Compendium of Auditory and Speech Tasks: Children's Speech and Literacy Difficulties, Book 4

Compendium of Auditory and Speech Tasks Children’s Speech and Literacy Difficulties Book 4 Joy Stackhouse, Maggie Vance...

Author: Joy Stackhouse | Maggie Vance | Michelle Pascoe | Bill Wells

22 downloads 466 Views 2MB 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

Compendium of Auditory and Speech Tasks Children’s Speech and Literacy Difficulties Book 4

Joy Stackhouse, Maggie Vance, Michelle Pascoe, and Bill Wells

Compendium of Auditory and Speech Tasks

Compendium of Auditory and Speech Tasks Children’s Speech and Literacy Difficulties Book 4

Joy Stackhouse, Maggie Vance, Michelle Pascoe, and Bill Wells

Copyright © 2007

John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England Telephone (44) 1243 779777

Email (for orders and customer service enquiries): [email protected] Visit our Home Page on www.wiley.com All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under the terms of the Copyright, Designs and Patents Act 1988 or under the terms of a licence issued by the Copyright Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & Sons Ltd, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to [email protected], or faxed to (44) 1243 770620. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The Publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the Publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought.

Other Wiley Editorial Offices John Wiley & Sons Inc., 111 River Street, Hoboken, NJ 07030, USA Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA Wiley-VCH Verlag GmbH, Boschstr. 12, D-69469 Weinheim, Germany John Wiley & Sons Australia Ltd, 42 McDougall Street, Milton, Queensland 4064, Australia John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809 John Wiley & Sons Canada Ltd, 6045 Freemont Blvd, Mississauga, ONT, L5R 4J3, Canada Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Anniversary Logo Design: Richard J. Pacifico Library of Congress Cataloging-in-Publication Data Compendium of auditory and speech tasks : children's speech and literacy difficulties / Joy Stackhouse ... [et al]. p. ; cm. Includes bibliographical references. ISBN 978-0-470-51659-1 (pbk. : alk. paper) 1. Speech disorders in children. 2. Language disorders in children. 3. Psycholinguistics. I. Stackhouse, Joy. [DNLM: 1. Language Disorders. 2. Child. 3. Psycholinguistics–methods. 4. Reading. 5. Speech Perception. WL 340.2 C737 2007] RJ496.S7C583 2007 618.92'855--dc22 2007014647 A catalogue record for this book is available from the British Library ISBN: 978-0-470-51659-1 Typeset by Thomson Press (India) Ltd. Printed and bound in Great Britain by TJ International Ltd, Padstow. This book is printed on acid-free paper responsibly manufactured from sustainable forestry in which at least two trees are planted for each one used for paper production.

For our parents, with many thanks for their support and interest Jean and Ron Stackhouse Eileen and Ron Holmes Jenny and Mike Pascoe Joan and Chris Wells

Contents

Preface Compendium CD-Rom Acknowledgements Conventions Phonetic Symbols and Diacritics Chapter 1: Chapter 2: Chapter Chapter Chapter Chapter Chapter Chapter Chapter

3: 4: 5: 6: 7: 8: 9:

Chapter 10:

xi xv xvii xix xxi

A Psycholinguistic Approach Development of the Assessment Tasks and Normative Data Auditory Discrimination Tasks Auditory Lexical Discrimination Tasks Speech Production: Single Words Speech Production: Connected Speech Speech Accuracy, Rate and Consistency Predicting Persisting Speech Difficulties Assessing Risk Factors through Questionnaires Using Auditory and Speech Tasks in Practice

References Appendix A.1: Appendix A.2: Appendix A.3: Appendix B: Appendix C.1:

Tasks for Each Question in the Psycholinguistic Assessment Framework Speech Processing Profile Speech Processing Model Standardised Assessments used in the Longitudinal Study Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words and Non-words – Reduced Version (from Bridgeman & Snowling, 1988) vii

1 27 47 73 95 131 153 175 191 209 231

243 247 249 251

253

viii Appendix C.2:

Appendix C.3: Appendix C.4: Appendix C.5: Appendix C.6: Appendix D.1:

Appendix D.2:

Appendix D.3:

Appendix D.4:

Appendix D.5:

Appendix D.6: Appendix E.1: Appendix E.2: Appendix E.3:

Appendix E.4:

Appendix E.5: Appendix E.6:

Contents Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words and Non-words – Shortest Form (from Bridgeman & Snowling, 1988) Auditory Discrimination Task 2: Complex Non-words (from Stackhouse, 1989) Auditory Discrimination Task 3: ABX Task – Full Version (Vance, 1996) Auditory Discrimination Task 3: ABX Task – Short Form (Nathan et al., 2004) Auditory Discrimination Task 4: Legal versus Illegal Non-words Auditory Lexical Discrimination Task 1: Mispronunciation Detection – Full Version (from Vance, 1995) Auditory Lexical Discrimination Task 1: Mispronunciation Detection – Short Form (from Nathan et al., 2004) Auditory Lexical Discrimination Task 2: Without Pictures (from Constable, Stackhouse & Wells, 1997) Auditory Lexical Discrimination Task 2: With Pictures (from Constable, Stackhouse & Wells, 1997) Auditory Lexical Discrimination Task 3: Mispronunciation Detection for Individual Children (after Locke, 1980) Auditory Lexical Discrimination Task 4: Words in Sentences (from Cassidy, 1994) Picture Naming Task 1 – Full Version (from Vance, Stackhouse & Wells, 2005) Picture Naming Task 1 – Short Form Guidelines for Scoring Full Versions and Short Forms of Speech Production Tasks Speech Production: Single Words – Picture Naming Task 2 (from Snowling, van Wagtendonk & Stafford, 1988) Word Repetition Task – Full Version (from Vance, Stackhouse & Wells, 2005) Word Repetition Task – Short Form (from Nathan et al., 2004a)

257 259 263 267 269

271

279

283

287

291 295 309 313

315

317 321 325

Contents Appendix E.7: Appendix E.8: Appendix E.9: Appendix E.10:

Appendix E.11: Appendix F.1:

Appendix F.2:

Appendix F.3:

Appendix G.1:

Appendix G.2:

Appendix G.3:

Appendix G.4:

Appendix G.5: Appendix Appendix Appendix Appendix

H.1: H.2: H.3: H.4:

Word Repetition Task: Low-Frequency Words (from Nathan et al., 2004a) Non-Word Repetition Task – Full Version (from Vance, Stackhouse & Wells, 2005) Non-Word Repetition Task – Short Form (from Nathan et al., 2004) Non-word Repetition Task: Matched to Low-Frequency Words (from Nathan et al., 2004a) Mispronunciation Self-Correction Connected Speech Task 1: Words in Sentences (from Vance, Stackhouse & Wells, 1995) Connected Speech Task 2: Connected Speech Processes (CSP) Repetition (from Newton, 1999) Connected Speech Task 3: Final Consonant Juncture Repetition (from Pascoe, Stackhouse & Wells, 2006) Speech Accuracy, Rate and Consistency Task 1: Oral Movements and Silent DDK Rates (from Williams & Stackhouse, 2000) Speech Accuracy, Rate and Consistency Task 2a: Repetition and Spoken DDK Rates: Words (from Williams & Stackhouse, 2000) Speech Accuracy, Rate and Consistency Task 2b: Repetition and Spoken DDK Rates: Non-Words (from Williams & Stackhouse, 2000) Speech Accuracy, Rate and Consistency Task 2c: Repetition and Spoken DDK Rates: Syllable Sequences (from Williams & Stackhouse, 2000) Speech Accuracy and Consistency Task 3 (from Coffield, 1994) Questionnaire 1: Developmental History Questionnaire 2: Family Information Questionnaire 3: Educational History Questionnaire 4: Speech and Language Therapy

Index Other Titles in this Series

ix

327 331 335

337 341

345

349

355

361

363

381

397 413 439 443 445 447 451 462

Preface

Book 1 of this series presented a psycholinguistic framework for practice and research (Stackhouse and Wells, 1997). It was based on our training courses in which principles of a psycholinguistic approach to investigating and promoting children’s spoken and written language skills were illustrated through a series of workshop activities. Since then, there has been a growing demand for procedures specifically designed for the framework, produced in an accessible format and located in one place. Book 4 therefore provides practitioners and researchers with a compendium of 30 auditory and speech assessment tasks (Chapters 3–7) which can be used to investigate underlying causes of speech and associated literacy difficulties, as well as 4 questionnaire formats to collect background information (Chapter 9). The tasks have not been designed for a specific client group. They have been used with children and adolescents with a range of difficulties associated with cleft lip and palate, dysarthria, dyspraxia, phonological impairment, Down syndrome, dyslexia, stammering, autism, semantic-pragmatic difficulties, general learning difficulties, and disadvantaged backgrounds. The tasks have also been used with normally developing children, particularly in the age range of 3–7 years (Chapter 2). As a result, Book 4 includes descriptions of typical performance so that the atypical can be identified more easily. In addition, as many of the procedures were used in a longitudinal study of children’s speech and literacy development between the age of 4 and 7 years, we can highlight those tasks that can help to predict persisting speech difficulties and associated literacy problems (Chapter 8).

xi

xii

Preface

However, Book 4 is not just a collection of tasks. Continuing the theme of training, which is a feature of this series, Book 4 aims to: • Serve as a practical handbook for investigating the speech and auditory skills of children with typical and atypical development. • Provide speech and auditory tasks that can be used with a range of children and adolescents with speech and literacy difficulties. • Incorporate both input and output tasks in investigations of children’s speech difficulties. • Explain how stimuli and procedures have been designed so that the reader can develop their own tasks for an individual child or particular client group as required. • Include matched stimuli across tasks to allow quantitative and qualitative comparison of performance across tasks. • Demonstrate a range of scoring techniques that can be used when assessing children. • Provide training in how to interpret findings with reference to the psycholinguistic framework presented in Book 1. • Illustrate typical performance on the tasks, as demonstrated by normally developing children in the age range of 3–7 years. • Indicate which tasks can help to predict which children are at risk for persisting speech and/or literacy problems. • Summarise research findings from a range of projects using these procedures. • Encourage users to develop their own tasks to investigate and support auditory and speech skills in their clients. • Facilitate appropriate planning and evaluation of intervention. Throughout Book 4 reference is made to Books 1, 2 and 3. It is assumed that users of the procedures included in Book 4 are already familiar with the psycholinguistic approach presented in Book 1 (summarised in Chapter 1 in this volume), and will refer to the principles and case studies presented in Books 2 and 3 when planning or evaluating intervention based on the data collected from the tasks presented here (see Chapter 10, this volume). It is also assumed that users of the speech tasks in particular will be trained in phonetic transcription and phonological analysis, skills which constitute “an indispensable foundation for the assessment of speech and language difficulties” (Book 1, p. 6). It is important to appreciate that the procedures included in this volume are not standardised tests, but age-typical performance data are provided as guidelines. For scores to be compared with typical children’s performance the tasks must be administered as specified. However, they can be used more flexibly when hypothesis testing with an individual child. It is hoped that this book will reach a wide-ranging audience, from student speech and language therapists/pathologists in training to

Preface

xiii

experienced practitioners and researchers in speech and language therapy/pathology and other disciplines. It aims to provide a ‘one-stop-shop’ for accessing speech and auditory procedures and their associated picture material for those who may not necessarily have the time to devise their own assessments. However, a key tenet of the psycholinguistic approach is that it is ‘carried in the head of the user and not in a case of tests’(Book 1, p. 49). In a recent conversation with one of the authors, a teacher stated that this psycholinguistic approach has provided her and her colleagues from different disciplines with a ‘conceptual framework’ for discussing children’s needs by developing a common understanding of a shared vocabulary. The procedures in this book are meant to be viewed in this way, rather than as an ‘off-the-shelf ’ assessment battery. Joy Stackhouse, Maggie Vance, Michelle Pascoe, and Bill Wells. May 2007

Compendium CD-Rom

Pictures and score-sheets for all the tasks described in this volume are provided on the accompanying CD-Rom. The score sheets and pictures needed for each task are available for printing out within the folder Task Score Sheets and Pictures. Within this folder there are separate folders for each task that are labelled with the appropriate appendix number. For example: ‘Appendix D.1’ contains material for the ‘Auditory Lexical Discrimination Task 1: Mispronunciation Detection Full Version’. The Appendix folder contains a file of the score sheet and a file of the picture set. These will need to be opened and printed out. The pictures needed for the task are presented in the correct order for task administration. They will appear 4 to a page, and can be cut up and mounted on card or made into a picture book for presentation. The relevant appendix code can be found within the description of each task in the book, in the section headed ‘Materials’. In addition, all of the pictures referred to in this volume are available for printing out individually in the folder A-Z pictures. Within this folder there is an individual file for each picture, labelled with the picture name. This is to enable you to use the pictures to devise your own procedures for individual children or specific groups that you are working with.

xv

Acknowledgements

We would like to acknowledge the significant contribution made by Liz Ison (née Nathan) who was the Research Assistant on the longitudinal study reported in this volume. We also acknowledge the contributions of our colleagues and collaborators in the development of the tasks for the psycholinguistic framework over a number of years; for this volume in particular: Alison Constable, Rosemary Morgan Barry, Caroline Newton, Rachel Rees, Maggie Snowling and Pam Williams. A number of undergraduate and postgraduate students have also contributed to the development of the procedures used and to our thinking about the psycholinguistic approach; specific to this volume are Anna Boase, Elizabeth Bridgeman, Phillipa Broadbent, Belinda Cassidy, Christine Coffield, Eydoxia Darili, Susan Dry, Michelle Habgood and Kerry Mathews. The adoption of the framework by non-English speaking users has been made possible through the translations and work of Professor Annette Fox, Blanca Schaefer and Silke Fricke in Germany, and Inge Benn Thomson in Denmark. The original photography for the stimuli used on the cross-sectional study was done by Ron Holmes; the original pictures for the connected speech task by Tessa Papier of Birdart Publications; and the illustrations included with this volume are by David Thompson. Funding for the two key studies in this volume was from the Joint Research Advisory Committee of the National Hospital, Queen Square, London for the cross-sectional study; and the North Thames Regional Health Authority Research and Development Programme for the longitudinal study.

xvii

Conventions

TIE

/stra/ [ta] “tie” → /k/ /t/ CVC N CA

spoken real word target/stimulus, or non-word presented with phonetic spelling, rather than phonetic transcription spoken non-word target/stimulus spoken response, where phonetic information is required spoken response, where phonetic information is not required ‘is realized as’ e.g. /stra/ → [ta] /k/ contrasted with /t/ word structure where C consonant; V vowel number of children in a group, or items in a task or condition Children’s chronological age given using the convention of years; months, e.g. 5;10 5 years and 10 months

xix

Phonetic Symbols and Diacritics

The phonetic symbols and diacritics that occur in this series are listed here. The reference accent used is that of the standard variety of Southern British English, as described for example by Cruttenden (1994), to whose description the reader is referred for further details. Each consonant and vowel symbol has a phonetic value broadly equivalent to that found in the corresponding word pronounced in the reference accent, unless otherwise indicated. If no such key word is provided, the sound type represented by the symbol is not found in that accent. Prosodic notation Nuclear tones in the reference accent, as follows:

` fall; rise-fall; ´ rise; fall-rise Relative pitch height and on-syllable pitch movement are represented impressionistically between staves, above the transcription. Pauses between or within utterances are represented in tenths of a second, e.g. (0.5). (.) indicates a pause of a tenth of a second or less; alternatively, – indicates a silent beat. : [ {} υ –

indicates a sustention of the preceding sound. represents the point at which simultaneous speech begins. represents notable loudness. rhythmically short. rhythmically long. indicates that in a real English word, e.g. BANANA, the following syllable bears the main lexical stress; in a non-word target, e.g. /bəkɑtə/, it indicates where the main stress is intended to be. In

xxi

xxii

Phonetic Symbols and Diacritics transcriptions of spoken utterances, it represents perceived prominence, mainly of loudness and pitch, e.g. [bənɑnə].

n

indicates that a consonant forms the nucleus of the syllable: e.g. BUTTON → [btn].

$ syllable boundary.

Vowel symbols Diphthongs have the quality of the two constituent vowel symbols pronounced in sequence, the first longer than the second. ə  ε æ

a ɒ υ ɵ u aυ əυ i e a ɒ ɑ ɔ ə εə

mid-central unrounded centralised close-mid front spread open-mid front spread raised open front unrounded open front unrounded raised open central unrounded open back rounded centralized close-mid back rounded mid central rounded close back rounded diphthong diphthong close central rounded close front spread diphthong diphthong diphthong open back unrounded open-mid back rounded mid central unrounded diphthong diphthong

SOFA, ABOUT, BUTTERCUP BIT BET BAT BAT

(Northern English)

BUG POT PUT BOOT, SHOE BOUT, COW BOAT, LOW

BEAT, BEE BAIT, BAY BITE, BUY NOISE, BOY PARK, BAR SORT, SAW BIRD, FUR BEARD, BEER BEAR

Consonant symbols: Plosives, affricates, nasals p b t d c

voiceless bilabial plosive voiced bilabial plosive voiceless alveolar plosive voiced alveolar plosive voiceless palatal plosive voiced palatal plosive

PIN BIN TIN DIN THANKYOU EGGYOLK

Phonetic Symbols and Diacritics k ʔ

voiceless velar plosive voiced velar plosive glottal stop

t d tʃ d

voiceless alveolo-palatal affricate voiced alveolo-palatal affricate voiceless postalveolar affricate voiced postalveolar affricate

xxiii

COAT GOAT

(Cockney pronunciation)

WATER

CHEESE, NATURE GEM, MAJOR

Consonant symbols: Fricatives φ β θ ð s z l ʃ ç x h

voiceless bilabial voiced bilabial voiceless dental voiced dental voiceless alveolar voiced alveolar voiceless alveolar lateral voiced alveolar lateral voiceless postalveolar voiced postalveolar voiceless alveolo-palatal voiced alveolo-palatal voiceless palatal voiceless velar voiceless glottal

THIN, MOTH THIS, MOTHER SEA, RICE ZOO, LOSE LLANDUDNO

(Welsh)

SHOE, NATION MEASURE

(German) BUCH (German) MICH

HAT

Approximants, secondary articulations etc. l w ɹ ɾ

voiced voiced voiced voiced voiced

alveolar lateral approximant labial-velar approximant labiodental approximant postalveolar approximant alveolar tap

LOCK WATER

RIGHT, CARRY BETTER

(American pronunciation) r

j j

cover symbol for different pronunciations of this phonological item in different varieties of English voiced palatal approximant palatal resonance/secondary articulation

YELLOW

xxiv

Phonetic Symbols and Diacritics

advanced tongue position/front resonance retracted tongue position/back resonance close articulation voiceless

h

aspirated unaspirated nasalised

→ [pəthetəυ] PET → [phεt] STAR →[st-ɑ] VIN (French) → [v ε] POTATO

Chapter One A Psycholinguistic Approach

Speech and language development needs to be viewed from various perspectives if the links between spoken and written language are to be understood and appropriate intervention planned for children with difficulties. These perspectives include educational, linguistic, medical, psycholinguistic and psychosocial. All are important and reference will be made to each of these throughout this volume, but the main focus will be on implementing a psycholinguistic approach to the investigation of children’s speech processing skills. A premise of the psycholinguistic approach is that children’s speech and literacy development is the product of an intact speech processing system, comprising input, output and storage skills. Although this development is influenced by each of the perspectives mentioned above, any difficulties with speech and literacy performance will arise from limitations within one or more of these speech processing domains. Thus, a key aspect of a psycholinguistic approach is to develop hypotheses about the nature of a child’s difficulties and then to test them out through specific assessment and intervention tasks. In their review of psycholinguistic models of speech development and the application to clinical practice, Baker et al. (2001) state that such an approach: can have important effects on clinical practice – not only in influencing assessment and intervention procedures, but in reshaping our thinking about the nature of speech impairment. (p. 686)

This volume includes psycholinguistic procedures for investigating children’s speech input, output and lexical storage skills. The aim of using these procedures is to identify a child’s speech processing strengths and weaknesses; use the data collected to plan appropriate

1

2


intervention if needed; establish if there are associated language and literacy difficulties; and predict if a child will have persisting speech and literacy difficulties. In order to use the procedures included in this volume, the reader needs to be familiar with the psycholinguistic framework presented in Book 1 of this series (Stackhouse & Wells, 1997). This chapter will present some background information about children’s speech and literacy difficulties and summarise the three main components of the psycholinguistic framework presented in Book 1: • speech processing profile • speech processing ‘box-and-arrow’ model • developmental phase model

Speech and Literacy Difficulties Speech difficulties in children are one of the most common communication problems and make up a high proportion of speech and language therapists’ caseloads (Enderby & Phillips, 1986). It is estimated that 5 % of primary school children have speech difficulties (Weiss, Gordon & Lillywhite, 1987) and that 3.8 % of children in the USA in the age range of 3–11 years old have what are described as ‘phonological problems’ (Shriberg, Tomblin & McSweeney, 1999). In the UK, Broomfield and Dodd (2004) estimate that 48,000 children per year are referred to speech and language therapy because of ‘primary speech disability’. Speech difficulties may be associated with language impairments. Prevalence rates of specific language impairments in young children are around 7 % (Tomblin, Smith & Zhang, 1997), but a systematic review of the literature by Law et al. (2000) indicates prevalence rates of up to 24.6 %. Although these rates are difficult to estimate, it is clear that there are a significant number of children who do not resolve their speech difficulties before they begin school and are not ready to take advantage of literacy instruction given. Predicting which children will not resolve their speech difficulties by 5 years of age is the subject of Chapter 8 in this volume. The lack of consensus in defining spoken language problems in the population is reflected in the diversity of labels used to describe children’s difficulties. When referring to a child’s ‘speech difficulties’ a number of labels may be applied, e.g. articulation difficulties, phonological disorder, phonological delay, expressive phonological impairment, speech disorder, speech sound difficulties, dyspraxia, stammering or non-fluent speech. Some labels indicate causation, e.g. ‘cleft palate speech’, ‘deaf speech’, dysarthria (see Book 1, pp. 3–8, for further discussion of the classification of speech problems in children). In this volume, the term speech difficulty is used to refer to children who have difficulties with producing speech segments in isolation, in single words or in connected

A Psycholinguistic Approach

3

speech, regardless of the origin of the difficulty. This term is preferred because children with such difficulties do not form a homogeneous group and a fundamental principle of the approach presented here is to investigate a child’s needs regardless of any label attached. Even if a possible cause of a speech difficulty is known or suspected, each child presents his or her speech difficulties in an individual way and with his or her own cluster of associated behaviours. These may include deficits in language skills (Shriberg, Kwiatkowski & Gruber, 1994; Bird, Bishop & Freeman, 1995), auditory discrimination (Bird & Bishop, 1992; Rvachew et al., 2003), oro-motor skills (Henry, 1990; Williams & Stackhouse, 1998, 2000), phonological awareness (Rvachew et al., 2003; Gillon, 2004) and literacy development (Bird, Bishop & Freeman, 1995; Larrivee & Catts, 1999; Nathan et al., 2004a; Stackhouse, 2006). Persisting speech, language and learning difficulties can also be associated with psychosocial difficulties such as bullying (Knox & Conti-Ramsden, 2003), behaviour problems (Botting & Conti-Ramsden, 2000) and school exclusion (Law & Sivyer, 2003). Often self-esteem is affected and intervention programmes need to adopt a total management approach involving the whole child (e.g. Nash, 2006; Nash et al., 2002). Given the heterogeneity of the population of children with speech difficulties, it is important to identify a profile of strengths and weaknesses for each child. For the child with persisting speech difficulties – that is, beyond 5 years of age – this becomes most important in order to support communication skills, psychosocial development, learning strategies and access to the school curriculum. (For further discussion of the management of children with persisting speech difficulties, see Book 3 of this series: Pascoe, Stackhouse & Wells, 2006.) However, these strengths and weaknesses may change as the child develops, or as different demands are made on him/her. It is, therefore, necessary to track the development of spoken and written language skills over time. This will increase our understanding of the way children with speech difficulties progress, which children are likely to resolve their difficulties, and how best to help them. Collecting the speech and language profiles of children whose speech difficulties resolve, compared to those who experience persisting difficulties, is important for both theoretical and practical reasons. Such knowledge enables appropriate service delivery and prioritisation. There have been several longitudinal studies of children with speech and language difficulties (e.g. Bishop & Edmundson, 1987; Conti-Ramsden et al., 2001; Leitao & Fletcher, 2004; Simkin & ContiRamsden, 2006), as well as longitudinal studies looking at the phonological awareness and/or literacy development of children with primary speech difficulties (e.g. Bird, Bishop & Freeman, 1995; Webster, Plante & Couvillion, 1997; Larrivee and Catts, 1999). However, few longitudinal group studies of children with primary speech difficulties have looked for predictors or markers of persisting speech difficulties

4


from a wide range of measures. Shriberg, Kwiatkowski and Gruber (1994) comment: Notwithstanding 60 years of research efforts to develop valid predictive instruments for developmental phonological disorders, there currently is no clinically effective procedure to predict which children will normalize with or without intervention. (p. 1129)

Shriberg, Kwiatkowski and Gruber’s (1994) follow-up study examined the short-term normalisation of 5-year-old children who had been classified as speech disordered one year earlier. A wide range of predictors was assessed including measures of phoneme accuracy and audiological, developmental, familial, psychosocial and therapeutic measures. Rather than make comparisons with typically developing children, Shriberg, Kiatkowski and Gruber (1994) compared children with speech difficulties who ‘normalised’ and those who did not. However, while the study offers detailed description of performance in the areas assessed and identifies the proportion of children whose speech difficulties resolved by age 5, the authors were unable to find any significant predictors of this short-term normalisation. In order to address this issue of prediction and to present procedures that are useful for investigating children’s speech and literacy difficulties, findings from two studies are reported throughout this volume. The first was a cross-sectional study of 100 normally developing children in the age range of 3 to 7 years (Vance, 1995; Vance, Stackhouse & Wells, 1994, 1995, 2005). A range of speech processing procedures was designed following the principles of the psycholinguistic framework presented in Book 1 of this series. These tasks were used in this study to investigate typical development in normally developing children, for later comparison with children presenting with speech difficulties. The second was a longitudinal study of 62 children in the age range of 4–7 years who had primary speech difficulties (Nathan et al., 2004a). Each child with speech difficulties was matched to a normally developing control child. This study allowed us to develop the procedures used in the first, cross-sectional study; make a direct comparison of the development of children with and without speech difficulties over time; and examine possible clinical markers for identifying children at risk for persisting speech difficulties and associated literacy problems. These two studies, together with a collection of smaller-scale projects, have enabled us not only to refine our assessment procedures, but also to indicate which tasks are the most helpful in predicting speech and literacy outcome. These auditory and speech procedures are presented in this volume. Some normative data have been collected for each of the tasks described. However, these are not standardised ‘tests’ since they have not undergone widescale normative data collection. Rather, they provide guidelines for interpreting the performance of a child with a speech difficulty across a range of tasks.


