHD Early Child Care Research Network (2000). The relation of child care to cognitive and language development. Child Development, 71, 960–980. O’Brien Caughy, M., DiPietro, J. A., & Strobini, D. M. (1994). Day-care participation as a protective factor in the cognitive development of low-income children. Child Development, 65, 457–471. Olson, S. L., Bates, J. E., & Bayles, K. (1984). Mother–infant interaction and the development of individual differences in children’s cognitive competence. Developmental Psychology, 20, 166–179. Olson, S. L., Bates, J. E., & Kaskie, B. (1992). Caregiver-infant interaction antecedents of children’s school-age cognitive ability. Merrill-Palmer Quarterly, 38, 309–330. Payne, A. C., Whitehurst, G. J., & Angell, A. I. (1994). The role of home literacy environment in the development of language ability in preschool children from low-income homes. Early Childhood Research Quarterly, 9, 427–440. Phillips, M., Crouse, J., & Ralph, J. (1998). Does the black-white test score gap widen after children enter school? In C. Jencks & M. Phillips (Eds.), The Black-White test score gap (pp. 229–272). Washington, DC: Brookings. Pianta, R. C., La Paro, K. M., Payne, C., Cox, M. J., & Bradley, R. (2002). The relation of kindergarten classroom environment to teacher, family, and school characteristics and child outcomes. Elementary School Journal, 102, 225–238.

446


Pressley, M., Wharton-McDonald, R., Mistretta-Hampston, J., & Echevarria, M. (1998). Literacy instruction in 10 fourth- and fifth-grade classrooms in upstate New York. Scientific Studies of Reading, 2, 159–194. Raviv, T., Kessenich, M. A., & Morrison, F. J. (2002). A mediational model of the association between socioeconomic status and preschool language abilities: The role of parent and child factors. Unpublished manuscript. Rayner, K., Foorman, B. R., Perfetti, C. A., Pesetsky, D., & Seidenberg, M. S. (2001). How psychological science informs the teaching of reading. Psychological Science in the Public Interest, 2, 31–74. Reznick, J. S. (1997). Intelligence, language, nature, and nurture in young twins. In R. J. Sternberg & E. L. Grigorenko (Eds.), Intelligence, heredity, and environment (pp. 483–504). New York: Cambridge University Press. Rimm-Kaufman, S. E., Pianta, R. C., & Cox, M. J. (2000). Teachers’ judgments of problems in the transition to kindergarten. Early Childhood Research Quarterly, 15, 147–166. Roberts, J. E., Burchinal, M., & Durham, M. (1999). Parents’ report of vocabulary and grammatical development of African American preschoolers: Child and environmental associations. Child Development, 70, 92–106. Roesner, R. W., Eccles, J. E., & Sameroff, A. J. (1998). Academic and emotional functioning in early adolescence: Longitudinal relations, patterns, and prediction by experience in middle school. Development and Psychopathology, 10, 321–352. Rogoff, B. (1998). Cognition as a collaborative process. In W. Damon (Series Ed.), D. Kuhn & R. Siegler (Vol. Eds.), Handbook on child psychology: Vol. 2. Cognition, perception, and language (5th ed., pp. 679–744). New York: Wiley. Rose-Krasnor, L. (1997). The nature of social competence: A theoretical review. Social Development, 6, 111–135. Rothbart, M. K., & Bates, J. E. (1998). Temperament. In W. Damon (Series Ed.) & N. Eisenberg (Vol. Ed.), Handbook of child psychology: Vol. 3. Social, emotional, and personality development (5th ed., pp. 105–176). New York: Wiley. Rothbart, M. K., Ahadi, S. A., & Hershey, K. L. (1994). Temperament and social behavior in childhood. Merrill-Palmer Quarterly, 40, 21–39. Rubin, K. H., Coplan, R. J., Fox, N. A., & Calkins, S. D. (1995). Emotionality, emotion regulation, and preschoolers’ social adaptation. Development and Psychopathology, 7, 49–62. Ruopp, R., Travers, J., Glantz, F., & Coelen, C. (1979). Children at the center. Cambridge, MA: Abt Associates. Shonkoff, J. P., & Phillips, D. A. (Eds.) (2000). From neurons to neighborhoods: The science of early childhood development. Washington, DC: National Academy Press. Siegel, L. S. (1981). Infant tests as predictors of cognitive and language development at two years. Child Development, 52, 545–557. Smith, S. S., & Dixon, R. G. (1995). Literacy concepts of low- and middle-class four-year-olds entering preschool. Journal of Educational Research, 88, 243–253. Smith, A., Flick, G., Ferriss, G., & Sellmann, A. (1972). Prediction of developmental outcome at seven years from prenatal, perinatal and postnatal events. Child Development, 43, 495–507. Snow, C. E., Burns, M. S., & Griffin, P. (Eds.) (1998). Preventing reading difficulties in young children. Washington, DC: National Academy Press. Steinberg, L. (1996). Beyond the classroom: Why school reform has failed and what parents need to do. New York: Simon & Schuster. Steinberg L., Dornbusch, S. M., & Brown, B. (1992). Ethnic differences in adolescent achievement: an ecological perspective. American Psychologist, 47, 723–729.


447

Steinberg, L., Lamborn, S. D., Darling, N., Mounts, N. S., & Dornbusch, S. M. (1994). Overtime changes in adjustment and competence among adolescents from authoritative, authoritarian, indulgent, and neglectful families. Child Development, 65, 754–770. Stevenson, H. W., Chen, C., & Lee, S. (1993). Mathematics achievement of Chinese, Japanese, and American children: Ten years later. Science, 259, 53–58. Stevenson, H. W., Chen, C., & Uttal, D. H. (1990). Beliefs and achievement: A study of Black, White, and Hispanic children. Child Development, 61, 508–523. Stevenson, H. W., Lee, S., & Stigler, J. W. (1986). Mathematics achievement of Chinese, Japanese, and American children. Science, 231, 693–699. Stevenson, H. W., Parker, T., Wilkinson, A., Hegion, A., & Fish, E. (1976). Longitudinal study of individual differences in cognitive development and scholastic achievement. Journal of Educational Psychology, 68, 377–400. Stipek, D. J., & Ryan, R. H. (1997). Economically disadvantaged preschoolers: Ready to learn but further to go. Developmental Psychology, 33, 711–723. Tamis-LeMonda, C. S., & Bornstein, M. H. (1989). Habituation and maternal encouragement of attention in infancy as predictors of toddler language, play, and representational competence. Child Development, 60, 738–751. Teale, W. H. (1986). Home background and young children’s literacy development. In W. Teale & E. Sulzby (Eds.), Emergent literacy: Writing and reading (pp. 173–205). Norwood, NJ: Ablex. Thelen, E., & Smith, L. B. (1998). Dynamic systems theories. In W. Damon (Series Ed.) & R. M. Lerner (Vol. Ed.), Handbook on child psychology: Vol. 1. Theoretical models of human development (5th ed., pp. 563–634). New York: Wiley. U.S. Department of Education (1991). America 2000: An education strategy. Washington, DC: U.S. Government Printing Office. Walker, D., Greenwood, C. R., Hart, B., & Carta, J. (1994). Prediction of school outcomes based on early language production and socioeconomic factors. Child Development, 65, 606–621. Weinert, F. E., & Helmke, A. (1998). The neglected role of individual differences in theoretical models of cognitive development. Learning & Instruction, 8, 309–323. Weinfurt, K. P. (2000). Repeated measures analyses: ANOVA, MANOVA, and HLM. In L. G. Grimm & P. R. Yarnold (Eds.), Reading and understanding more multivariate statistics (pp. 317–361). Washington, DC: American Psychological Association. Wentzel, K. R. (1991). Relations between social competence and academic achievement in early adolescence. Child Development, 62, 1066–1078. Wentzel, K. R. (1993). Does being good make the grade? Social behavior and academic competence in middle school. Journal of Educational Psychology, 85, 357–364. White, K. R. (1982). The relation between socioeconomic status and academic achievement. Psychological Bulletin, 91, 461–481. Whitehurst, G. J., Epstein, J. N., Angell, A. L., Payne, A. C., Crone, D. A., & Fischel, J. E. (1994). Outcomes of an emergent literacy intervention in Head Start. Journal of Educational Psychology, 86, 542–555. Wise, B. W., Ring, J., & Olson, R. K. (1999). Training phonological awareness with and without attention to articulation. Journal of Experimental Child Psychology, 72, 271–304.


Author Index

A Aadland, J. 186, 204, 217 Aber, J.L. 53, 97 Achenbach, T.M. 45, 95 Acredolo, L. 189, 217 Adams, M.J. 416, 440 Adams, S.E. 199, 218 Adler, B. 381, 404 Affleck, G. 338, 347, 356 Agostin, T.M. 414, 440 Aguayo, V. 396 Aguirre, D.M. 404 Ahadi, S.A. 419, 420, 446 Ahmed, A. 190, 217 Ainsworth, M.D.S. 142, 144, 168, 425, 440 Aitken, D.H. 359 Akaho, R. 403 Aksan, N. 159, 160, 169 Albert, M.L. 282, 320 Alberts, D.M. 187, 219 Albonico, M. 407 Alexander, G.E. 180, 221 Alexander, K. 95, 414, 415, 417, 418, 427, 440 Aligne, A. 400 Allen, C.N. 192, 194, 196, 217 Allen, D.A. 356 Allen, L. 381, 395 Allen, R. 360 Allison, T. 225 Allport, A. 275, 286, 320 Allred, E. 402 Alp, E. 204, 217 Alsaker, F. 69, 95 Alsop, D.C. 219 Amato, J. 366, 399 Ames, L.B. 235, 266 An, C. 223 Anderson, A. 280, 290, 323, 377, 395 Anderson, R.C. 280, 320 Anderson, V. 373, 395

Andrew, J. 398 Angell, A.I. 421, 445 Angell, A.L. 447 Ani, C. 386, 400 Applebee, A.N. 427, 440 Applegate, B. 101 Arber, M. 186, 204, 221 Arendt, R. 406 Armstrong, T. 343, 356 Arsenio, W.F. 442 Asarnow, J.R. 97 Asarnow, R. 406 Asher, S.R. 45, 49, 50, 54, 58, 60, 61, 62, 63, 65, 67, 68, 69, 70, 72, 73, 80, 95, 96, 102, 103 Ashworth, A. 400 Aslin, R.N. 214, 217 Assuncao, A. 400 Astington, J.W. 289, 323 Atkinson, R.C. 176, 217 Atlas, J. 373, 396 Atlas, S. 219 Attig, M. 320 Auinger, P. 400 Aumiller, K. 92, 96 Awh, E. 223, 226 Ayers, P. 401 Ayles, R. 409 Azuma, R. 307, 319, 325

B Baak, K. 103 Bachman, H.J. 413, 414, 417, 427, 431, 432, 433, 437, 440 Backscheider, A. 296, 326 Baddeley, A.D. 176, 177, 179, 201, 207, 214, 217, 221 Baghurst, P. 403 Bagigian, H. 97 449

450

Author Index

Bagwell, C.L. 68, 95 Baillargeon, R. 190, 217, 260, 266 Bain, S.K. 414, 440 Baker-Ward, L.E. 206, 217 Baksh, M. 406 Baldwin, J.M. 176, 217 Bamford, J. 409 Bank, L. 366, 404 Barad, N. 134 Barclay, J. 280, 320 Barker, D. 381, 396 Barnard, K.E. 440 Barnes, R. 391, 392, 402 Barr, R. 374, 396 Barr, R.G. 359 Barraza, V.H. 354, 357 Barresi, J. 241, 268 Barrett, D. 383, 387, 388, 396 Barrett, L.R. 384, 399 Barrett, P. 404 Barrouillet, P. 206, 217 Barsalou, L.W. 280, 320 Bartels, B. 350, 358 Bartolini, R. 396 Bartsch, K. 143, 168, 185, 203, 226 Bassily, N. 403 Basso, A. 282, 325 Bates, E. 213, 217, 220, 232, 266, 298, 324 Bates, J. 99, 368, 396, 414, 419, 420, 422, 424, 426, 436, 442, 445, 446 Bauer, P.J. 285, 309, 322 Bauer, R.H. 180, 221 Baumrind, D. 423, 425, 440 Bayles, K. 414, 422, 445 Beals, D.E. 305, 321 Beardsall, L. 146, 147, 154, 168 Beatty, W.W. 186, 204, 217 Becker, B. 347, 359 Becker, D. 406 Becker, P. 376, 401 Bee, H.L. 422, 440 Beeman, M. 274, 282, 321 Behrman, J. 373, 396 Beilin, H. 5, 12, 33, 38, 39, 41 Belger, A. 181, 217 Bell, M.A. 199, 209, 217, 218, 221 Bell, R.Q. 335, 361 Bell, S.M. 197, 217 Bellinger, D. 378, 379, 380, 396, 402

Belsky, J. 141, 153, 168 Bem, D.J. 48, 56, 60, 87, 96, 100 Bender, B. 374, 396 Beninni, L. 217 Benning, J.J. 52, 98 Benson, B.A. 337, 356 Benson, J.B. 233, 266, 267 Bentin, S. 225 Bently, M. 394, 396, 406 Berch, D. 122, 131, 135 Berger, J. 371, 396 Bergman, L.R. 437, 444 Berkman, D. 376, 396 Berndt, T.J. 61, 68, 70, 93, 95, 100, 104 Berney, B. 377, 396 Bernstein, D. 397 Berry, J. 389, 406, 409 Berry, J.C. 58, 103 Bertenthal, B.I. 198, 200, 218 Berthier, N.E. 201, 218 Best, K. 366, 403 Bhan, M. 387, 406 Bidell, T.R. 206, 220 Bierman, K. 92, 95 Bigelow, A.E. 198, 218 Bihrle, A.M. 321 Bijur, P. 405 Birch, H.G. 350, 360 Birch, S.H. 46, 62, 69, 70, 76, 78, 89, 91, 95, 100, 418, 444 Bird, H.R. 101 Birsh, H. 405 Bishop, D.V. 313, 321 Bishry, Z. 403, 407, 408 Bithoney, W. 380, 396 Bjko¨rkqvist, K. 69, 96 Bjork, E. 203, 218, 219 Bjorklund, D. 364, 396 Bjorkqvist, K. 51, 101 Blachman, B.A. 417, 440 Black, J. 366, 383, 400 Black, M. 382, 385, 393, 394, 396, 397 Blair, E. 324 Blake, R. 222 Blamire, A.M. 223 Blascovich, J. 333, 356, 360 Blehar, S.K. 440 Bloch, G. 223 Bloom, P. 304, 321 Blue, J. 356

Author Index Boag, P.T. 360 Bodnoff, S.R. 359 Boettcher, W. 119, 135 Bogden, J. 377, 397, 401 Bohlin, G. 386, 394, 400 Bohlmann, N. 296, 321 Boivin, M. 65, 71, 76, 96, 376, 394, 397 Bond, L. 395 Bonett, D. 403 Booth, A. 198, 218 Borden, M.G. 102 Bordieu, P. 9, 39 Bornschein, R. 399 Bornstein, M. 365, 367, 397, 407, 422, 442, 447 Borwick, D. 55, 103 Bosquet, M. 196, 210, 225 Boukydis, C.F.Z. 350, 356 Boulton, M.J. 64, 69, 71, 96 Bowe, T.B. 290, 321 Bowker, A. 92, 99 Bowlby, J. 94, 96, 139, 145, 168, 425, 441 Bowman, L.L. 287, 324 Bowman, M. 125, 134, 135 Bowman, P.H. 99 Boyce, W. 374, 396, 397 Bradbury, T. 373, 397 Bradley, R. 364, 365, 366, 397, 414, 422, 424, 441, 442, 445 Brainerd, C.J. 52, 104 Branca, F. 372, 373, 397 Brandenberg, A. 154, 168 Brandimonte, M. 265, 266 Brandt, J. 397 Bransford, J.D. 320 Braun, C. 292, 321 Braver, T.D. 219 Bream, L. 55, 103 Brennan, S.E. 280, 290, 321 Brenneman, K. 291, 322 Brentano, F. 216, 218 Breslau, N. 350, 360 Bretherton, I. 140, 145, 165, 168, 213, 217 Brieger, W. 376, 409 Britner, P.A. 142, 144, 170 Britto, P.R. 426, 441 Broadbent, D.E. 176, 218 Broderick, V. 117, 119, 135 Brody, L.R. 192, 195, 218

Bronfenbrenner, U. 276, 279, 321, 364, 366, 368, 369, 373, 381, 389, 391, 397, 413, 441 Bronson, M.B. 414, 418, 420, 441 Brooks, P.H. 199, 220 Brooks-Gunn, J. 374, 403, 426, 441, 442 Brown, A.L. 188, 219, 236, 266 Brown, B. 366, 367, 397, 407, 428, 446 Brown, J. 146, 147, 154, 168, 169, 344, 356, 370, 397 Brown, J.R. 168 Brown, M. 356 Brown, N.B. 305, 323 Brown, P.J. 59, 76, 103 Brown, R. 280, 321 Brownell, H. 281, 282, 283, 286, 321 Brownlee, E. 443 Bruce, C.J. 180, 195, 206, 221 Bruner, A.B. 382, 397 Bruner, J. 33, 41, 218 Bryant, D. 441 Bryant, F.B. 413, 414, 427, 428, 429, 433, 440, 442, 445 Bryk, A.S. 438, 441 Buchbinder, S.B. 357 Buczowska, E. 269 Bugental, D.B. 334, 338, 339, 347, 349, 351, 354, 356, 357, 358 Buhrmester, D. 61, 62, 68, 96 Buhs, E. 53, 62, 69, 70, 76, 78, 89, 91, 96, 100, 101, 418, 444 Buka, S. 373, 397 Bukowski, W.M. 62, 63, 65, 68, 71, 72, 76, 95, 96, 102 Bullock, J. 63, 103 Bullock, M. 260, 266 Burchinal, M. 374, 405, 414, 428, 429, 441, 446 Burger, L.K. 282, 322 Burgess, K.B. 45, 48, 54, 55, 56, 57, 60, 76, 78, 80, 81, 87, 101, 103 Burgy, L. 68, 95 Burnette, M. 378, 398 Burns, M.S. 106, 135, 416, 446 Burruss, G. 58, 102 Bus, A.G. 421, 441 Busch-Rossnagel, N.A. 335, 360 Bushnell, E.W. 200, 218 Butterworth, G. 202, 218