5

Working within the psycholinguistic framework, however, is not done in isolation. It is also important to take on board the other perspectives mentioned above in the introduction. Collecting information about a child’s developmental history, psychosocial development and family background is equally important. In this volume we therefore include some of the questionnaires used in the longitudinal study and a discussion of the subsequent findings. Before making use of the procedures devised, readers should understand the rationale behind the psycholinguistic framework. The remainder of this chapter will provide a brief summary of this. Throughout this chapter and this volume, the reader is referred to Book 1 in this series for further information.

What is a Psycholinguistic Approach? A psycholinguistic approach to assessment explores how a child receives information of different kinds (e.g. spoken or written utterances), remembers and stores it within lexical representations (information about words) within the lexicon (a store of words) and how he or she selects and produces spoken and written words. Figure 1.1 illustrates the basic structure of a psycholinguistic model of speech processing. On the left of the figure there is a channel for the input of information via the ear and on the right a channel for the output of information through the mouth. At the top of the model there are the lexical representations that store previously processed information, while at the bottom

Figure 1.1 The basic structure of the speech processing system Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

6


there is no such store. A lexical representation is considered to comprise the following information about a word: • Semantic representation – information about what the word means, the attributes of the word, what category it is in, e.g. animal, food. • Phonological representation – information about how the word sounds, allowing discrimination of the target word from other similar words. • Motor programme – a stored set of instructions for how to say the word, i.e. the pronunciation of the word. • Grammatical representation – information about the class of word, e.g. noun, how it can be used in a sentence, and whether there is a plural form that can be derived from a rule, e.g. adding ‘s’ at the end of it. • Orthographic representation – information about what the word looks like in its written form, thus enabling automatic recognition when reading. See Activity 1.2, Book 1, pp. 9–10 for further information. In psycholinguistic terms, top-down processing refers to speech activity that makes use of previously stored information (i.e. in the lexical representations). A bottom-up processing activity requires no such prior knowledge and can be completed without accessing stored linguistic knowledge from the lexical representations. It is helpful to identify four key anchor points in the model above: top and bottom; left and right (see Figure 1.2). A number of models use the same basic structure of input/output and representations as depicted in Figure 1.1 (e.g. Dodd, 1995, 2005; Stackhouse & Wells, 1997; Hewlett, Gibbon & Cohen-McKenzie, 1998; Chiat, 2000). Although different in their presentations, these models share the premise that children’s speech development is dependent on the normal functioning of this system and that children’s speech

Figure 1.2 The four key anchor points in a speech processing model Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.


7

difficulties arise at one or more points in a faulty speech processing system (see Baker et al., 2001, for a review of psycholinguistic models). The aim of the psycholinguistic approach in practice and research is to find out exactly where, within a chosen model, a child’s speech processing skills are breaking down, and how this might be affecting their speech, lexical and literacy development. The information gathered through this approach informs understanding of a child’s communication difficulties and the impact of this on other aspects of his or her development and educational performance.

Principles of a Psycholinguistic Assessment For assessment purposes, Stackhouse and Wells (1997) analysed the psycholinguistic properties of commonly used tests of auditory discrimination, speech and phonological awareness to address the question What do tests really test? (see Chapters 2 and 3 in Book 1). By asking questions about each task, they were classified within the simple speech processing model (depicted in Figure 1.1) as being input or output and top-down or bottom-up. The first question to ask about a task would be: Is this an input or output task? That is not as straightforward as it sounds! A task may require a verbal response, but still be an input task. For example, if the child is asked to say ‘same’ or ‘different’ in response to two words presented by the tester, this is not counted as an output response. The child could perform the task equally well by nodding or shaking his or her head or pointing to a symbol. An output task, by definition, requires the child to generate different answers for each item, for instance naming a series of pictures or producing a string of rhyming words to a given target. The second question is: Does the child have lexical representations for the stimuli used in the task? If the answer to this question is ‘Yes’ then the task is more of a top-down than bottom-up task. However, another question also needs to be asked: Does the child have to access these lexical representations in order to complete the task? If ‘Yes’ then the task involves the higher levels of processing, if ‘No’ then it is not reliant on top-down processing. A clearer perspective on what aspects of processing a task is really testing allows for more systematic interpretation of results and a more precise understanding of the nature of a child’s difficulties, thus supporting more appropriate planning of intervention. If you are familiar with these concepts then read on. If not, it is essential to read Chapter 2 of Book 1 and carry out the activities in that chapter, before administering and interpreting the procedures included in Chapters 3–7 of this volume. By analysing tasks in this way, a framework for collating and interpreting assessment results emerged. Tasks clustered together at hypothetical points on the model, depending on their processing demands

8


Figure 1.3 Crosses on the simple speech processing model to mark the levels at which tests cluster. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

(see Figure 1.3). However, this clustering was not always as might be expected. Some of the tests that could be considered to be assessing different areas because of their ‘title’ (e.g. ‘auditory discrimination’, or ‘phonological awareness’ or ‘speech’) clustered together. They were tapping the same level of processing, for example auditory discrimination of spoken minimal pairs and rhyme detection of spoken words (without pictures). Other tests that all had the same title (e.g. auditory discrimination) were tapping different levels of processing because of the nature of the stimuli involved, for instance discrimination of real or non-words. A clear message was: ‘Do not believe what you read in a test title or description!’ Before using a task, it should be analysed in terms of how it is presented; the stimuli involved; and the response required. (See Rachel Rees, Chapter 3 in Book 2, Stackhouse & Wells, 2001, for how to apply these principles to intervention tasks.) Using the illustration in Figure 1.3, we recognised the need to assemble a profile for summarising a child’s speech processing skills (see Book 1, Chapter 4). To develop such a profile, we posed a question about each


9

level of processing marked by a cross on the model. For example, to assess the effectiveness of processing at point A, nearest the ear, we asked: Does the child have adequate auditory perception? This could be answered by doing a hearing test. At point K, nearest to the mouth, we asked: Does the child have adequate sound production skills? This could be answered by an oral examination. Thus, appropriate investigations can be selected and/or devised to answer the questions posed at each level of processing (see Figure 1.4 for a complete list of assessment questions). Questions placed on the left of the model (A–F) address input; questions on the right

Figure 1.4 Questions for the speech processing profile. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

10


(G–K) address output; questions near/at the top of the model on both sides are addressed by tasks that require access to lexical representations; and questions near/at the bottom of the model on both sides are addressed by tasks that do not require lexical access. We have added a question ‘L’ between the mouth and the ear to remind us of the importance of observing if a child can monitor their own speech output, and, if so, can change it to produce a more accurate pronunciation. Appendix 2 of Book 1 lists examples of existing tests that can be used to answer the questions posed in Figure 1.4 and to complete a speech processing profile on a child. Appendix A.1 of this volume supplements this list by including additional, specifically designed speech and auditory tasks. These may help fill in some gaps in assessment material, and provide useful examples for how to design a series of tasks to assess a child’s individual speech patterns. However, the framework of questions is not restricted to only the tasks presented here; any task that involves speech processing can be classified within this framework. Indeed, an aim of this series is to encourage readers to devise their own procedures, based on a hypothesis testing approach, for investigating children’s speech and literacy difficulties. Answering each question about a child’s speech processing skills will normally involve administering more than one procedure (Book 1, pp. 99–101). Few individual tasks provide the answer to a question directly. An exception might be a test of hearing to answer Question A. However, problems identified at one level, such as a hearing impairment, do not necessarily mean that the child will perform poorly or as expected at all the other levels (see Ebbels, 2000 for a psycholinguistic case study of a child with a hearing impairment). Another point to remember is that the two dimensions of the model (left/right and top/bottom) are not organised in terms of difficulty. Tasks tapping lower levels of processing on the model are not necessarily easier than tasks involving higher levels of processing. Similarly, tasks placed on the left-hand side of the model are not necessarily easier than those placed on the righthand side. What is easier or more difficult for an individual child will depend on his or her own profile of skills, not on the structure of the framework. Identification of a child’s strengths and weaknesses is best done by contrasting results from different procedures within and between levels of processing. Assessing within levels establishes whether or not the child can demonstrate adequate processing skills. Investigations of different complexity can be carried out at the same level of processing. For example, for Question B auditory discrimination tasks using non-words are used. These tasks can comprise simple stimuli, for example VOS/VOT; articulatorily complex stimuli, for example VOST/VOTS; or multisyllabic stimuli, for example IKIBUS/IBIKUS. Tasks may have increased memory demands such as listening to two similar-sounding words such as


11

VESH/FESH and deciding if a third word heard, such as FESH, was the first or second of the original pair. Comparison of performance across these tasks establishes if the child has a specific difficulty at this level; that is, if he or she cannot perform age appropriately on the simple auditory discrimination task, or whether the ‘discrimination’ problem is on complex words only, or only when there is an increased memory load. Testing between levels involves unpicking why a child may have difficulties at one of the levels. Naming is a good example of this. A child may make errors on a picture-naming task for a number of reasons. The difficulty may be related to inaccuracies within the lexical representations, and/or at the level of assembling the word for spoken output, and/or at a lower level of articulatory difficulty that affects the pronunciation of the word. Comparison of performance across a number of output levels will allow the precise nature of the difficulty to be identified (see Book 1, p. 100, for more details). Identifying which levels of processing are involved will have implications for the design and delivery of an intervention programme (see Constable, 2001, Chapter 10 in Book 2).

The Speech Processing Profile The speech processing profile provides a structure for organising a wide variety of assessment procedures in such a way that a child’s performance on a range of tasks can be interpreted from a psycholinguistic perspective and over time. The purpose of collecting assessment information is to arrive at a greater understanding of a child’s needs. However, the results of an assessment procedure are limited in value if viewed in isolation. A bit of assessment information is like one piece of a jigsaw puzzle; it is necessary to collate results from different procedures and tasks in a systematic way in order to get the whole picture of the child’s needs. The aim of the speech processing profile is to fit together all the pieces to form a picture of the child’s strengths and weaknesses in terms of input, lexical representation and output skills. On the basis of this profile a comprehensive individual intervention programme can be designed, which takes into account the nature of the child’s speech processing difficulties. Without this, less obvious factors, such as subtle auditory processing deficits, can easily be overlooked, particularly when there appears to be a very obvious explanation for the child’s speech problems, such as a physical abnormality of the oral structure or cerebral palsy. For example, Pascoe, Stackhouse and Wells (2005) present a 6-yearold girl whose speech processing deficits belied the expectations one might have had given the child’s diagnosis of ataxia. Based on her speech processing profile, intervention did not focus on lower-level articulatory skills per se but rather on her auditory discrimination and

12


connected speech. It is only by profiling a child’s speech processing skills that the true extent of the underlying difficulties can be confirmed. The individualised nature of this approach to assessment does not, however, preclude the subsequent therapy or teaching being carried out in groups or within the classroom (see Popple & Wellington, 2001, Chapter 9 in Book 2). A child’s speech processing profile is constructed by providing answers to at least some of the questions A–L presented in Figure 1.4. For the purpose of recording a child’s profiling skills succinctly, a profiling sheet has been devised that incorporates these questions (see Figure 1.5, and Appendix A.2, which may be photocopied). This profiling sheet allows space to indicate a child’s performance at each level. This may be in the form of ticks (✓) for age-appropriate performance or above, and crosses (x) for performance that is less than age appropriate. Where the child’s performance can be compared to normative data, we can mark the degree of severity in terms of distance from the mean by standard deviations (sd), for example 1 cross (x) 1sd; 2 crosses (xx) 2sd; and 3 crosses (xxx) 3sd. Where normative data is not available, a more subjective indication of degree can be given. There is also space for administrative information about the child (e.g. name, date of birth, age, date of profile and name of profiler) and for general comments. Using this profile in practice and research has demonstrated how children given the same diagnostic label (e.g. ‘phonological impairment’ or ‘verbal dyspraxia’) can perform very differently (Stackhouse & Snowling, 1992; Williams & Stackhouse, 1998). Profiles of Tom and Zara, both 4 years of age and described as having phonological delay, are presented in Book 2 and in Figures 1.6 and 1.7 of this volume. These illustrate how Zara’s difficulties were confined to the right-hand side of the profile – that is, she had a specific speech output difficulty – while Tom’s were more pervasive, involving both input and output difficulties. The case of 5-year-old Alan, presented by Waters (2001, Chapter 6 of Book 2), demonstrates how a child can have severe speech output difficulties but still have intact input skills (see Figure 1.8). Interestingly, the approach taken by Waters in intervention was to work on speech input; that is, using Alan’s strengths. For further discussion of intervention case studies carried out within this framework, see chapters in Book 2 by Corrin, Waters, Dent, and Nathan and Simpson; Stackhouse, Pascoe & Gardner (2006); and Book 3 of this series, which focuses on school-age children. Similar intervention approaches are described by Dodd (2005); Crosbie, Holm & Dodd (2005); Crosbie & Dodd (2001); Dodd & Bradford (2000); and Ebbels (2000). The psycholinguistic framework presented here comprises not only the speech processing profile in Figure 1.5 but also two associated theoretical speech processing models: a box-and-arrow model, and a developmental phase model. For many practical purposes of assessment and


13

SPEECH PROCESSING PROFILE Name: Age:

Comments: d.o.b:

Date: Profiler:

Figure 1.5 A speech processing profile summary sheet. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

14

Compendium of Auditory and Speech Tasks SPEECH PROCESSING PROFILE

Name: Zara

Comments:

Age: 4;3

Non-verbal SS II

Date: T 1 Profiler:

Figure 1.6 Zara’s speech processing profile CA 4;3

A Psycholinguistic Approach SPEECH PROCESSING PROFILE Name: Tom Age: 4;0 Date: T 1 Profiler:

Figure 1.7 Tom’s speech processing profile CA 4;0.

Comments:

15

16


Name: Alan

Comments:

Age: 5;0 Date: Profiler:

Figure 1.8 Alan’s speech processing profile at age 5 years


17

intervention, using the speech processing profile format and procedures to identify a child’s speech processing skills will suffice. However, using a psycholinguistic model can add another dimension to our understanding of a child’s needs. Some users of the framework find models helpful for practice and research, for developing theoretical understanding and for communicating with others about using a psycholinguistic approach.

Theoretical Speech Processing Models Box-and-Arrow Model The conventional way of representing levels of processing and routes between them is through an information processing model in the form of boxes (levels of processing) and arrows (flow of information): Such models make explicit the hypothesized information-processing activities as carried out in a particular cognitive function (such as language), in a manner analogous to computer flowcharts that depict the processes and decisions carried out by a computer program. (Baker et al., 2001, p. 687)

The box-and-arrow model used in the framework described in this book is presented in Figure 1.9 (and in Appendix A.3 for photocopying). Plain boxes represent levels of processing, boxes in bold represent stored knowledge and shaded boxes represent off-line processing. Arrows show the route of processing of spoken material, and the broad arrows show how information flows between boxes as part of the learning process; that is, off-line processing. Such a model allows us to be more explicit about the levels of processing and processing routes that tasks are tapping. To understand many of the popular procedures, such as picture naming, it is informative to follow the journey taken through the model as the task is performed (see Book 1, pp. 173–187 for examples of routes). The model can, therefore, be used to illustrate the processing demands made by individual tasks and tracking routes through the model allows the demands of different tasks to be compared. The following is a summary of the components of the model in Figure 1.9: • Peripheral auditory processing refers to general auditory ability, not specifically related to speech. • Speech/non-speech discrimination is a pre-linguistic level of recognising that what is heard is speech rather than non-speech sounds and noises.

18


Figure 1.9 The box-and-arrow model. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

• Phonological recognition describes the level at which a listener recognises the language heard as familiar as compared to a language that is not. The listener is using knowledge about language-specific structures and will recognise, for example, that BLICK is a possible English word but BNICK is not. • Phonetic discrimination is the recognition of phonetic distinctions that are new to the listener, and is used in early stages in language learning as a child begins to learn which segments are contrastive. It is also used in second-language learning when unfamiliar segments are heard. • Phonological representations contain enough information for a heard word to be recognised as distinct from other similar-sounding words (e.g. TAP vs. CAP; CAP vs. CAT). • Motor programmes are a series of stored gestural targets (or articulatory instructions) for the production of known words.


19

• Motor programming is a facility to create new motor programmes. It is conceived as a store of phonological units that are selected and assembled into new combinations so that new productions of words can be articulated. • Motor planning assembles the gestural targets (articulatory instructions) into the correct sequence in real time, taking into account the context, for example assimilations and the appropriate intonation for a question form. • Motor execution refers to the peripheral production of speech at the level of the vocal tract. If this material is not familiar to you, you should read Chapter 6 in Book 1 and work through the activities, as this model will be used throughout this volume to illustrate the processing routes of the tasks presented. The model can also be used as a summary sheet to show a child’s specific speech processing deficits. For example, in Book 2, Waters (2001) interprets Alan’s speech processing profile at 5 years of age through the box-and-arrow model and summarises his performance by using ticks for strengths and crosses for weaknesses on the model itself (see Figure 1.10). She concluded that Alan’s internal representations were accurate, but that he had an inability to devise motor programmes to reproduce either stored phonological representations or auditorily presented familiar or novel combinations of phonemes. As with the profile, the model can also be used to track a child’s development over time. For example, although speech intelligibility and performance on speech processing tasks change, a core deficit may remain that can be depicted on the model and used as evidence for why a child may still need support even though progress has been made. The case of Zoë between the ages of 3;09 and 9;08 presented in detail in Chapters 9 and 10 in Book 1 is an illustration of this. Zoë’s speech was unintelligible at 3;09 and at 5;11. She had pervasive speech processing difficulties affecting both input and output, as demonstrated in the profile and the model. Although her speech was intelligible by 9;08 there were still residual speech problems, not obvious to her teachers, and also associated spelling difficulties. Figure 1.11 shows her persisting core deficits in the context of other processing strengths. One criticism of using box-and-arrow models with children has been their failure to account for developmental change (Stackhouse & Wells, 1996). This is addressed in Book 1 by building up the speech processing model developmentally from Cries to Words (Chapter 7) and from Speech to Literacy (Chapter 8). From this exercise, a second model was developed to account for developmental phases and change over time.

20


Figure 1.10 Speech processing model indicating Alan’s strengths and weaknesses at age 5 years

Developmental Phase Model The developmental phase model presented comprises five phases: • The prelexical phase includes babbling, up to about 1 year of age. • The whole word phase relates to the learning of first words as gestalts, up to about 2 years of age. • The systematic simplification phase is characterised by the emergence of simplifying processes in speech output between 2 1/2 and 4 years of age. • The assembly phase describes the mastering of connected speech at about 3 to 4 years of age. • The metaphonological phase relates to the breakthrough to phonological awareness by about 5 years of age.


21

Figure 1.11 Speech processing model for Zoë at CA 9;08. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

It is proposed that typically developing children move through these phases smoothly and, as a consequence, develop the skills necessary to perform phonological awareness and literacy tasks. In contrast, children with speech difficulties have trouble at one or more of these phases. Further, the precise nature of their speech difficulties will depend on which particular phase (or phases) is troublesome for them. For example, children stuck at the whole word phase, who are not yet using phonological simplifying processes, may present with classic symptoms of apraxia of speech. Children who are delayed reaching and/or moving on from the systematic simplification phase present with persisting phonological simplifying processes and may be described as having a ‘phonological impairment’ or ‘delay’. Associated with a delay moving on from the systematic simplification phase in particular are difficulties with connected speech production (in the assembly phase) and with developing phonological awareness; thus putting the child at risk of

22

Compendium of Auditory and Speech Tasks SPEECH TYPICAL DEVELOPMENT

DIFFICULTY

Pre-lexical

Hearing Structural Dysarthria

Whole Word

Dyspraxia

Systematic Simplification

Phonological Impairment/ Delay

Assembly

Stammering Prosodic ‘Mumbly’

Metaphonological

Literacy Dyslexia

Normal Development

Figure 1.12 A developmental phase perspective on speech difficulties. Typical development is listed on the left of the diagram and the subsequent difficulty arising from arrested or delayed development at each phase is indicated on the right. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

literacy problems. Figure 1.12 presents these phases and the difficulties associated with them if development does not follow normal lines. Katy, the 6-year-old girl described by Pascoe, Stackhouse and Wells (2005), had speech that showed characteristics of the whole word phase, with most productions having a Consonant Vowel syllable structure. Her intervention programme aimed to expand her syllable structures to include CVC, enabling her to make a much greater range of lexical contrasts. Although she had some patterns characteristic of the systematic simplification phase (e.g. stopping), it was developmentally appropriate to focus intervention on the earlier whole word phase. Following intervention it was noted that Katy’s speech was more characteristic of the


23

systematic simplification phase; she had been successfully helped to progress from the earlier whole word phase and could take advantage of intervention aimed at producing phonological contrasts.

Using the Psycholinguistic Framework in Practice and Research The speech processing profile and models in the psycholinguistic framework are a useful way of systematically collating assessment results. The profile can be used to make two types of comparison: intra-child and inter-child. Intra-Child Comparison When an individual child’s assessment results are collated on a speech processing profile, they can be used to determine the child’s personal strengths and weaknesses to aid in intervention planning. One of the principles of the psycholinguistic approach is that assessment should reveal a child’s processing strengths as well as weaknesses. Another is that it can be helpful to adapt assessments to individuals. To this end, we have tried to develop a set of procedures that can be used in a flexible and informative way: The psycholinguistic framework for assessment . . . does not constitute a fixed and immutable set of procedures, and thus differs from a test battery. (Stackhouse & Wells, 1997, p. 307)

Central to the development of assessment procedures are considerations relating to stimuli design (see Chapter 11 in Book 1). It was suggested here that stimuli should be carefully selected based on both lexical (i.e. real or non-words; words with high or low imageability) and phonological criteria (e.g. words with CCVC phonotactic structure; words with /s/ onset). Rees (2001, Chapter 2 of Book 2) suggested that stimuli should be designed to reflect a child’s pattern of errors, and gives the example of Robert, a 7-year-old boy with mild cerebral palsy who had difficulty in producing some specific consonant clusters (e.g. syllable initial /pl/) in real and non-words. In a mispronunciation detection task, where Robert was presented with both correct and incorrect forms of target words, his performance varied from word to word: he showed particular difficulties in discriminating the correct and incorrect forms of words such as PLATE and AEROPLANE. These items may not have been included in a generically designed mispronunciation detection task, but were incorporated in Robert’s assessment based on his own speech difficulties with /pl/ in a naming task. Bryan and Howard (1992), McGregor and Leonard (1989), and Ebbels (2000) also provide good examples of ways in which

24


assessment stimuli were modified in light of their participants’ specific areas of difficulty. Book 3 provides further examples of how stimuli for segment, cluster and connected speech tasks were designed for individual children with persisting speech difficulties. A child’s progress can be monitored using such specifically designed tasks; in this way the child acts as his or her own control in terms of whether he or she is making the same or fewer errors as time goes on. Inter-Child Perspective A child’s performance on a particular task can be compared to the performance of age-matched peers on the same task. Any task with norms can be used in this way, entering appropriate ticks or crosses to indicate degree of difference from the normative data on the speech processing profile. The procedures included in this compendium all have some normative data so that an individual child’s performance can be interpreted, and his or her progress monitored, in relation to typically developing peers. Using this inter-child perspective does not preclude the adoption of an intra-child perspective: the child will still have his or her own areas of relative strengths and weaknesses. For example, 1 cross (X) at a particular level of processing within the profile may be considered a relative strength for the child, even though performance is 1 s.d. below the mean, when compared to a level that has three crosses (XXX), indicating performance that is 3 s.d.s below the mean. There is a tension between the provision of normative data for tasks and sets of stimuli to allow inter-child comparison, and the need to identify individual children’s particular profiles. These profiles will be more accurate when the stimuli used in input and output tasks reflect the speech errors that the individual child makes (see the example from Rees, 2001 above). It is not possible in this volume to present normative data on every possible type of speech error that children make. Rather, normative data can be provided for tasks with generic sets of stimuli that might indicate whether this is an area of potential concern. Assessment of individual children may then require adaptation of the tasks to use stimuli that target the child’s own errors in order to be clearer about the nature of the difficulty.

Conclusion The psycholinguistic approach to the investigation and management of children with speech and literacy difficulties should not be used in isolation from other approaches, but it does help to provide a foundation on


25

which to develop ‘intervention based on sound rationales’ (Gourlay, Joffe & Levin, 2005). However, Gourlay, Joffe and Levin state that a barrier to using the psycholinguistic approach in everyday practice has been the lack of specifically designed tasks with normal control data. The compendium of assessment tasks presented in this volume aims not only to fill this gap, but also to indicate which procedures can predict speech and literacy outcome. It is hoped that this will facilitate the use of a psycholinguistic approach in practice and research.

Summary • Young children with speech difficulties make up a high proportion of speech and language therapists’ caseloads and need to be prioritised for management, in order to prevent long-term and associated difficulties. • Persisting speech difficulties are complex in nature and require careful investigation and management. • This chapter summarises the three components of the psycholinguistic framework: the speech processing profile, the box-andarrow model and the phase model. • The speech processing profile is a useful way of systematically ordering a child’s assessment results and can be repeated for comparison over time. • Referring to the box-and-arrow and phase models adds another dimension to our theoretical understanding of a child’s needs. • The aims of a psycholinguistic assessment are to identify a child’s processing strengths and weaknesses; the impacts these have on communication and literacy development; and how they can be used in the child’s intervention programme and to predict children at risk for persisting speech and literacy difficulties. • Completing a speech processing profile on a child allows two comparisons to be made: (a) intra-child – the child’s own strengths and weaknesses and (b) inter-child – how well the child performs in comparison to typically developing peers. • The psycholinguistic framework can be used to monitor progress in intervention and to uncover hidden difficulties beyond surface speech errors and diagnostic labels. • Although this volume (Book 4) revises principles of the psycholinguistic approach, it is assumed that readers will already have some knowledge of the approach, as presented in Books 1–3 in this series. • If the psycholinguistic approach is unfamiliar to readers, they should refer to Book 1 before trying to use and interpret findings from the procedures in this volume.

Chapter Two Development of the Assessment Tasks and Normative Data

Prior to adopting the psycholinguistic framework in their school in 1991, Popple and Wellington (2001) report that their routine assessment procedure for each child examined the following areas: 1 2 3 4 5 6 7 8 9 10 11

Medical background and developmental history. Expressive and receptive language. Use of syntax. Hearing. Auditory discrimination of simple real words. Oral skills, including mobility, coordination and muscle tone and the functioning of lips, tongue and palate. Ability to repeat sounds in isolation and within syllables. Ability to sequence single sounds and syllables. Diadochokinetic rates (the speed and sequencing of sounds in a repeated response). Naming pictures to examine sound production skills in single-syllable words and in words with increasing articulatory complexity. Spontaneous speech.

Indeed, this was a comprehensive assessment at that time carried out in a situation where the children could be seen on a regular basis. However, after plotting their assessment procedure on the speech processing profile (see Figure 2.1), Popple and Wellington report their surprise at the gaps. Typically, the focus was more on output than on input (four of their assessments target motor execution and lower-level speech output skills), with little formal investigation of children’s lexical representations and phonological awareness. Further, they report that as assessments had not been collated within a common framework, there was a tendency to

27

28


Name: Age:

Comments: d.o.b:

Date: Profiler:

Figure 2.1 A typical pre-psycholinguistic framework assessment Source: Popple & Wellington, 2001, p. 327

Development of the Assessment Tasks and Normative Data

29

view test results in isolation and not to examine the relationship between them, for example that between input and output processing skills. This segregation of tests and bias towards output was reflected in the therapy used and targets set for children (for further discussion see Chapter 9 by Popple and Wellington in Book 2, Stackhouse & Wells, 2001, p. 302). One aim of this volume is to supplement existing resources for completing a child’s speech processing profile by providing procedures for the investigation of auditory and speech skills. Ready-made procedures, such as those included in Chapters 3–7, are not tailor-made for in-depth assessment of an individual child’s phonological system, but they do have the advantage of having normative data available for comparison purposes, and are therefore useful for initial/diagnostic assessments and monitoring a child’s progress. They also have the advantage of carefully designed stimuli to allow cross-task comparisons, for example between different levels of input or different levels of output tasks; or between a matched input and an output task. The information derived from these will highlight a child’s strengths and weaknesses and therefore contribute to decision making about intervention strategies. However, further individually designed investigations based on hypotheses about possible levels of breakdown generated from a linguistic analysis may be needed where there is an opportunity to study a child in more depth (see Book 3, Pascoe, Stackhouse & Wells, 2006, for examples of such case studies). In summary, the psycholinguistic profile tells you how to go about the intervention, while the linguistic analysis provides the information about what to use in intervention, in particular the stimuli and targets to select. As stated in Chapter 1, the procedures included in this volume are not standardised, as the samples of typical data presented are small. However, they have been used in a number of group and case studies. As a result, we can provide information about sensitivity at different ages, and which procedures are the best predictors of speech and literacy outcome at the age of around 7 years. However, there is no one assessment that stands alone as a predictor of persisting speech difficulties and associated risk of literacy problems. Rather, it is the collective findings from the procedures and the monitoring of type and rate of progress that form the most important clinical marker of outcome (see Chapter 8 for further discussion). The information gathered from the studies has also provided some insight into typical development and the nature of atypical development in children with speech and literacy difficulties. Guidelines on typical performance on the procedures are given for comparison purposes throughout. This chapter presents the background information to two key projects that have contributed the most to our understanding of both performance on speech processing tasks in typical development and the trajectory of children with persisting speech difficulties. The first of these projects was a cross-sectional study of 100 typically developing children in the age range of 3 to 7 years. It gives us a ‘snap-shot’ view of children’s task

30


performance at the ages of 3, 4, 5, 6 and 7 years of age. The second was a longitudinal study of 47 children with primary speech difficulties, together with matched controls without speech difficulties, from the age of 4 to almost 7 years of age. This gives us a developmental perspective on how children with normal and atypical speech development perform on speech processing tasks over the early school years, a critical time for transferring spoken language skills to written language.

The Cross–Sectional Study The cross-sectional study was funded by the Joint Research Advisory Committee (JRAC) of the National Hospital for Neurology and Neurosurgery at Queen Square in London. It investigated typically developing children’s performance on a range of tasks devised for the speech processing profile. These included assessments of auditory discrimination, speech output and phonological awareness. The stimuli were carefully designed to allow comparison across tasks. The main aims of the project were to: • Collate and develop assessment materials for the psycholinguistic investigation of children with speech difficulties. • Collect normative data from children in the age range of 3–7 years on these assessments. • Use the materials in selected case studies of children with speech difficulties Participants One hundred children in the age range of 3–7 years participated in this study. They were attending mainstream nurseries and primary schools and met the following criteria: • No history of speech and language difficulties. • No history of a significant hearing loss. • Reported by their teacher to be within average range for their year group. • No evidence of general or specific learning difficulty. • English was the language spoken at home. • No significant medical or neurological condition. • Speech production within normal limits – with a standard score of above 85 on the Edinburgh Articulation Test (Anthony et al., 1971). There were 20 children, ten boys and ten girls, in each of five age groups: 3, 4, 5, 6 and 7 year olds (see Table 2.1). Children from at least two different schools were represented within each age group.


31

Table 2.1 Details of participants in the cross-sectional study

Age group 3 4 5 6 7

year year year year year

olds olds olds olds olds

Number (10 boys and 10 girls per group) 20 20 20 20 20

Mean (s.d.) age in years; months 3;07 4;05 5;06 6;06 7;05

Age range in years; months

(0.26) (0.27) (0.27) (0.29) (0.27)

3;01–3;11 4;01–4;11 5;01–5;11 6;01–6;11 7;01–7;11

Stimuli Most of the tasks used stimuli based on a list of 60 core words: 20 1-syllable words; 20 2-syllable words; and 20 3- or 4-syllable words. These words were selected using the following criteria: • • • •

Within the expected vocabulary of 3-year-old children. Able to be represented by a picture. Of appropriate word length. Representing a range of consonants and clusters within the list at each word length. • Half of the items within the 2- and 3/4-syllable word lists able to be manipulated to result in a metathetic change; that is, segments could be transposed within the word as in CATERPILLAR to CAPERTILLAR. Phonetically matched non-words were then derived from these real word stimuli by changing the vowels. Each vowel was replaced by one of a similar length; that is, short, long or diphthong. The usual stress pattern of each word was maintained for the matching non-word. See Table 2.2 for a full list of real and non-word stimuli. The Tasks Two auditory tasks were designed using these stimuli: • Mispronunciation detection, an auditory discrimination task with pictures. A picture was presented of one of the targets, for example PLATE, and a monkey soft toy ‘said’ the name of the picture correctly or wrongly (e.g. PLATE or PATE). The child was asked to say whether the monkey got the word right or wrong. (We have previously called this same task auditory lexical decision and ‘Picture Yes/No auditory discrimination’ (e.g. in Book 1, Stackhouse & Wells, 1997).)

32


Table 2.2 The 60 core word stimuli and their matched non-words used in the cross-sectional study 1-syllable words Real words

Non-words

brush

sponge glove duck leaf sock cat book torch mouse knife snake train van watch plate roof fish chair thumb

2-syllable words

3- and 4-syllable words

Real words

Non-words

Real words

Non-words

/brIʃ/

sandwich

toilet money feather yellow kitchen ladder flower dustbin jelly tractor fishing biscuit scooter parrot seesaw slipper sausage guitar spider

/sImwəd/ /teIlət/ /mεnə/ /fæðI/ /jælɔI/ /kɒtʃn / /lεdI/ /fluwI/ /dæsbən/ /dzlə/ /trεktI/ /fɒʃIŋ/ /bɒskεt/ /skitə/ /prIt/ /sɑsi/ /slɔpə/ /sεsəd/ /gεtɔ/ /speIdI/

caterpillar

/spænd/ /glεv/ /dæk/ /lɔf/ /sɔk/ /kεt/ /bɔk/ /tutʃ/ /mɔIs/ /nəυf/ /snaIk/ /trɔIn/ /vIn/ /wυtʃ/ /pləυt/ /rɔf/ /fεʃ/ /tʃi/ /ɒm/

/kItəpælə/ /spgItə/ /ælIfɒnt/ /krɒvIn / /krIkədaυl/ /æmbrælI/ /hIləkæptə/ /kɒŋgIrɑ/ /tυləvæn / /hæspətIl/ /tɒləfaIn/ /pεrəʃit/ /bætəfləυ/ /kεmpjaυtI/ /rəυndəbaIt/ /hɔIdræsI/ /ɒrəpləυn/ /pdɔmIz/ /hImbɑgI/ /deInIsɑ/

spaghetti elephant caravan crocodile umbrella helicopter kangaroo television hospital telephone parachute butterfly computer roundabout hairdresser aeroplane pyjamas hamburger dinosaur

• Non-word ABX task. Two soft toys were used, one ‘said’ a nonword, for example PLEUT (A stimulus), and the second ‘said’ a minimally different non-word, for example PEUT (B stimulus). The child was then asked, ‘Now show me which one said PEUT’ (X stimulus). In addition a third auditory task was administered: • Same/different task with words and non-words. Paired stimuli taken from Bridgeman and Snowling (1988) were presented. Children were required to make same/different judgements about minimal pairs, for example KISS KIT; KES KET. Three speech output tasks were designed using the same stimuli: • Naming pictures of the items. • Repetition of the words. • Repetition of the non-words. The aims of the assessments were to: • Identify the age range in which changes in speech processing and production might be expected.


33

Table 2.3 Order of presentation of the six assessment tasks in the crosssectional study (and chapter in which task is presented in detail in this volume)

Non-word repetition (Chapter 5) Word repetition (Chapter 5) Naming (Chapter 5) Mispronunciation detection (Chapter 4) ABX task (Chapter 3) Same/different task (from Bridgeman & Snowling, 1988) (Chapter 3)

Session 1

Session 2

List A List B List A List A List A All

List B List A List B List B List B

• Describe the profile of performance across different tasks and how this changes with age. • Evaluate effects of length of stimuli on task performance. • Collect normative data for the purposes of comparison with data gathered from children with speech and literacy difficulties. Procedure The tasks were administered over two sessions. The stimuli were divided into two lists: List A contained all one-syllable words and non-words and 10 of the two-syllable word or non-words; List B contained 10 of the two-syllable words or non-words and all three- and four-syllable words or non-words. Table 2.3 summarises the order of task presentation that took place for each child over two sessions. This order ensured that children named pictures before they repeated the words and did not repeat the words before the matched non-words, thus preventing priming effects. Results The results from this study suggested the following: • Children generally improved on the tasks between the ages of 3 and 7 years, which confirmed that the tasks were developmentally sensitive. • In typically developing children, performance across the tasks was significantly correlated, suggesting that dissociations between tasks may indicate specific difficulties. For the auditory discrimination tasks, testing indicated that the 3- and 4-year-old children could not respond reliably to the non-word ABX task or the same/different task. The differing demands of auditory discrimination tasks will be discussed in Chapter 3.

34


The findings revealed the following about speech production skills in typically developing children (see Vance, Stackhouse & Wells, 2005 for further discussion): • Stimulus length had an effect on speech production. • There was no significant difference between 6 and 7 year olds on the speech production tasks, suggesting that typically developing children reach a ceiling on these tasks by about 6 years of age. • Typically developing 3 and 4 year olds were significantly more accurate on repetition tasks than on naming tasks. • There was no significant difference between 3 year olds’ performance on real and non-word repetition tasks. • By 4 years of age, real word repetition was significantly better than non-word repetition, and this difference increased with age. • By 5 years of age, there was no difference in performance between word repetition and naming performance. This study formed the basis for many of the procedures described in this volume. Having established that the tasks were developmentally sensitive, it was possible to go on to use them, or adaptations of them, in a longitudinal investigation of children’s speech and literacy skills.

The Longitudinal Study The longitudinal study was funded by a research and development grant from the North Thames Regional Health Authority (NTRHA). It explored the relationships between children’s speech development, language development, phonological awareness and literacy skills. The purpose of the study was to examine why children with speech difficulties may be at risk for literacy problems, and to establish when and how children who will go on to have persisting speech difficulties, and associated literacy problems, may be identified. The project had the following aims: • To compare the development of speech, language and literacy skills in children with speech difficulties with those of typically developing children in the age range 4–7 years. • To identify which children referred to speech and language therapy with speech difficulties are at risk for literacy difficulties and why. • To evaluate an assessment procedure that could contribute to the identification of the at-risk child. • To develop our understanding of the relationship between speech and literacy difficulties.


35

• To examine the implications of the findings for the management of children with primary speech difficulties. Participants Children with primary speech difficulties who met the following criteria were referred to the project by their speech and language therapists: • Chronological age 4–5 years. • Obvious speech difficulties with no evident physical cause (e.g., not children with cleft lip and palate, or cerebral palsy). • No hearing impairment. • No associated medical condition (e.g. epilepsy, a named syndrome). • No severe receptive or pragmatic language difficulties. • Monolingual speakers of English. Eighty-two children were referred of whom 47 met the following criteria: • Significant speech difficulties; that is, scoring more than one standard deviation below the mean on the Edinburgh Articulation Test (Anthony et al., 1971). • Non-verbal IQ within normal limits, on two subtests of the WPSSI-R: Block Design and Picture Completion (Wechsler, 1990). • Non-reader or beginning reader. All children scored below the 60th centile on the British Ability Scales (BAS) single-word reading test (Elliott, Murray & Pearson, 1983). Most children could not read any words, four children could read one or two words, and one child could read seven words. Of this group of 47 children with primary speech difficulties, 31 were boys and 16 were girls. All participants were retained for the duration of the study. Each of these children was matched to a peer from within their own school or nursery where possible. Class teachers were asked to refer children to the study as potential controls for the individual speech cases, on the basis of: • • • • • • • • •

Same gender as the child with a speech problem. No speech or language difficulties. No history of speech and language problems. No speech and language therapy appointments. No known hearing impairments. No specific or general learning difficulties noted. No obvious medical condition (e.g. epilepsy). Monolingual speaker of English. Similar home background/mother’s educational level.

36


Table 2.4 Mean (standard deviation) of scores on two non-verbal measures of children with speech difficulties and the typically developing control group

Non-verbal IQ

Speech-disordered group Mean (s.d.)

Typically developing control group Mean (s.d.)

Block Design (ss) Picture Completion (rs)

10.28 (2.29) 12.11 (2.11)

10.68 (2.73) 12.21 (2.22)

This putative control group was tested on the WPSSI-R Block Design and Picture Completion Tests (Wechsler, 1990). There was no difference between the children with speech difficulties and their controls on nonverbal IQ measures (see Table 2.4). They were also tested on the singleword reading test of the British Ability Scales (BAS) (Elliott, Murray & Pearson, 1983). Thirty-six children were non-readers (raw score of 0) and 11 were beginning readers (i.e. reading between one and three words and scoring below the 60th centile on the BAS). In 25 cases, the typically developing control child attended the same nursery or school as the matched child with speech difficulties in order to control for teaching environment. Where this was not possible, the control was drawn from the same pool of nurseries/schools. Each of the 47 typically developing children was, therefore, matched to one of the children with speech difficulties on the basis of chronological age (within a six-month range), gender and non-verbal IQ (within 2 points on the averaged standard score of Block Design and Picture Completion; Wechsler, 1990). There were also no differences between the two groups on social class or level of parents’ education. A range of backgrounds was sampled and all children received state education from a range of Local Education Authority schools (see Table 2.5). The children were assessed at three points in time (T): • T1 at age 4 years (mean age of group 4;06) • T2 at age 5 years (mean age of group 5;08) • T3 at age 6 years (mean age of group 6;08) There was no attrition from the group with speech difficulties. All typically developing controls were retained at T2, but three were not traceable at T3. Seventeen pairs (36%) of the sample remained matched by school at age 6 years 7 months. Assessments included a range of expressive language, receptive language, speech output skills, speech input skills, phonological awareness and literacy measures. The assessment battery included standardised tests and procedures devised for the psycholinguistic framework. In addition, medical, family, psychosocial, educational and therapy information was collected through questionnaires from parents, teachers


37

Table 2.5 Descriptive data of children with speech disorders and their typically developing control group Speech-disordered group % Gender

Male Female

Typically developing control group %

70 34

70 34

*Parental occupation I II III IV V

12.5 20 57.5 10 0

8.8 23.5 61.8 5.9 0

Father’s education No school qualifications CSEs/O Levels/GCSEs A Levels Further qualifications

36.4 36.4 27.3 48.4

37.9 51.7 10.3 48.3

Mother’s education No school qualifications CSEs/O Levels/GCSEs A Levels Further qualifications

23.1 59 17.9 37.5

22.9 60 17.1 45.7

Note: *Parental Occupation Categories (HMSO, 1995): I Professional etc. occupations; II Managerial and technical occupations; III Skilled occupations (N) Non-manual (M) Manual; IV Partly skilled occupations; V Unskilled occupations; if both parents were working, the profession rating the higher was noted

and speech and language therapists (see Chapter 9). Structured interviews were also carried out with the children themselves about their attitudes towards reading. The basic premise of this study was a psycholinguistic one: that the skills necessary for successful spoken and written language development arise from an intact speech processing system. The hypothesis was that compared to typically developing matched controls, the children with specific speech difficulties would have a deficit at one or more points in their speech processing system (as depicted in Figure 1.1 in Chapter 1). To test this hypothesis we needed a battery of assessments that would allow us to assess different levels of the system. Assessments At T1 our speech processing tasks included: • picture naming • real word repetition • non-word repetition

38

Compendium of Auditory and Speech Tasks • speech rate • mispronunciation detection (an auditory discrimination task using pictures, as described in the previous section on the cross-sectional study) • auditory discrimination of non-words, ABX task • auditory discrimination of words and non-words, same/different task • rhyme production • rhyme detection using pictures • phoneme completion

However, a challenge for a longitudinal study is to devise tasks that will be sensitive throughout the duration of the study. For example, in our study it was necessary to make the speech repetition tasks, designed at T1, more challenging as the children got older. More complex stimuli were therefore included at T2 and T3. The same principle applied to the phonological awareness and literacy tasks. Table 2.6 lists the assessments administered. See Appendix B. for full details of the published assessments used as part of the test battery. The speech and auditory procedures used will be presented in detail in Chapters 3–7 in this volume. Data from the Typically Developing Children Although data from this control sample was originally collected for a match with the children with speech difficulties, it is possible to use this as normative data for comparison with other cases. This sample demonstrated a range of non-verbal skills within the normal range, and the children came from a range of socio-economic backgrounds. The higher proportion of males mirrors the effect of gender on prevalence rates of spoken language difficulties in children (Shriberg, Tomblin & McSweeney, 1999; Locke, Ginsborg & Peers, 2002). Their performance was also within the range of published normative data on assessments that were standardised. In order to confirm that this control group was representative of the population, data was collected from a further 20 5-year old children, attending a reception class in a mainstream primary school in North London (Broadbent, 2000). There were no exclusion criteria for this study, and all children whose parents gave permission were included. Thus, this further sample included six bilingual children and one trilingual child; 45% were boys and 55% girls. Three of the children performed less well than the others. With the exception of these outliers, the children were very similar to the typically developing control group in terms of non-verbal IQ, and performance on both speech input and output tasks. We can therefore confidently include data from this typically developing control group in the following chapters as normative data.