451

452

Author Index

Bwibo, N. 406, 408 Byrne, B. 124, 134

C Cairns, B.D. 48, 51, 52, 53, 93, 96 Cairns, R.B. 48, 51, 52, 53, 93, 96, 412, 413, 441 Caldwell, B.M. 414, 422, 424, 441, 442 Calkins, S.A. 97 Calkins, S.D. 446 Callaghan, T.C. 37, 40 Callanan, M.A. 154, 168 Callender, G. 220, 322 Calzi, F. 385, 407 Camaioni, L. 217 Camos, V. 206, 217 Campbell, H. 350, 358 Campbell, R.N. 290, 321 Campos, J. 198, 222 Canfield, R. 398 Canino, G. 101 Capuni-Paracka, S. 408 Cardoso-Martins, C. 127, 128, 134 Carey, S. 201, 222 Carlson, S.M. 302, 321 Carroll, K.A. 342, 348, 359 Carta, J. 447 Carter, A. 227 Carver, C.S. 347, 357 Carver, L.J. 224 Casassa, R. 409 Case, R. 176, 177, 212, 218, 263, 266 Casella, J.F. 397 Casey, B.J. 181, 183, 218 Casey, P. 397 Caskie, G.I.L. 443 Caspi, A. 48, 56, 74, 87, 90, 92, 96 Cassel, J. 405 Cassidy, D.J. 188, 219 Cassidy, J. 68, 96, 141, 168 Cassidy, K.W. 114, 134 Castellino, N. 377, 398 Castellino, P. 377, 398 Casterline-Sabel, J. 381, 395 Cataldo, M. 372, 398 Cather, A. 399 Catroppa, C. 395 Caulfield, L. 396, 403

Ceci, S.J. 276, 279, 317, 320, 321, 364, 397, 413, 441 Cejka, M. 284, 326 Cepeda, N.J. 320, 321 Cervantes, C.A. 154, 168 Chacon, M. 402 Chamberlain, P. 50, 102 Chang, S. 387, 388, 399, 400, 403 Chao, R.K. 423, 428, 441 Chase-Landsdale, P. 367, 398, 428, 429, 442 Chavez, A. 383, 386, 398 Chen, C. 415, 427, 428, 447 Chen, L. 358 Cheng, K. 218 Chesney, R. 402 Chin-Quee, D.S. 428, 441 Chorev, Z. 275, 324 Christensen, L. 385, 398 Christian, M.K. 428, 429, 442 Christopoulos, C. 96 Chromiak, W. 320 Chugani, H.T. 211, 218, 219 Chukovsky, K. 291, 321 Church, J. 33, 40 Chwaya, H. 407 Cicchetti, D. 347, 359 Cillessen, A.H.N. 63, 96 Cillessen, T. 97 Clark, B. 440 Clark, E.V. 280, 281, 287, 321 Clark, H.H. 280, 290, 321 Clark-Carter, D. 170 Clarke-Klein, S. 374, 405 Clark-Lempers, D. 426, 444 Claussen, A. 381, 401 Clements, W. 201, 219 Clifton, R. K. 218 Coates, D.L. 414, 422, 424, 442 Coatsworth, J.D. 346, 359 Coelen, C. 446 Cohen, J.D. 181, 218, 219 Cohen, L. 282, 286, 321 Cohen, P. 356 Coie, J.D. 44, 45, 50, 51, 52, 53, 54, 63, 67, 70, 73, 74, 75, 92, 94, 96, 97, 98, 100 Coldham, C. 405 Cole, D.A. 56, 97 Cole, H.I. 144, 169

453

Author Index Coleman, C. 62, 69, 70, 71, 72, 100 Collins, D.L. 420, 441 Collins, F. 377, 401 Collins, W. 366, 398 Collins, W.A. 420, 442 Collison, C. 186, 204, 224 Collister, E.G. 60, 100 Colombo, M. 387, 388, 398 Colvin, M. 409 Committee on Environmental Health 376, 398 Conger, J.J. 50, 97 Connell, J.P. 53, 97 Connolly, K. 375, 397 Conrad, R. 111, 134 Cook, S. 326 Cooney, R.R. 413, 414, 420, 421, 425, 427, 428, 442, 445 Cooper, D.H. 417, 418, 442 Coplan, R.J. 55, 56, 57, 59, 60, 97, 446 Coppotelli, H. 63, 97 Corley, R. 203, 210, 225 Cornell, E.H. 187, 219 Corrin, L.G. 11, 40 Cortez, V. 356, 357 Corwyn, R. 365, 397 Cosden, C. 10, 40 Cosgrove, J.M. 291, 321 Costello, E.J. 412, 413, 441 Cotton, B. 236, 268 Couchoud, E.A. 146, 168 Courtney, S.M. 181, 219 Cowan, N. 177, 178, 198, 219 Cowen, E.L. 67, 97 Cox, M.J. 417, 443, 446 Coy, K. 336, 357 Craik, K. 139, 168 Crawford, P. 401 Creed-Kanashiro, H. 396 Crick, N.R. 51, 52, 68, 97, 98, 104 Crnic, K. 153, 168 Crnic, L.S. 208, 219 Crone, D.A. 447 Cross, D. 143, 171, 185, 203, 226 Crouse, J. 427, 445 Crouter, A. 366, 369, 397, 398 Crowe, T. 373, 398 Crowley, K. 318, 326 Cruse, D. 280, 321

Cruttenden, L. 202, 203, 220 Cueto, S. 387, 404, 408 Cummings, E. 203, 218, 219, 221 Cupples, L. 399

D Dabholkar, A.S. 210, 211, 222 Daigneault, S. 292, 321 Dale, P. 220 Daltabuit, M. 372, 407 Daly, M. 336, 340, 344, 345, 357 Damasio, A.R. 274, 282, 321, 326 Damasio, H. 282, 326 Damm, D. 398 Danet, B. 280, 322 Daniels, D. 346, 357 Daniels-Bierness, T. 55, 103 Danks, J.H. 274, 318, 323 Dann, S. 94, 98 Darling, N. 367, 406, 423, 442, 447 Dasen, P. 406 Dauber, S.L. 417, 418, 440 Davidson, C. 370, 398 Davis, J.N. 343, 357 Davison, L. 42 Dawber, T. 372, 403 Day, R.H. 189, 224 de Abreu, M.D. 127, 134 de Groot, C. 350, 357 de Groot, L. 350, 357 de la Parra, A. 387, 388, 397 Deacon, T.W. 318, 322 Dea´k, G.O. 278, 279, 285, 287, 289, 291, 296, 298, 303, 304, 305, 306, 307, 309, 311, 312, 313, 315, 316, 319, 320, 322, 323, 325 Dean, R. 383, 391, 399 Deane, K.E. 151, 171 DeBlois, S. 218 Defeyter, M. 274, 323 DeFries, J. 364, 398 Dehaene, S. 282, 286, 321 Del Dotto, J.E. 350, 360 Dell, G.S. 282, 322 DeLoache, J.S. 33, 40, 188, 219 DeLuca, H. 402 DeMarie-Dreblow, D. 324 Demissie, T. 397

454

Author Index

Dempster, F.N. 212, 219, 286, 300, 322 Denham, S. 146, 152, 164, 168 Deregowski, J. 6, 40 DeRosier, M.E. 63, 67, 70, 81, 85, 87, 98, 103, 417, 442 Desai, S. 428, 429, 442 D’Esposito, M. 181, 219 Detre, J.A. 219 DeVos, J. 190, 217 DeVries, M. 386, 398 Dewdney, A. 9, 10, 40 Dhingra, P. 387, 406 Diamond, A. 182, 185, 186, 192, 195, 196, 199, 200, 202, 203, 207, 208, 210, 211, 219, 220, 274, 292, 302, 322, 323, 393, 398 Dickson, W.P. 442 Dietrich, K. 378, 379, 390, 398, 399, 404 DiLalla, L.F. 414, 445 DiModugno, M. 409 DiPietro, J.A. 403, 428, 429, 445 Dishion, T.J. 417, 442 Dixon, R.G. 426, 446 Doar, B. 185, 186, 192, 195, 196, 207, 210, 220 Dobkin, P.L. 104 Dodge, K.A. 50, 51, 52, 54, 63, 92, 96, 97, 98, 99, 100, 426, 442 Dohrenwend, B.P. 80, 98 Dohrenwend, B.S. 80, 98 Dolcourt, J. 403 Donahoe, A. 225 Donaldson, M. 285, 322 Donayre, M. 399 Donnelly, M. 222 Dorea, J. 404 Dornbusch, S. 366, 367, 407, 423, 428, 443, 444, 447 Douglas, L.W.B. 102 Dove, H. 372, 401 Downs, R.M. 4, 6, 33, 40, 41 Doxey, P.A. 199, 220 Dreyfuss, M. 407 Driver, J. 275, 325 Drotar, D. 399 Druin, D.P. 220, 322 Ducette, J.P. 360 Duggan, A. 351, 357, 397 Dulcan, M.K. 101

Dumaret, A. 366, 399 Duncan, G.J. 426, 441, 442 Duncker, K. 274, 277, 322 Duner, A. 50, 54, 102 Dunn, J. 65, 98, 146, 147, 154, 162, 164, 168, 169 Dunst, C.J. 199, 220 Durham, M. 414, 446 Durrell, D.D. 106, 134 Duvvuri, R. 135 Dziurawic, S. 223

E Earle, F. 104 Eastman, P.D. 290, 322 Easton, D. 99 Ebbert, R. 398 Eberly, S. 377, 401 Eccles, J.E. 437, 446 Echevarria, M. 446 Eckerman, C.O. 65, 98 Edelbrock, C.S. 44, 95 Edwards-Hawver, C. 398 Egan, S.E. 69, 98 Ehri, L.C. 106, 124, 134 Einstein, G.O. 265, 266 Eisenberg, J. 12, 40 Eisenberg, N. 169 Ekman, K. 69, 96 Elardo, R. 422, 442 Elder, G. 48, 56, 87, 96, 367, 399, 412, 413, 441 Ellerson, P.C. 349, 356 Elliott, T.R. 331, 358 Ellis, A.E. 199, 213, 223 Ellis, H. 223 Elman, J. 211, 214, 220, 232, 266 Emerson, E.S. 65, 101 Emory, E.K. 350, 358 Emslie, H. 221 Endriga, M.C. 360 Engle, P. 387, 393, 394, 396, 399, 404 Engle, R.W. 177, 179, 201, 206, 220 Ensminger, M.C. 48, 50, 52, 98 Entwisle, D.R. 53, 95, 414, 415, 417, 418, 427, 440 Epel, E.S. 347, 357 Epstein, J.N. 447

Author Index Ernhart, C. 377, 399 Ernst, J.M. 360 Errera, J. 402 Escher, M.C. 37, 40 Eschevarria, M. 445 Eslinger, P.J. 274, 286, 323 Espinosa, M. 382, 406 Estrada, P. 414, 424, 442 Evans, G. 368, 399 Ewald, W. 10, 40 Eyres, S.J. 440

F Faber, B. 401 Fabes, R.A. 152, 169 Fabricius, W.V. 317, 322 Factor-Litvak, P. 408 Faier-Routman, J. 331, 358 Farina, A. 358 Farkas, K. 406 Farmer, J. 372, 399 Farran, D.C. 417, 418, 442 Farrington, D.P. 50, 51, 98, 102, 104 Fauconnier, G. 281, 322 Fazio, R.H. 326 Feitelson, D. 106, 133, 134 Feldhusen, J.F. 52, 98 Feldman, J.F. 191, 216, 226 Feldman, R. 220 Felt, B.T. 385, 399 Fenson, L. 213, 220 Ferguson, L.L. 96 Fernald, L. 381, 382, 384, 400 Fernandez, T. 404 Fernyhough, C. 170 Ferreiro, E. 130, 134 Ferriss, G. 446 Ferro-Luzzi, A. 397 Field, T.M. 345, 350, 357 Fieselman, S.D. 193, 204, 225 Figueroa, A. 403 Finke, R.A. 274, 277, 284, 322 Finn, J.D. 78, 98 Finney, J. 398 Fiore, S.M. 274, 322 Fischel, J.E. 447 Fischer, K.W. 198, 206, 220, 222 Fischhoff, B. 254, 267

Fish, E. 447 Fisher, P.A. 360 Fishkin, A.S. 274, 324 Fivush, R. 146, 149, 154, 169, 171, 233, 268, 421, 442 Flavell, E.R. 287, 288, 289, 323 Flavell, J.H. 143, 169, 206, 220, 265, 268, 287, 288, 289, 323 Fleck, K. 356 Flemming, D. 50, 98 Fletcher, J.M. 442 Fletcher, K.E. 346, 357 Flick, G. 446 Foley, M.A. 267 Folkman, S. 342, 357 Foorman, B.R. 436, 442, 446 Foreman, D. 204, 221 Foreman, N. 186, 204, 221 Forssberg, H. 182, 223 Forsyth, J. 409 Foulks, B. 417, 442 Fox, N.A. 97, 209, 218, 221, 446 Fox, R. 198, 221 Fozard, J.L. 236, 268 Fraleigh, M. 442 Francis, D.J. 135, 442 Francis, J. 414, 442 Frank, D. 369, 370, 399 Frank, R.G. 331, 358 Frankenfield, A. 324 Franks, J. 320 Frazee, H.E. 58, 67, 98 Freeman, H. 443 Freeman, N.H. 29, 40 Freeman, W. 335, 358 Freese, M.K. 413, 414, 430, 431, 442 French, R. de S. 358 Freud, A. 94, 98 Frick, P.J. 101 Friebert, S.E. 265, 268 Friedman, S. 366, 399 Friedman, W.J. 229, 230, 235, 236, 237, 238, 239, 240, 242, 243, 244, 245, 246, 247, 248, 249, 250, 251, 252, 253, 254, 255, 256, 257, 260, 264, 265, 266, 267 Fries, K. 333, 358 Frith, U. 124, 132, 134 Fromhoff, F.A. 149, 154, 169 Frosch, C.A. 414, 443

455

456

Author Index

Frye, D. 227, 278, 285, 286, 293, 294, 295, 296, 298, 313, 319, 323, 327 Fuddy, L. 357 Fueser, J.J. 196, 210, 225 Fukuda, R. 403 Fulker, D. 364, 398 Fullmer, C. 379, 399 Fulton, M. 407 Funahashi, S. 180, 195, 206, 221 Fung, H. 147, 149, 170 Furman, W. 61, 62, 68, 69, 72, 81, 96, 98 Fuster, J.M. 180, 221 Fyer, A.J. 358

G Gabrieli, J.D.E. 211, 221 Gage, F.H. 225 Galal, O. 403, 407, 408 Galanter, E. 177, 179, 224 Galen, B.R. 51, 98 Galler, J. 384, 399 Gallistel, C.R. 218 Garber, J. 76, 102 Garcia, L. 399 Gardner, A.G. 238, 239, 252, 254, 267 Gardner, H. 29, 40, 321 Gardner, J. 29, 40, 379, 399 Gariepy, J.L. 96 Garmezy, N. 74, 99, 359, 366, 396, 403 Garralda, E. 374, 399 Garrod, S. 280, 290, 323 Gathercole, S.E. 214, 221 Gatsonis, C. 378, 403 Gauvain, M. 38, 40 Gavinski-Molina, M.H. 97 Gazelle, H. 55, 56, 57, 58, 59, 76, 88, 93, 99 Geber, M. 383, 391, 399 Gell, A. 230, 267 Gelman, R. 260, 266 Gelman, S.A. 316, 324 George, S. 387, 406 Georgieff, M. 381, 385, 405 German, T.P. 323 Gershkoff-Stowe, L. 282, 323 Gerstadt, C.L. 292, 302, 323 Gersten, N. 103 Gest, S.D. 359

Giammarino, M. 87, 104 Gibeault, S. 187, 204, 223 Gibson, J.J. 34, 40 Giedd, J. 218 Gilbert, S.J. 275, 323 Gillberg, C. 378, 399 Gilman, R. 396 Gilmore, R.O. 193, 195, 208, 221, 223 Giordini, B. 394, 397 Giuffre, K. 9, 40 Glanz, F. 446 Glasgow, K.L. 423, 443 Glassman, R.B. 186, 221, 224 Gluckman, P. 396 Glucksberg, S. 274, 291, 318, 323 Glueckauf, R.L. 331, 358 Godfrey, K. 396 Goetz, C.G. 221 Goffman, E. 340, 358 Golden, M. 428, 443 Goldman, B.D. 225, 226, 443 Goldman, P.S. 180, 221 Goldman-Rakic, P.S. 180, 195, 206, 208, 217, 220, 221, 222, 223, 227 Golinkoff, R.M. 216, 222 Gombert, J.E. 291, 323 Gonez, C. 371, 399 Gonzales, G. 371, 399 Gonzalez de Sather, J.C. 320, 321 Goodman, J.C. 304, 323 Gopnik, A. 289, 323 Gore, J.C. 217, 225 Goren, C.C. 198, 221 Gorman, K. 382, 400, 404 Gorman-Smith, D. 51, 54, 104 Goswami, U. 132, 135 Gottfried, A.W. 414, 443 Gottlander, K. 343, 358 Gottlieb, A. 40 Gottlieb, G. 81, 99, 364, 400, 412, 413, 443 Gottman, J.M. 61, 95 Gough, P.B. 124, 134 Goyer, R. 370, 377, 378, 400 Graber, M. 190, 217 Grabie, E. 405 Graef, J. 380, 406 Graham, S. 68, 69, 92, 99, 100 Grandjean, P. 380, 398, 402 Grant, L. 403