39

Table 2.6 Assessment batteries at Times 1, 2 and 3 in the longitudinal study Time 1 Mean CA 4:06 Non-verbal IQ

Output phonology

Input phonology

Expressive language

Receptive language

Phonological awareness

Literacy

Block Design (WPSSI-R) Picture Completion (WPSSI-R) Word repetition Non-word repetition Naming scored for articulation Speech rate and consistency *Edinburgh Articulation Test Mispronunciation detection ABX non-word discrimination Discrimination same/ different Naming scored for vocabulary Renfrew Bus Story Renfrew Action Picture Test

British Picture Vocabulary Scales Test for Reception of Grammar Rhyme production Rhyme detection PAT Phoneme completion

Letter names Word reading (BAS)

Time 2 Mean CA 5;08

Time 3 Mean CA 6;08 Block Design (WPSSI-R) Picture Completion (WPSSI-R)

Word repetition Non-word repetition Naming scored for articulation Speech rate and consistency

Mispronunciation detection ABX non-word discrimination Discrimination same/ different Naming scored for vocabulary Renfrew Bus Story Renfrew Action Picture Test CELF-R Recalling Sentences Semantic fluency British Picture Vocabulary Scales Test for Reception of Grammar Rhyme production Rhyme detection PAT Phoneme completion Alliteration fluency Phoneme deletion Letter names Word reading (BAS) Spelling (BAS) Non-word reading (graded non-word reading test) Reading comprehension (Neale) Spelling from pictures

Note: *EAT administered to speech-disordered group only.

Naming scored for articulation Speech rate and consistency

Mispronunciation detection ABX non-word discrimination

Naming scored for vocabulary Renfrew Bus Story Renfrew Action Picture Test CELF-R Recalling Sentences Semantic fluency British Picture Vocabulary Scales Test for Reception of Grammar Rhyme production Rhyme oddity PAT Phoneme completion Alliteration fluency Phoneme deletion Letter names Word reading (BAS) Spelling (BAS) Non-word reading (Graded non-word reading test) Reading comprehension (Neale) Spelling from pictures Non-word spelling

40


Results Generally, the children with speech difficulties performed significantly less well than the typically developing children on the majority of speech, language and phonological awareness measures. At T2 (CA 5;08) and T3 (CA 6;08) they also performed significantly less is omitted well on literacy tasks. However, performance on all tasks showed considerable variability, with some children with speech difficulties showing age-appropriate skills on some tasks. This highlights the heterogeneity of this group of children, and the need to investigate individual children’s strengths and weaknesses. Change in performance over time also occurred as some of the children’s speech difficulties resolved. However, sometimes the improvement in a child’s speech was at a surface level only and masked the persisting underlying difficulties that may manifest in more complex utterances, connected speech and on literacy tasks, particularly on spelling (see Nathan & Simpson, 2001, Chapter 8 of Book 2, for discussion of such a case).

Using Psycholinguistic Procedures It is important to screen children presenting with speech difficulties for input and output processing difficulties and, in selected cases, it is appropriate to draw up a full speech processing profile. The following chapters present some assessment materials, from the two studies described above, which will help to do this. They are by no means comprehensive or finite. The tasks included in this volume address questions B to E on the input side of the speech processing profile and questions G and I to L on the output side (see Appendix A.1). Questions F and H are addressed by phonological awareness tasks that are not included here. However, suggestions for appropriate tasks for Questions F and H are given in Appendix 2 of Book 1 and of Book 3. Answering Question A on the profile involves hearing tests and acoustic investigations. Chapters 3 and 4 of this volume (Book 4) will focus on input (auditory discrimination) procedures and Chapters 5, 6 and 7 on output (speech production) procedures. Chapter 8 describes how these tasks may be used to predict which children will resolve their speech difficulties and which will have persisting speech difficulties and associated literacy problems, and Chapter 9 presents the questionnaires used to collect background information on the children in the longitudinal study. However, the user is not tied to using only the tasks presented here: other materials can also be used to answer the questions on the speech processing profile if they have been subjected to the What do tests really test? exercise presented in Chapters 2 and 3 of Book 1. It is assumed that users of these procedures will have undergone some professional training in the principles of administering assessments,


41

including standardised tests and non-standardised procedures, or that they will be undergoing this training with supervision. They will also need to be familiar with the psycholinguistic framework presented in this and previous volumes in the series, in order to make appropriate interpretations of a child’s performance. Accent Warning! Even though users of the procedures may be presenting the tasks in the same language, in this case in English, they should be aware that differences in regional accents can affect their administration and the interpretation of a child’s performance, particularly if the child speaks with a different accent from the investigator. Stimuli that are particularly vulnerable to accent changes are marked in the tasks: for example, in Chapter 5 STRAWBERRY is presented as a three-syllable word but may be produced as a two-syllable word by some speakers (‘straw-bre’). Similarly, consider how many syllables are in your own production of POLICEMAN (two or three?) and in BALLOON (one or two?) and compare your production to the typical production of the children with whom you work. Variations in accent should also be considered when scoring a child’s speech production as correct or not. For example, a child who uses a glottal stop at the ends of words would not be marked as ‘wrong’ on a speech task if this was part of his or her regional accent. Who Can the Procedures be Used With? The procedures included in this volume have been used with both typically developing children and children with a range of difficulties (e.g. children with dyslexia, dyspraxia, phonological impairment, cleft lip and palate, specific language impairment, dysfluency, dysarthria, cerebral palsy, Down’s syndrome, acquired language problems or semantic-pragmatic difficulties). The procedures have not been designed for a particular clinical entity or population – far from it. Rather, they investigate speech processing skills in any child regardless of a diagnostic label that might have been given. As a result, the approach has uncovered individual differences between children given the same diagnostic label, for example dyspraxia. It has located more than one level of difficulty within children suspected of having a specific level of difficulty: for example, some children with dysarthric difficulties or with dysfluency have been found to have auditory discrimination problems. The procedures have also helped to locate different levels of difficulty for different aspects of a child’s speech output: for example, voicing errors might reflect a different underlying difficulty to cluster reduction within the same child. They have contributed to theoretical understanding of the nature of disorders such as word-finding difficulties.

42


When Not to Use the Procedures The procedures are not advocated for young, pre-school children or older children with marked attention difficulties. They should be discontinued if a child is unwell or unhappy in the assessment situation. Although the procedures have been used with bilingual children, the stimuli presented in this volume are monolingual English and will not keep their original phonetic/phonological properties when directly translated to another language. With this in mind, phonological awareness tasks have been specifically developed for German speakers by Blanca Schaefer and Silke Fricke at the University of Sheffield. Information about the psycholinguistic framework is published in German (Fox, 2004), Danish (Thompson, 2000), French (Wells, Stackhouse & Vance, 1999) and Portuguese (Vance, 1996).

Interpreting Children’s Responses The normative data reported in this volume provides the mean performance on tasks and stimuli sets for groups of typically developing children across a range of ages. Standard deviations are also given, where these are available. The data can be used to interpret the performance of children with speech disorders on the tasks presented. A comparison of a child’s score with the mean score for age-matched typically developing children can be made, but a more robust conclusion about a child’s profile of skills can be drawn if z scores are calculated. A z score gives an indication of how discrepant a score is from the normal range, a z score of plus or minus 1.65 being significantly different from the normal range. The larger the z score, the more discrepant the child’s performance. The process is illustrated in the chapter by Juliette Corrin (Chapter 4 of Book 2) in which she presents data from Anna, a 7 year-old girl with a severe speech difficutly. She reports Anna’s z scores on different tasks and different stimuli sets and uses this to complete a speech processing profile for her (see Corrin, 2001a, pp 105–107). Z scores can be calculated by taking the child’s score, subtracting the mean from the normative data and then dividing by the standard deviation. For example, Table 2.7 presents normative data on an auditory discrimination task that is reported in Chapter 3 of this volume. The mean scores and standard deviation are given for groups of children aged 5, 6 and 7 years. If a 7-year-old child with speech difficulties scores 40 on this task, then the z score is calculated as follows: Child’s score (40) Group mean for 7 year olds (52.3) Standard deviation for 7 year olds (4.65)

2.64


43

Table 2.7 The mean (standard deviation) scores on the Full Version of Auditory Discrimination Task 3: ABX task (data from cross-sectional study) Age group

Mean score (s.d.)

5 yrs (N 20)

6 yrs (N 20)

7 yrs (N 20)

40.25 (9.16)

45.65 (7.86)

52.30 (4.65)

(N 60)

This z score is, therefore, significantly discrepant from that of the typically developing children. Where no data is available for the age of the child, then comparison can be made with the group of children who are closest in age. If this group is younger than the child being compared, the z score is likely to be an under-estimate and, therefore provides a conservative value. If this group is older than the child being compared, the z score is likely to be an over-estimate of the amount of difficulty, and should be treated with caution. For some tasks and stimuli sets there is a wide variation of performance within the typically developing group. This will have the effect of lowering z scores, and children with lower scores may still fall within the normal range. For some of the auditory tasks presented, the child has a choice of only two responses. This means that the child is likely to get over 50% of the items correct just by chance. Thus, a child might appear to have a high score but might not fully understand the task or be able to complete it correctly. Where possible, chance level on the tasks is indicated in the following chapters.

Outline of the Following Chapters Chapters 3–7 describe procedures for assessing input and output speech processing skills. For each task the following information is given: • • • • • • • • •

aims description of the task use of the task psycholinguistic analysis of the task level on the speech processing profile stimuli materials (supplied on CD or required) procedure scoring

44

Compendium of Auditory and Speech Tasks • normative data • questions to ask • other tasks and/or standardised assessments tapping the same level of processing

Conclusion The psycholinguistic perspective is an important addition to understanding a child’s needs if it is linked with investigations from other perspectives: medical, linguistic, psycho-social and educational. It should not be carried out in isolation from a holistic consideration of the child and how he or she functions in the world around him or her. When a child is unable to complete the practice items on a particular task, the investigator should analyse why the child cannot embark on this procedure: for example, is it because of hearing problems, verbal comprehension problems or an attention deficit? For some children, poor performance on a task may not be a reflection of their typical speech processing skills. Other cognitive or language limitations may affect performance, or the child may be feeling unwell or tired that day or lack concentration.

Summary • The aim of this volume is to supplement existing resources for completing the speech processing profile by providing specifically designed assessment procedures for the investigation of auditory and speech production skills. • This chapter presents the background information on two key projects that have contributed to our understanding of speech processing development: a cross-sectional study of 100 normally developing children in the age range of 3 to 7 years, and a longitudinal study of 47 children with primary speech difficulties and their typically developing matched controls from the age of 4 to 7 years of age. • An overview of the assessments administered in these studies is given, and the speech production and auditory tasks used are presented in detail in the chapters that follow. • The detail of each task will be presented using 12 subheadings, enabling the reader to access relevant information more easily. • It is possible to compare the performance of a child with speech processing difficulties with the normative data provided by calculating z scores.


45

• There is no one task that stands alone as a predictor of persisting speech difficulties and associated risk of literacy problems. Rather, it is the collective findings from the procedures collated onto the speech processing profile, and the monitoring of a child’s progress, that are the most important indications of outcome. • Users of the procedures presented in this volume should have training in principles of testing children, be familiar with a range of both standardised and non-standardised assessment procedures, and be familiar with the psycholinguistic approach used in this volume and in Book 1 of the series. • Users of the procedures should be aware of their own speech production, in terms of accent, and how this might affect interpretation of a child’s performance. • The stimuli developed in English for the tasks presented in this volume will not keep their key phonetic/phonological properties when translated to other languages. Although the procedures can be used globally, language-specific stimuli for the individual tasks will need to be designed for users who are not working in English.

Chapter Three Auditory Discrimination Tasks

Hearing is one of the first areas to check when a child presents with a speech difficulty. However, a child may pass a hearing test perfectly well, but still not be able to detect differences between similar-sounding words or speech sounds. Tests of auditory discrimination that involve the perception of ‘small differences in the properties and qualities of sounds that fall within the same type’ (Morgan Barry, 1989, p. 24) should therefore be carried out routinely when investigating children’s speech, language and literacy difficulties. Locke (1980a, b) reviewed auditory perception assessments commonly used by clinicians at that time, and suggested some specific ways of assessing speech perception. One of these is the ABX task, presented later in this chapter, where two stimuli (A and B) are presented and then one of them is presented again (X); the task is to decide whether X is the same as A or B. This illustrates how an auditory discrimination task can incorporate complex speech processing, conceptual and memory demands. Schissel (1980) identifies the variables of memory, attention and processing time, as well as other cognitive factors such as conceptual understanding, which may also affect performance on auditory tasks. Kamil and Rudegeair (1972) found that the performance of 5- and 6-year-old children on an ABX task and a same/different task improved significantly on repeated testing. They suggested this was not the result of improved discrimination but of an ‘increased facility with the task, or to factors of attention, or to both’ (Kamil & Rudegair, p. 1090). It can be difficult to measure auditory discrimination skills and preschool children often perform inconsistently on tasks. A child’s performance on a same/different task, for example, will depend on his or her ability to recognise and make judgements about the concepts of same and not the same, or different. Beving and Eblen (1973) found that 4-year-old children were more able to repeat non-word pairs than to judge whether the non-word

47

48


pairs were the same or different. They concluded that these young children could obviously discriminate the relevant sound differences sufficiently well to achieve correct repetitions of the non-word pairs, but were unable to make the relevant cognitive judgements to do the same/different task. Task performance improved with age and for 6- and 8-year-old children same/different performance was a better reflection of discrimination skill. Successful performance on same/different tasks also requires short-term memory skills, as the child retains the two stimuli for long enough to make the necessary comparison and judgement (Morgan Barry, 1994). There is a greater short-term memory demand on an ABX task than on a same/ different task, as there are three stimuli to be compared rather than two. These issues need to be taken into account, and some tasks and stimuli sets would not be appropriate for younger children or for children whose language and/or conceptual development is very delayed. Their performance on auditory discrimination tasks might be affected by factors other than input speech processing skills; that is, they may ‘fail’ an auditory discrimination test because of memory or conceptual limitations rather than because of auditory discrimination difficulties per se. Two published examples of same/different tasks are Wepman’s Auditory Discrimination Test (Wepman & Reynolds, 1987) and the auditory discrimination subtest of the Aston Index (Newton & Thomson, 1982). In both of these tests the tester presents two spoken words, for example WEB~WED, LACK~LACK (from Wepman’s Auditory Discrimination Test), LET~NET; BUN~BUN (from the Aston Index) and the child has to say if the two spoken words are the same or different. Both of these tests comprise discrimination of simple Consonant Vowel Consonant (CVC) real words and are a good starting point. A child who is unable to perform age appropriately on these may have a hearing loss (perhaps undiagnosed) and/or poor listening and attention skills or a serious speech input processing difficulty. However, many children can perform well on these tests, but still have auditory difficulties with unfamiliar words, more complex words or words in connected speech. Bridgeman and Snowling (1988) devised an auditory discrimination task to investigate the perception of sound sequences in children, aged 7 to 11 years, with ‘verbal dyspraxia’ compared to younger typically developing reading age-matched controls. There was no difference between the two groups when they were asked to discriminate differences between pairs of words without clusters (e.g. LOSS~LOT); in fact, all were at ceiling with these stimuli. However, the children with speech difficulties performed less well than the reading age control group on a cluster reversal condition (LOST~LOTS), particularly when non-words were used (e.g. VOST~VOTS). The ability to detect such cluster sequence changes is important not only for clear speech production but also for accurate spelling: if a child cannot detect the correct sequence of segments in a word, he or she will not be able to allocate the appropriate letters (graphemes) to each segment.

Auditory Discrimination Tasks

49

A study by Stackhouse (1989) extends the non-word condition of Bridgeman and Snowling’s task, by incorporating more challenging stimuli requiring syllable sequence discrimination, for example IBIKUSIKIBUS. Forty-two children in the age range of 3–8 years were tested. There was a significant main effect of chronological age on the total number of correct responses, indicating that this test was developmentally sensitive across this age range. The ‘different’ responses – that is, the performance on the items that were different from each other (IBIKUS~IKIBUS) rather than the same (WESP~WESP) – were examined separately and also correlated significantly with the children’s chronological ages. Given that children may use orthographic cues to help them with tasks such as this, the children’s reading ages were examined and these also correlated significantly with their percentage of total correct scores. As the ten youngest children had been tested on the Edinburgh Articulation Test (Anthony et al., 1971), it was possible to look at the relationship between performance on this test and articulation age in these children. There was a significant correlation between children’s percentage of total correct responses on this task and articulation age. These findings suggest that typically developing children’s performance is correlated across reading, speech production and auditory discrimination tasks and dissociations between these skills may suggest an atypical pattern of development. This chapter describes tasks that extend the range of stimuli presented to the child to supplement existing auditory discrimination assessments. Included in this range is a preliminary procedure for investigating auditory discrimination of language-specific stimuli. When children are first exposed to language in their environment, they need gradually to build up a repertoire of what are acceptable vs unacceptable segments and phonotactic structures (combinations of segments for the language they are learning). For example, when learning English there is a tacit awareness that words do not begin with the clusters /bn/ or /ts/, or the segment /], as in the Welsh place name LLANDUDNO. This language-specific knowledge is accrued early on in typical development. Children with significant auditory processing problems may be delayed in developing this language-specific awareness: for example, /ts/ vs /st/ may sound the same to them and they may not detect the difference between STAR and TSAR. For other children, it may be that their speech output is interfering with their discrimination abilities: for example, an English-speaking child may say // for /s/ and not be able to detect the difference between these sounds, particularly if they try to say them. Therefore, it may be necessary, with some children, to include stimuli in auditory discrimination tasks that tap legal (i.e. acceptable pronunciations within the language) vs illegal (not acceptable pronunciations) non-words; such a task addresses the question at Level C on the Speech Processing Profile: Does the child have language specific representations of word structures? This procedure can also be useful to investigate the separate language stores of children learning more than one language to establish if

50


they are recognising and storing what is and what is not acceptable appropriately for each of the languages they are learning. A task designed to investigate Level C is, therefore, quite specific to the child in question and needs to be based on the languages they are learning and/or the specific speech errors they are making. We are including some preliminary work on Level C in this chapter as guidelines for users to develop their own procedures for individual children and for different languages. It would not be appropriate to complete all of the tasks and stimuli sets in this chapter on one child. Rather, the task(s) and the stimuli set(s) should be selected to suit the child, taking into account factors such as age, likely severity of any auditory difficulties and which tasks/stimuli would most usefully be compared to tasks tapping other levels of processing, for intra-child comparison. Some of the tasks have both a long and a short form. It is not intended that both versions should be used with every child. Use of the longer version will provide more qualitative detail, in that the child’s responses to a wider range of contrasts will be available. The short form can obviously be completed more quickly, will provide an indication of whether or not this is an area of concern, and may lead to further investigation if appropriate. The reader is referred to Book 1 (Stackhouse & Wells, 1997), Chapter 2, pp. 28–38 for background reading and is encouraged to complete Activity 2.1 on p. 32 in order to understand the rationale for the design and interpretation of auditory discrimination tasks. The following are included in this chapter: • Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words and Non-words (from Bridgeman & Snowling, 1988) • Auditory Discrimination Task 2: Same/Different, Complex Nonwords (from Stackhouse, 1989) • Auditory Discrimination Task 3: ABX, Non-words (from Vance, 1995) • Auditory Discrimination Task 4: Legal vs Illegal Non-words (from Dry, 1997 and Darili, 1994)

Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words and Non-words (from Bridgeman & Snowling, 1988) Aims • To assess a child’s ability to discriminate similar-sounding pairs of words and non-words from auditory presentation only. • To extend already published auditory discrimination tests by including stimuli comprising cluster sequence discrimination in real and non-words.


51

• To allow a comparison to be made between real and non-word performance. Description of the Task The child is asked to listen to a pair of words or non-words spoken by the tester and to decide if both items in each pair are the same or different. Use of the Task This task has been useful in identifying children who may pass auditory discrimination tasks comprising simple CVC words but who have specific auditory discrimination problems with sound sequences (e.g. able to discriminate between LOSS ~ LOT but not between LOST~LOTS). Children with persisting speech difficulties and associated literacy problems have performed less well on this task compared to typically developing children and to children who resolve their speech difficulties by the age of around 5;06. The task has also been useful in identifying more subtle auditory discrimination difficulties in both children and adults with specific language impairment and with dyslexia. The inclusion of non-words allows a comparison to be made between performance on known (real words) and novel (nonword) stimuli. Children who have difficulty in discriminating between non-word stimuli may take longer than their peers to learn new words, for example in the classroom. Psycholinguistic Analysis of the Task The child does not need to access the meaning of the words in order to perform this task and therefore the word and non-word stimuli might be processed in a similar way (see Route 1 in Figure 3.1). This task does not, therefore, necessarily rely on the child’s prior knowledge about the individual items. However, there may be a word advantage if lexical representations of familiar words are accessed and this supports performance (see Route 2 in Figure 3.1). This is why discrimination of real words is located nearer the top of the simple speech processing model, and above discrimination of non-words (see Activity 2.1, pp. 32–50 in Book 1). Generally, typically developing children perform similarly on both real and non-word tasks, suggesting that processing following Route 1 (in Figure 3.1) may be usual for most children. For some children with lower-level auditory discrimination problems, top-down lexical support when discriminating words may compensate for these difficulties to some extent, and processing for word discrimination may follow Route 2. Thus, these children may perform better when discriminating words than non-words. However, children who have stored fuzzy phonological representations of the items presented may be disadvantaged by using

52


Figure 3.1 Non-lexical (Route ) and lexical (Route ) through the speech processing model for auditory discrimination of real and non-word minimal pairs

Route 2 to process the words, and may perform better on non-words than real words. The child who has both lower-level auditory discrimination difficulties and stored fuzzy representations is likely to perform poorly on this task with both real and non-words. This task, therefore, allows intra-child comparison across the two types of stimuli: real and nonwords. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – Yes for real words; No for non-words. • Does the child have to access these lexical representations in order to complete the task? – No.


53

Table 3.1 One example of each stimuli pair used in Auditory Discrimination Task 1: Same/Different (from Bridgeman & Snowling, 1988)

Real words Non-words

CVC (feature contrast)

CVCC (sequence contrast)

15 items, e.g. LOSS~LOT 15 items, e.g. VOSS~VOT

15 items, e.g. LOST~LOTS 15 items, e.g. VOST~VOTS

Level on Speech Processing Profile • Non-word condition: Level B • Real word condition: Level D Stimuli The pairs of words presented are either real or non-word pairs and differ either by a feature change (e.g. /s/ – /t/, as in LOSS/LOT or VOS/VOT) or a sequence change (e.g. /st/ – /ts/, as in LOST/LOTS or VOST/VOTS). There are therefore four sets of stimuli words (see Table 3.1). In the original study, Bridgeman and Snowling (1988) presented 30 pairs of real words and 30 pairs of non-words. Twenty-four pairs were different and six pairs were the same for each real and non-word set. We used a sample of these stimuli in our studies as follows: • Reduced version: For our cross-sectional study, described in Chapter 2, a reduced set of 18 real word and 18 non-word pairs, together with a further 4 non-word pairs to provide practice items, were selected from the original Bridgeman and Snowling stimuli (see Table 3.2).

Table 3.2 Stimuli for reduced version of Auditory Discrimination Task 1: Same/Different (adapted from Bridgeman & Snowling, 1988) used in the crosssectional study

Words feature

Non-words feature

Words cluster sequence

Non-words cluster sequence

plate – place miss – mitt guess – get hiss – hit toss – tot race – rate met – met goat – goat kit – kit

/bleIs/ – /bleIt/ /dIt/ – /dIs/ /kεs/ – /kεt/ /vIs/ – /vIt/ /fɒs/ – /fɒt/ /jeIs/ – / jeIt/ /zεt/ – /zεt/ /pəυt/ – /pəυt/ /bIs/ – /bIs/

placed – plates mitts – missed guessed – gets hissed – hits tossed – tots rates – raced messed – messed goats – goats kissed – kissed

/bleIst/ – /bleIts/ /dIts/ – /dIst/ /kεst/ – /kεts/ /vIst/ – /vIts/ /fɒst/ – /fɒts/ /jeIts/ – /jeIst/ /zεts/ – /zεts/ /pəυts/ – /pəυts/ /bIst/ – /bIst/

54


Table 3.3 Stimuli for shortest form of Auditory Discrimination Task 1: Same/Different (adapted from Bridgeman & Snowling, 1988) used in the longitudinal study

Words feature

Non-words feature

Words cluster sequence

Non-words cluster sequence

plate – place miss – mitt race – rate goat – goat met – met

/bleIs/ – /bleIt/ /dIt/ – /dIs/ /jeIs/ – / jeIt/ /pəυt/ – /pəυt/ /zεt/ – /zεt/

placed – plates mitts – missed rates – raced goats – goats messed – messed

/bleIst/ – /bleIts/ /dIts/ – /dIst/ /jeIts/ – /jeIst/ /pəυts/ – /pəυts/ /zεts/ – /zεts/

• Shortest form: For our longitudinal study (see Chapter 2) an even shorter form of the task was created, which may be more useful for screening children in practice. In this shortest form 10 pairs of real words and 10 pairs of non-words were selected (see Table 3.3). Materials • Score sheets (on the CD-Rom, Appendix C.1 for the reduced version; Appendix C.2 for the shortest form). Procedure The child is asked to say if a pair of stimuli (words or non-words) spoken by the investigator sound the same or different. The investigator’s mouth should be casually obscured to prevent lip reading by the child. In cases where it is suspected that a child is reliant on lip reading, the practice items can be administered with or without lip reading and further made-up items can be given to investigate if lip cueing helps. However, the task items should be presented without lip reading if a comparison is to be made with the normative data provided below. The investigator must ensure that the child understands the concept of ‘same/different’ before starting the task. Where this is problematic, ‘same/not the same’ can be used or any alternative means of signalling a reliable response, such as yes/no or use of symbols. The task is demonstrated using the child’s name: for example, if the child’s name is BEN, the child is asked if ‘BEN/BEN’ are the same or different; then if ‘BEN/OLIVER’ are the same or different – that is, deliberately choosing a name that sounds quite different; then if ‘BEN/PEN’ are the same or different – that is, choosing a name or an object that sounds similar. Intersperse ‘BEN/BEN’ with these practice items. This not only checks the concept of same/different, but also that the child understands the presentation of the task and knows what to expect. Following this introduction the four


55

practice items are administered more formally, during which as much corrective feedback as is needed is given. The aim is to reduce the corrective feedback over the practice items. If having demonstrated the task using the child’s name and then administered the four practice items the child appears to be confused, two further practice items can be added; if the child is still confused the indications would be not to continue with the task, but that in itself provides useful information about the child. The stimuli are presented in blocks of non-words and blocks of real words as indicated on the score sheet. General encouragement can be given during the main part of the task, but no specific help should be given with the non-practice items. One repetition of each item is allowed if the child requests it or if the child was not attending. The child’s response to each item should be written onto the score sheet at the time of the assessment. Scoring The task is scored by number of items correctly identified as same or different out of 36 for the reduced version, and out of 20 for the shortest form. When interpreting results it is important to bear in mind that a child can get over half the items correct by guessing. On the shortest form, for example, a score of 15 or more out of 20 items can be considered above chance (calculated using a binomial table). Any score below this could be due to the effects of chance. Normative Data In the original study by Bridgeman and Snowling (1988), typically developing children were at ceiling on this task by 7 years of age. The typically developing children in our own studies have been at ceiling by 6 to 7 years of age. There was a significant difference between the 5 and 6 year olds on this task, but not between the 6 and 7 year olds. In the cross-sectional study, using the reduced version of the task, we collected data from typically developing children aged 5 to 7 years. This data is provided in Table 3.4 and allows comparison of real versus nonword performance, and between feature and cluster sequence change pairs, as well as interpretation of total scores. There were significant differences between 5- and 6-year-old children on non-word cluster sequence changes, on non-word total scores, on cluster sequence change total scores and on the overall total score. Seven year-old children were at ceiling on all scores. The typically developing children in our longitudinal control sample were at ceiling on the shortest form of the task by 6 years of age. Table 3.5 presents the mean and standard deviation scores for 4 and 5 year olds on the shortest form.

56


Table 3.4 Mean (standard deviation) correct scores for the reduced version of Auditory Discrimination Task 1: Same/Different (data from the cross-sectional study) Age group

Non-word cluster sequence (N 9) Non-word feature (N 9) Real word cluster sequence (N 9) Real word feature (N 9) Real word total (N 18) Non-word total (N 18) Cluster sequence total (N 18) Feature total (N 18) Total score (N 36)

5 yrs (N 20)

6 yrs (N 20)

7 yrs (N 20)

7.1 (1.45)

8.25 (1.21)

8.6 (0.5)

7.5 (1.5)

8.4 (1.47)

9 (0)

7.35 (1.53)

8.0 (1.52)

8.8 (0.49)

7.85 (1.39)

8.35 (1.27)

8.85 (0.49)

15.2 (2.57)

16.35 (2.66)

17.65 (.75)

14.6 (2.46)

16.65 (2.48)

17.6 (.50)

14.45 (2.78)

16.25 (2.44)

17.4 (.60)

15.35 (2.58)

16.75 (2.59)

17.95 (.6)

29.8 (4.81)

33 (4.83)

35.25 (0.79)

Table 3.5 Mean (standard deviation) correct scores for the shortest form of Auditory Discrimination Task 1: Same/Different (data from the longitudinal study) Age group

Total score (N 20)

4 yrs (N 47)

5 yrs (N 47)

15.9 (3.6)

18.11 (2.38)

Questions to Consider • Does the child have difficulties with auditory discrimination generally or only when stimuli comprise clusters? • Is there a difference between real and non-word performance? If so is this: • specific to auditory tasks, or • typical of the child’s performance generally? • How does performance on this input task relate to the child’s cluster production in speech and spelling? Other Same/Different Assessments Tapping Processing at Level D, Auditory Discrimination of Real Words • Aston Index: Auditory Discrimination (Newton & Thomson, 1982) • Test of Auditory Processing Skills (TAPS – 3): Word Discrimination (Martin & Brownell, 2005)


57

• Test of Language Development (TOLD – P:3): Word Discrimination (Newcomer & Hammill, 1997) • Wepman’s Auditory Discrimination Test (Wepman & Reynolds, 1987) Other Same/Different Assessments Tapping Processing at Level B, Auditory Discrimination of Non-words • Auditory Discrimination Task 2: Same/Different – Complex Nonword Discrimination (see below). • Auditory Discrimination Task 3: ABX Non-word Task (see below).

Auditory Discrimination Task 2: Complex Non-words (from Stackhouse, 1989) Aims • To extend the non-word condition of the Auditory Discrimination Task 1 above by incorporating more challenging stimuli, including syllable sequence discrimination. • To observe if a child uses speech for rehearsal on this task and to consider whether this strategy is helpful or not, particularly where speech output problems exist. Description of the Task The child is asked to listen to pairs of spoken non-words and to decide if the non-words in each pair are the same or different. Use of the Task This task should not be used with children under 5 years of age but has been helpful in detecting auditory discrimination problems in older children, who may well be able to process real word stimuli but have difficulties with new and complex material. By observing the child’s strategies for completing this task the investigator can evaluate whether speech output is used to perform the task and, where there are speech output difficulties, whether these interfere with performance on the task. For some children the primary cause of poor performance on this task may, in fact, be speech output difficulties that prevent them rehearsing the items. The material can help identify a range of possible difficulties and is therefore a useful assessment for school-age children with speech, language and literacy difficulties. Children with apraxia of speech and dyslexia have performed less well than typically developing children on this task (Stackhouse & Snowling, 1992). Children who have difficulties with this task may well find dealing with new words in the

58


classroom problematic. They may need more exposure to novel items in order to assimilate them into their lexicons. Psycholinguistic Analysis of the Task This task assesses a child’s ability to deal with novel words. The child has to rely on ‘bottom-up’ skills since all the stimuli are non-words, of which he or she does not have prior knowledge. The route this task takes through the speech processing model is the same as for the non-words in the Auditory Discrimination Task 1 (see Figure 3.1). However, the stimuli are more complex and make more demands on a child’s memory and rehearsal strategies. Performance on this task may, therefore, be dependent on intact articulatory skill for rehearsal of the stimuli for segmentation and matching purposes. Orthographic support may also be an advantage. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – No. • Does the child have to access these lexical representations in order to complete the task? – Not applicable. Level on Speech Processing Profile • Level B Stimuli Forty articulatory complex non-word pairs were devised. The pairs differed in either place of articulation (e.g. SPOBE/SPODE), voicing (e.g. BATE/PATE), cluster sequence (e.g. WESP/WEPS), vowel (e.g. TANT/TINT) or as a result of metathesis (e.g. IBIKUS/IKIBUS). Fourteen pairs of words were the same and twenty-six were different. See Table 3.6 for the list of stimuli. For presentation the list is pseudo-randomised into four parts so that there is a balance of same/different pairs in each quarter. This enables the list to be presented in sections, as appropriate for the attention span of the child. Materials • Score sheet on the CD-Rom (Appendix C.3) Procedure The child is told that he or she will hear some funny words made up by the investigator. He or she is to listen to two words each time and decide


59

Table 3.6 Stimuli for Auditory Discrimination Task 2: Complex Non-word Auditory Discrimination (from Stackhouse, 1989) Stimuli

Nature of contrast Set A

/wεsp/ /wεps/ /snmɒn/ / snmɒn/ /wb/ /jb/ / ləυthas/ / ləυthaf/ /dks/ /dks/ / skkre/ / ktre/ / ətɔ/ /tə ɔ/ / pnsəl/ / pnsəl/ / bkəs/ / kbəs/ / beskət/ / beksət/

Nature of contrast Set B

Cluster (Same) Place Place (Same) Place Metathesis (Same) Metathesis Cluster

Set C /b / /b / / slepə/ / sletə/ / pεpi/ / tεpi/ / ksl/ / ksn/ /tnt/ /tnt/ /tʃps/ /tʃps/ / bskɔts/ / pskɔts/ /dl/ /dl/ /dsk/ /dks/ / tʃkləυt/ / tʃkləυt/

Stimuli

/stεmp/ /stεmp/ /bet/ /pet/ / rεkət/ / rεtək/ /smak/ /smak/ /rliskəυts/ /lriskəυts/ / dr ən/ /dr ən/ / bkt/ / btk/ / krvn/ / krvm/ / rnd / / r / /spəυdə/ /spəυdə/

(Same) Voicing Metathesis (Same) Metathesis (Same) Metathesis Place Place (Same)

Set D (Same) Place Place Place Vowel (Same) Voicing (Same) Cluster (Same)

/sti/ /ski/ /be/ /be/ /spəυb/ /spəυd/ / tr ɑ/ / trðɑ/ /krεb/ /krb/ / bɑ li/ / bɑdli/ /tʃsp/ / tʃps/ /tɒlɒv n/ /tɒlɒv n/ /spəυd/ /spəυd/ /mtbəυk/ /mkbəυt/

Place (Same) Place Place Vowel Place Cluster sequence (Same) (Same) Metathesis

if they have been said the same or not. Younger children may respond ‘Yes’ to same items and ‘No’ to different items. Older children may prefer to say ‘same’ or ‘different/not the same’. As for Auditory Discrimination Task 1 above, the child needs to understand the same/different concept and the requirements of the task before it is administered. The same introductory activity using the child’s name outlined in Auditory Discrimination Task 1 above can be used, before moving on to the prescribed practice items. One repetition of each pair of stimuli is permissible for a comparison with the normative data available. If a child requires more than one repetition of the items, then other areas to follow up might be hearing and attention. All of the items should be presented, for comparison with the normative data, but not necessarily at one sitting. The task can be administered in quarters, but where possible half should be administered in one session and the other half soon afterwards, with all of the items being completed within the same week, if possible.

60


Scoring Record the child’s response on the score sheet. The number of correct responses are calculated for total correct responses for all items, out of 40; number of correct responses to same items, out of 14; number of correct responses to different items, out of 26. Scores are then converted into percentage of items correct to compare to the table of normative data (Table 3.8) by doing the following calculation: Child’s score 100 % correct Total number of items For example, if the task has a total of 40 items and a child scores 30 correct responses, the percentage correct score is (30 / 40) 100 75 %. Where there are errors, the nature of the stimuli can be examined, for example voicing contrast or metathesis, to look for any patterns in the error types. Normative Data Data was collected from 42 typically developing children (20 girls, 22 boys) in the age range of 3;3–8;11 (see Table 3.7 for further details). Their percentages correct for total responses and for same and different items are presented in Table 3.8. Table 3.8 shows that the 3- to 4-yearold children find this a difficult task; many of them guess the response. Thus, this task should only be used for children aged 5 years and over. Questions to Consider • How does performance on this task compare to the non-word condition in Auditory Discrimination Task 1 (Bridgeman & Snowling, 1988) or any other non-word auditory tasks comprising less complex stimuli?

Table 3.7 Details about typically developing children who completed the Auditory Discrimination Task 2: Complex Non-words (from Stackhouse, 1989) Age group 3–4 yrs (N 12) Mean (s.d.) Range 5–6 yrs (N 15) Mean (s.d.) Range 7–8 yrs (N 15) Mean (s.d.) Range

Chronological age

Reading age

Vocabulary age

4;03 (0.5) 3;03–4;10

–

4;05 (1.52) 2;06–7;02

6;02 (0.57) 5;01–6;11

6;10 (0.61) 6;02–8;06

5;05 (1.42) 2;06–7;02

7;11 (0.65) 7;01–8;11

8;05 (1.12) 7;03–11;03

6;08 (1.52) 4;00–8;11


61

Table 3.8 Mean (standard deviation) percentage correct for typically developing children on the Auditory Discrimination Task 2: Complex Non-words (from Stackhouse, 1989) Age group

Total % correct Same % correct Different % correct

3–4 yrs (N 12)

5–6 yrs (N 15)

7–8 yrs (N 15)

61.25 (18.18) 77.77 (18.71) 54.32 (24.16)

82.5 (13.32) 93.33 (9.26) 72.28 (18.83)

90.66 (7.5) 96.92 (4.7) 87.65 (10.96)

• Did the child use speech to rehearse the items before making a decision? If so, did this help? • What qualitative information can be obtained from the auditory discrimination errors? • Is there any relationship between these errors and errors noted in the child’s speech output? Other Same/Different Assessments Tapping Processing at Level B, Auditory Discrimination of Non-words • Auditory Discrimination Task 1: Same-Different, S-Cluster Sequences, Non-words (from Bridgeman & Snowling, 1988) • Auditory Discrimination Task 3: ABX Auditory Discrimination Non-word Task (see below)

Auditory Discrimination Task 3: ABX Non-word Task (from Vance, 1995) Aims • To assess auditory discrimination of similar sounding non-words. • To assess auditory discrimination skills using a task with a higher memory load than the same/different paradigm described above. • To compare speech input performance on this task with speech output performance on the Naming, Word Repetition and Nonword Repetition, using the same or matched stimuli, presented in Chapter 5. Description of the Task This assessment procedure was described by John Locke in 1980. A pair of non-words is presented orally by the investigator (non-word A and non-

62


word B). This is followed by a repetition of one of these non-words: X (i.e. X is the same as either non-word A or non-word B). The child is then required to identify whether X matches A or B. The task can be presented using soft toys, for example toy monkeys, each of which ‘says’ one of the pair (Monkey 1 ‘says’ non-word A, e.g. FESH; Monkey 2 ‘says’ non-word B, e.g. VESH) and the child is asked which monkey said for example FESH. Use of the Task The ABX task is a different paradigm for assessing auditory discrimination of spoken non-words (although words could also be used). Children who do not understand the same/different task, described above, may be able to complete the ABX task, as the conceptual requirements are a little different, and it can be explained to the child without using the words same and different. However, the ABX task has a greater demand on short-term memory skills and this could affect a child’s performance. For example, if a child performs appropriately in the non-word condition of Auditory Discrimination Task 1 (Bridgeman & Snowling, 1988) or on Auditory Discrimination Task 2 (Stackhouse, 1989) but less well on the ABX task, the child may not have auditory discrimination problems per se but rather an auditory memory problem. Comparing performance on the ABX task with performance on the speech production tasks that use the same or matched stimuli may be useful to indicate whether a child’s speech production errors are reflecting an auditory discrimination difficulty with specific sounds. Children with persisting speech difficulties and associated literacy problems have performed less well on this task than typically developing children and than children who resolve their speech difficulties by the age of around 5;06 in our longitudinal study (Nathan et al., 2004a). Psycholinguistic Analysis of the Task This task is similar to the auditory discrimination of non-words in Auditory Discrimination Tasks 1 and 2 above (see Figure 3.1 for route through the model). The child cannot draw on stored knowledge about the task items since all of the stimuli are novel (i.e. non-words). However, it carries more of a short-term memory loading than the other tasks. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – No. • Does the child have to access these lexical representations in order to complete the task? – Not applicable.


63

Level on Speech Processing Profile Level B, but note the increased short-term memory load on this task as compared to the same/different non-word tasks above. It is, therefore, a potentially harder task but still taps the same level of processing, since the answers to the questions above are the same for each of the nonword auditory discrimination tasks. This is an example of how complexity can be manipulated within a level, so that tasks can be developed for older children. Stimuli Full version The 60 non-word stimuli devised for the cross-sectional study tasks (see Table 2.2 in Chapter 2) were used. Further non-word stimuli were developed by changing either place of articulation, manner of articulation or voicing feature of one consonant, or by changing the relative position of two consonants – that is, metathesis – or by reducing a cluster. The 60 items of stimuli are shown in Table 3.9. The stress pattern Table 3.9 Stimuli for the full version of Auditory Discrimination Task 3: ABX (used in the cross-sectional study) Derived from

Stimulus A

Stimulus B

Stimulus C

1

plate

/pəυt/

/pləυt/

/pləυt/

2 3 4 5

sock van roof snake

/sɔk/ /vIn/ /rɔf/ /naIk/

/zɔk/ /zIn/ /rɔs/ /snaIk/

/zɔk/ /vIn/ /rɔs/ /naIk/

6

train

/tɔIn/

/trɔIn/

/tɔIn/

thumb fish watch torch brush mouse cat sponge mouth leaf book duck

/ðɒm/ /vεʃ/ /wυtʃ/ /dutʃ/ /brIs/ /mɔIt/ /tεt/ /spænd / /məυ/ /lɔf/ /mɔk/ /dæk/

/ɒm/ /fεʃ/ /rυtʃ/ /tutʃ/ /brIʃ/ /mɔIs/ /kεt/ /spændz/ /nəυ/ /jɔf/ /bɔk/ /gæk/

/ɒm/ /fεʃ/ /wυtʃ/ /tutʃ/ /brIʃ/ /mɔIt/ /kεt/ /spændz/ /nəυ/ /lɔf/ /bɔk/ /dæk/

Nature of contrast

1 syllable

7 8 9 10 11 12 13 14 15 16 17 18

Cluster reduction Voicing Place Place Cluster reduction Cluster reduction Voicing Voicing Place Voicing Place Manner Place Place Place Manner Manner Place (Continued )

64


Table 3.9 (Continued ) 19 20

chair glove

/tʃi/ /glεb/

/tsi/ /glεv/

/tʃi/ /glεb/

Place Place and manner

parrot spider biscuit toilet dustbin jelly seesaw tractor flower fishing sandwich kitchen feather money ladder guitar scooter yellow sausage slipper

/ prIt/ / speIdI/ / bɒksεt/ / deIlət/ / dæsbn/ / dzlə/ / sɑsi/ / drεktI/ / fluwI/ / ʃɒfIŋ/ / sImwəd / / kɒtʃn/ / fæðI/ / nεmə/ / jεdI/ /tεgɔ/ / stikə/ / læjɔI/ / sεd əs/ / slɔpə/

/ brIt/ / steIbI/ / bɒskεt/ / teIlət/ / bæsdn/ / d lə/ / sɑti/ / trεktI/ / sluwI/ / fɒʃIŋ/ / fImwəd / / gɒtʃn/ / ðæfI/ / mεnə/ / lεdI/ /gεtɔ/ / skitə/ / jælɔI/ / sεsəd / / lɔpə/

/ brIt/ / steIbI/ / bɒskεt/ / teIlət/ / bæsdn/ / d lə/ / sɑti/ / trεktI/ / sluwI/ / fɒʃIŋ/ / sImwəd / / kɒtʃn/ / fæðI/ / mεnə/ / jεdI/ /gεtɔ/ / skitə/ / læjɔI/ / sεd əs/ / slɔpə/

Voicing Place Cluster Voicing Metathesis Place Manner Voicing Place Place Place Voicing Metathesis Metathesis Place Metathesis Metathesis Metathesis Metathesis Cluster reduction

41

butterfly

/ bætəfləυ/

/ bætəfəυ/

/ bætəfləυ/

42 43 44 45 46 47 48

dinosaur hairdresser spaghetti parachute crocodile elephant aeroplane

/ deInIzɑ/ / hɔIdræsI/ /spgItɘ/ / pεrəsit/ / krIdəkaυl/ / æfIlɒnt/ / ɒrəprəυn/

/ deInIsɑ/ / hɔIdrætI/ /gspItɘ/ / pεrəʃit/ / krIkədaυl/ / ælIfɒnt/ / ɒrəpləυn/

/ deInIsɑ/ / hɔIdrætI/ /spgItɘ/ / pεrəʃit/ / krIdəkaυl/ / ælIfɒnt/ / ɒrəprəυn/

49 50 51 52 53 54 55 56 57 58 59 60

pyjamas telephone hamburger roundabout caravan caterpillar computer helicopter hospital umbrella kangaroo television

/ / / / / / / / / / / /

/ pId ɔbIz/ / tɒləfaIn/ / hImbɑdI/ /wəυndəbaIt/ / krɒvIn / / kIpətælə / / kεmpjaυtI/ / hIləkæptə/ / hæsptIl/ / æmblærI/ / gɒŋkIrɑ/ / tυlɒvæ n/

/ / / / / / / / / / / /

Cluster reduction Voicing Manner Metathesis Place Metathesis Metathesis Cluster reduction Manner Voicing Place Place Place Metathesis Voicing Metathesis Metathesis Metathesis Metathesis Metathesis

2 syllable 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 3–4 syllable

pId ɔmIz/ dɒləfaIn/ hImbɑ I/ rəυndəbaIt/ kvɒrIn / kItəpælə / gεmpjaυtI/ hIlətæpkə/ hæstpIl/ æmbrælI/ kɒŋgIrɑ/ tυvɒlæ n/

pId ɔmIz/ dɒləfaIn/ hImbɑdI/ rəυndəbaIt/ krɒvIn / kIpətælə / kεmpjaυtI/ hIlətæpkə/ hæsptIl/ æmblærI/ kɒŋgIrɑ/ tυlɒvæ n/


65

for each non-word is the same as for its matched real word. The 60 items are divided into two lists for presentation. List A contains all onesyllable non-words and 20 of the two-syllable non-words. List B contains 10 of the two-syllable non-words and all three- and four-syllable non-words (see Appendix C.4); these lists can be presented in different assessment sessions.

Short form For the longitudinal study 24 pairs of non-word stimuli were selected from the full version of the task (see Table 3.10). A further reduced version comprising half of these non-words can also be used; these items are indicated by an asterisk in Table 3.10, and on the score sheet in Appendix C.5.