Author Index Grantham-McGregor, S. 366, 374, 381, 382, 383, 384, 386, 387, 388, 391, 394, 400, 403, 405, 406 Grattan, L.M. 274, 286, 323 Gratz, R. 372, 400 Gray, C.A. 440 Graziano, J. 408 Green, C. 405 Green, E. 91, 99 Green, F.L. 287, 288, 289, 323 Green, L. 414, 442 Green, R.W. 102 Greenough, W. 366, 400 Greenwood, C.R. 447 Griffin, E.A. 413, 421, 442 Griffin, P. 106, 135, 416, 446 Griffith, E.M. 414, 415, 416, 445 Grimwood, K. 395 Gross, A.M. 356 Gross, C.G. 180, 222 Gross, R.T. 350, 358 Grossi, M. 443 Grossman, E. 199, 222 Grossman, M. 219 Grotpeter, J.K. 51, 52, 97 Gruen, G. 364, 408 Gruendel, J. 143, 170 Grundman, M. 367, 400 Guilarte, T. 385, 400 Guilford, J.P. 274, 323 Gunnar, M. 367, 400 Gunnoe, C. 404 Guntheroth, W. 372, 400 Gyaltsen, P. 401

H Haden, C.A. 421, 442 Hadzialjevic, S. 408 Haegerich, T.M. 186, 221 Hagekull, B. 386, 394, 400 Haggerty, R. 396 Haith, M.M. 233, 241, 266, 267 Hall, L.K. 177, 223 Halle, D. 9, 40 Halterman, J. 382, 400 Hammad, T. 379, 400 Hammond, M.A. 440 Hammond, W.A. 102

Han, S. 377, 401 Hanley, G. 16, 41 Hansen, E.E. 419, 444 Hansen, O. 380, 398, 402 Happaney, K. 338, 339, 356 Harding, C.G. 216, 222 Harding, J. 396 Hardway, C.L. 445 Harkness, S. 394, 407 Harlow, H.F. 180, 224 Harner, L. 230, 232, 233, 235, 267 Harnishfeger, K.K. 286, 323 Harper, L. 373, 401 Harris, J. 365, 371, 387, 388, 401 Harris, P.L. 143, 169, 185, 192, 195, 202, 222 Harris, R. 409 Harrison, G. 403, 407, 408 Harrist, A.W. 55, 56, 57, 59, 60, 99 Hart, B. 413, 414, 416, 426, 428, 442, 447 Hart, C.H. 444 Hart, E.L. 101 Hartup, W.W. 61, 96, 99 Haselager, G.J.T. 63, 96 Hasher, L. 320 Hashimoto, O. 403 Hastings, P.D. 55, 56, 103 Hastorf, A.H. 358 Hauser, P. 401 Havighurst, R.J. 52, 99 Havlik, B.S. 421, 442 Hawk, B. 380, 406 Hawker, D.S.J. 64, 65, 99 Hawkins, J.D. 97 Haxby, J.V. 208, 219, 226 Haymes, C.W. 350, 358 Haynes, V. 324 Hazler, R.J. 71, 99 Hebl, M.R. 332, 358 Hedlund, M. 331, 358 Hegion, A. 447 Heise, D. 284, 285, 326 Helenius, H. 373, 406 Helmke, A. 414, 447 Hembree, S.E. 70, 72, 104 Hepburn, W. 407 Hermer-Vazquez, L. 317, 318, 323 Herrera, C. 164, 169 Herschkowitz, N. 210, 222 Hershey, K.L. 420, 446

457

458

Author Index

Hertz-Pannier, L. 218 Hess, R.D. 414, 422, 424, 442 Hesselbrock, V.M. 58, 99 Heth, C.D. 187, 219 Hetherington, E. 397, 442 Hildreth, G. 119, 134 Hillinger, M.L. 124, 134 Himes, J. 403 Hinde, R.A. 65, 99, 141, 169 Hindle, R.A. 99 Hirsh, K.W. 282, 323 Hitch, G.J. 176, 177, 179, 200, 207, 217 Hite, S.J. 428, 444 Hocevar, D. 274, 323 Hodapp, R.M. 336, 358 Hodgson, K. 225 Hoffman, M.L. 159, 169 Hoffman, S.L. 350, 358 Hofstadter, M. 193, 199, 202, 222 Hogan, K. 331, 359 Hohjoh, H. 403 Holloway, S.D. 442 Holmes, D.H. 413, 417, 418, 419, 425, 435, 445 Holmes, D.L. 359 Homatidis, S. 401 Hong, Y. 292, 302, 323 Hood, B.M. 201, 222 Hoogstra, L. 170 Hoover, J.H. 71, 99 Hopkins, B. 350, 357 Hopkins, R. 404 Hornbaker, A.V. 424, 445 Horst, R. 379, 401 Houde´, O. 286, 300, 323 Houston-Price, C. 326 Hoven, C. 401 Howard, C. 377, 401 Howard, L.W. 342, 360 Howe, M.J.A. 317, 320, 321 Howes, C. 62, 65, 99, 428, 442 Hoza, B. 62, 65, 72, 96 Hsieh, S. 275, 286, 320 Hu, H. 396 Huang, C.F. 401 Huang, J. 404, 406 Hubbard, J.J. 359 Hudes, M. 401 Hudson, J.A. 143, 147, 149, 169, 233, 241, 267 Hughes-Wagner, J. 289, 323

Huie, K. 373, 401 Hulme, C. 212, 222 Humphreys, G. 282, 286, 326 Hunter, R. 407 Hunter, S.B. 356 Hunter, W.S. 184, 185, 192, 194, 203, 210, 222 Huntsinger, C.S. 442 Huntsinger, P.R. 442 Hurtado, E. 381, 401 Husaini, M. 404, 408 Hutchinson, M.K. 374, 401 Huttenlocher, J. 188, 190, 222, 224, 238, 239, 267 Huttenlocher, P.R. 210, 211, 222 Huttunen, M. 403 Hyder, F. 223 Hyman, C. 97 Hymel, S. 59, 65, 67, 68, 70, 71, 76, 92, 95, 96, 99, 102

I Ickovics, J.R. 347, 357 Igou, B. 6, 40 Ihsen, E. 189, 224 Inhelder, B. 262, 268 Iny, L.J. 359 Ironsmith, M. 63, 103 Isager, T. 403 Isherwood, S. 9, 40 Ittleson, W.H. 12, 40 Ivenz, S.H. 135 Iversen, I.A. 106, 135 Izzo, L.D. 97

J Jackson, E. 198, 222 Jackson, J.L. 236, 267 Jacobs, J. 402 Jacobsen, B. 401 Jacobsen, C.F. 180, 222 Jacobson, J. 370, 401 Jacobson, S. 370, 401 Jacoby, E. 387, 404, 408 Jacques, S. 294, 323

Author Index Jahari, A. 404, 408 Jahn, S. 401 James, W. 175, 222, 381, 401 Janes, C.L. 58, 60, 67, 99 Jankowski, J.J. 191, 226 Janoff, E. 404 Jansson, L. 401 Jarvis, L.H. 288, 324 Jason, L. 401 Jemerin, J. 374, 397 Jencks, C. 413, 427, 428, 442 Jeng, S. 358 Jenkins, L.B. 414, 445 Jensen, P.S. 101 Jerome, N. 403, 407, 408 Jersild, A. 91, 100 Jest, C. 371, 404 Jezzard, P. 218 Jha, A.P. 181, 191, 222 Jimenez, E. 377, 380, 409 Joanette, Y. 274, 326 Joesch, J.M. 336, 337, 358 Joffe, A. 397 John, R.S. 58, 100 Johnson, A.S. 274, 324 Johnson, J.G. 356 Johnson, J.H. 80, 85, 100 Johnson, M. 379, 401 Johnson, M.H. 193, 195, 198, 208, 211, 220, 221, 223, 224 Johnson, M.K. 231, 254, 267 Johnson, R.J. 106, 134 Johnson-Laird, P.N. 140, 169 Johnston, C. 335, 338, 356, 358 Jones, E.E. 358 Jones, K. 336, 357 Jonides, J. 181, 219, 223, 226 Jordan, K. 374, 401 Jose, P.E. 428, 442 Juel, C. 436, 442 Jusczyk, P.W. 214, 217 Juvonen, J. 68, 69, 92, 99, 100, 332, 358

K Kaczorowski, J. 400 Kagan, J. 55, 103, 192, 203, 210, 222, 223, 414, 419, 442 Kagitcibasi, C. 395, 401

459

Kail, R. 177, 212, 223 Kalish, C.W. 316, 324 Kanashiro, H. 408 Kane, M.J. 177, 179, 201, 220 Kaplan, H. 372, 401 Karabenick, J.D. 291, 326 Karmiloff-Smith, A. 220, 277, 285, 317, 324, 418, 442 Karp, R. 370, 401 Kaskie, B. 424, 445 Kastens, K.A. 20, 41 Katch, L. 413, 430, 436, 445 Kato, T. 403 Kaysen, D. 218 Kearsley, R. 203, 223 Keefe, K. 68, 70, 93, 95, 100 Keele, S.W. 324 Keil, K. 219 Keir, E. 395 Kellam, S.G. 50, 98 Kelleher, K. 397 Kellum, G. 356 Kelly, M.H. 114, 134 Kemp, F. 401 Kemp, S. 238, 239, 241, 252, 255, 267 Kerr, M. 60, 100 Kessenich, M.A. 414, 415, 416, 422, 424, 431, 434, 442, 446 Keusch, G. 386, 401 Kibler, J. 360 Kieras, D. 275, 324 Kiernan, K. 398 Kinney, D. 373, 401 Kirkham, N. 323 Kirksey, A. 403, 407, 408 Klebanhoff, M. 397 Klebanov, P.K. 426, 442 Kleck, R.E. 332, 358 Klein, N. 386, 402 Klein, R. 387, 388, 396 Kleinman, J. 382, 401 Klem, L. 432, 444 Kline, J. 408 Kline, R.B. 432, 444 Klingberg, T. 182, 223 Knutson, J.F. 334, 336, 339, 360 Kobak, R.R. 144, 169 Kochanska, G. 144, 147, 159, 160, 162, 165, 169, 170, 368, 401

460

Author Index

Kochenderfer, B.J. 62, 64, 69, 70, 71, 72, 100, 101 Kochenderfer-Ladd, B. 63, 65, 71, 81, 100, 101 Koeppe, R.A. 223, 226 Kohlberg, L. 45, 50, 100 Kohnstamm, G. 396 Koinis, D. 325 Kokotovic, A. 356 Konieczna, E. 269 Konstantareas, M. 373, 401 Kopeikin, H. 356, 357 Kopp, C.B. 420, 444 Koriat, A. 254, 267 Kose, G. 12, 33, 38, 40, 41 Koskelainen, M. 373, 406 Kovalenko, P. 373, 401 Koyama, T. 403 Kraft, K. 399 Kramer, A.F. 320, 321 Kramer, U. 377, 378, 409 Krantz, J. 357 Krasner, D.V. 336, 358 Krauss, R.M. 291, 323 Kress, G. 9, 41 Kreutzer, M.A. 265, 268 Krippner, S. 6, 41 Krishnakumar, A. 394, 397 Kroncke, A. 225 Krystal, J.H. 217 Kubota, K. 180, 211, 223, 226 Kuhlen, R. 60, 100 Kuipers, B. 229, 268 Kupersmidt, J.B. 50, 52, 53, 67, 75, 92, 97, 98, 100, 417, 442 Kusel, S.J. 62, 103 Kusumi, I. 403 Kvalsvig, J. 375, 376, 398, 401, 407

L La Paro, K.M. 445 LaCrosse, J. 45, 50, 100 Ladd, G.W. 45, 46, 48, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 65, 68, 69, 70, 71, 72, 74, 75, 76, 78, 80, 81, 82, 87, 88, 89, 91, 92, 93, 94, 95, 96, 97, 99, 100, 101, 417, 418, 444 Lagace-Seguin, D.G. 97

Lagerspetz, K. 51, 69, 96, 101 Laible, D.J. 144, 158, 160, 161, 162, 163, 164, 165, 170 Lakoff, G. 281, 324 Lalor, G. 399 Lamb, M.E. 139, 170, 336, 358 Lambert, N.A. 52, 60, 101 Lambert, W.W. 60, 100 Lamborn, S.D. 423, 444, 446 Landry, S.H. 422, 444 Lang, A.E. 292, 326 Lang, B. 294, 297, 300, 302, 310, 315, 319, 325 Langenberg, P. 379, 400, 402 Langer, J.A. 427, 440 Lanphear, B. 377, 401 Larsen, J.M. 428, 444 Laub, J. 368, 405 Lauber, E.J. 223 Lavigne, J.V. 331, 358 Lawrence, A.L. 222 Laxen, D. 407 Lazarus, R.S. 80, 101 Learmonth, A. 190, 224 Ledingham, J.E. 56, 59, 60, 61, 101 Lee, C. 97 Lee, K. 287, 325 Lee, M. 441 Lee, S. 415, 428, 447 Leftwich, M. 377, 401 Legerstee, M. 223 Lehey, B.B. 51, 101 Lehtonen, L. 359 Leiderman, P. 442 Lekic, V. 408 LeMare, L. 55, 99, 103, 104 Lemmon, K. 241, 268 Lempers, J.D. 426, 444 Leniek, K.M. 186, 221 Lennon, M.C. 360 Lennox, C. 401 Leonard, C. 265, 268 Lescarno, A. 396 Lester, B. 383, 401 Lester, B.M. 350, 356, 358 Lester, D. 350, 351, 358 Lester, M. 379, 383, 402 Levin, I. 106, 128, 129, 134 Levine, L.J. 149, 170 Levine, M. 16, 41 LeVine, R. 369, 402

461

Author Index LeVine, S. 369, 402 Levine, S. 347, 358 Leviton, A. 378, 379, 396, 402, 404 Levitsky, D. 385, 391, 392, 402, 407 Lewis, J.C. 339, 356, 357 Lewis, M. 350, 358, 380, 402, 414, 422, 424, 442 Lezak, M.D. 292, 324 Lhotska, L. 394, 399 Li, P. 232, 266 Liao, H. 358 Liaw, F.R. 442 Liben, L.S. 4, 5, 9, 12, 20, 31, 33, 36, 37, 38, 40, 41, 42 Lickel, B. 356 Lickliter, R. 81, 99, 412, 413, 443 Liddle, G.P. 99 Light, R. 406 Liittschwager, J. 289, 324 Lilienthal, H. 377, 378, 409 Lin, E. 338, 356, 357, 358 Lin, J. 224 Linakis, J. 377, 395 Lindenfelser, K.S. 359 Lindsay, C. 398 Lindstrom, M. 375, 396 Lin-Fu, J. 377, 402 Lira, P. 400 Lister, M. 10, 40 Liu, H. 401 Liu, X. 408 Liwag, M.D. 149, 170 Llewellyn, A. 331, 359 Lochman, J.E. 75, 96, 97, 101 Lockitch, B. 377, 402 Loeber, R. 51, 54, 101, 102 Lohman, R. 372, 400 Lollis, S. 55, 103, 104 London, B. 8, 41 Long, B. 97 Lopez, I. 387, 388, 397 Lopez, S. 396 Louria, D.B. 377, 397 Lovelace, L. 414, 445 Lozoff, B. 377, 380, 382, 383, 384, 385, 386, 387, 388, 393, 399, 402, 409 Lucariello, J. 233, 268 Lucas, S. 402 Luchins, A.S. 277, 324 Luciana, M. 208, 224

Lujan, C. 396 Lumaa, V. 344, 359 Luria, A.R. 292, 324 Luthar, S.S. 347, 359 Lynch, S. 402 Lyngbye, T. 380, 398, 402 Lyon, J.E. 356, 357

M MacArthur, R.S. 274, 325 MacCarthy, G. 225 Maccoby, E. 162, 170, 397, 423, 442, 444 MacDonald, D. 198, 218 MacDonald, L. 198, 218 MacDonald, S.E. 187, 204, 223 MacDougall, P. 58, 67, 73, 75, 102 Machon, R. 403 MacWhinney, B. 272, 298, 324 Maestripieri, D. 342, 348, 359 Magnusson, D. 50, 54, 102, 364, 402, 437, 444 Maguire, M. 147, 169 Mahaffey, K. 379, 402 Maier, N.R.F. 203, 223 Maier, R.A. 359 Main, M. 425, 444 Maki, R.H. 186, 204, 217 Maky, M. 397 Mandell, D. 401 Mann, J. 341, 344, 345, 359 Manuel, M. 402 Maratsos, M. 287, 288, 298, 320, 322, 324 Marazita, J.M. 288, 324 Marcoen, A. 144, 171 Marchon, I. 16, 41 Marcovitch, S. 185, 203, 223 Margalit, T. 134 Marin, R. 396 Markey, F. 91, 100 Markman, E.M. 288, 289, 291, 324 Markman, H.J. 97 Markowitz, M. 405 Marks, D. 331, 359 Marks, J. 359 Markus, H. 358 Marlowe, M. 377, 402 Marolla, F. 406 Mars, K.T. 101

462

Author Index

Martin, J. 423, 444 Martin, J.M. 97 Martinez, C. 383, 386, 397 Martino, G. 281, 283, 286, 321 Martorell, G.A. 354, 357 Martorell, R. 381, 402, 404 Marvin, R.S. 142, 144, 170 Marx, J. 370, 403 Mash, C. 409 Mason, C.A. 360 Massar, B. 226 Massaro, E. 378, 379, 403 Massaro, T. 378, 379, 403 Masten, A. 59, 74, 98, 102, 346, 359, 366, 403 Matheny, A. 372, 403 Mathews, M.E. 236, 268 Mathews, C.V. 99 Mathiesen, B. 394, 409 Matorell, G.A. 358 Matsumura, M. 211, 226 Matsuo, K. 403 Matthews, A. 199, 213, 223 Mayr, U. 324 Mazziotta, J.C. 211, 219 McBride-Chang, C. 118, 120, 134 McBurnett, K. 101 McCabe, A. 149, 170 McCabe, G. 393, 403, 407, 408 McCall, R. 409 McCarrell, N. 320 McCarthy, G. 181, 191, 217, 222, 223, 225 McCartney, K. 428, 429, 444 McCarton, C. 374, 403 McCleary, C. 406 McClelland, J.L. 224 McClelland, M.M. 413, 414, 416, 417, 418, 419, 420, 425, 433, 435, 444, 445 McClish, D. 402 McCormack, M. 402 McCormack, T. 236, 238, 268 McCormick, S.E. 169 McCullough, A. 385, 388, 403 McDaniel, C. 198, 221 McDaniel, M.A. 265, 266 McDonald Connor, C. 413, 430, 436, 445 McDonald, M. 406, 408 McDonough, L. 184, 186, 199, 224, 305, 323 McEwen, B.S. 347, 357, 359 McFarlane, E.C. 357