Materials • Score sheets on the CD-Rom Appendix C.5 for short form). • Not supplied: two soft toys.

(Appendix C.4 for full version;

Procedure Two soft toys are placed in front of the child (e.g. two different monkeys) and practice items are presented by pointing to one toy and saying ‘He says /səυf/’, pointing to the second toy and saying ‘He says / təυf /’ and then asking ‘Who said /səυf /?’. The child is encouraged to point to one of the toys. Corrective feedback is given as required for the practice items. During the practice items the cue phrases are phased out, so that just the non-word stimuli are spoken as each toy is pointed to, for example /grIb/ ~ /sIb/, and then the child is asked ‘Who said X?’. Finally all the cue phrases can be omitted, so that the X stimulus is also presented as a single non-word /spəυd/ /pəυd/; /spəυd/. The task items are presented with general encouragement as required but no feedback. The investigator’s mouth should be casually covered to prevent lip reading by the child. As in auditory tasks 1 and 2, the effect of plus/minus lip reading could be assessed, but if lip reading is allowed then the normative data provided will not apply. If some, or all, of the single-word speech production tasks that use the same or matched stimuli (naming, word repetition, and non-word repetition, described in Chapter 5) are also going to be administered to the child, then the order of presentation of tasks and stimuli lists should be as shown in Table 2.3 in Chapter 2.

66


Table 3.10 Stimuli for the short form of the Auditory Discrimination Task 3: ABX (used in the longitudinal study)

Derived from Stimulus A

Stimulus B

Stimulus X

Nature of contrast

1 syllable 1

snake

/naIk/

/snaIk/

/naIk/

2* 3

van plate

/vIn/ /pəυt/

/zIn/ /pləυt/

/vIn/ /pləυt/

4* 5*

fish train

/vεʃ/ /tɔIn/

/fεʃ/ /trɔIn/

/fεʃ/ /tɔIn/

6 7 8*

brush sponge glove

/brIs/ /spænd / /glεb/

/brIʃ/ /spændz/ /glεv/

/brIʃ/ /spændz/ /glεb/

9 10 11

guitar sausage slipper

/tεgɔ/ / sεd əs/ / slɔpə/

/gεtɔ/ / sεsəd / / lɔpə/

/gεtɔ/ / sεd əs/ / slɔpə/

12* 13* 14 15* 16*

yellow tractor jelly spider flower

/ / / / /

/ / / / /

/ / / / /

17 18 19

elephant umbrella butterfly

/ æfIlɒnt/ / æmbrælI/ / bætəfləυ/

/ ælIfɒnt/ / æmblærI/ / bætəfəυ/

/ ælIfɒnt/ / æmblærI/ / bætəfləυ/

20* 21* 22* 23 24*

parachute aeroplane caterpillar pyjamas spaghetti

/ pεrəsit/ / ɒrəprəυn/ / kItəpælə / / pId ɔmIz/ /spgItə/

/ pεrəʃit/ / ɒrəpləυn/ / kIpətælə / / pId ɔbIz/ /gspItə/

/ pεrəʃit/ / ɒrəprəυn/ / kIpətælə / / pId ɔmIz/ /spgItə/

Cluster reduction Place Cluster reduction Voicing Cluster reduction Place Place Manner

2 syllable

læjɔI/ drεktI/ dzlə/ speIdI/ fluwI/

jælɔI/ trεktI/ d lə/ steIbI/ sluwI/

læjɔI/ trεktI/ dzlə/ steIbI/ sluwI/

Metathesis Metathesis Cluster reduction Metathesis Voicing Place Metathesis Place

3–4 syllable Metathesis Metathesis Cluster reduction Place Place Metathesis Manner Metathesis

Note: * Items can be selected for a reduced version of this task.

Scoring The child’s responses are noted on the score sheet. When interpreting results it is important to bear in mind that a child can score more than 50 % correct by guessing. A score of 37/60 on the long version, 17/24 on the short version or 10/12 on the reduced version can be considered above chance; that is, any score below these scores could be due to the effects of chance and may reflect a marked auditory discrimination deficit, or a lack of understanding of the task, or short-term memory overload.


67

Table 3.11 The mean (standard deviation) correct scores on the full version of Auditory Discrimination Task 3: ABX task (data from cross-sectional study) Age groups

Mean (s.d.) (N 60)

5 yrs (N 20)

6 yrs (N 20)

7 yrs (N 20)

40.25 (9.16)

45.65 (7.86)

52.30 (4.65)

Table 3.12 The mean (standard deviation) correct scores on the short form (CA5 and 6) and the reduced version (CA4) of Auditory Discrimination Task 3: ABX task (data from longitudinal study) Age group

Mean (s.d.) (N 24)

4 yrs (N 47)

5 yrs (N 47)

6 yrs (N 44)

*7.87 (20.1)

17.93 (3.5)

20.45 (2.46)

Note: *4 year olds completed the reduced version with just 12 items.

Normative Data Table 3.11 summarises the performance of typically developing 5–7year-old children on the full version of the ABX task. Table 3.12 summarises the performance of typically developing 4–6-year-old children on the short form of the ABX task. Table 3.12 shows that this ABX task is not suitable for children under the age of 5 years. Children aged 4 years were performing at a chance level, as was the case in Auditory Discrimination Task 2 (above), and generally found the more complex auditory discrimination tasks too difficult. Questions to Consider • Is there a difference in performance on this task (which adds a memory load) to other non-word auditory discrimination tasks? • What qualitative information is provided by the errors made? • Is there a relationship between these errors and errors noted in the child’s speech output? Other Tasks • Speech Input Processing in Children (SIPc) (Vance, Rosen & Coleman, in preparation) includes a computer-presented XAB non-word task (i.e. the same as ABX, but stimuli presented in a different order).

68


Auditory Discrimination Task 4: Legal vs Illegal Words (from Dry, 1997 and Darili, 1994) Aims • To investigate the child’s awareness of the segments and phonotactic structure of the language(s) being learned. • To facilitate development of this task across other languages.

Description of the Task This task can take different forms depending on the age and the needs of the child. A general procedure is that the child hears a series of nonwords, some of which accord with English phonology and phonotactic structures and some of which are illegal or exotic, containing nonEnglish clusters or segments. The child indicates in some way whether the word was acceptable or not, for example was it said by an Englishspeaking puppet or by a Non-English speaking puppet. There are no pictures for this task.

Use of the Task Children need to be able to recognise the phonological and phonotactic structure of the language they are learning, and to distinguish between legal (e.g. /pləυt/) and illegal non-words (e.g. /pnəυt/). They also need to distinguish between sounds that are phonemes in the language they are learning and sounds that are not. Some children will use non-native sounds as substitutions, for example in English using // for /s/. Most children, including most of those with speech disorders, only use the phonology and phonotactic structures of their own language. This task is, therefore, only likely to be useful for children who have severe speech input processing difficulties, who may not have developed sufficient knowledge of the language, and for children who use non-native language sounds in their own speech production. Ideally, stimuli should include the errors that the child makes and investigators are encouraged to devise stimuli for individual children’s needs. This task can also be used with children who are bi- or multilingual to investigate the child’s language-specific knowledge for two or more languages. Appropriate stimuli need to be selected: it may be that these children will process the speech input in a different way from monolingual speakers. A third use of the task is for the investigation of how and when young, typically developing children learn language-specific information.


69

Psycholinguistic Analysis of the Task This task targets the question at Level C of the speech processing profile: Does the child have language-specific representations of word structures? Within the model the level of speech input processing is described as ‘phonological recognition’. This is the point at which speech input is recognised as being in a familiar language and sent on for further decoding, while unfamiliar speech is not processed further. When a child hears a word in an unfamiliar language, or is presented with one in a madeup task, he or she is assumed to match the input against an inventory of familiar phonotactic patterns. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – No, but he or she will have representations that are more similar to legal non-words than to illegal non-words, which may make a difference in his or her responses. • Does the child have to access these lexical representations in order to complete the task? – No. Level on Speech Processing Profile • Level C Stimuli The following two examples of stimuli are presented to facilitate users developing their own stimuli for specific children. Example 1 (from Darili, 1994) The stimuli consist of 40 non-words, 20 that would be legal words in English, 10 that contain non-English consonant clusters and 10 that contain non-English segments. The non-words were derived from 20 English words by changing one or two features in the initial consonant or consonant cluster (see Table 3.13). Example 2 (from Dry, 1997) Eight Greek words with non-English phonotactic structures and eight Greek words with non-English phonemes were selected. A phonetically matched non-word that accorded to English phonotactic and phonological structure was created for each Greek word (e.g. Greek word /cefI/, legal non-word /tefI/). These are given in Table 3.14.

70

Compendium of Auditory and Speech Tasks Table 3.13 Stimuli used in the Auditory Discrimination Task 4: Legal versus Illegal Non-words (from Darili, 1994) Words

Legal non-words

Illegal non-words

duck knife book sock mouth soap juice house train plate bread snake school fruit star glove brush dress

/ðk/ /baIf/ /zuk/ /pɒk/ /gaυ/ /fəυp/ /rus/ /aυs/ /reIn/ /fleIt/ /gred/ /sleIk/ /snul/ /prut /snɑ/ /grv/ /grʃ/ /res/

/k/ /xaIf / /ŋuk/ /t’ɒk/ /k’aυ/ /cəυp / /γus/ / aυs/ /reIn/ /xleIt/ /zred/ /bneIk/ /sul/ /xrut /sfɑ/ /gmv/ /bnʃ/ /zres/

Materials • An example of a score sheet is given on the CD-Rom (Appendix C.6). Procedure An example of a procedure is as follows. Two puppets are introduced to the child, one called Tommy (a boy) and one called Kivos /xvɒs/ (a robot). ‘These are my two puppets that can talk but they talk differently. Listen to them talk. Tommy is talking like all the boys and girls, he’s saying: Hello little boy/girl. My name is Tommy and I am very happy that you are going to play this game with us. Listen to Kivos now: (produce a non-word string that uses nonEnglish sounds and clusters). They talk differently don’t they? I’m going to say some words, some you will know and some of which you don’t know, and you point to who you think said the word. Was it Tommy or was it Kivos?’ Other suggestions are to use a boy/girl toy and an alien toy, or to introduce two puppets, one as an ‘English-speaking’ boy/girl and one as a ‘non-English-speaking’ boy/girl. Scoring Percentages of correct responses to words, to legal non-words and to illegal non-words with non-English segments or non-English clusters can be calculated.


71

Table 3.14 Stimuli used in the Auditory Discrimination Task 4: Legal vs illegal non-English words (from Dry, 1997). Greek words (non-English segment)

Greek words (non-English cluster)

Legal non-words (English)

/ftinəυs/ /fəυnəυs/ /ptIsI/ /ðɹəυməυs/ /vgenəυ/ /htIzəυ/ /psIɹa/ /ksæɹəυ/

/tIma/ /nata / /ena/ /zati/ /tefi/ /seɹI/ /zeɹəυs/ /fIna/ /flinəυs/ /fɹəυnəυs/ /plISI/ /ɹəυməυs/ /skenəυ/ /stIzəυ/ /plIɹa/ /klæɹəυ/

/cma/ / ata/ /e a/ /ati/ /cefi/ /çeɹI/ /eɹəυs/ /cIna/

Normative Data Typically developing children recognise illegal segments and phonotactic structures by about 4 years of age. Questions to Ask • How does performance compare with other auditory discrimination tasks in this chapter that only include legal stimuli? • Is the child having more difficulty in recognising non-English (or other language) segments than in recognising non-English (or other language) structures? • Is the child having difficulty learning a specific language?

Summary • This chapter presents four auditory discrimination tasks. Psycholinguistic analysis shows that each one is different either in the level of processing it taps (depending on whether stimuli are words or non-words) or in the complexity of the stimuli presented within a level, or in the task demands generally. • When real words are presented in Task 1, the same/different task from Bridgeman and Snowling (1988), the task is tapping Level D on the speech processing profile.

72

Compendium of Auditory and Speech Tasks • When non-word stimuli are presented in Tasks 1, 2 and 3, the task is tapping Level B on the speech processing profile. • Contrasting legal and illegal stimuli taps level C on the profile. • Intra-child comparison can be made for word versus non-word stimuli. • Within Level B, the tasks differ in the demands made on the child. The non-words used in Task 2 are more complex than those in Task 1, and the memory demands in Task 3 are higher than for the other two tasks. • The more complex tasks at level B (Task 2 and Task 3) are not suitable for children under 5 years of age, since typically developing 4 year olds do not perform above chance level on these tasks. • By presenting more challenging stimuli for discrimination, children are more likely to repeat the items (sub-vocally or out loud) in order to hold on to them while a same/different decision is made (e.g. in Tasks 2 and 3). By observing this behaviour it becomes apparent if a child does this automatically and, if so, whether it helps or interferes with their decision making. • Qualitative analysis of a child’s errors on auditory discrimination tasks can be compared with his or her speech production errors to establish if there is a direct relationship between the two.

Chapter Four Auditory Lexical Discrimination Tasks

The tasks presented in the previous chapter all used verbally presented stimuli. This chapter introduces auditory discrimination tasks that include picture stimuli as well. An auditory discrimination task using pictures typically involves a child looking at pictures of two similarsounding words, such as PEAR and BEAR, and pointing to the word spoken by the tester. This type of task is often called a minimal pair picture task. To complete this task the child has to access his or her own lexical representations, first to access the name of the pictures involved, and second to match what he or she has heard to his or her phonological representation for that word. Thus, auditory discrimination tasks with pictures tap a higher level on the left-hand side of the speech processing profile than auditory discrimination tasks without pictures, such as those presented in Chapter 3. The Auditory Discrimination and Attention Test by Rosemary Morgan Barry (1988), for children in the age range of 3;06 to 12 years, is a good example of a discrimination test involving picture stimuli. Having checked that the items are in the child’s vocabulary, the tester presents the child with pairs of pictures covering a range of different contrasts, including clusters, for example COAT~GOAT, SUM~SUN, CROWN~CLOWN. The tester names each item in the pair in a random order on six occasions (fewer at the lower end of the age range). The child is asked to identify which item has been named by posting a counter underneath the appropriate picture. An alternative method of testing phonological representations of lexical items is to present a familiar word alongside similar-sounding non-words to see if a child can detect the non-words or if he or she accepts them as the matched real word. For example, a picture of a SOCK is shown, and the child is asked if the following productions are right or

73

74


wrong: SOCK; TOCK; SOT (see Locke, 1980b for further discussion of this task). This presentation allows a wider range of items to be tested since only one known word and picture pair is required. It is also a useful way of confronting children with their own productions of a word to see if they are aware or not of their own mispronunciations (see Task 3 in this chapter). Accepting similar-sounding non-words as the real word implies that the child’s phonological representation of that word is ‘fuzzy’ or inaccurate. However, it is important to use challenging stimuli when investigating this area, particularly with older children, as familiar words may be well learned and practised (German, 1998a). For example, children with speech difficulties have found sound sequence changes, such as LOST~LOTS, more difficult to discriminate than feature changes, such as LOT~LOSS, when compared to normal controls (Bridgeman & Snowling 1988; Stackhouse & Snowling, 1992; and see Chapter 3, this volume). It is, therefore, important to incorporate sequential changes in test stimuli particularly if working with school-age children. Constable, Stackhouse and Wells (1997) did this in their investigations of word-finding difficulties in children. They followed the procedure outlined above using multisyllabic stimuli, such as MICROPHONE. They demonstrated that children’s word-finding difficulties may be specifically related to the fuzziness of their phonological representations and not necessarily to problems within the semantic representations. One of the cases presented was of a 7-year-old boy with word-finding difficulties who, for example, when looking at a picture of a HOSPITAL, accepted the incorrect production ‘hostipal’ (/ hɒstpəl/) as correct. The items on which he did this were also words he had difficulty accessing in his speech production. This suggests that there is a direct relationship between the precision of phonological representations and word-finding skill. The procedures described in this chapter are sometimes known as auditory lexical discrimination tasks, because they check not only if the child can detect differences between similar-sounding words at an auditory level but also how those words have been stored in the child’s lexical representations. In effect, the child is accepting what he or she believes to be real words and rejecting what he or she considers to be nonsense words. This can be done with or without pictures. It would not be appropriate to complete all of the tasks and stimuli sets presented in this chapter with each child. Rather the task(s) and the stimuli set(s) should be selected to suit the child, taking into account factors such as age, likely severity of any auditory difficulty, and which tasks/stimuli would be most usefully compared to tasks tapping other levels of processing, for intra-child comparison. For example, one of the tasks has a longer version and a short form, but it is not intended that both versions should be used with every child. Use of the longer version will provide more qualitative detail, in that the child’s responses to a wider range of contrasts will be available. The short form can obviously

Auditory Lexical Discrimination Tasks

75

be completed more quickly and will provide an indication of whether or not further investigation is appropriate. The reader is referred to Book 1 (Stackhouse & Wells, 1997), Chapter 2, pp. 28–38 for background reading and is encouraged to complete Activity 2.1 on p. 32 of that book in order to understand the rationale for the design and interpretation of the auditory lexical discrimination tasks. The following are included in this chapter: • Auditory Lexical Discrimination Task 1: Mispronunciation Detection (from Vance, 1995) • Auditory Lexical Discrimination Task 2: With Pictures and Without Pictures (from Constable, Stackhouse & Wells, 1997) • Auditory Lexical Discrimination Task 3: Mispronunciation Detection for Individual Children (after Locke, 1980b) • Auditory Lexical Discrimination Task 4: Words in Sentences (from Cassidy, 1994)

Auditory Lexical Discrimination Task 1: Mispronunciation Detection (from Vance, 1995) Aims • To investigate the precision of phonological representations without a verbal response being required. • To compare performance on this input task with speech output performance on tasks where the same stimuli are used (see naming, word repetition and non-word repetition tasks in Chapter 5). Description of the Task The procedure used in this task was described by John Locke in 1980. The child is asked to look at a picture and decide if a spoken word is the correct name of that picture (e.g. a picture of a PLATE is presented and the child hears PATE or PLATE). Use of the Task Current theories of both speech and reading difficulties postulate that children with unclear speech, who may have associated literacy problems, have under-specified phonological representations, at least for some items. This may cause problems when creating a motor programme for speech output or when mapping phonological information onto orthographic representations (see Snowling, 2006 for a review of language skills and learning to read, and also Book 1 in this series, Stackhouse &

76


Wells, 1997). Identifying the clarity of a child’s representations as early as possible (i.e. before a child is failing to learn to read) is essential for appropriate intervention and possible prevention of persisting speech difficulties and later literacy problems. As this task taps the precision of a child’s phonological representations by requiring only a simple yes/no response, it is well suited to children with speech output difficulties. Psycholinguistic Analysis of the Task Unlike the auditory discrimination tasks presented in Chapter 3, this mispronunciation detection task requires the child to draw on his or her own knowledge about a word. Detecting whether the name of a picture is produced correctly or not can only be done by comparison with a stored representation of that word; that is, semantic representations to access the name of the picture and phonological representations to reflect on how the name of the picture ‘sounds’. The task therefore comprises lexical decision (is it a known word or not?) as well as auditory discrimination and it assesses the precision of a child’s phonological representations. See Figure 4.1 for a comparison of the routes taken by auditory discrimination and auditory lexical discrimination tasks. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – Yes. Level on Speech Processing Profile • Level E Stimuli Full version The list of 60 words devised for the cross-sectional study comprises the full version of this task. A mispronunciation was created for each word by changing one of the consonants to one with similar features, or by changing the order of two consonants (see Table 4.1). For each picture, the child either hears two stimuli in turn, the word and the matched non-word, or three stimuli, with either the word or the non-word repeated. This is to ensure that a child does not get into a pattern of expecting one correct and one incorrect stimulus for each picture, as this might affect his or her responses. The response to the repeated stimulus is not scored. The order of presentation of the two or three stimuli is varied.


77

Figure 4.1 Routes through the speech processing model for auditory discrimination and auditory lexical discrimination tasks.

Short form As for the Auditory Discrimination Task 3: ABX (see Chapter 3), 24 words and their matched non-words were selected to provide a shortened form for the longitudinal study (see Table 4.2). Materials • Pictures and Score sheets on CD-Rom. (Appendix D.1 for full version and Appendix D.2 for short form). • Not supplied: soft toy.

78


Table 4.1 Stimuli for the full version of Auditory Lexical Discrimination Task 1: Mispronunciation Detection (used in the cross-sectional study) 1-syllable words

2-syllable words

3-syllable words

Real words

Non-words Real words

Non-words

Real words

Non-words

brush sponge glove duck leaf sock cat book torch mouse knife snake train van watch plate roof fish chair thumb

/brs/ /spndz/ /glb/ / k/ /jif/ /zɒk/ /tt/ /mυk/ /dɔtʃ/ /maυt/ /maf/ /nek/ /ten / /zn/ /rɒtʃ/ /pet/ /rus/ /vʃ/ /tseə/ /ðm/

/ fmwd/ / dɔlət/ / nm/ / ðεfə/ / lεləυ/ / gtʃn/ • / jdə/ / slaυwə/ / bsdn/ / dzεl/ / drktə/ / ʃfŋ/ / bkst/ / stukə/ / brət/ / stɔ/ / lpə/ / sɒds/ /t gɑ/ / stapə/

caterpillar spaghetti elephant caravan crocodile umbrella helicopter kangaroo television hospital telephone parachute butterfly computer roundabout hairdresser aeroplane pyjamas hamburger dinosaur

/ kpətlə/ / gəspεt/ / εflənt/ / kvərn/ / krɒdəkal/ / mblεrə/ / hεltɒpkə/ / ŋkəru/ / tεvəln / / hɒstpl/ / dεlfəυn/ / prəsut/ / btəfa/ / gəmpjutə/ / waυndəbaυt/ / hεədrεtə/ / εərəpren/ /pədɑbəz/ / hmbdə/ / danəzɔ/

sandwich toilet money feather yellow kitchen ladder flower dustbin jelly tractor fishing biscuit scooter parrot seesaw slipper sausage guitar spider

Procedure The child should be asked to name each of the pictures first. This provides a vocabulary check and allows the child’s speech production to be recorded as for a naming task (see Chapter 5) and compared with his or her performance on this auditory task. Table 4.2 Stimuli for the short form of Auditory Lexical Discrimination Task 1: Mispronunciation Detection (used in the longitudinal study) 1-syllable words

2-syllable words

Word

Non-word

Word

brush sponge glove snake train van plate fish

/brs/ /spndz/ /glb/ /nek/ /ten / /zn/ /pet/ /vʃ/

yellow flower jelly tractor slipper sausage guitar spider

Non-word / lεləυ/ / slaυwə/ / dzεl/ / drktə/ / lpə/ / sɒds/ /t gɑ/ / stapə/

3-syllable words Word caterpillar spaghetti elephant umbrella pyjamas parachute butterfly aeroplane

Non-word / kpətlə/ / gəspεt/ / εflənt/ / mblεrə/ /pədɑbəz/ / prəsut/ / btəfa/ / εərəpren/


79

The child is asked to look at each picture in turn and to decide if the name of the picture has been said correctly or not. A soft toy, for example a monkey, can be used to explain the task. The child is told that the monkey is going to say some words, sometimes he is ‘clever’ and sometimes ‘silly’, or sometimes the monkey says the words ‘right’ and sometimes ‘wrong’ (the investigator can decide what best suits the child). There are two practice items during which corrective feedback can be given. Only general encouragement is given during the main part of the task. One repetition of an item is allowed if the child fails to respond to a stimulus or he or she requests a repetition. If some, or all, of the single-word speech production tasks that use the same or matched stimuli (naming, word repetition and non-word repetition, described in Chapter 5) are also going to be administered to the child, then the order of presentation of tasks and stimuli lists should be as shown in Table 2.3 in Chapter 2.

Scoring The task is scored by calculating the total number of correct responses out of 120 in the full version and 48 in the short form (i.e. correctly accepting words and correctly rejecting non-words). For each picture, the child’s response to one correct and one incorrect production is scored. Where a third, correct or incorrect pronunciation is given this is not scored (starred (*) items on the score sheet). When interpreting results it is important to bear in mind that a child can score more than 50% correct by guessing. A score of at least 70/120 or 31/48 can be considered above chance.