McGourty, S. 152, 171 McGrade, B.J. 356 McHale, S. 366, 398 McKenzie, B.E. 189, 224 McLaughlin, M. 366, 407 McMichael, A. 379, 403 McQueeney, M. 356 McSwain, R. 372, 403 Meaney, M.J. 347, 359 Medin, D.L. 280, 324 Mednick, S. 58, 100, 373, 403 Medrano, Y. 399 Meeks-Gardner, J. 386, 403 Meher, C. 404 Mehta, P. 442 Meier, N.R. 274, 277, 324 Meijer, A. 381, 404 Meins, E. 144, 170 Meiran, N. 275, 324 Melgar, P. 402 Mendes, W.B. 356 Menzel, E.W. 187, 204, 224 Mercer, L. 102 Merialdi, M. 383, 384, 403 Merriman, W.E. 287, 288, 324 Messer, S.C. 356 Meyer, D.E. 275, 324 Meyer, D.H. 180, 224 Meyer, D.J. 336, 358 Meyer, K. 387, 388, 409 Meyers, A. 399 Meyers, D.G. 99 Meyerson, L. 332, 359 Michael, C.M. 58, 102 Michael, R.T. 428, 429, 442 Michael, W. 274, 323 Michelow, D. 321 Michon, J.A. 236, 267 Miguel, J. 396 Milar, C. 390, 403 Millard, R.T. 135 Miller, D.T. 358 Miller, G. 373, 378, 379, 397, 403 Miller, G.A. 177, 179, 224 Miller, P. 401 Miller, P.H. 143, 147, 169, 317, 324 Miller, P.J. 149, 170 Miller, W.C. 50, 97 Miller-Loncar, C.L. 444 Milner, B. 286, 325

463

Author Index Minde, K. 350, 359 Minden-Cupp, C. 436, 442 Minnes, S. 406 Minoshima, S. 223, 226 Mintz, J. 147, 149, 170 Mirsky, A.F. 226 Mishkin, M. 226 Mistretta-Hampson, J. 446 Mock, D.W. 343, 359 Moffit, T.E. 51, 54, 102, 104 Molfese, V.J. 414, 444, 445 Mollnow, E. 380, 407 Monk, C.S. 210, 211, 224, 226 Monsell, S. 275, 307, 319, 325 Monteior, A. 404 Montresor, A. 407 Moon, C. 402 Moore, C. 241, 268 Moore, D.R. 50, 102 Moore, K. 225 Moore, R. 401 Moore, T. 225 Moreau, T. 350, 360 Morgan, V. 222 Moriana, M. 408 Morna, N. 408 Morris, C. 324 Morris, D.P. 58, 60, 102 Morris, S. 400, 415 Morrison, F.J. 413, 414, 415, 416, 417, 418, 419, 420, 421, 422, 424, 425, 427, 428, 429, 430, 431, 432, 433, 434, 436, 437, 438, 439, 440, 443, 444, 445, 446 Morrison, P. 59, 102 Morrongiello, B. 372, 403 Morrow, J. 225, 226 Morrow, R.D. 417, 442 Morrow-Tlucak, W.A. 399 Morton, J. 198, 223, 224 Moscovitch, M. 224 Moses, L.J. 302, 321 Mosier, C.E. 216, 224 Mosotofsky, D. 409 Moss, H.A. 48, 102 Mounts, N.S. 444, 447 Mouridsen, S. 373, 403 Moussa, W. 408 Mouzaki, A. 135 Muir, C. 222 Muir, D. 287, 325

Mukai, L.H. 50, 102 Muller-Brettel, M. 395, 406 Mulligan, G. 50, 102 Mullis, I.N.S. 427, 440 Mullis, I.V.S. 414, 445 Munakata, Y. 201, 224, 296, 325 Mundform, D. 397 Munholland, K.A. 140, 165, 168 Munn, P. 164, 169 Murphy, S. 359 Murray, A.D. 424, 445 Murray, P. 204, 221 Musabegovic, A. 408 Mushak, A. 403 N Nadel, L. 174, 190, 224, 226 Nadolny, T. 226 Nairne, J.S. 176, 224 Nakamichi, M. 344, 359 Nando, S. 403 Napoleone, M. 399 Narasimham, G. 296, 299, 305, 311, 312, 322, 325 Narita, K. 373, 403 National Research Council 365, 366, 403 Ndanga, K. 397 Neckerman, H.J. 52, 96 Neebe, E. 441 Needleman, H. 378, 396, 402, 403, 404 Neiderman, D. 202, 203, 220 Neighbor, G. 222 Nelson, C. 366, 393, 404 Nelson, C.A. 174, 181, 199, 208, 210, 211, 213, 223, 226 Nelson, E. 402 Nelson, K. 143, 145, 146, 147, 170, 230, 232, 233, 234, 235, 265, 268 Neufeld, L. 402 Neumann, C. 406, 408 Newbury, D. 10, 41 Newcomb, A.F. 63, 68, 95, 96, 102 Newcombe, N. 188, 190, 222, 224 Newport, E.L. 214, 224 NICHD Early Child Care Research Network 414, 415, 428, 429, 445 Nichols, M. 396 Nicola, R.M. 350, 359 Nielsen, S. 403

464

Author Index

Niki, H. 180, 223 Ninio, A. 33, 41 Nitsch, K. 320 Nobre-An, C. 223 Nolan, T. 395 Noll, D.C. 218, 219 Novak, M. 218 Novotny, T. 370, 404 Nsamenang, A. 369, 372, 404 Nurmi, J.E. 241, 268 Nye, R. 38, 42 Nystrom, L.E. 219

O O’ Brien-Caughy, M. 428, 429, 445 O’ Bryan, K.G. 274, 325 O’ Connor, J. 12, 38, 41 O’ Connor, J.M. 41 O’ Connor, R.C. 186, 224 O’ Connor, T.G. 141, 171 O’ Hara, N. 356 O’ Neill, D.K. 281, 298, 325 O’ Scalaidhe, S.P. 208, 227 Oakhill, J. 314, 327 Okabe, Y. 403 Okamoto, Y. 263, 266 Okazaki, Y. 403 Oleske, J.M. 377, 397 Oliva, A. 284, 326 Olkin, R. 331, 360 Ollendick, T.H. 60, 67, 102 Olson, D.R. 314, 325 Olson, R.K. 417, 447 Olson, S.L. 414, 422, 424, 445 Olster, D.H. 358 Olton, D.S. 186, 204, 206, 224 Olweus, D. 48, 62, 68, 69, 102 Ontai, L.L. 144, 154, 158, 170 Oppenheim, D. 151, 171 Orkow, B. 359 Ornstein, P.A. 206, 217, 413, 445 Ortony, A. 280, 320 Oswald, D.P. 102 Otero, G.A. 382, 404 Overman, W.H. 186, 204, 225 Owens, J. 396 Owens, S. 221

P Paditch, R. 378, 404 Paine, P. 387, 388, 404 Palfai, T. 323 Palti, H. 381, 404 Panak, W.F. 76, 102 Pandya, D.N. 274, 325 Pang, J.C. 187, 204, 223 Papagno, C. 282, 325 Papas, B.C. 224 Pape, D.A. 333, 360 Parad, H.W. 87, 104 Paris, S.G. 291, 325 Parisi, D. 220 Park, Y. 223 Parker, G.A. 343, 359 Parker, J.G. 45, 49, 50, 54, 55, 58, 60, 62, 65, 67, 69, 70, 72, 73, 74, 80, 94, 102, 103 Parker, T. 447 Parmelee, G. 370, 404 Parsons, M.J. 29, 40, 41 Pascual-Leone, J. 176, 225 Pasore, G. 397 Pasquali, J. 404 Passmore, L.A. 356 Pate, B.J. 225 Patel, S. 134 Patterson, C.J. 98, 291, 321, 417, 442 Patterson, G. 366, 404 Pawson, I. 371, 404 Payne, A.C. 421, 445, 447 Payne, C. 445 Pearson, D. 378, 379, 404 Pedersen, A. 97 Peeke, L.G. 97 Peeler, J. 402 Pellegrini, A. 364, 396, 421, 441 Pelphrey, K.A. 193, 199, 204, 205, 225, 445 Peltonen, T. 51, 101 Penniman, J.H. 99 Pennington, B.F. 178, 201, 208, 219, 225, 267 Penny, M. 396 Peresie, H. 404 Perfetti, C.A. 125, 135, 446 Perlstein, W.M. 219

465

Author Index Perner, J. 201, 219, 294, 297, 300, 302, 310, 319, 325 Perry, D.G. 53, 62, 65, 69, 92, 98, 103 Perry, K.E. 45, 69, 103 Perry, L.C. 62, 103 Pesetsky, D. 446 Peterson, C. 149, 170 Peterson, K. 399 Peterson, L. 372, 399 Pethick, S.J. 220 Petit, G.S. 426, 442 Petrides, M. 181, 183, 187, 225, 286, 325 Pettit, G.S. 99 Pettit, J. 287, 289, 316, 322 Peuster, A. 225 Phelps, L. 378, 404 Phelps, M.E. 211, 219 Phillips, D.A. 414, 415, 416, 418, 420, 446 Phillips, M. 413, 427, 428, 442, 445 Phillipsen, L.C. 65, 99 Piaget, J. 38, 41, 91, 103, 185, 225, 262, 268, 324, 325 Pianta, R.C. 417, 430, 446 Picard, D. 317, 325 Picard, K.M. 350, 360 Pickens, D. 222 Pien, D.L. 325 Pierce, J.V. 99 Pierroutsakos, S.L. 40 Pihl, R.O. 104 Pine, D.S. 408 Pinõn, D.E. 292, 295, 302, 327 Pinto, J. 198, 218 Pinzo, L. 401 Pisoni, D.B. 214, 217 Ple´h, C. 298, 324 Plomin, R. 346, 357, 364, 368, 398, 408 Plunkett, K. 220 Poirier, C. 218 Policare, H. 443 Pollitt, E. 371, 375, 382, 383, 384, 385, 386, 387, 388, 404, 407, 408 Poortinga, Y. 406 Pope, S. 397 Popovac, D. 408 Porcayo, R. 404 Posner, J. 368, 407 Potasova, A. 377, 405 Poteat, G.M. 63, 103

Potter, H.H. 321 Potts, R. 170 Powell, C. 366, 384, 394, 400, 405 Powelson, J. 321 Prabucki, K. 386, 402 Prasada, S. 201, 222 Pressley, M. 52, 104, 177, 226, 430, 446 Preteni-Redjepi, E. 408 Prevor, M.B. 220, 322 Pribram, K.H. 177, 180, 224, 225 Price, D. 6, 38, 41, 53, 93 Price, G.G. 442 Price, J.M. 101, 103, 417, 444 Price, R. 222 Principe, G. 206, 217 Prinz, R.J. 55, 104 Puce, A. 199, 217, 223, 225 Pueschel, S. 377, 395

Q Qiao, X. 401 Quay, H.C. 101

R Raab, G. 407 Rabinowitz, M. 370, 380, 396, 398, 402, 405 Rabinowitz, S. 409 Radke-Yarrow, M. 383, 387, 388, 396 Raikes, H.A. 141, 151, 171 Rainey, B. 356 Ralph, J. 427, 445 Ramakrishnana, U. 402 Ramey, C. 366, 405, 428, 441 Ramey, S. 97, 366, 405 Ramirez, M. 359 Ramsay, D. 402 Rangel, L. 374, 399 Rankin, M.P. 37, 40 Rao, R. 381, 385, 405 Rapport, J.L. 218 Rapus, T. 286, 294, 313, 319, 327 Raudenbush, S.W. 438, 441 Raviv, T. 415, 434, 446 Ray, S.D. 291, 322 Raye, C.L. 267

466

Author Index

Rayner, K. 414, 416, 417, 446 Read, C. 129, 135 Redbond, J. 326 Reece, C. 129, 135 Reed, M.A. 325 Reed, R. 404 Reese, E. 149, 161, 171, 233, 268, 421, 442 Reich, J.N. 359 Reif, L. 346, 350, 360 Reiss, C. 356 Renshaw, P.D. 59, 62, 68, 76, 95, 103 Resende, S.M. 127, 128, 134 Revelle, G.L. 291, 325 Reynold, M.C. 331, 360 Reynolds, D. 10, 40 Reznick, J.S. 55, 103, 186, 193, 196, 198, 199, 202, 203, 204, 210, 215, 216, 220, 222, 225, 226, 227, 279, 292, 295, 302, 317, 327, 421, 446 Ricciuti, H. 386, 405 Rice, C. 289, 325 Rich, B. 403 Richardson, S. 383, 405 Richman, G. 398 Richman, J. 401 Richmond-Welty, E.D. 129, 135 Ricks, D. 45, 50, 58, 100, 103 Riggins, R. 441 Riley, A. 398 Rimm-Kaufman, S.E. 417, 446 Ring, J. 417, 447 Risley, T.R. 413, 414, 416, 426, 442 Ritter, P. 443 Rivera, J. 404 Roazzi, A. 413, 441 Robbins, L.N. 55, 58, 67, 81, 103, 104 Robbins, P. 62, 68, 69, 72, 98 Robbins, T. 274, 325 Roberts, A. 274, 325 Roberts, D. 442 Roberts, J. 374, 405, 414, 441, 446 Roberts, R.R. 267 Robertson, E. 403 Robinson, E. 38, 42 Robinson, J. 203, 210, 225, 396 Robinson, R.J. 343, 356 Rock, S.L. 422, 440 Rodrigues, L.A. 127, 128, 134 Rodriguez, K. 124, 135 Roediger, H.L. 200, 226

Roesner, R.W. 437, 446 Roff, J.D. 51, 67, 103 Roff, M. 67, 103 Rogers, J.E. 12, 42 Rogoff, B. 145, 146, 171, 216, 224, 413, 446 Rohde, C.A. 357 Roos, N. 399 Rose, S. 191, 226, 384, 405 Rose-Krasner, L. 418, 446 Rosen, J. 402, 405 Rosenberg, A. 103 Rosenberg, L.A. 357 Rosenbluth, L. 443 Rosengren, K.S. 40 Roskind, W. 287, 326 Rosser, R. 190, 227 Rosvold, H.E. 180, 221, 226 Rothbart, M.K. 325, 414, 420, 436, 446 Rowat, W.L. 187, 219 Rowden, L. 99 Rowe, D. 365, 405 Rubin, B.R. 50, 98 Rubin, K.H. 55, 56, 59, 97, 99, 102, 103, 104, 420, 446 Rudman, P. 42 Ruff, H. 379, 380, 405 Ruffman, T. 190, 217 Runco, M. 274, 325 Ruopp, R. 429, 446 Rushton, J.P. 52, 104 Russell, J. 170, 236, 238, 268 Rutter, M. 74, 93, 94, 104, 141, 171, 346, 360, 367, 396, 405 Rvachew, S. 374, 405 Ryan, A. 390, 396 Ryan, R.H. 416, 427, 447 Rydell, A. 386, 394, 400

S Sachs, J. 232, 233, 235, 268 Saco-Pollitt, C. 371, 375, 404, 405 Saenger, G. 406 Saint-Cyr, J.A. 326 Salkever, D.S. 357 Salmona, M. 385, 407 Saltz, E. 287, 326 Salzinger, S. 356 Sameroff, A. 364, 405, 437, 446

Author Index Sampson, R. 368, 405 Samuels, S.J. 106, 135 Samuelson, R.J. 186, 204, 206, 225 Sanchez, G. 405 Sandall, S. 374, 401 Sandberg, E. 188, 222 Sandiford, P. 393, 405 Sandson, J. 282, 320 Sapir, A. 275, 324 Sapp, F. 287, 325 Sarty, M. 198, 221 Sasaki, T. 403 Sasson, N.J. 225 Satz, P. 372, 406 Savage, J. 186, 204, 221 Savin-Williams, R.C. 68, 104 Savioli, L. 407 Sawaguchi, T. 211, 226 Sazawal, S. 387, 406 Scarr, S. 365, 406, 428, 441 Schacter, D.L. 174, 200, 226 Schaffer, S. 377, 406 Schank, R.C. 140, 171 Schatschneider, C. 265, 268, 442 Scheper-Hughes, N. 345, 360, 386, 393, 406 Schmidt, C.R. 291, 325 Schmitz, K. 346, 350, 360 Schneider, W. 177, 203, 226 Schneirla, T.C. 223 Schober, M.F. 281, 290, 325 Schoenbaum, S. 402 Schoepflin, U. 395, 406 Scholnick, E.K. 265, 267 Schooler, J.W. 274, 322 Schore, A. 366, 406 Schroeder, D. 402 Schroeder, S. 380, 403, 406 Schulman, R.G. 224 Schulsinger, F. 58, 100 Schumacher, E.H. 223, 226 Schwartz, B.B. 186, 193, 196, 198, 226 Schwartzman, A.E. 56, 60, 61, 101 Schyns, P.G. 284, 326 Scola, P. 374, 406 Scott, K. 381, 401 Scott, R.A. 358 Segall, M. 369, 406 Seid, M. 53, 101 Seidenberg, M.S. 446

Seidman, S. 12, 41 Sellmann, A. 446 Semecesen, T. 323 Seppanen, P.S. 414, 418, 441 Seroczynski, A.D. 97 Sethuraman, K. 381, 382, 384, 400 Settlage, P.H. 180, 224 Seuss, D. 114, 135 Sexton, M. 379, 400, 402 Shaffer, D. 101 Shaheen, F. 408 Shaheen, S. 379, 406 Shallice, T. 275, 323 Shannon, M. 380, 406 Shantz, C.U. 91, 104 Shapiro, L.R. 241, 267 Shapp, L. 338, 360 Share, D. 121, 123, 135 Sharoni, L.J. 114, 134 Sharples, M. 42 Shatz, M. 296, 326 Shennum, W. 339, 357 Sherman, L. 370, 397 Sherrod, L. 396 Sherry, D.F. 174, 200, 226 Shiffrin, R.M. 176, 217 Shillam, P. 404 Shin, R.K. 219 Shoben, E.J. 280, 324 Shonkoff, J. 374, 406, 414, 415, 416, 418, 420, 446 Short, E.J. 265, 268 Shrout, P. 408 Shulka, R. 399 Shultz, T.R. 260, 268 Shure, M.B. 97 Sia, C.J. 357 Siegal, M. 285, 326 Siegel, I.S. 287, 288, 326 Siegel, L.S. 414, 416, 422, 446 Siegler, R.S. 224, 263, 268, 276, 286, 318, 326 Sigel, I.E. 12, 42 Sigman, M. 383, 406, 408 Silansky, M. 360 Silberberg, M.C. 106, 135 Silberberg, N.E. 106, 135 Silva, P.A. 104 Silverman, J.L. 241, 246, 252, 268 Simeon, D. 381, 384, 406

467

468 Simons, R.L. 426, 444 Simpson, T.L. 321 Singer, L. 368, 406 Singerman, J.D. 225 Singh, J. 221 Singleton, L.C. 95 Skuse, D. 394, 409 Slavkovich, V. 408 Slawinski, E. 405 Slaymaker, F.L. 359 Slee, P.T. 104 Sloman, J. 396 Slomkowski, C. 169 Sloper, P. 334, 335, 360 Slusarcick, A.L. 52, 98 Smith, A. 414, 416, 446 Smith, C. 402 Smith, E.E. 181, 208, 219, 223, 226 Smith, E.R. 284, 326 Smith, K.E. 444 Smith, K.R. 336, 337, 358 Smith, L.B. 202, 226, 284, 285, 309, 326, 412, 413, 447 Smith, P.K. 64, 96 Smith, S.M. 274, 277, 284, 322 Smith, S.S. 426, 446 Smoot, D.L. 92, 96 Snidman, N. 55, 103 Snow, C.E. 106, 135, 416, 446 Snyder, C. 440 Snyder, J. 213, 226 Sobhy, A. 408 Solomon, J. 425, 444 Soloway, R.M. 236, 269 Soma, S. 398 Sonnenschein, S. 291, 326 Sontag, S. 10, 42 Soong, W. 380, 404 Sophian, C. 202, 203, 226, 317, 322 Soroker, E. 58, 102 Sosa, B.S. 241, 267 Sotak, L. 125, 135 Sourander, A. 373, 406 Speece, D.L. 417, 442 Speer, J.R. 310, 326 Speltz, M.L. 336, 357 Spelz, M.L. 336, 360 Spencer, M.B. 53, 97 Spiers, P. 372, 400 Spietz, A.L. 440

Author Index Spiker, D. 350, 358 Spinazzola, J. 215, 227, 317, 327 Spritz, B. 153, 168 Sroka, I. 12, 33, 42 Staghezza-Jaramillo, B. 408 Stattin, H. 50, 54, 102 St.Clair, C.C. 343, 360 St.Clair, R.C. 360 Stebbins, G.T. 221 Steele, C.M. 337, 360 Stein, M.R. 65, 98 Stein, N.L. 149, 170 Stein, Z. 406, 408 Steinberg, L. 366, 367, 397, 406, 407, 423, 428, 443, 444, 447 Stemberger, J.P. 282, 326 Stemmer, B. 274, 326 Stern, D.N. 143, 171 Stevenson, H.W. 414, 415, 427, 428, 447 Stevenson, L.M. 20, 41 Stewart, S.L. 97 Stigler, J.W. 428, 447 Stipeck, D.H. 416, 427, 447 Stolzfus, R. 370, 375, 381, 385, 407 Straus, M.A. 351, 360 Strobini, D.M. 428, 429, 445 Strupp, B. 385, 391, 402, 407 Stuart, M. 384, 400 Stummer, S. 325 Styles, E.A. 275, 286, 320 Succop, P. 399 Suengas, A.G. 267 Sugar, M. 350, 360 Sullivan, H.S. 61, 104 Sullivan, K. 325 Sullivan, P.M. 334, 336, 339, 360 Suozzo, M. 401 Super, C. 394, 407 Super, D. 399 Susman, E.J. 48, 102 Susser, M. 406 Svaib, T.A. 321 Svec, W.R. 282, 322 Swank, P.R. 444 Sweifi, E. 408 Szarkowski, J. 8, 42 Szechter, L.E. 9, 12, 16, 31, 37, 41, 42 Szilagyi, P. 377, 400, 406 Szwarcbart, M.K. 226

469

Author Index T Tabors, P.O. 305, 321 Tacconi, M. 385, 407 Tager-Flusberg, H. 325 Talbert, J. 366, 407 Talmi, A. 233, 266 Tamis-LeMonda, C. 367, 407, 422, 447 Tamplin, A. 99 Tarvydas, V.M. 333, 360 Tatareswicz, J.E. 359 Taylor, A.E. 326 Taylor, C. 322 Taylor, K. 401 Teale, W.H. 447 Teberosky, A. 130, 134 Teichmann, R. 230, 268 Tellegen, A. 74, 98, 359 Tellez, W. 396 Tennen, H. 347, 356 Terry, R. 96, 97 Tesla, C. 169 Teti, D. 152, 171 Thal, D. 220 Thatcher, R. 379, 401 Thelen, E. 202, 226, 285, 326, 412, 413, 447 Thomas, G.V. 12, 33, 38, 42 Thomas, K.M. 224 Thomas, R. 372, 407 Thompson, C. 241, 268 Thompson, G. 379, 407 Thompson, R.A. 141, 142, 144, 151, 154, 158, 159, 160, 162, 163, 164, 165, 170, 171 Thomson, N. 222 Thurman, S.K. 360 Thurston, J.R. 52, 98 Thyer, N. 409 Tielsch, J. 407 Tincoff, R. 129, 135 Tinsley, B.R. 95 Titelbaum, L. 396 Titmus, G. 99 Tivnan, T. 414, 417, 418, 441 Tizard, J. 102 Todd, P.M. 343, 357 Tokunaga, K. 403 Tolan, P.H. 51, 54, 104 Tomaka, J. 339, 360 Tomasello, M. 145, 171, 327 Tonascia, J. 374, 403

Tong, S. 403 Torrance, E.P. 274, 326 Torrey, E.F. 397 Towse, J.N. 294, 296, 326 Trainor, R.J. 218 Traissac, P. 396 Tramer, S. 401 Tranel, D. 282, 326 Travers, J. 446 Treiman, R. 106, 117, 119, 120, 121, 122, 124, 125, 129, 130, 131, 134, 135 Tremblay, R.E. 51, 104 Triana, N. 408 Trillingsgaard, A. 398 Trivers, R. 340, 341, 360 Tronick, E. 372, 407 Troop, W.P. 63, 81, 82, 101 Troseth, G.L. 37, 42 Trost, M.A. 97 Troyer, L. 443 Truglio, R. 97 Truwit, C.L. 224 Tsuang, M. 397 Tuholski, S.W. 177, 179, 201, 220 Tulving, E. 174, 200, 226 Turek, V. 358 Turkewitz, G. 350, 360 Turnbull, D. 6, 42 Turner, P. 6, 38, 42 Turner, R. 218 Turner, S. 334, 335, 360 Tzeng, O.J.L. 236, 268

U Underwood, K. 69, 71, 96 Underwood, M.K. 51, 98 Ungerleider, L.G. 208, 219, 226 UNICEF 381, 407 Upton, J. 8, 41 U.S.Department of Education 447 Uttal, D.H. 40, 427, 447

V Vaillancourt, T. 102 Valenzuela, M. 386, 407

470

Author Index

van IJzendoorn, M.H. 140, 171, 421, 441 van Leeuwen, T. 9, 41 Vandell, D. 69, 72, 104, 368, 407 Vandeven, A. 390, 396 Vandiver, T. 386, 407 Varma, S. 409 Venezky, R.L. 122, 135 Vermeeren, A. 236, 267 Verschueren, K. 144, 171 Villaume, S.K. 119, 135 Villena, A. 371, 399 Vimpani, G. 403 Vinter, A. 317, 325 Vitaro, F. 104 Vitkovitch, M. 282, 286, 326 Volterra, V. 217 vom Saal, F.S. 342, 360 von Hofsten, C. 200, 218 von Wright, J.M. 236, 268 Vorhees, C. 380, 407 Vrenezi, N. 408 Vuorisalo, T. 359 Vurpillot, E. 291, 326 Vutchov, M. 399 Vygotsky, L. 38, 42, 278, 327

W Waas, J.R. 360 Wachs, T.D. 364, 365, 367, 368, 381, 382, 383, 384, 385, 387, 388, 389, 390, 392, 393, 394, 396, 402, 403, 407, 408 Wachtel, G.F. 287, 324 Waddington, S.R. 335, 360 Wahlsten, D. 81, 99, 412, 413, 443 Wainright, P.E. 385, 386, 409 Walberg, H.J. 331, 360 Waldman, I. 101 Walka, H. 383, 387, 408 Walker, D. 414, 426, 447 Walker, J.A. 225 Walker, S. 366, 387, 388, 394, 399, 400, 405 Wall, S. 440 Walther, T. 10, 42 Wang, J. 380, 404 Wang, M.C. 331, 360 Wanlass, R.L. 55, 104 Ward, T.B. 274, 277, 284, 322

Wardrop, J. 65, 71, 81, 100 Warry, R. 204, 221 Wasserman, G. 336, 360, 379, 380, 408 Waternaux, C. 396 Waters, E. 143, 151, 171, 440 Waters, H.A. 171 Watkins, W. 375, 385, 408 Watson, J. 171 Watt, N.F. 97 Watts, S. 376, 409 Wauben, P.M. 385, 386, 409 Wayland, K.K. 75, 101 Weatherston, S. 122, 130, 135 Webb, R.A. 192, 226 Webb, S.J. 210, 211, 226 Webster, A. 409 Weinert, F.E. 414, 447 Weinfurt, K.P. 438, 447 Weinstein, R.S. 45, 53, 69, 103 Weisberg, R. 291, 323 Weiskrantz, L. 180, 222, 274, 325 Weiss, M. 334, 361 Weist, M.D. 102 Weist, R.M. 232, 233, 235, 268, 269 Weitzman, M. 377, 406 Welch-Ross, M.K. 143, 147, 171 Wellman, H. 143, 168, 171, 185, 203, 226, 291, 317, 322, 325, 327 Wells, L. 6, 38, 41 Wentzel, K.R. 78, 104, 418, 447 Werner, E. 346, 361, 366, 409 Werner, N.E. 52, 104 Werz, M.A. 186, 204, 224 West, D.J. 104 West, M. 401 West, S.G. 97 Westerberg, H. 182, 223 Whaley, S. 382, 406 Wharton-McDonald, R. 446 Whitaker, H. 292, 321 White, J.L. 51, 104 White, K.R. 426, 447 Whitehurst, G.J. 291, 326, 421, 445, 447 Whiteside, L. 397 Wichmann, C. 97 Wicks, J. 401 Wigg, N. 403 Wilcox, T. 190, 226 Wilkie, D.M. 187, 204, 223 Wilkinson, A. 447

471

Author Index Willatts, P. 215, 216, 226, 382, 409 Williams, L. 409 Williams, M. 405 Williams, S. 442 Williamson, G. 414, 415, 416, 445 Willis, C. 221 Wilson, F.A.W. 208, 227 Wilson, L.C. 119, 135 Wilson, M. 336, 340, 345, 357 Wilson, M.S. 169 Wimmer, H. 132, 135 Windhan, A.M. 357 Winick, M. 387, 388, 409 Winneke, G. 377, 378, 379, 409 Winner, E. 29, 42, 325 Winograd, E. 236, 269 Wirt, D. 51, 67, 103 Wise, B.W. 417, 447 Witkin, H. 389, 409 Witkowska-Stadnik, K. 269 Wolf, A. 377, 380, 409 Wolke, D. 394, 409 Woody, E. 92, 99 Woody-Ramsey, J. 324 Worden, P.E. 119, 135 Worobey, J. 402 Wright, J.C. 87, 104 Wright, V. 97 Wroble, M. 409 Wu, P.Y.K. 198, 221 Wysocka, H. 269

Y Yacoob, M. 376, 409 Yamashita, T. 406

Yangzom, Y. 401 Yaschine, J. 386, 397 Yau, K.T. 358 Yehuda, S. 409 Yen, L. 287, 289, 313, 316, 322, 327 Yengo, L. 202, 226 Yerys, B. 296, 325 Yeterian, E.H. 274, 325 Yeung, N. 307, 319, 325 Yip, R. 409 Yoder, K. 405 Youngblade, L. 169 Yuill, N. 314, 327 Yunis, F. 407, 408

Z Zaia, A.F. 99 Zaitchik, D. 38, 42 Zansen, A. 406 Zaucha, K. 406 Zavelata, N. 403 Zeedyk, M.S. 216, 227 Zeisel, S.A. 441 Zeitlin, M. 399 Zelazo, P.D. 185, 215, 223, 227, 278, 279, 285, 286, 292, 293, 294, 295, 296, 298, 302, 313, 317, 319, 323, 327 Zelazo, P.R. 203, 223 Zeltzer, L. 374, 396 Zeno, S.M. 118, 125, 135 Zilles, K. 210, 222 Zirpoli, T.J. 335, 361 Zoller, D. 146, 168 Zubin, N.R.E. 238, 239, 252, 254, 267


Subject Index

Note: Please note that the letter f or t following a page reference indicates figures and tables respectively

A

preterm vs. full-term infants, 213 Anthropology, child development, 364 Antisocial conduct, aggressive behavior and, 50–52 Apgar score, 350 Aphasia, language studies, 282 Appearance impairment, social stimulus features (effect on others), 334, 336–337 Art as representations, 4–5 Attachment Q-Sort (AQS), 151–152 Attachment security, 139 assessment, 144–145, 151–152 changes in, 143 conscience development, 161, 162–163 emotional understanding and, 151–153, 157–158 impact, 165–168 mother–child conversations, 161 Attachment theory, 138–139 internal working models, 139–141 language role, 145 personality development, 140 problems/limitations, 141 Attention, 176 hearing impairment, 335 working memory and, 178 Autobiographical events, past–future confusion, 258 Autobiographical memory, 143

Abstract thought, development, 144 Abuse of children with MPDs, 338, 339, 344 infanticide, 336, 340 socio-economic status and, 344–345 unresponsive children, 335–336, 339 Aesthetics, 9–10 Affective disorders child by environment models, 76–77, 77f withdrawn behavior and, 59 Aggressive behavior aggressive-withdrawn, 57, 59–60, 61 maladjustment and, 47–54 antisocial conduct/psychopathology, 50–52 internalization, 51 predictive nature, 49–50, 50–51, 54 prevention programs, 53 school adjustment, 52–54, 78–80 stability coefficients, 48–49, 49t types of, 51–52 peer relationships, 53, 78–80, 81–84 pre-school, 53–54 Altitude, developmental impact, 371–372 American Disabilities Act (ADA), 332–333 Animal studies memory, 180, 186, 208–209 human studies vs., 206 parental investment theory, 342–344 Annual events, sense of, 242–244 daily events vs., 248–250 past–future confusion, 256–257 A-not-B paradigm, 185–186 gaze vs. reach discrepancy, 199–201

B Bacterial infections, developmental impact, 373–375 Baldwin, J.M., working memory, 176 473

474

Subject Index

Behavioral genetics, child development, 364 Behavioral internalization, 160, 161 Behavioral style and adjustment child by environment models, 74–80 duration of effects, 93–95 main effects models, 47–61 Belief systems, internal working models, 149 Bio-ecological environment, 368–371 altitude, 371–372 bacterial infections, 373–375 climate/seasonality, 370, 372–373 definition, 363, 369 developmental impact, 371–388 active covariance, 376, 386t moderating factors, 379–380, 387t, 388 passive covariance, 371–372, 374, 386t evidence, 370–371 individual differences in exposure, 370 nutritional status, 380–388, 382t, 383t parasitic infections, 375–376 physical hazards, 372 psychosocial environment similarity, 394 psychosocial framework integration, 388–395, 390f functional isolation, 391–394, 392f research implications, 394–395 toxin exposure, 370, 376–380 covariates, 380 moderating factors, 379–380 see also specific toxins types of variation, 369–370 viral infections, 373–375 see also Psychosocial environment Birth defects, impact of, 336 Blame, medical and physical disorders (MPDs), 338 Blood–brain barrier, lead effects, 377 Brain damage, language defects, 282–283 Bullying see Victimization in peer relationships

C Calendar systems, past–present–future distinction, 231 ‘‘Cartographic eye’’, 6 Categorization, future events, 250 Centrated thinking, 286

Cerebral asymmetry, language, 282–283 Child by environment models, 73–84 additive models, 74, 75, 87–88, 91 adjustment and, 74–80 peer relationship duration, 80–84, 83f psychopathology, 75–77, 77f school adjustment, 78–80, 79f assumptions, 86 continuity models, 86–87 other models vs., 90–91 debate, 92–93 history, 44–45 mediator models, 75, 76, 88–89, 91 moderator models, 74–75, 76, 86, 88, 91 resources, 74 risk factors, 74 support for, goodness of fit, 85–86 validity, 91–92 see also Main effects models Child care, literacy development, 428–430 Child effects models, 44 Chinese culture, parenting style, 423 Cigarette smoke, developmental impact, 376–377 Climate, developmental impact, 372–373 Clock systems, past–present–future distinction, 231 Cognition development see Cognitive development flexible see Flexible cognition inflexible, 278–279, 286, 288–289, 310–311 see also Perseveration metacognitive abilities, 317 Cognitive ability lead level correlation, 378 literacy development, 416–417 nutritional status and, 380–388, 382t Cognitive Complexity and Control (CCC) theory, 296–301, 303 Cognitive development, 138 flexibility, 277–279 internal working models, 141–150 language role, 145–148 mental representations, 142–145 preschool, 315–316 temporal orientation and, 262–264 working memory, 212 effects, 213–216