Normative Data Table 4.3 summarises the performance of typically developing children on the full version of the task. There was more variability in the performance of 3 and 4 year olds on this task than in the other age groups. There were no significant differences between the total scores of 3 and 4 year olds, 4 and 5 year olds, 5 and 6 year olds and 6 and 7 year olds. However, there is a significant difference between the scores of 3 and 5 year olds and 5 and 7 year olds, suggesting that the tasks are developmentally sensitive (Vance, 2001, pp. 114–115). Means and standard deviations are also given for scores at different word lengths. On the short form of the task, the 6 year olds performed at ceiling. Table 4.4 summarises the mean score and standard deviation of the 4 and 5 year olds.

80


Table 4.3 Mean (standard deviation) correct scores for the full version of Auditory Lexical Discrimination Task 1: Mispronunciation Detection (data taken from the cross-sectional study) Age group

1-syllable words N 40 Mean (s.d) 2-syllable words N 40 Mean (s.d) 3–4-syllable words N 40 Mean (s.d) Total words N 120 Mean (s.d.)

3 years (N 17)

4 years (N 18)

5 years (N 20)

6 years (N 20)

7 years (N 20)

34.76 (4.45)

37.63 (2.21)

37.05 (1.75)

37.05 (1.83)

38.2 (1.63)

34.18 (5.25)

35.74 (5.86)

38.05 (1.11)

38.6 (1.56)

38.8 (1.12)

30.94 (5.98)

33.89 (5.87)

36.4 (2.27)

37.8 (1.94)

37.65 (1.52)

113.5 (4.78)

114.7 (3.17)

99.9 (14.94)

108.5 (10.3)

111.5 (4.1)

Questions to Ask • How does the child’s performance compare over 1-, 2- and 3- or 4-syllable words? Does increasing the syllable length of the stimuli present problems for the child? • How does performance on this task compare with that on auditory discrimination tasks presented in Chapter 3? • How does performance on this task compare with naming, word and non-word repetition tasks in Chapters 5, where the same stimuli are used? Other Assessments Tapping Processing at Level E, Accuracy of Phonological Representations • Auditory Discrimination and Attention Test (Morgan Barry, 1988) • Goldman-Fristoe-Woodcock Auditory Skills Battery (Goldman, Fristoe & Woodcock, 1975)

Table 4.4 Mean (standard deviation) correct scores on the short form of Auditory Lexical Discrimination Task 1: Mispronunciation Detection (data taken from the longitudinal study) Age group Total scores N 48 Mean (s.d.)

4 years

5 years

(N 47)

(N 47)

42.94 (4.34)

45.59 (3.83)


81

• Speech Assessment and Interactive Learning System (SAILS). A computer-presented mispronunciation detection task: www.avaaz.com • Quality of Phonological Representations task (Claessen et al., submitted) • Speech Input Processing in Children (SIPc) (Vance, Rosen & Coleman, in preparation) also includes a computer-presented mispronunciation detection task.

Auditory Lexical Discrimination Task 2: With and Without Pictures (from Constable, Stackhouse & Wells, 1997) Aims • To investigate the precision of phonological representations without a verbal response being required. • To allow direct comparison between auditory lexical discrimination tasks with and without pictures. Description of the Tasks The first task requires the child to listen to real and non-words. After each word the child says ‘Yes’ if he or she has heard a real word and ‘No’ if he or she has heard a nonsense or ‘made-up’ word. The second task follows the same procedure as in Task 1 above. The child is asked to look at a picture and decide if a spoken word is the correct name of that picture. Both of these tasks, first without pictures and then with pictures, use the same stimuli. The words selected are more complex than in Auditory Lexical Discrimination Task 1: Mispronunciation Detection. For example, for MICROPHONE the child hears the stimuli: • • • •

MICROPHONE MICROKONE MIFROKONE TELEPHONE

/ makrəkəυn/ / mafrəkəυn/ (a real word distractor)

Use of the Task If the auditory lexical discrimination task is administered with and without pictures, a direct comparison can be made between an auditory-only presentation and an auditory plus picture presentation. Constable, Stackhouse and Wells (1997) suggest that this comparison allows three possible outcomes in performance to be considered, enabling identification of the precise location of a child’s difficulty: • Performance on the picture task is better than on the auditory-only task. This suggests that the visual/semantic information facilitates

82

Compendium of Auditory and Speech Tasks access to the phonological representation, which results in rejection of the non-words. The implication is that, when pictures are not present, incorrect responses are the result of poor access to an accurate (and not fuzzy) representation. • Performance is poor, but the same, on both tasks. This suggests that the stored phonological representation is not accurate. • Performance on the picture task is worse than on the auditory-only task. The presence of the visual information may have encouraged the child to rely on top-down processing, rather than attending to the auditory information, and so non-words are accepted more readily.

Psycholinguistic Analysis of the Task Whether the spoken stimuli are presented with or without pictures, this task requires the child to draw on his or her lexical knowledge. Detecting whether a spoken word is a known word or is the name of a picture produced correctly requires the child to compare his or her own stored phonological representations of that word with the stimulus. The task therefore assesses the precision of a child’s phonological representations. Speech processing in this task follows the same route through the model as Task 1 (see Figure 4.1), but this task includes more complex stimuli. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – Yes. Level on Speech Processing Profile • Level E Stimuli These stimuli were originally selected from the word-finding errors of two individual children. They may be useful for assessing skills with multi-syllabic words, and similar stimuli sets can be created to examine individual children’s errors. There are 10 target lexical items. Two closely matched non-words were derived from each of these items (i.e. 20 nonwords in total). To control for possible response bias, a further 10 real words similar in phonological structure to the targets are included. The non-words are of two types in which the sound sequence of each real word has been systematically altered as follows:


83

• Non-word Type A is formed by altering one phoneme at the beginning of the third or final syllable to imitate a perseverative error, e.g. ESCALATOR → ESCALACOR (/ eskəlekə/). • Non-word Type B is formed by transposing two phonemes, e.g. ESCALATOR → ESTALACOR (/ / εstəlekə/). The 40 words are presented in a randomised order. See Table 4.5 for a complete list of stimuli. Materials • Pictures and score sheets on CD-Rom (Appendix D.3-without pictures; D4-with pictures). • Not supplied: soft toy. Procedure Auditory Lexical Discrimination Task (without pictures) This task is given first to prevent any priming by pictures. Prior to presenting the items the child is given instructions along the following lines: ‘I’m going to say some words, some of the words you will know, but some you won’t: they will sound strange or silly. When you hear a word that you think is said right, you can say “yes”, but if you don’t know the word, you can say “no”. OK, let’s have a practice. What if you heard the word “butterfly”?…….. / desənɔ/? ………… / sεnttid/?’ Auditory Lexical Discrimination Task (with pictures) The 10 picture cards should first be presented in a randomised order for the child to name. Responses should be audio-recorded for checking later Table 4.5 Stimuli used on the Auditory Lexical Discrimination Task: with and without pictures (from Constable, Stackhouse & Wells, 1997) Target Items

Paired non-word type A

hospital elephant crocodile microphone octopus escalator binoculars helicopter television caterpillar

/ / / / / / / / / /

hɒsppl/ εllənt/ krɒkəkal/ makrəkəυn/ ɒʔtətəs/ εskəlekə/ bnɒkjunəz/ hεlkɒpkə/ tεlln/ • ktətlə/

Paired non-word type A

Real word distractor

/ / / / / / / / / /

eskimo president porcupine telephone octagon alligator rhinoceros radiator competition calculator

hɒstpl• / εflənt/ krɒdəkal/ mafrəkəυn/ ɒʔpətəs/ εstəlekə/ blɒkjunəz/ hεltɒpkə/ tεvln• / kpətlə/

84


and transcribed in phonetic script at the time. Using four copies of each picture, the 40 pictures are assembled in the order given on the score sheet and then turned over one at a time, as the child hears one of the word or non-word stimuli. The following instructions are given: ‘Now you are going to hear some more words. I will show you a picture and say a word. I want you to tell me if you think the word I say goes with the picture, and you can say “Yes”. If you think I say it wrong, or it sounds silly, then you can say “No”.’ For a younger child a soft toy could be introduced to ‘say’ the words, as in Task 1. Scoring For both tasks (with and without pictures), scores can be calculated for each type of stimuli for comparison with the normative data. Percentages should be calculated for correct responses to real words (accepting words), to non-words Type A (rejecting non-words) and to non-words Type B (rejecting non-words). Responses to the distractor word stimuli are not scored (and are starred (∗) items on the score sheet). When interpreting results it is important to bear in mind that a child can score more than 50% correct by guessing. A percentage score of 73% correct can be considered above chance on a total of 30 items. Normative Data The study showed that 7 year olds were between 98 and 100% successful on this task (Constable, Stackhouse & Wells, 1997). In fact children are generally at ceiling on this task by 6 years of age (see Table 4.6) and there was no significant difference in their responses to words and nonwords. There was also no significant difference in performance between the two sets of non-words. Questions to Ask • How does performance compare on the task with and without pictures? • Is there a difference in the number of errors in items from nonword Type A and non-word Type B? • Is there a relationship between errors on this task and the child’s ability to spontaneously name the picture, in terms of accuracy of production or any word-finding difficulty? Other Assessments Tapping Processing at Level E, Accuracy of Phonological Representations As listed for Task 1 above.


85

Table 4.6 Mean (standard deviation) percentage correct scores of typically developing 6-year-old children (N 12) on the Auditory Lexical Discrimination Task 2: With and Without Pictures (data from Constable, Stackhouse & Wells, 1997).

Without pictures % correct mean (s.d.) With pictures % correct mean (s.d.)

Non-words Type A

Non-words Type B

Real words

91.67 (8.35)

92.5 (9.65)

95 (5.22)

89.17 (9.0)

91.67 (9.37)

96.67 (6.51)

Auditory Lexical Discrimination Task 3: Mispronunciation Detection for Individual Children (from Locke, 1980b) Aim To investigate the relationship between a child’s output difficulties and the precision of their phonological representations, using stimuli targeting the child’s own speech errors. Description of the Task The task procedure is the same as in Auditory Lexical Discrimination Task 1. The difference is that no stimuli are supplied for this task, as the investigator will select his or her own stimuli depending on the child’s own speech production errors. The procedure was described by John Locke in 1980, and called the Speech Production-Perception Task (Locke, 1980a). Use of the Task The advantage of this procedure is that it allows an investigation of the nature of speech errors specific to a child. If a child produces /t/ for /s/, for example, it is important to know if this is because he or she cannot detect the difference between words like TOE SEW, or HITHISS, or if he or she cannot produce the t s contrast in speech output. To investigate this, the child is presented with a picture of a SOCK, for instance, and asked if different productions of the word are right or not: for example SOCK, the correct production; TOCK, the child’s error; or POCK, a neutral distractor. If the child rejects TOCK as correct for SOCK, but says ‘No, it’s a TOCK’ then this suggests that the child’s speech input and stored phonological representation of the word are clear enough to reject the tester’s inaccurate production of the word. However, the fact that the child still says ‘it’s a TOCK’ implies that the speech difficulty is somewhere on the output side of the

86


profile. In contrast, if a child accepts ‘TOCK’ as correct for SOCK, then it suggests that the stored form of the target word is imprecise; that is, lacking some important phonological information. This will lead to difficulties when the child wants to produce the word in speech or spelling, since the foundation for that production is faulty. Thus this procedure, using the child’s own errors, can confirm if a child can recognise the correct form even if he or she is unable to produce it. In summary, if the child accepts the correct production and rejects the inaccurate productions, it suggests that he or she has an accurate phonological representation of the word. If the child accepts his or her own error form but rejects the neutral mispronunciation, it suggests that he or she has an inaccurate phonological representation that encodes the features of the child’s error. If the child accepts both his or her own error form and the neutral mispronunciation, it suggests that he or she has an imprecise phonological representation of the word, or maybe has not fully understood the task. This procedure can also be used to investigate fine auditory discrimination between similar phonetic realisations of a sound. For example, if a child can produce a fricative at the beginning of SOCK, albeit incorrect, such as // or //, the child can be presented with different phonetic realisations of the /s/ in SOCK to establish if this is an input- or output-only difficulty. For example, the investigator can ask if the following productions are right or not: /ɒk/, /ɒk/, /sɒk/, /ʃɒk/, to see if the child can detect the correct realisation of the target. Use of this procedure has shown that there can be different sources of speech errors within the same child. For example, in the case of Zoë, presented in Chapters 7 and 8 of Book 1, her voicing errors were related to an inability to discriminate voice~voiceless sounds but this was not true of other errors, such as stopping or cluster reduction, which arose from an articulatory output problem. Psycholinguistic Analysis of the Task The psycholinguistic nature of this task is the same as for the Auditory Lexical Discrimination Tasks 1 and 2 discussed above. Processing will follow the same route through the model as in Figure 4.1. In summary, three outcomes are possible: • The child rejects his or her errors, indicating that input processes are intact, and that production of that item does not reflect difficulties in stored phonological representation. Problems are likely to be in the speech output system. • The child accepts his or her errors as correct, indicating that either his or her production of those items reflects inaccurate stored phonological representation, or he or she may have auditory discrimination problems, or a combination of both.


87

• The child accepts some of his or her errors and a control mispronunciation as correct, suggesting that the child has more gross auditory discrimination difficulties, or he or she is not attending closely to the stimuli. Level on Speech Processing Profile • Level E Stimuli Stimuli are selected by the investigator based on a child’s speech output errors. The child is asked to name pictures. If the child’s production is inaccurate, the word and picture are then used in the auditory discrimination task. They are presented to the child (a) correctly, for example /ki/ (target); (b) as he or she said it, for example /ti/ (error); and (c) in a third form, also incorrect but not as the child has said it, for example / i/ (control). Locke (1980b) advocates that each stimulus is presented six times and reports that giving 18 presentations for each word only takes about a minute. Materials • Blank score sheet format on CD-Rom (Appendix D.5). • Not supplied: pictures of stimuli (to be selected by the investigator to suit the child); soft toy. Procedure The child is asked to name the chosen pictures. His or her production is transcribed and, if inaccurate, entered onto the score sheet, together with the adult production of the word and with a perceptually similar, but also incorrect, production. The procedure for Auditory Lexical Discrimination Task 1 can be followed to see if the child can reject the mispronunciations. It is also possible to present the words and mispronunciations without pictures as for Auditory Lexical Discrimination Task 2. Scoring The tester scores the child’s yes/no response for each item. The totals of correct responses to the target production, to his or her own error production and to the neutral, control mispronunciation are calculated. Normative Data None is available for this individually designed task. However, the investigator should consider at what age children could complete this

88


type of task successfully, and an indication of this is given in the normative data for Tasks 1 and 2 above. Questions to Ask • How do the child’s responses to his or her own productions compare to his or her responses to the correct target productions and to the neutral, control, inaccurate productions? • Does the child have difficulties with all types of stimuli or only on specific items, for example when voicing, fronting, stopping or cluster reduction is involved? Other Tasks Assessing a Child’s Own Mispronunciation Errors • SAILS, described above, provides a range of mispronunciations for each target and these reflect more common errors in children’s speech. This can be useful in selecting stimuli that mirror the child’s own speech errors.

Auditory Lexical Discrimination Task 4: Words in Sentences (from Cassidy, 1994) Aims • To assess auditory discrimination skills in connected speech. • To examine auditory discrimination performance with and without semantic support. Description of the Task The child is shown two pictures that illustrate a minimal pair, for example PEAR~BEAR, and is asked to point to one of the pictures in response to either a single word, such as PEAR, or a sentence with one of the words in it, for instance ‘The girl put the PEAR in her bag’. Three types of sentence context are used: neutral sentences, biased sentences and nonsense sentences. Use of the Task This task is unusual in that it requires the child to discriminate minimal pair words presented in a sentence, thus allowing assessment of a child’s auditory discrimination skills in a more ‘real life’ context; that is, in connected speech. The task can differentiate between children who rely on top-down vs bottom-up skills when processing auditory information and provide information that can be used when planning appropriate


89

intervention, for example to work on targets in or out of context depending on the child’s needs. Psycholinguistic Analysis of the Task The single word condition follows the same route through the model as in Figure 4.1. Again, the task is tapping the accuracy of phonological representations, but there may be differences in a child’s performance on the sentence-level task as opposed to single words. The sentence condition involves sentence processing. Longer chunks of speech need to be processed and remembered. The presence of a semantic context may facilitate access to phonological representations and top-down processing. The interaction between bottom-up and top-down processing can be particularly observed in the child’s responses to the different sentence types. Children who make relatively more errors (as compared to the normative data) on the sentences that don’t make sense than on the neutral sentences may be relying too heavily on top-down processing. ‘What do tests really test’ questions • Is this an input or output task? – Input. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – Yes. Level on Speech Processing Profile • Level E Stimuli The stimuli are selected from the Auditory Discrimination and Attention Test (ADAT) by Rosemary Morgan Barry (1989). They reflect different contrasts, for example voicing, place, manner and cluster changes in initial and final position, and are presented in Table 4.7. Each pair of these stimuli is then used in three different types of sentence: neutral, biased and nonsense (see Table 4.8). To control for assimilations that occur when producing words in connected speech, the target stimuli are preceded by a vowel or by the fricative /z/. For example, the following illustrates the sentences for the pair TEA~KEY. Neutral sentence Both words in the minimal pair make equal sense within the sentence: • Mum put the

TEA

on the table.

90


Table 4.7 Stimuli used in Auditory Lexical Discrimination Task 4: Words in Sentences (from Cassidy, 1994)

Practice items: PEAR ~ BEAR TIN~ BIN Test items: COAT~ GOAT LOCK~ LOG TEA~ KEY MOUSE~ MOUTH BAT~ MAT HEAD~ HEN GLASS~ GRASS CLOWN~ CROWN

• Mum put the

KEY

Target position in word

Contrast made

Initial Initial

Voicing Place and voicing

Initial Final Initial Final Initial Final Initial Initial

Voicing Voicing Place Place Manner Manner Cluster Cluster

on the table.

Biased sentence Both words make sense within the sentence, but one is more appropriate within the context provided: • I dropped my • I dropped my

TEA KEY

on the biscuits. on the biscuits.

Nonsense sentences Only one of the words in the minimal pair makes sense within the sentence: • The girl used the • The girl used the

TEA KEY

to open the door. to open the door.

Materials • Pictures and score sheet on CD-Rom (Appendix D.6). Procedure Before administering the task, ask the child to name each picture and record his or her responses. If a child is unable to name any of the pictures, establish if he or she can point to the correct picture. There are four different conditions of the task: single word, neutral sentences, biased sentences and nonsense sentences. It may be appropriate to spread testing


91

Table 4.8 Sentence stimuli for Auditory Lexical Discrimination Task 4: Words in Sentences, from Cassidy (1994) Practice sentences 1. Pear/bear (a) Neutral: The girl put the PEAR/BEAR in her bag. (b) Biased: I put the BEAR/PEAR on my bed. (c) Nonsense: The boy ate the PEAR/BEAR for lunch. 2. Tin/bin (a) Neutral: The TIN/BIN is empty. (b) Biased: We keep the BIN/TIN outside the door. (c) Nonsense: I ate the TIN/BIN of fruit. Test sentences 1. Coat/goat (a) Neutral: I think your COAT/GOAT is lovely. (b) Biased: I liked the COAT/GOAT with long fur. (c) Nonsense: Mum put her COAT/GOAT in the cupboard. 2. Lock/log (a) Neutral: I put the LOCK/LOG in the cupboard. (b) Biased: The man put the LOCK/LOG beside the key. (c) Nonsense: The woman used the LOG/LOCK to light the fire. 3. Tea/key (a) Neutral: Mum put the KEY/TEA on the table. (b) Biased: I dropped the TEA/KEY on the biscuits. (c) Nonsense: The girl used the KEY/TEA to open the door. 4. Mouse/mouth (a) Neutral: My teacher drew a MOUSE/MOUTH in my book. (b) Biased: The boy’s MOUTH/MOUSE was full of food. (c) Nonsense: The cat chased the MOUSE/MOUTH around the house. 5. Bat/mat (a) Neutral: I threw the MAT/BAT down the stairs. (b) Biased: I put the MAT/BAT outside the door. (c) Nonsense: The boy used the BAT/MAT to hit the ball. 6. Head/hen (a) Neutral: The girl saw the HEN/HEAD in the picture. (b) Biased: The boy rested his HEAD/HEN on the pillow. (c) Nonsense: The farmer’s HEN/HEAD has run away. 7. Glass/grass (a) Neutral: The girl dropped the GLASS/GRASS on the ground. (b) Biased: The man sat on the GRASS/GLASS in the garden. (c) Nonsense: The girl filled the GLASS/GRASS with lemonade. 8. Clown/crown (a) Neutral: We looked at the CLOWN/CROWN in the picture book. (b) Biased: The children watched the CLOWN/CROWN on television. (c) Nonsense: The king put the CROWN/CLOWN on his head.

92


over two sessions to maintain attention. Children of 3 years old should only complete the single-word task and the neutral sentences. For both the single-word and the sentence condition, each stimuli is presented three times. The following instructions summarise the procedure: • Present a pair of pictures and ask the child to name them – record the response. If the child is unable to name a picture, name the picture for the child and ask him or her to point to the one named. • For the single-word task, present a pair of pictures to the child and ask him or her to point to the one you name. The order of presentation is varied, with each one of the pair being presented three times. • For the sentence-level task, present a pair of pictures in the same way, but tell the child that he or she will hear the name of one of the pictures in a sentence and to point to the picture that he or she hears. Ask the child to listen to the whole sentence before pointing to one of the pictures. Again the order is varied, with each stimulus presented three times. When including the biased and nonsense sentences, tell the child that some of the sentences may sound a bit funny, but ask him or her to listen carefully and point to the picture being named. Support can be given with the practice items to ensure that the child understands the task. On the test items, however, only one repetition is allowed and no help should be given other than general encouragement. It does not matter in what order you give the single-word or sentence task. However, if you are using the ADAT as part of your investigation, administer that before carrying out either the single word or sentence level of the present task. Scoring A score of two points is given for a correct response, one point for a correct response following a repetition and no points for an incorrect response either with or without repetition. The scores for each sentence type should be recorded. There is a maximum score of 96 for each condition. Normative Data Data was collected from 41 normally developing children aged 3–5 years. Table 4.9 summarises the mean performance of the children on the single-word and word-in-sentence tasks. The children rarely asked for clarification during the task and their auditory discrimination skills


93

Table 4.9 Mean correct scores for the Auditory Lexical Discrimination Task 4: Words in Sentences (data from Cassidy, 1994) Age group Condition (N 96)

3 years (N 13)

4 years (N 15)

5 years (N 13 )

Single words Neutral sentences Biased sentences Nonsense sentences

85 74.5 – –

87.5 81 71 68.5

92.5 87.5 82.5 80

increased with age, confirming that the tasks are developmentally sensitive. All of the children found the single-word condition easier than the sentence condition, and were approaching ceiling by 5 years of age. At the sentence level the children found the neutral sentences easier than the biased or nonsense sentences. In the biased sentences the children made more errors on the word the sentence was biased against, for example in the sentence ‘I dropped my TEA/KEY on the biscuits’, KEY was selected less frequently than TEA. Similarly, in the nonsense sentences the children made more errors on the words that did not make sense, for instance in ‘The girl used the TEA/KEY to open the door’, TEA was selected less frequently than KEY. Normally developing children are therefore processing the meaning of the sentence and this context influences their auditory discrimination. Questions to Ask • How does the child’s discrimination of the minimal pairs compare with his or her naming of the pictures? • Is there a difference between performance on single words and in sentences? • Is there a difference on performance across the different sentence types? • What does the child’s performance tell you about his or her sentence processing skill? Other Assessments of Connected Speech • Speech Production: Connected Speech Tasks, see Chapter 6 of this volume. • Chapters 7 and 8 of Book 3 in this series (Pascoe, Stackhouse & Wells, 2006) provide further information on designing and using connected speech tasks and stimuli.