475

Subject Index Cognitive neuroscience flexible cognition, 274–275, 281, 282–283 working memory studies, 179–182, 208–210 Cognitive semantics, 281–282 Cognitive stimulation, literacy development, 422–423, 424 Collaborators, reading style, 421–422 Committed compliance, 160 Communication parent–child see Parent–child communication photographic representations, 8–11 Comprehenders, reading style, 421–422 Conceptual flexibility, 287 Conceptual knowledge, dynamic, 283–284 Conceptual mappings, 281–282, 290 symbolic/lexical, 295–296 Conflict resolution strategies, 164 Conscience, development, 158–163 assessment of, 160 conflict resolution strategies and, 164 prediction, 161–162 Consonants, 109–110 discriminability, 111 Continuants, discriminability, 111 Creativity, flexible cognition, 274 Cross-cultural psychology, bio-ecological environment, 369 Cryptosporidium parvum, developmental impact, 376 Culture developmental role, 413 parenting differences response to physical hazards, 372 styles, 423 past–present–future distinction, 230, 261 role in photographic comprehension, 9 see also Ethnicity

age-related changes, 205–206, 205f gaze vs. reach discrepancy, 199–201 methodological challenges, 203–204 monkeys vs., 208–209 perseveration, 201–202 oculomotor, 195 see also Working memory Dependent children, social stimulus features (effect on others), 334–335 Depressive disorders, child by environment models, 76–77, 77f Describers, reading style, 421–422 Developmental neuroscience thriving/resilience, 346–347, 355 working memory in infants, 208–210 Developmental psychology bio-ecological environment, 369–370 child development, 364 dynamic systems of literacy development, 413 Dimensional Change Card-Sort test (DCCS), 292–294, 293f, 294t complex rules, 296–301, 297f three rule variant (3DCCS), 296, 300f, 299–301, 301f see also Flexible cognition Disability see Medical and physical disorders (MPDs) Disability rights activists, 333 Discourse flexible processing adults, 280–281 children, 290–303 parent–child see Parent–child communication Discrimination, medical and physical disorders (MPDs), 332 Divorce, medical and physical disorders (MPDs), 335, 337 Drawings as representations, 4–5 Dynamic systems, literacy development, 413

D Daily events, sense of, 244–247 annual events vs., 248–250 past–future confusion, 257 Delayed-response (DR) tasks, 174 animal studies, 180 infants, 184–185, 194–196

E Egocentrism, temporal concepts, 262 Ego-deictic representations, 37–38 Elaborative communication style, 149, 156, 161

476

Subject Index

Electroencephalography (EEG), infant working memory, 209–210 Electrophysiology, working memory, 180, 209–210 Embedded Model, photographic comprehension development, 38 Emotion early understandings, 137–172 assessment, 152 attachment security and, 151–153, 157–158 conscience and, 158–163 family conflict, 163–164 mother–child conversation and, 147, 150, 154–158, 156f, 165–167 relationships impact, 165–167 valence role, 153 parenting style, 424 photographic communication, 9–10 Verbal Description task, 24–27, 25f regulation and social skills, 420 temporal orientation and, 260–261 Environment, 363–410 bio-ecological see Bio-ecological environment concept, 364 definition, 365 development role, 363–365, 413 integrated framework, 388–395 memory, 213 school behavior/relationship effects, 44–104 psychosocial see Psychosocial environment Episodic memory, 174 Ethnicity literacy development, 423, 427–428 parenting styles, 423 see also Culture Event representation, mother–child conversation, 147 Evolutionary psychology, child development, 364 Executive function, problem solving, 213 Explicit reasoning, 277 Externalization child by environment models, 75 main effects model, 59–60, 67 peer relationships and, 67 withdrawn behavior and, 59–60

F Faces, infant working memory, 198–199 Family conflict medical and physical disorders (MPDs), 335 parent–infant discourse, 163–164 Family learning environment definition, 420–421 literacy development, 420–422, 426 prediction, 421 reading styles, 421–422 statistical analysis, 437–439 Family Stress Checklist, 350 Feeding strategies, parental investment, 343, 345 Flexible cognition, 275–279 adaptive behavior vs., 276 adults, 279–283 brain damage and, 282–283 discourse and shared meaning, 280–281 updating conceptual mappings, 281–282, 290 assessment measures, 275, 276–277, 315–316 appearance–reality test, 287 classification tasks, 295 Dimensional Change Card-Sort tests, 292–294, 293f, 296–301, 297f, 300 Flexible Induction of Meaning test, 305–310, 305f Indeterminacy Detection task, 291–292 methodological problems, 318–319 response times (RT), 275, 319 Stroop Day/Night test, 302 see also specific tests children, 285–317 assumptions about, 283–284 common factors, 314–316 developmental ecology, 277–279, 314–316, 319–320 discourse flexibility, 290–303 flexible naming, 286–289 indeterminacy detection, 291–292 inhibition of prior thoughts, 278, 286, 300–303, 311–313, 317–320

477

Subject Index language development and, 278, 286, 287 narrative role, 290–291 rule-following, 292–303, 294t, 301f, 301t, 316 word meanings, 303–315, 308f, 315t see also Perseveration cognitive neuroscience, 274–275, 281, 282–283 conceptual flexibility, 287 creativity and, 274 definition, 275–277 function, 277–278, 319 Gestaltist approach, 274 as hallmark of intelligence, 272 historical aspects, 273–275 language processing, 271–328 Lava Lamp metaphor, 284–285 logical/metacognitive ability and, 317 partial flexibility, 310 pragmatic flexibility, 282–283 representations, 280, 284–285 limiting factors, 288–289 model, 283–285, 316, 317–318 within-subjects variable, 276 working memory role, 300–303 Flexible Induction of Meaning (FIM) test, 305–310 animates (FIM-An), 305 method, 305–306 object (FIM-Ob), 305, 305f, 311–313 results, 307–310, 308f, 311–313 use, 306–307 Foraging tasks, 174, 187 methodological challenges, 204–205 Friendship definition, 62 psychopathology modification, 68, 82 school adjustment and, 70 stability, 65–66 Frontal cortex inhibitory role, 302 language role, 282 see also Prefrontal cortex Functional isolation, 391–394, 392f definition, 392 evidence, 393–394 validity, 394 Functional neuroimaging, working memory, 181–182, 211

Future events adult’s understanding, 259–260 children’s understanding, 260–261 concepts of, 259–261 development of sense, 240–251 annual events, 242–244 annual events vs. daily events, 248–250, 251 categorization, 250 daily events, 244–247 day vs. year timescales, 249–250 importance, 265 language role, 232–234, 240 past–future confusion, 254–259 past sense vs., 251–252 picture-pointing tasks, 241–242, 243–244 time-pattern dependent, 243 representations, 244, 245 age-related changes, 252–253 flexibility, 246–247, 250, 251 privileged reference point, 248 see also Past–present–future distinction Futurity, concept, 259

G

Gender, lead exposure, 379 Gestalt psychology, flexible cognition, 274 Giardia lamblia, developmental impact, 376 Glucose utilization, working memory development, 211

H

Head Start program, 421 Healthy Start model, 351 Hearing impairment, social impact, 335 Hebrew letter names literacy acquisition and, 128, 129 phonological patterning, 113 spelling, 130 Hide-and-seek tasks, infant working memory, 188–189

478

Subject Index

Hierarchical linear modeling (HLM), literacy development, 431, 437–438, 439 HOME Inventory checklist, 422, 424

Intrauterine growth retardation (IUGR), 381 Iron deficiency, 381, 385 anemia, 375, 384, 385, 393 Isolated behavior, 55

I J Iconicity, letters, 109–110, 120, 128, 133 Infanticide, children with MPDs, 336, 340, 344 Infants anticipation, perceptual–motor tasks, 241 attachment, 144–145 premature, 350 relationship with parent see Parent–infant relationship working memory, 173–231 Infections, developmental impact, 373–376 Information-processing cognitive development, 212 short-term storage models, 176–177 Insight, 142, 143 Intelligence flexible cognition, 272 literacy development, 416 see also Cognitive ability Intelligence quotient (IQ) literacy development, 416, 433, 437–439 socio-economic status and, 426 Intentionality, 216 Internalization aggressive behavior and, 51 child by environment models, 75 main effects models, 51, 57–59, 67 peer relationships and, 67 withdrawn behavior and, 57–59 Internal working models (of emotion, morality and self), 137–172 cognitive growth, 141–150 development of, 141, 150–164 hierarchical organization, 142 impact, 165–168 relationships, 139–141 Interpersonal relationships see Relationships

James, W, working memory, 175–176, 216

L Language acquisition errors, 286 flexible cognition and, 278, 286, 287 literacy and, 416–417 otitis media and impairment, 374 temporal language, 232–234, 240 see also Letter learning ambiguity, 291–292 cerebral asymmetry, 282–283 cognitive development, 145–148 internal representations, 143 working memory role, 212–213 cognitive flexibility, 271–328 see also Flexible cognition; Word meaning comprehension, temporal language, 232–233 neuropsychology, 274–275, 281, 282–283 production, temporal language, 232–233 see also Literacy Latin, phonological patterning, 112–113 Lava Lamp metaphor of flexible cognition, 284–285 Lead exposure, 377–380 behavioral effects, 378 cognitive effects, 378 covariates, 380 models of, 378–379 moderating factors age, 379 gender, 379 nutritional status, 379–380 neurodevelopmental effects, 377

Subject Index Learning-related social skills definition, 417–418 literacy/academic development, 417–419, 425, 433–435 parenting style and, 425 Letter learning, 113–123 age of, 114 case effects, 114, 115–116, 117, 117t, 118 distinctiveness, 115–116, 116t, 117t exposure effects, 118–119 frequency effect, 118 name–sound relationships, 121–123 predictor variables case similarity, 116, 116t, 117t phonological similarity, 116–117, 116t, 117t visual similarity, 116, 116t, 117t principled learning, 107 rote memorization, 107, 119 shape–name associations, 114–119, 115t, 116t, 117t teaching practice, 123 see also Literacy Letter names, 108–113 discriminability, 110–111 iconicity, 109–110, 120, 128, 133 inter-language differences, 106–107, 109, 112–113, 127–128, 130, 132, 133 learning see Letter learning legality, 113 literacy acquisition role, 123–128 indirect effects, 128, 130–131 positive/negative effects, 122–123, 133 reading, 123–128 reliability, 125–127, 126t, 129 spelling, 129–132 name–shape relationship, 114, 132–133 name–sound relationship, 121 phonetics, 107, 108, 131 phonological patterning, 112–113, 114 pre-school knowledge, 105–106 Letter shapes iconicity, 110 learning see Letter learning Letter sounds learning, 119–123 name–sound relationship, 121 phonetics, 107, 108, 131 pre-school knowledge, 120–121

479

Lexical mapping, flexible cognition, 295–296 Linguistic motivation, 110 Literacy, 411–447 child factors, 413, 416–420 general social skills, 417 IQ, 416, 433, 437–439 language/phonological skills, 416–417 learning-related social skills, 417–419, 433–435 temperament, 419–420, 435 dynamic systems perspective, 413 early variation, 414, 415 letter names in acquisition, 105–138 parenting factors, 413, 420–426 cognitive stimulation, 422–423, 424 family learning environment, 420–422, 426, 437–439 impact, 425–426 parenting style, 423–425, 435 reading style, 421–422 pre-school, 415 principled learning, 107, 124 relationships between factors, 431–439 age-related changes, 435 child–schooling interactions, 436–437 complex model (Bachman and Morrison), 431–433, 432f interacting relations, 433–434 intervening relations, 434 rote memorization, 107, 119, 124 sociocultural factors, 413, 414, 426–430, 434 child care, 428–430 ethnicity, 427–428 parental education, 426, 429, 433, 434 schooling, 430–431, 436–437 socio-economic status, 426–427 spelling, 129–132 statistical methods, 437–439 hierarchical linear modeling (HLM), 431, 437–438, 439 structural equation modeling (SEM), 431, 432, 439 see also Language; Letter learning; Letter names; Reading Locutions, 280, 283 Loneliness, bullying and, 71

480

Subject Index M

Main effects models, 47–73, 364 behavioral style and adjustment, 47–61 aggressive behavior, 47–54, 49t ‘‘risky behavior,’’ 73 withdrawn behavior, 54–61 criticism, 44, 73 lead exposure, developmental impact, 378–379 peer relationships and adjustment, 61–73 ‘‘risky relationships’’, 73 support for, 85 see also Child by environment models Malnutrition epidemiology, 381 functional isolation and, 391–394, 392f postnatal, 382t, 383t moderate–severe, 384 severe micronutrient deficiency, 384–385 prenatal, 381–382, 382t, 383t, 384 Maps as representations, 4, 6 Maternal language see Mother–child conversations Medical and physical disorders (MPDs) anti-discrimination legislation, 332–333 bio-social-cognitive approach, 329–362 child health outcomes, 349, 352–354, 354f community-based home visitation, 349 deficit model, 329–330 etiology, 331 integrated approach, 355–356 interactive model, 337–346, 348 animal studies, 342–344 human studies, 344–346, 353, 353f parental cognitions, 337–339 parental investment, 339–346, 348 management, 331 medical model, 331, 337 model comparison/integration, 347–348 optimizing outcomes, 349–355 parental abuse, 338, 339 geographical variations, 345 infanticide, 336, 340, 344 preventative strategies, 349–355 risk assessment, 350 socio-economic status and, 344–345, 350 unresponsive children, 335–336, 339

parental conflict/divorce, 335, 337 parental distress, 334–335 parental empowerment, 349, 351–355 child health effects, 352–354, 354f harsh parenting/abuse effects, 351–352, 352f Healthy Start model, 351 parental investment effects, 352–353, 353f parental neglect, 335, 336 parental resource holding potential (RHP), 349 psychological effects, 332 resilience model, 346, 348 risk assessment, 350 social model, 332–337, 348 social stigma, 332–333 social stimulus features (effect on others), 332, 333–337, 348 appearance impairment, 334, 336–337 dependent children, 334–335 unresponsive children, 334, 335–336, 339 temperament and, 346 thriving model, 346–347 Memory autobiographical, 143 definitions, 173–174 episodic, 174 short-term, 174, 177 temporal adults, 235–236 children, 237–238, 251–252, 261, 264, 265 see also Temporal orientation working memory see Working memory Mental representations, 142 flexible cognition, 280, 284–285 metarepresentational, 144 short-term visual, 191 temporal, 231, 241, 244, 245, 248 age-related changes, 252–253, 262, 263 flexibility, 246–247, 250, 251 Mercury, developmental impact, 377 Metacognition flexibility and, 317 metarepresentational development, 144 Metamemory, development, 265 Metaphor, 272

481

Subject Index Micronutrient deficiency, 382t, 383t, 384, 393 severe, 384–385 Morality, early understandings, 137–172 family conflict, 163–164 mother–child conversation, 147, 159–163, 165–167 relationships impact, 165–167 Mother–child conversations attachment security, 161 discipline encounters/conflict, 159, 163–164 emotional understanding, 147, 150, 154–158, 156f internal working model development, 146–148 moral understanding, 147, 159–163 narrative style, 148–150, 155–157, 156f Motor control, development, 200 Multi-aspect representation (MAR), flexible cognition, 284–285, 315 Multi-aspect representational medium (MARM), flexible cognition, 284–285, 317–318

N

Narrative, flexible cognition development, 290–291 Narrative style, 148–150 emotional understanding, 155–157, 156f n-back task, 174 functional imaging, 181 Neurodevelopment lead effects, 377 motor control, 200 nutritional status effects, 385–388 visual system, 200 working memory development, 210–212 Neuroimaging, working memory, 181–182, 209–210 development, 211 Neuropsychology, language ability, 274–275, 281, 282–283 Nutritional status and development, 380–388, 382t, 383t chronic undernutrition, 385 cognitive impact, 382t

covariance/interactions, 386–388, 386t–387t functional isolation, 391–394, 392f iron deficiency, 375, 381, 384, 385, 393 lead exposure, moderating factor, 379–380 moderate–severe postnatal malnutrition, 384 moderating factors, 387t, 388 neurodevelopmental effects, 385–388 prenatal malnutrition, 381–382, 384 socio-emotional function, 383t temporal consequences, 387 vitamin/mineral deficiency, 384–385, 393 O Open communication style, 148, 161 Organism–environment covariance, 367–368 lead exposure, 380 nutritional status, 386t, 387–388 Organism–environment interaction, 368, 380 P Paintings as representations, 4–5 Parasitic infections, developmental impact, 375–376 Parental cognitions, effect of children with MPDs, 337–339 Parental conflict, effect of children with MPDs, 335 Parental distress, effect of children with MPDs, 334–335 Parental empowerment, children with MPDs, 349, 350–355 Parental investment theory, 340–342, 348 computer simulations, 343 cost–benefit analysis, 340 evidence animal studies, 342–344 human studies, 344–346, 352–353, 353f reproductive potential, 340, 341 Parental neglect, effect of children with MPDs, 335 see also Abuse of children with MPDs

482

Subject Index

Parental resource holding potential (RHP), 349 Parent–child communication, 146–148, 150–151 book reading, 421, 421–422 developing internal working models, 146–148 emotional understanding, 147, 150, 154–158, 156f family conflict, 163–164 impact, 165–166 moral understanding, 147, 159–163 narrative style, 148–150, 155–156 temporal language development, 233–234, 240 working memory development, 213 see also Mother–child conversations Parent–infant relationship attachment, 139, 158, 161 conscience development, 161–162 impact on emotion, morality and self, 165–168 Parenting, 420 control/discipline, 425 cultural differences, 372, 423 investment theory see Parental investment theory literacy development, 413, 420–426, 435 style, 423–425 definition, 423 warmth/sensitivity/responsivity, 424 see also entries beginning parental Past events adults’ memory for timing, 235–236, 238, 240 adults’ understanding, 259–260 children’s memory for timing, 237–238 adults’ versus, 239–240 children’s understanding, 260–261 concepts of, 259–261 development of sense, 235–240 future sense vs., 251–252 past–future confusion, 254–259 studies, 236–239 distance-based processes, 235, 253 development, 236–237, 238 spatial representation, 239 language, 232–234 location-based processes, 235, 236

development, 236–237, 240 representations, 239, 252–253 see also Past–present–future distinction Past–future confusion, 254–259 annual events, 256–257 daily events, 257 memory for time/future differentiation studies, 254–255 non-cyclic events, 257–258 past-future differentiation studies, 255–258 Pastness, concept, 259 Past–present–future distinction, 230, 253–261 clock/calendar systems, 231 concepts, 259–261 cyclical patterns and, 254 development, 251–253 developmental studies, utility, 231 mechanism, 253 parent–child discourse, 234 past–future confusion, 254–259 as social construct, 230, 254, 261 Peer relationships adaptive significance, 85 child by environment models, 74–84 continuity, 66 duration effects, 80–84, 83f, 93–95 main effects models, 61–73 maladjustment and, 61–84 differential contributions, 72–73 psychopathology, 67–69 school adjustment, 53, 69–72, 78 measures, 65, 66t stability, 63–67, 64t, 66t types, 62 see also specific relationships Peer-report measures, 65 Perceptual endowment, photographic representations, 2, 2f Perseveration, 279, 286 brain damage, 282 naming errors, 287 switching errors, 292–303, 294t complexity effects, 296–301, 301f, 301t inhibition and, 300–303 mapping errors, 295–296 word meaning errors, 310–311 inhibition and, 311–313 working memory tasks, 201–202

Subject Index Personality development, attachment theory, 140 temperament see Temperament Person-based methods, 437 literacy analysis, 437–439 Phenomenology, working memory, 216 Phenylketonuria inhibitory processes, 302 working memory development, 211 Phonemes, 108 missing, 131 Phonetics, 107, 108 Phonological awareness, literacy development, 416–417 Phonological patterning children’s learning, 114 definition, 112 letter names, 112–113 Photographer, communicative role, 10 Photographic comprehension, 1–43 adults vs. children, 22–27, 23f, 28f, 29 cultural variables, 9 development, 33–39 age-related components, 3, 7 constructive processes, 38 experience-related components, 3, 36–38 referential meaning, 33–36, 34t, 35t developmental psychology use, 3, 38–39 factors affecting perception, 1–3, 2f, 5 investigations, 12–33 contrasting pairs, 12–17, 14f explaining preferences, 27–32, 28f, 31f photographic model task, 17–20, 18f, 20f, 21f Verbal Description task, 20–27, 23f, 24f, 25f see also specific tasks social variables, 9 verbalization ability, 13, 17 ‘‘Photographic eye’’, 6 Photographic model task, 17–20, 27 rationale, 17–19, 18f results, 19–20, 20f, 21f Photographic pairs task, 12–17, 14f, 27 communicative/media-specific qualities, 16–17 rationale, 12 vantage-point, 13–16 Photograph preferences, 27–32, 31f

483

children vs. adults, 28f, 29 explanations, 29–30 Photographic representations, 4–6, 4f communicative, 8–11 archival functions, 9 developmental questions, 11 development of understanding, 16–17, 33–36, 34t, 35t emotional communication, 9–10 keeper/curator’s role, 10–11 photographer’s role, 10 viewer’s role, 11 completeness/redundancy, 5 comprehension see Photographic comprehension media-specific, 7–8, 16–17 perceptual similarity, 5 spatial, 6–7, 33–36, 34t, 35t view-specific nature, 6–7 Physical hazards, developmental impact, 372 Physical illness, developmental impact, 374 Picture-pointing task, temporal orientation, 241–242 annual events, 243–244 Portuguese letter naming in literacy acquisition, 127–128 letter-system iconicity, 109 Possible/impossible events paradigm, 190–191 Pragmatic communication style, 149, 156 Pragmatic flexibility, 282–283 Prefrontal cortex development, nutrition and, 385 inhibition and cognitive flexibility, 300 working memory, 180, 181, 183, 208 development, 210–211 see also Frontal cortex Premature infants, medical and physical disorders, 350 Present, development of sense, 237 Principle components analysis, emotion understanding, 156 Problem solving development of skills, 316 word meanings, 303–315 working memory role, 213–214 Prosody, word meaning, 304 ‘‘Prospective memory’’, 265

484

Subject Index

Psychopathology aggressive behavior and, 50–52 child by environment models, 75–77, 77f main effects model, 50–52, 57–60, 67–69 peer relationships and, 67–69 withdrawn behavior and externalization, 59–60 internalization, 57–59 Psychosocial environment, 365–368 bio-ecological environment similarity, 394 bio-ecological framework integration, 388–395, 390f functional isolation, 391–394, 392f Bronfenbrenner’s hierarchical model, 366, 389, 390f definition, 365 development role, 365–366, 367–368 organism–environment covariance, 367–368, 380, 386t organism–environment interaction, 368, 380, 387t specificity, 367 temporal processes, 367, 387t family, 366 moderating factors, 366–367 non-family, 366–367 see also Bio-ecological environment

R Radial arm maze, 186–187 methodological challenges, 204 Reading IQ and learning-related skills, 433–434, 438 letter names and, 123–128 phonological awareness role, 416–417 Rejection in peer relationships, 62 aggressive children, 79–80, 82 psychopathology and, 67–68, 75 school adjustment and, 53, 69–70, 78, 79–80 stability, 64t Relationships aggressive behavior and, 53, 78–80 family conflict, 163–164 impact on emotion, morality and self, 165–168

internal working models, 138, 139–141 representation development, 144 learning opportunities, 138 school adjustment, 46 secure vs. insecure attachment, 139, 140 withdrawn behavior and, 57–59 see also Attachment theory; specific relationships Remembered events, adults memory of timing, 235–236 Repetitive communication style, 149, 156 Representations development of understanding, 33–36, 34t, 35t ego-deictic, 37–38 experience-related components, 3, 36–38 maps, 4, 6 mental see Mental representations paintings/drawings, 4–5 photographs see Photographic representations Resistance to temptation task, 160 Response times (RT), flexible cognition, 275, 318, 319 Reticent behavior, 55 Rule-following and flexible cognition, 292–303, 294t assessment see Dimensional Change Card-Sort test (DCCS) complexity effects, 296–301, 297f, 300 inhibition/memory role, 300–303 symbolic/lexical mapping, 295–296 ‘‘uncertainty stance’’, 316

S Sandbox task, 190 Schizophrenia early withdrawn behavior, 58 seasonality of birth, 373 School adaptation to see School adjustment developmental context, 45 literacy development, 430–431, 436–437 psychosocial environment, 366 socialization context, 45 School adjustment, 44–104 aggressive behavior and, 52–54 child by environment models, 74–80

485

Subject Index definition, 45 interpersonal antecedents, 46 literacy and, 417, 418, 419 main effects perspective, 47–73 peer relationships and, 53, 69–72, 78 ‘‘risk and resilience’’, 74 temperament, 419 withdrawn behavior and, 60–61 Seasonality, developmental impact, 370, 372–373 Self early understandings, 137–172 relationships impact, 167–168 negative perceptions, 58 Self-awareness, 143 Self-conscious reasoning, 277 Self-esteem, friendship role, 68 Self-reflection, 143 Semantic inflexibility, 288–289 Semantic representations, 280 Short-term memory, 174 working memory vs., 177 Short-term visual representations, 191 Siegler’s theory of cognitive development, 263 Social avoidance, appearance impairment, 336 Social constructs, past–present–future distinction, 230, 254, 261 Social experience developmental role, 413 dynamics, 93–94 representational understanding, 36–37 see also Relationships; Social skills Social skills development, 138 critical periods, 94 emotion regulation, 420 general, 417 learning-related, 417–419 literacy development, 417–419 temperament role, 419 Social stigma, children with MPDs, 332–333 Socialization effects models, 44, 61 Sociocultural factors, literacy development, 413, 414, 426–430 Socio-economic status (SES) abuse of children with MPDs, 344–345 developmental effects

physical illness, 374–375 seasonality and, 373 indices, 426 literacy development, 426–427 child care, compensatory effect, 428–430 Socio-emotional functioning, nutritional status effects, 383t Solitary-active behavior, 55 Solitary-passive behavior, 55 Spanish, letter names, 130 Spatial memory, 186, 187, 188, 204–205 Spatial-orientation tasks, 189–190 Spatial representations development of understanding, 33–36, 34t, 35t photographs, 6–7 time, 239, 241 Speech locutions, 280, 283 temporal language, 232–233 see also Discourse; Language Spelling, letter names and, 129–132 Stability coefficients aggressive behavior, 48–49, 49t peer relationships, 63–67, 64t, 66t withdrawn behavior, 56–57, 58t Stop consonants, 111 Strange Situation, 144–145 Stress ‘‘immunization’’, 347 Stroop Day/Night test, flexible cognition, 302 Structural equation modeling (SEM), literacy development, 431, 432, 439 Suicide, early aggressive behavior and, 51 Symbolic knowledge, flexible cognition, 318 Symbolic mapping, flexible cognition, 295–296 Synaptogenesis, working memory development, 210–211

T Task-switching methods, flexible cognition, 275

486

Subject Index

Temperament definition, 419 literacy development, 419–420, 435 parasitic infection and, 376 resilience to MPDs, 346 school adjustment, 419–420 Temporal cortex, language role, 282 Temporal language, development, 232–234, 240 Temporal orientation cognitive development and, 262–264 concepts of past/future, 259–261 definition, 229–230 development, 229–270 components required for, 264 language acquisition, 232–234 memory role, 251–252, 261, 262 past–future differentiation, 251–253 sense of the future, 240–251 sense of the past, 235–240 emotions/feelings and, 260–261 past–future confusion, 254–259 see also Past–present–future distinction Tense, children’s comprehension and production, 232–233 Theory of mind, 143, 157 Time adult’s memory for, 235–236 children’s memory for, 237–238 egocentrism, 262 orientation in see Temporal orientation representations, 244, 245, 248 age-related changes, 252–253, 262, 263 flexibility, 246–247, 250, 251 linguistic, 231 spatial, 239, 241 systems, 231, 235 TORCH disorders, 374 Toxin exposure, developmental impact, 370, 376–380

U Undernutrition, chronic, 385 Unresponsive children, social stimulus features (effect on others), 334, 335–336, 339

V Vantage-point in photographic representation, 6–7 children vs. adults, 23–24, 23f understanding, 13–16, 14f viewing angle/azimuth, 15–16 Variable-based methods, 437 literacy analysis, 437–439 Verbal Description task, 20–27 affective tone, 24–27, 25f children vs. adults, 23f methodology, 21–22 research goal, 21 vantage-point studies, 22–24, 23f, 24f Victimization in peer relationships, 62 psychopathology and, 68–69 school adjustment and, 70–71 stability, 63–65, 64t, 66t Viral infections, developmental impact, 373–375 Visual system, development, 200 Vitamin deficiency, developmental impact, 382t, 383t, 384–385, 393 Vowels, discriminability, 111 W Way-finding tasks, 187 Windows and curtains procedure, 196 Withdrawn behavior affective difficulties, 59 aggressive-withdrawn, 57, 59–60, 61 features, 55 maladjustment and, 54–61 externalization, 59–60 internalization, 57–59 predictive value, 61 school adjustment, 60–61 stability, 56–57, 58t subtypes, 55 Word meaning Adults’ inference, 280–281 children’s inference, 303–315, 315t age-related changes, 303–310, 308f context-dependent, 311 flexible induction, 314–315 individual differences, 310–311

Subject Index inhibition and, 311–313 memory role, 313–314 verbal knowledge, 313–314 predicate role, 304–305, 308f order, 307–308 specificity, 308–310 prosody role, 304 shape bias hypothesis, 308–310 Working memory, 175–182 age-related changes, 196–197, 197f, 203, 205–206, 205f, 210–213 assessment, 174 challenges, 202–208 in infants, 184–191, 202–208 performance-based, 207 previous response memory, 186–187 see also specific tasks/paradigms cognitive flexibility, 300–303 cognitive neuroscience, 179–182, 208–210 competence vs. performance, 206–207 content, 207–208 controlled attention model (Engle et al.), 179 definitions, 174, 182, 202–206 development, 210–213 durability, 194–196, 203

487

embedded process model (Cowan), 178–179 experience effects, 213 historical aspects, 175–176 in infants, 173–231 language acquisition role, 212–213 multicomponent models, 177–178 neurobiology, 180–182, 183, 208, 209–210, 210–211 perseveration, 201–202 phenomenology, 216 primary–secondary distinction, 175–176 problem solving role, 213–214 processing limits, 178 recognition–recall distinction, 183 research criteria, 182–184 research results, 191–196, 192–193t response modality effect, 199–201 short-term memory vs., 177 stimulus type effects, 197–199 tripartate model (Baddeley and Hitch), 176–177, 207 central executive, 176, 177 phonological loop, 176 visuo-spatial sketchpad, 176 verbal children vs., 202 word meaning, 313–314


Contents of Previous Volumes Volume 1 Responses of Infants and Children to Complex and Novel Stimulation Gordon N. Cantor Word Associations and Children’s Verbal Behavior David S. Palermo Change in the Stature and Body Weight of North American Boys during the Last 80 Years Howard V. Meredith Discrimination Learning Set in Children Hayne W. Reese Learning in the First Year of Life Lewis P. Lipsitt Some Methodological Contributions from a Functional Analysis of Child Development Sidney W. Bijou and Donald M. Baer The Hypothesis of Stimulus Interaction and an Explanation of Stimulus Compounding Charles C. Spiker The Development of ‘‘Overconstancy’’ in Space Perception Joachim F. Wohlwill Miniature Experiments in the Discrimination Learning of Retardates Betty J. House and David Zeaman AUTHOR INDEX—SUBJECT INDEX

Selected Anatomic Variables Analyzed for Interage Relationships of the Size-Size, Size-Gain, and Gain-Gain Varieties Howard V. Meredith AUTHOR INDEX—SUBJECT INDEX

Volume 3 Infant Sucking Behavior and Its Modification Herbert Kaye The Study of Brain Electrical Activity in Infants Robert J. Ellingson Selective Auditory Attention in Children Eleanor E. Maccoby Stimulus Definition and Choice Michael D. Zeiler Experimental Analysis of Inferential Behavior in Children Tracy S. Kendler and Howard H. Kendler Perceptual Integration in Children Herbert L. Pick, Jr., Anne D. Pick, and Robert E. Klein Component Process Latencies in Reaction Times of Children and Adults Raymond H. Hohle AUTHOR INDEX—SUBJECT INDEX

Volume 2 The Paired-Associates Method in the Study of Conflict Alfred Castaneda Transfer of Stimulus Pretraining to Motor PairedAssociate and Discrimination Learning Tasks Joan H. Cantor The Role of the Distance Receptors in the Development of Social Responsiveness Richard H. Walters and Ross D. Parke Social Reinforcement of Children’s Behavior Harold W. Stevenson Delayed Reinforcement Effects Glenn Terrell A Developmental Approach to Learning and Cognition Eugene S. Gollin Evidence for a Hierarchical Arrangement of Learning Processes Sheldon H. White

Volume 4 Developmental Studies of Figurative Perception David Elkind The Relations of Short-Term Memory to Development and Intelligence John M. Belmont and Earl C. Butterfield Learning, Developmental Research, and Individual Differences Frances Degen Horowitz Psychophysiological Studies in Newborn Infants S. J. Hutt, H. G. Lenard, and H. F. R. Prechtl Development of the Sensory Analyzers during Infancy Yvonne Brackbill and Hiram E. Fitzgerald The Problem of Imitation Justin Aronfreed AUTHOR INDEX—SUBJECT INDEX

489

490

Contents of Previous Volumes

Volume 5 The Development of Human Fetal Activity and Its Relation to Postnatal Behavior Tryphena Humphrey Arousal Systems and Infant Heart Rate Responses Frances K. Graham and Jan C. Jackson Specific and Diversive Exploration Corinne Hutt Developmental Studies of Mediated Memory John H. Flavell Development and Choice Behavior in Probabilistic and Problem-Solving Tasks L. R. Goulet and Kathryn S. Goodwin AUTHOR INDEX—SUBJECT INDEX

Volume 6 Incentives and Learning in Children Sam L. Witryol Habituation in the Human Infant Wendell E. Jeffrey and Leslie B. Cohen Application of Hull-Spence Theory to the Discrimination Learning of Children Charles C. Spiker Growth in Body Size: A Compendium of Findings on Contemporary Children Living in Different Parts of the World Howard V. Meredith Imitation and Language Development James A. Sherman Conditional Responding as a Paradigm for Observational, Imitative Learning and VicariousReinforcement Jacob L. Gewirtz AUTHOR INDEX—SUBJECT INDEX

Volume 7 Superstitious Behavior in Children: An Experimental Analysis Michael D. Zeiler Learning Strategies in Children from Different Socioeconomic Levels Jean L. Bresnahan and Martin M. Shapiro Time and Change in the Development of the Individual and Society Klaus F. Riegel

The Nature and Development of Early Number Concepts Rochel Gelman Learning and Adaptation in Infancy: A Comparison of Models Arnold J. Sameroff AUTHOR INDEX—SUBJECT INDEX

Volume 8 Elaboration and Learning in Childhood and Adolescence William D. Rohwer, Jr. Exploratory Behavior and Human Development Jum C. Nunnally and L. Charles Lemond Operant Conditioning of Infant Behavior: A Review Robert C. Hulsebus Birth Order and Parental Experience in Monkeys and Man G. Mitchell and L. Schroers Fear of the Stranger. A Critical Examination Harriet L. Rheingold and Carol O. Eckerman Applications of Hull–Spence Theory to the Transfer of Discrimination Learning in Children Charles C. Spiker and Joan H. Cantor AUTHOR INDEX—SUBJECT INDEX

Volume 9 Children’s Discrimination Learning Based on Identity or Difference Betty J. House, Ann L. Brown, and Marcia S. Scott Two Aspects of Experience in Ontogeny: Development and Learning Hans G. Furth The Effects of Contextual Changes and Degree of Component Mastery on Transfer of Training Joseph C. Campione and Ann L. Brown Psychophysiological Functioning, Arousal, Attention, and Learning during the First Year of Life Richard Hirschman and Edward S. Katkin Self-Reinforcement Process in Children John C. Masters and Janice R. Mokros AUTHOR INDEX—SUBJECT INDEX

Contents of Previous Volumes Volume 10 Current Trends in Developmental Psychology Boyd R. McCandless and Mary Fulcher Geis The Development of Spatial Representations of Large-Scale Environments Alexander W. Siegel and Sheldon H. White Cognitive Perspectives on the Development of Memory John W. Hagen, Robert H. Jongeward, Jr., and Robert V. Kail, Jr. The Development of Memory: Knowing, Knowing About Knowing, and Knowing How to Know Ann L. Brown Developmental Trends in Visual Scanning Mary Carol Day The Development of Selective Attention: From Perceptual Exploration to Logical Search John C. Wright and Alice G. Vlietstra AUTHOR INDEX—SUBJECT INDEX

Volume 11 The Hyperactive Child: Characteristics, Treatment, and Evaluation of Research Design Gladys B. Baxley and Judith M. LeBlanc Peripheral and Neurochemical Parallels of Psychopathology: A Psychophysiological Model Relating Autonomic Imbalance to Hyperactivity, Psychopathy, and Autism Stephen W. Porges Constructing Cognitive Operations Linguistically Harry Beilin Operant Acquisition of Social Behaviors in Infancy: Basic Problems and Constraints W. Stuart Millar Mother-Infant Interaction and Its Study Jacob L. Gewirtz and Elizabeth F. Boyd Symposium on Implications of Life-Span Developmental Psychology for Child Development: Introductory Remarks Paul B. Baltes Theory and Method in Life-Span Developmental Psychology: Implications for Child Development Aletha Huston-Stein and Paul B. Baltes The Development of Memory: Life-Span Perspectives Hayne W. Reese Cognitive Changes during the Adult Years: Implications for Developmental Theory and Research Nancy W. Denney and John C. Wright

491

Social Cognition and Life-Span Approaches to the Study of Child Development Michael J. Chandler Life-Span Development of the Theory of Oneself: Implications for Child Development Orville G. Brim, Jr. Implications of Life-Span Developmental Psychology for Childhood Education Leo Montada and Sigrun-Heide Filipp AUTHOR INDEX—SUBJECT INDEX

Volume 12 Research between 1960 and 1970 on the Standing Height of Young Children in Different Parts of the World Howard V. Meredith The Representation of Children’s Knowledge David Klahr and Robert S. Siegler Chromatic Vision in Infancy Marc H. Bornstein Developmental Memory Theories: Baldwin and Piaget Bruce M. Ross and Stephen M. Kerst Child Discipline and the Pursuit of Self: An Historical Interpretation Howard Gadlin Development of Time Concepts in Children William J. Friedman AUTHOR INDEX—SUBJECT INDEX

Volume 13 Coding of Spatial and Temporal Information in Episodic Memory Daniel B. Berch A Developmental Model of Human Learning Barry Gholson and Harry Beilin The Development of Discrimination Learning: A Levels-of-Functioning Explanation Tracy S. Kendler The Kendler Levels-of-Functioning Theory: Comments and an Alternative Schema Charles C. Spiker and Joan H. Cantor Commentary on Kendler’s Paper: An Alternative Perspective Barry Gholson and Therese Schuepfer Reply to Commentaries Tracy S. Kendler

492


On the Development of Speech Perception: Mechanisms and Analogies Peter D. Eimas and Vivien C. Tartter The Economics of Infancy: A Review of Conjugate Reinforcement Carolyn Kent Rovee-Collier and Marcy J. Gekoski Human Facial Expressions in Response to Taste and Smell Stimulation Jacob E. Steiner AUTHOR INDEX—SUBJECT INDEX

Volume 14 Development of Visual Memory in Infants John S. Werner and Marion Perlmutter Sibship-Constellation Effects on Psychosocial Development, Creativity, and Health Mazie Earle Wagner, Herman J. P. Schubert, and Daniel S. P. Schubert The Development of Understanding of the Spatial Terms Front and Back Lauren Julius Harris and Ellen A. Strommen The Organization and Control of Infant Sucking C. K. Crook Neurological Plasticity, Recovery from Brain Insult, and Child Development Ian St. James-Roberts AUTHOR INDEX—SUBJECT INDEX

Volume 15 Visual Development in Ontogenesis: Some Reevaluations Ju¨ri Allik and Jaan Valsiner Binocular Vision in Infants: A Review and a Theoretical Framework Richard N. Aslin and Susan T. Dumais Validating Theories of Intelligence Earl C. Butterfield, Dennis Siladi, and John M. Belmont Cognitive Differentiation and Developmental Learning William Fowler Children’s Clinical Syndromes and Generalized Expectations of Control Fred Rothbaum AUTHOR INDEX—SUBJECT INDEX

Volume 16 The History of the Boyd R. McCandless Young Scientist Awards: The First Recipients David Palermo Social Bases of Language Development: A Reassessment Elizabeth Bates, Inge Bretherton, Marjorie Beeghly-Smith, and Sandra McNew Perceptual Anisotrophies in Infancy: Ontogenetic Origins and Implications of Inequalities in Spatial Vision Marc H. Bornstein Concept Development Martha J. Farah and Stephen M. Kosslyn Production and Perception of Facial Expressions in Infancy and Early Childhood Tiffany M. Field and Tedra A. Walden Individual Differences in Infant Sociability: Their Origins and Implications for Cognitive Development Michael E. Lamb The Development of Numerical Understandings Robert S. Siegler and Mitchell Robinson AUTHOR INDEX—SUBJECT INDEX

Volume 17 The Development of Problem-Solving Strategies Deanna Kuhn and Erin Phelps Information Processing and Cognitive Development Robert Kail and Jeffrey Bisanz Research between 1950 and 1980 on Urban–Rural Differences in Body Size and Growth Rate of Children and Youths Howard V. Meredith Word Meaning Acquisition in Young Children: A Review of Theory and Research Pamela Blewitt Language Play and Language Acquisition Stan A. Kuczaj II The Child Study Movement: Early Growth and Development of the Symbolized Child Alexander W. Siegel and Sheldon H. White AUTHOR INDEX—SUBJECT INDEX

Volume 18 The Development of Verbal Communicative Skills in Children Constance R. Schmidt and Scott G. Paris

Contents of Previous Volumes Auditory Feedback and Speech Development Gerald M. Siegel, Herbert L. Pick, Jr., and Sharon R. Garber Body Size of Infants and Children around the World in Relation to Socioeconomic Status Howard V. Meredith Human Sexual Dimorphism: Its Cost and Benefit James L. Mosley and Eileen A. Stan Symposium on Research Programs: Rational Alternatives to Kuhn’s Analysis of Scientific Progress—Introductory Remarks Hayne W. Reese, Chairman World Views and Their Influence on Psychological Theory and Research: Kuhn-Lakatos-Laudan Willis F. Overton The History of the Psychology of Learning as a Rational Process: Lakatos versus Kuhn Peter Barker and Barry Gholson Functionalist and Structuralist Research Programs in Developmental Psychology: Incommensurability or Synthesis? Harry Beilin In Defense of Kuhn: A Discussion of His Detractors David S. Palermo Comments on Beilin’s Epistemology and Palermo’s Defense of Kuhn Willis F. Overton From Kuhn to Lakatos to Laudan Peter Barker and Barry Gholson Overton’s and Palermo’s Relativism: One Step Forward, Two Steps Back Harry Beilin AUTHOR INDEX—SUBJECT INDEX

493

Effects of the Knowledge Base on Children’s Memory Strategies Peter A. Ornstein and Mary J. Naus Effects of Sibling Spacing on Intelligence, Interfamilial Relations, Psychosocial Characteristics, and Mental and Physical Health Mazie Earle Wagner, Herman J. P. Schubert, and Daniel S. P. Schubert Infant Visual Preferences: A Review and New Theoretical Treatment Martin S. Banks and Arthur P. Ginsburg AUTHOR INDEX—SUBJECT INDEX

Volume 20 Variation in Body Stockiness among and within Ethnic Groups at Ages from Birth to Adulthood Howard V. Meredith The Development of Conditional Reasoning: An Iffy Proposition David P. O’Brien Content Knowledge: Its Role, Representation, and Restructuring in Memory Development Michelene T. H. Chi and Stephen J. Ceci Descriptions: A Model of Nonstrategic Memory Development Brian P. Ackerman Reactivation of Infant Memory: Implications for Cognitive Development Carolyn Rovee-Collier and Harlene Hayne Gender Segregation in Childhood Eleanor E. Maccoby and Carol Nagy Jacklin Piaget, Attentional Capacity, and the Functional Implications of Formal Structure Michael Chapman INDEX

Volume 19 Volume 21 Response to Novelty: Continuity versus Discontinuity in the Developmental Course of Intelligence Cynthia A. Berg and Robert J. Sternberg Metaphoric Competence in Cognitive and Language Development Marc Marschark and Lynn Nall The Concept of Dimensions in Developmental Research Stuart I. Offenbach and Francine C. Blumberg

Social Development in Infancy: A 25-Year Perspective Ross D. Parke On the Uses of the Concept of Normality in Developmental Biology and Psychology Eugene S. Gollin, Gary Stahl, and Elyse Morgan Cognitive Psychology: Mentalistic or Behavioristic? Charles C. Spiker

494


Some Current Issues in Children’s Selective Attention Betty J. House Children’s Learning Revisited: The Contemporary Scope of the Modified Spence Discrimination Theory Joan H. Cantor and Charles C. Spiker Discrimination Learning Set in Children Hayne W. Reese A Developmental Analysis of Rule-Following Henry C. Riegler and Donald M. Baer Psychological Linguistics: Implications for a Theory of Initial Development and a Method for Research Sidney W. Bijou Psychic Conflict and Moral Development Gordon N. Cantor and David A. Parton Knowledge and the Child’s Developing Theory of the World David S. Palermo Childhood Events Recalled by Children and Adults David B. Pillemer and Sheldon H. White INDEX

Volume 22 The Development of Representation in Young Children Judy S. DeLoache Children’s Understanding of Mental Phenomena David Estes, Henry M. Wellman, and Jacqueline D. Woolley Social Influences on Children’s Cognition: State of the Art and Future Directions Margarita Azmitia and Marion Perlmutter Understanding Maps as Symbols: The Development of Map Concepts Lynn S. Liben and Roger M. Downs The Development of Spatial Perspective Taking Nora Newcombe Developmental Studies of Alertness and Encoding Effects of Stimulus Repetition Daniel W. Smothergill and Alan G. Kraut Imitation in Infancy: A Critical Review Claire L. Poulson, Leila Regina de Paula Nunes, and Steven F. Warren AUTHOR INDEX—SUBJECT INDEX

Volume 23 The Structure of Developmental Theory Willis F. Overton

Questions a Satisfying Developmental Theory Would Answer: The Scope of a Complete Explanation of Development Phenomena Frank B. Murray The Development of World Views: Toward Future Synthesis? Ellin Kofsky Scholnick Metaphor, Recursive Systems, and Paradox in Science and Developmental Theory Willis F. Overton Children’s Iconic Realism: Object versus Property Realism Harry Beilin and Elise G. Pearlman The Role of Cognition in Understanding Gender Effects Carol Lynn Martin Development of Processing Speed in Childhood and Adolescence Robert Kail Contextualism and Developmental Psychology Hayne W. Reese Horizontality of Water Level: A Neo-Piagetian Developmental Review Juan Pascual-Leone and Sergio Morra AUTHOR INDEX—SUBJECT INDEX

Volume 24 Music and Speech Processing in the First Year of Life Sandra E. Trehub, Laurel J. Trainor, and Anna M. Unyk Effects of Feeding Method on Infant Temperament John Worobey The Development of Reading Linda S. Siegel Learning to Read: A Theoretical Synthesis John P. Rack, Charles Hulme, and Margaret J. Snowling Does Reading Make You Smarter? Literacy and the Development of Verbal Intelligence Keith E. Stanovich Sex-of-Sibling Effects: Part I. Gender Role, Intelligence, Achievement, and Creativity Mazie Earle Wagner, Herman J. P. Schubert, and Daniel S. P. Schubert The Concept of Same Linda B. Smith Planning as Developmental Process Jacquelyn Baker-Sennett, Eugene Matusov, and Barbara Rogoff AUTHOR INDEX—SUBJECT INDEX


495

Volume 25

Volume 27

In Memoriam: Charles C. Spiker (1925–1993) Lewis P. Lipsitt Developmental Differences in Associative Memory: Strategy Use, Mental Effort, and Knowledge Access Interactions Daniel W. Kee A Unifying Framework for the Development of Children’s Activity Memory Hilary Horn Ratner and Mary Ann Foley Strategy Utilization Deficiencies in Children: When, Where, and Why Patricia H. Miller and Wendy L. Seier The Development of Children’s Ability to Use Spatial Representations Mark Blades and Christopher Spencer Fostering Metacognitive Development Linda Baker The HOME Inventory: Review and Reflections Robert H. Bradley Social Reasoning and the Varieties of Social Experiences in Cultural Contexts Elliot Turiel and Cecilia Wainryb Mechanisms in the Explanation of Developmental Change Harry Beilin

From Form to Meaning: A Role for Structural Alignment in the Acquisition of Language Cynthia Fisher The Role of Essentialism in Children’s Concepts Susan A. Gelman Infants’ Use of Prior Experiences with Objects in Object Segregation: Implications for Object Recognition in Infancy Amy Needham and Avani Modi Perseveration and Problem Solving in Infancy Andreá Aguiar and Reneé Baillargeon Temperament and Attachment: One Construct or Two? Sarah C. Mangelsdorf and Cynthia A. Frosch The Foundation of Piaget’s Theories: Mental and Physical Action Harry Beilin and Gary Fireman

AUTHOR INDEX—SUBJECT INDEX

Volume 26 Preparing to Read: The Foundations of Literacy Ellen Bialystok The Role of Schemata in Children’s Memory Denise Davidson The Interaction of Knowledge, Aptitude, and Strategies in Children’s Memory Performance David F. Bjorklund and Wolfgang Schneider Analogical Reasoning and Cognitive Development Usha Goswami Sex-of-Sibling Effects: A Review Part II. Personality and Mental and Physical Health Mazie Earle Wagner, Herman J. P. Schubert, and Daniel S. P. Schubert Input and Learning Processes in First Language Acquisition Ernst L. Moerk AUTHOR INDEX—SUBJECT INDEX


Volume 28 Variability in Children’s Reasoning Karl S. Rosengren and Gregory S. Braswell Fuzzy-Trace Theory: Dual Processes in Memory, Reasoning, and Cognitive Neuroscience C. J. Brainerd and V. F. Reyna Relational Frame Theory: A Post-Skinnerian Account of Human Language and Cognition Yvonne Barnes-Holmes, Steven C. Hayes, Dermot Barnes-Holmes, and Bryan Roche The Continuity of Depression across the Adolescent Transition Shelli Avenevoli and Laurence Steinberg The Time of Our Lives: Self-Continuity in Native and Non-Native Youth Michael J. Chandler AUTHOR INDEX—SUBJECT INDEX

Volume 29 The Search for What is Fundamental in the Development of Working Memory Nelson Cowan, J. Scott Saults, and Emily M. Elliott

496


Culture, Autonomy, and Personal Jurisdiction in Adolescent–Parent Relationships Judith G. Smetana Maternal Responsiveness and Early Language Acquisition Catherine S. Tamis-Lemonda and Marc H. Bornstein Schooling as Cultural Process: Working Together and Guidance by Children from Schools Differing in Collaborative Practices Eugene Matusov, Nancy Bell, and Barbara Rogoff Beyond Prototypes: Asymmetries in Infant Categorization and What They Teach Us about the Mechanisms Guiding Early Knowledge Acquisition Paul C. Quinn Peer Relations in the Transition to Adolescence Carollee Howes and Julie Wargo Aikins

Volume 30



Learning to Keep Balance Karen Adolph Sexual Selection and Human Life History David C. Geary Developments in Early Recall Memory: Normative Trends and Individual Differences Patricia J. Bauer, Melissa M. Burch, and Erica E. Kleinknecht Intersensory Redundancy Guides Early Perceptual and Cognitive Development Lorraine E. Bahrick and Robert Lickliter Children’s Emotion-Related Regulation Nancy Eisenberg and Amanda Sheffield Morris Maternal Sensitivity and Attachment in Atypical Groups L. Beckwith, A. Rozga, and M. Sigman Influences of Friends and Friendships: Myths, Truths, and Research Recommendations Thomas J. Berndt and Lonna M. Murphy

Advances in Child Development and Behavior Volume 31

Advances in Child Development and Behavior Volume 30

Advances in Child Development and Behavior Volume 21

Advances in Child Development and Behavior Volume 12

Advances in Child Development and Behavior Volume 30

Advances in Child Development and Behavior Volume 15

Advances in Child Development and Behavior Volume 26

Advances in Child Development and Behavior Volume 28

Advances in Child Development and Behavior Volume 37

Advances in Child Development and Behavior, Volume 36

Advances in Child Development and Behavior Volume 1

Advances in Child Development and Behavior Volume 16

Advances in Child Development and Behavior Volume 22

Advances in Child Development and Behavior Volume 8

Advances in Child Development and Behavior Volume 23

Advances in Child Development and Behavior Volume 35

Advances in Child Development and Behavior Volume 24

Advances in Child Development and Behavior Volume 17

Advances in Child Development and Behavior Volume 10

Advances in Child Development and Behavior Volume 33

Advances in Child Development and Behavior Volume 5

Advances in Child Development and Behavior, Volume 4

Advances in Child Development and Behavior Volume 34

Advances in Child Development and Behavior Volume 11