94


Summary • This chapter presents four tasks of auditory lexical discrimination. • These tasks involve auditory discrimination of acoustically similar words. However, as they also require children to access their own lexical representations, they are tapping a higher level of speech input processing than the tasks presented in Chapter 3. • In Auditory Lexical Discrimination Task 1: Mispronunciation Detection (from Vance, 1995) the child looks at a picture and decides if a word is the correct name of that picture. • In Task 2 (from Constable, Stackhouse & Wells, 1997) the child listens to real and non-words and decides whether or not he or she has heard a real word. Then, using the same stimuli, the child is asked to look at a picture and decide if a word is the correct name of that picture. • In Task 3, a picture task for individual children is administered using the same format as in Tasks 1 and 2b (after Locke, 1980a). The advantage of this procedure is that it allows an investigation of the speech errors specific to the child. • Task 4, the words-in-sentences task from Cassidy (1994), examines discrimination of single words and the stimuli are then used in sentences with varying semantic contexts. This allows consideration of the interaction between top-down and bottom-up processing. • These tasks all tap into Level E of the speech processing profile, although task 4 involves sentence processing with greater contextual support and increased language and memory demands. • Performance on these tasks indicates the precision of the child’s stored phonological representations in particular, and this information is necessary for planning intervention. • Unlike other auditory discrimination tasks (e.g. ABX) these picture tasks can be used with younger children.

Chapter Five Speech Production: Single Words

Investigation of a child’s speech production through naming and repetition tasks is important when assessing a child with a speech difficulty. This chapter presents stimuli and data for single-word speech production tasks and demonstrates how comparison of performance across different tasks can inform our understanding of children’s speech difficulties.

Picture Naming Picture naming is a popular method of investigating children’s lexical and speech difficulties. It is a complex task and needs to be carefully examined in terms of lexical accuracy (i.e. does the child know the word, can the child access the word) and articulatory accuracy (i.e. can the child say the word accurately). Some assessments that use picture naming are designed to investigate children’s expressive vocabulary and whether or not word-finding difficulties are present, and other assessments are designed to assess speech production skills. A key difference between the two functions of the task is in how a child’s responses are scored: whether any recognisable attempt to produce the target word is scored as correct; or whether the child has to produce the word with an accurate pronunciation to be scored as correct. If a child has a word-finding difficulty, the target word is present in his or her receptive vocabulary but he or she is unable to produce the word easily. It is not a ‘disorder’ in itself but a sign that something is wrong within the speech processing system (see German & Newman, 2004 for further discussion of lexical factors that affect word-finding difficulties). Word-finding errors on a picture-naming task might include

95

96


the following types (from Alison Constable, Chapter 10 of Book 2 in this series, Stackhouse & Wells, 2001): • Semantic error, e.g. SCREWDRIVER : “tool”. • Phonological error, e.g. BINOCULARS : “nokunars, nokumilars”. • Mixed semantic/phonological error, e.g. SADDLE : “handle, horse, handle, don’t know”. • Circumlocution (a description of the meaning), e.g. HAMMOCK : “a net where you sleep on”. Identification of word-finding difficulties can begin informally through observation of the child in different communicative contexts. A picturenaming task designed to identify word-finding difficulties can be administered. For example, the Word Finding Vocabulary Test (Renfrew, 1995a) asks the child to name pictures. Errors are then classified into whether the child did not know the picture (DKP) or did not know the name of the picture (DKN) – a simple but useful distinction to make when using a picture-naming task. A more comprehensive battery comprising picture-naming and connected speech tasks is the Test of Word Finding (German, 1989). In these tasks the child’s response is scored as correct if it is recognisable as the target word, even if there is some mispronunciation. See Constable, Chapter 10, Book 2 (Stackhouse & Wells, 2001), for an assessment procedure for the investigation of word-finding difficulties and implications for intervention. Other picture-naming tests have been designed not to assess lexical access per se but to collect information on how a child pronounces words; that is, as a speech assessment. The Edinburgh Articulation Test (EAT; Anthony et al., 1971) is one of the few UK standardised tests of articulatory maturity covering an age range of 3–6 years. An articulation age can be calculated together with a standard score, which allows an objective cut-off point below which further investigation is indicated (i.e. when the standard score is 85). It can therefore be used for screening children for speech difficulties and also in research projects to match individuals or groups of children on articulation age. Other picture-naming assessments have different aims, for example to collect sufficient information about the child’s production of a wide range of sounds in different positions in the words (Phonological Assessment of Child Speech, PACS; Grunwell, 1985) in order to undertake an analysis of his or her phonological system. Other assessments investigate the child’s use of specific phonological processes, for example Weiner’s Phonological Process Analysis (1979) and the Metaphon Resource Pack (Dean et al., 2000). The Diagnostic Evaluation of Articulation and Phonology (DEAP; Dodd et al., 2002) combines the above speech assessment aims in one test battery. It includes five assessments, four of which involve picture naming:

Speech Production: Single Words

97

• A 5-minute diagnostic screen (picture naming), following which one or more of the other four assessments can be administered. • An articulation assessment (picture naming) samples all English speech sounds (consonants and vowels) in initial and/or final position. When a speech sound is not produced correctly the child is asked to imitate it in a CV syllable or in isolation to establish if the child can produce an acceptable sound or not. • An oro-motor assessment investigates isolated and sequenced movements. • A phonological assessment (picture naming) identifies the use of simplifying processes, such as fronting or stopping. For comparison with the naming of single words condition, a picture description task is included that elicits the target words in connected speech. • An inconsistency assessment (picture naming) samples the production of 25 words on three separate trials and examines the stability of a child’s phonological system. Another useful resource for investigating phonological difficulties is the Hodson Assessment of Phonological Patterns (HAPP-3; Hodson, 2004), which has the advantage of including toys rather than pictures to elicit speech productions. Structured score sheets facilitate phonological analysis and intervention planning. The assessment also includes two screening procedures: a pre-school phonological screening and a multisyllabic word screening. This standardised assessment includes normative data in the age range 3 years to 7 years 11 months so that phonological ‘ability’ scores and percentile ranks can be calculated. Like the EAT above it can, therefore, be useful for matching children on phonological performance. The manual includes a chapter on phonological intervention and provides a clear link between assessment and intervention planning.

Repetition Tasks Children’s speech production can also be investigated by asking them to repeat words; these may be known words or unfamiliar words or nonwords. Published assessments of speech production tend to be picturenaming tasks, with the instruction that if the child does not recognise the word, the tester can say the word for the child to repeat (e.g. EAT, Anthony et al., 1971; DEAP, Dodd et al., 2002). This repetition response is then recorded. However, within a psycholinguistic perspective on speech production, a response to repetition of a word may not mirror a child’s naming of a picture of the same word (see the psycholinguistic analyses of these tasks below). Repetition tasks are suitable for all children including the very young and children with poor language or cognitive skills. Roy and

98


Chiat (2004) found that children from the age of 2 years were able to repeat both words and non-words with high levels of response for both types of stimuli. Words were repeated more accurately than nonwords and one-syllable items were repeated more accurately than longer items. Non-word repetition has recently received a lot of attention, as research indicates that poor performance may be a marker for specific language impairment (Bortolini et al., 2006; Conti-Ramsden, Botting & Faragher, 2001) or one of the predictors of persisting speech and literacy difficulties (Nathan et al., 2004a). Some researchers describe nonword repetition as a measure of ‘phonological working memory’ or ‘phonological short-term memory’, and interpret difficulties with nonword repetition as indicating phonological short-term memory deficits. The Children’s Test of Non-word Repetition (Gathercole & Baddeley, 1996) is a standardised assessment of non-word repetition designed to measure phonological working memory. However, children’s responses are scored for accuracy of production of the non-words. Other researchers have highlighted the speech processing demands involved in non-word repetition. These include speech perception, analysis or segmentation of the stimulus, and execution of motor programmes for speech (Snowling, Chiat & Hulme, 1991; Dollaghan, Biber & Campbell, 1995; Leitao, Hogben & Fletcher, 1997). In practice, non-word repetition tasks contribute most to our understanding of a child’s speech difficulties when they are compared with performance on naming and real word repetition tasks.

Task Comparisons Children with speech difficulties have shown different patterns of performance on naming and repetition tasks, and there may be some diagnostic value to identifying these patterns. Williams and Chiat (1993) found that a group of children described as having ‘phonological delay’ had similar error scores across naming, word and non-word repetition. In contrast, a group of children described as having ‘phonological disorder’ performed significantly less well on the naming task (for which access of lexical representations is required) but better on the word repetition tasks (for which access to lexical representations is not necessarily required). Bradford and Dodd (1994) found that children with phonological disorder who made inconsistent errors were significantly poorer on a non-word repetition task than children whose speech errors were consistent (see Dodd, 2005 and Broomfield & Dodd, 2004 for further discussion of subgroups based on nature of speech output errors). A similar pattern emerged from a study of 6–7-year-old children described as ‘dyspraxic’ (Thoonen et al., 1997). The children with dyspraxia performed less well


99

on both real and non-word repetition compared to a group of children with normally developing speech, and they showed particularly poor performance on non-word repetition. However, even children given the same diagnostic label, for example ‘dyspraxia’, can perform differently from each other on speech production tasks (Stackhouse & Snowling, 1992) and case studies have made an important contribution to our understanding of speech difficulties in children. For example, different profiles of performance that have been reported in children with speech difficulties include: • More accurate non-word repetition than word repetition or naming (Bryan & Howard, 1992.) • More accurate word repetition than naming (Stackhouse & Wells, 1993). • Poor but similar performance across repetition and naming tasks (Chiat & Hunt, 1993; Pascoe, Stackhouse & Wells, 2005). These authors use the different profiles of speech production performance to infer a particular speech processing deficit that underlies a child’s speech difficulties and draw implications from these for intervention. To illustrate the value of making task comparisons, the following procedure uses the speech processing profile to develop understanding of the nature of a child’s speech production difficulties. The letters in brackets denote the level on the speech processing profile being targeted (see Appendix A.2). It does not have to be followed in any prescribed order: • Compare performance on repetition of real words (I) and of nonwords (J). If real word repetition is better than non-word repetition, this suggests that the child may have a difficulty in assembling new motor programmes. If non-word repetition is better than real word repetition, the child may have some established, inaccurate motor programmes that do not reflect current speech production abilities. If real and non-word repetition are equally poor, it is likely that the child has poor articulatory skills, affecting all speech output. • Results from assessment of oral structure and function (K) will establish if the child has articulatory difficulties (see Chapter 7). • Comparison of real word repetition (I) with picture-naming performance (G) will indicate whether motor programmes are inaccurately stored, in which case repetition is likely to be more accurate than naming, because the child does not have to rely on his or her own existing motor programmes. However, auditory difficulties may affect repetition performance more than naming, as the child has to identify the spoken stimulus presented for

100


repetition, and these need to be investigated if a child is better at naming than at repetition. • Comparison of performance on picture naming (G) with the child’s responses to auditory lexical decision tasks (E) (as described in Chapter 4) will establish if the child’s phonological representations are not sufficiently well specified or inaccurate. This can include presenting naming errors back to the child for identification as right or wrong, for example: Mispronunciation Detection for individual children. • Information from other auditory discrimination tasks (see Chapter 3) will also help to identify any auditory discrimination problems that might result in phonological representations being underspecified. • Observing the child’s self-monitoring skills and attempts (or not) to correct his or her own productions (L) will indicate awareness of the output errors and the ability to correct them. Attempting to correct an output error also demonstrates that the child knows how the word should sound, and so his or her phonological representation is likely to be accurate. (See the mispronunciation selfcorrection task in this chapter for a procedure that addresses Level L of the speech processing profile.) Comparisons across different speech processing tasks are most useful when the stimuli used in different tasks are similar in terms of phonetic content and complexity. More details of possible outcomes from administering naming and repetition tasks and suggestions for how to follow up on these are in Book 1 (Stackhouse & Wells, 1997, pp. 44–48). However, task comparisons will only be informative if the stimuli used are challenging enough to the child. This chapter, therefore, includes stimuli at different word lengths for naming and repetition. Two sets of stimuli are described for repetition tasks: the first set comprises highfrequency words likely to be familiar to younger children and useful for older children with severe speech difficulties; the second set comprises low-frequency words more suitable for school-age children. As indicated in previous chapters, it is not intended that all the tasks and stimuli sets presented in this chapter be used with each child. A selection should be made to suit the child, taking into account factors such as age, severity of speech difficulties and whether a particular task or stimuli set can be usefully compared to tasks/stimuli tapping other levels of processing. The reader is referred to pp. 38–49 of Book 1 and encouraged to complete Activities 2.2 and 2.3 before administering the tasks presented in Chapters 5, 6 and 7 of this volume that describe speech output tasks at successive levels of the speech processing profile. This chapter (Chapter 5) focuses on single-word speech production tasks. It includes picture naming, in which processing is high up on the right-hand side of the speech


101

processing profile, and repetition tasks reflecting lower levels of speech output processing. Chapter 6 investigates production within connected speech and Chapter 7 oro-motor tasks at the bottom of the output side of the profile. The following tasks are included in this chapter: • Picture Naming Task 1: from Vance, Stackhouse & Wells (2005). • Picture Naming Task 2: from Snowling, van Wagtendonk & Stafford (1988). • Word Repetition Task: from Vance, Stackhouse & Wells (2005) and from Nathan et al. (2004a). • Non-word Repetition Task: from Vance, Stackhouse & Wells (2005) and from Nathan et al. (2004a). • Mispronunciation Self-Correction.

Speech Production: Single Words – Picture Naming Task 1 (from Vance, Stackhouse & Wells, 2005) Aims • To assess a child’s speech accuracy on a naming task. • To allow a direct comparison to be made between performance on naming and repetition, and/or auditory discrimination of the same or matched stimuli. Description of the Task The child is asked to name a series of pictures. The responses are then scored for articulatory accuracy. Use of Task This task was originally devised for the cross-sectional study described in Chapter 2. This study used the same core set of stimuli (60 words and matched non-words; see Table 2.2 in Chapter 2) across auditory, naming and repetition tasks. Thus, a comparison can be made of a child’s performance across tasks and a clearer idea developed about the nature of his or her speech difficulties. The complete set of stimuli is presented in the full version so that items can be selectively examined across tasks. Psycholinguistic Analysis of the Task Naming pictures requires the child to access his or her own semantic representation of the word, in order to activate his or her motor programme

102


Figure 5.1 Speech processing route for picture naming. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

for production (see Figure 5.1 for the processing route through the model for naming). Motor planning and execution are also implicated in the articulation of the word. If the word is present in the child’s receptive vocabulary (i.e. one with which the child is familiar) but he or she does not have an existing motor programme for it, he or she may reflect on the phonological representation of that word in order to create a new motor programme (see Figure 5.2). ‘What do tests really test’ questions • Is this an input or output task – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – Yes.


103

Figure 5.2 Naming a word for which the child has no stored motor programme. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

Level on Speech Processing Profile • Level G Stimuli Full version The 60 words from the cross-sectional study are used for this task. There are 20 words of one syllable, for example BRUSH; 20 words of two syllables, for example SANDWICH; and 20 words of three or four syllables, for example CATERPILLAR (see Table 2.2 Chapter 2). Short form Twenty-four words selected from the original set (see Table 4.2, Chapter 4) make up the short form of this naming task. These 24 words can also be used for word repetition so that qualitative comparisons can be made across tasks, but no normative data is available for naming of this subset of words. Comparisons can also be made between naming these words and performance on the Auditory Lexical Discrimination Task 1: Mispronunciation Detection, presented in Chapter 4, which uses the same words. Materials • Pictures and score sheets on the CD-Rom (Appendix E.1 for full version and Appendix E.2 for short form).

104


Procedure Each picture is presented in turn and the child asked to name it. If a child fails to name the picture or responds with a different lexical item, the investigator can use cues such as semantic, gap-fill or first sound to prompt him or her. If a cue is used then, after a brief pause, the child should be asked to say the word again and the second response recorded and scored. No further help should be given. If some, or all, of the auditory tasks (Auditory Discrimination Task 3 or Auditory Lexical Discrimination Task 1: Mispronunciation Detection) that use the same or matched stimuli are going to be administered to the child with the full version or short form of the naming task, and/or other speech production tasks described in this chapter, then the order of presentation of tasks should be as shown in Table 2.3 in Chapter 2. Scoring The child’s responses should be audio-recorded and transcribed in phonetic script at the time and then checked against the recording made. The production of the consonant sounds in each response is examined. For each word to be scored as correct the production of the consonants should be an accepted adult realisation, taking into account any regional variation in accent. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any words in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Normative Data Table 5.1 presents the mean number of words produced correctly by children aged 3 to 7 years. The children in the 3-year-old group were the most variable on this task and 4 year olds performed significantly less well than 5 year olds. However, there was no significant difference in performance between 6 and 7 year olds on articulatory accuracy, suggesting that children were approaching ceiling on this task by 6 years of age. Table 5.1 also shows the children’s mean performance on words of different lengths. Children’s speech production was less accurate as syllable length increased. There is a greater difference in performance between two-syllable and three- or four-syllable words than between one- and two-syllable words. Questions to Consider • Is the child able to pronounce the words accurately – in line with his or her chronological age group?


105

Table 5.1 Mean (standard deviation) correct scores for the full version of the Picture Naming Task 1 (data from the cross-sectional study, Vance, Stackhouse & Wells, 2005) Age group 3 years (N 16)

4 years (N 20)

5 years (N 20)

6 years (N 20)

7 years (N 20)

14.38 (2.83)

15.30 (1.81)

17.55 (1.79)

18.35 (1.31)

18.8 (1.20)

11.81 (3.90)

14.85 (1.90)

17.70 (1.30)

17.50 (1.50)

18.45 (1.28)

8.80 (3.61)

12.35 (3.23)

15.85 (3.18)

15.50 (3.07)

16.95 (2.33)

34.05 (8.68)

42.5 (5.61)

51.1 (5.67)

51.35 (4.22)

54.2 (3.93)

1 syllable N 20 Mean (s.d) 2 syllable N 20 Mean (s.d) 3–4 syllable N 20 Mean (s.d) Total N 60 Mean (s.d)

• How does performance on this naming task compare with performance on the matched real and non-word repetition tasks? • How does performance on this naming task compare with the child’s discrimination of the same items (see Auditory Lexical Decision: Mispronunciation Detection Task, Chapter 4)? Other Naming Assessments Tapping Processing at Level G, Indicating Accuracy of Motor Programmes • Assessment of Vowels Summary (Watts, 2004) • Diagnostic Evaluation of Articulation and Phonology (DEAP; Dodd et al., 2002) • Edinburgh Articulation Test (EAT; Anthony et al., 1971) • Goldman-Fristoe Test of Articulation 2 (Goldman & Fristoe, 2000) • Hodson Assessment of Phonological Patterns (HAPP-3; Hodson, 2004) • Metaphon Resource Pack (Dean et al., 2000) • Phonological Assessment of Child Speech (PACS; Grunwell, 1985) • PACS Toys (Grunwell & Harding, 1995) • South Tyneside Assessment of Phonology (STAP; Armstrong & Ainley, 1988)

106


Speech Production: Single Words – Picture Naming Task 2 (from Snowling, van Wagtendonk & Stafford, 1988) Aims • To assess lexical access and production. • To assess articulatory performance on picture naming. • To include more challenging stimuli for older children. Description of the Task The child is asked to name a series of pictures. The responses are then scored for access of the correct lexical item, and pronunciation of that item. Use of the Task This task was originally devised as a lexical naming task and it allows a comparison to be made with performance on receptive vocabulary tasks (e.g. British Picture Vocabulary Scale, Dunn et al., 1997). By making this comparison the tester can establish if the child has a specific word-finding difficulty at the output level or if the child has poor vocabulary development affecting both receptive and expressive performance. The original task has been extended here by not only scoring for naming of the lexical item, but also for how well the word was produced in terms of articulatory accuracy. By examining both lexical and articulatory skill on the same items, the tester can be clearer about the source of a child’s difficulties. The articulatory accuracy score on this naming task is one of the speech output tasks that has differentiated not only between children with and without persisting speech difficulties, but also between children with delayed and typical literacy development at the age of 6;08 in our longitudinal study (Nathan et al., 2004a, and see Chapter 8 for further discussion). Psycholinguistic Analysis of the Task When the child’s response is being scored for naming of the lexical item, the task is tapping his or her semantic representations and access to the lexical representation. Although a motor programme is activated for speech output, this may or may not be accurate. When the child’s response is being scored for articulatory accuracy, the stages of output processing discussed for Picture Naming Task 1, above, are also being explored.


107

‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – Yes. Level on Speech Processing Profile • Level G Stimuli Twenty pictures illustrating words of between two and five syllables are used: 6 2 syllables, 9 3 syllables, 4 4 syllables and 1 5 syllables (see Table 5.2 for the list of stimuli). There is one semantically and one phonologically related word for each word selected, for instance SADDLE: REINS, STIRRUP. (See Appendix E.4 for a full list of these.) These are used to establish if the child can recognise the target word from similar words, if he or she does not produce the word spontaneously. Materials Required • Pictures and score sheet on the CD-Rom (Appendix E.4).

Table 5.2 Stimuli for Picture Naming Task 2: (from Snowling, van Wagtendonk & Stafford, 1988) 2 syllables

3 syllables

4 syllables

5 syllables

acorn giraffe seahorse puppet hammock saddle

screwdriver strawberry* microscope elephant umbrella stethoscope pyramid octopus unicorn

ballerina helicopter binoculars rhinoceros*

hippopotamus

Note: *Accent warning! Depending on the accent of the speaker, STRAWBERRY may be produced with two or three syllables; that is, if BERRY is pronounced as ‘bree’ then the word will have two syllables not three. Similarly, RHINOCEROS may be produced with three or four syllables; that is, if CEROS is pronounced a ‘sres’ then the word will have three syllables not four.

108


Procedure The child is asked to name each picture. If he or she is unable to name it or provides a different lexical item, three choices are given to the child. For example, if the child does not produce the item MICROSCOPE, the tester can ask ‘Is it a TELESCOPE, a MICROSCOPE or a MICROPHONE?’ It is then indicated on the score sheet if the child’s production was spontaneous or as a result of this prompting. All responses should be audiorecorded using a good-quality recorder and microphone as well as transcribed live in phonetic script at the time. The transcription can then be checked after assessment from the audio recording. Scoring The child’s responses are scored twice. Lexical score The number of spontaneous naming responses that are recognisable as the target word is calculated. Forced-choice (prompted) responses are also noted, as this indicates that the child knew the word, but was unable to access it immediately. However, these are not included in the total lexical score. Speech score The percentage of consonants correct in each naming response is calculated. Correct responses following a forced choice are also included in this score. If the child failed to make a correct response following a forced choice or made no response for another reason, the percentage should be calculated on the number of consonants in words attempted. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 86 consonants in this task, but this can vary depending on the child’s naming response to the items HIPPOPOTAMUS and RHINOCEROS as follows: •

HIPPOPOTAMUS:

either ‘hippopotamus’ or ‘hippo’ can be accepted as a correct lexical response. If ‘hippo’ is realised, for the speech score the percentage is calculated on a possible total of four fewer consonants; that is, 82. • RHINOCEROS: either ‘rhinoceros’ or ‘rhino’ can be accepted as a correct lexical score. If ‘rhino’ is realised, for the speech score the percentage is calculated on a possible total of three fewer consonants; that is, 83. • If a child produces both ‘hippo’ and ‘rhino’, the total possible number of consonants correct is 79.


109

Table 5.3 Mean (standard deviation) and range for lexical scores (number of items named spontaneously) for Picture Naming Task 2 (data from the longitudinal study) Age group Lexical score

4 years (N 47)

5 years (N 47)

6 years (N 44)

Mean (s.d.) N 20 Range

4.45 (3.08) 3–17

11.32 (3.13) 6–19

12.98 (3.0) 7–18

Normative Data Table 5.3 summarises the lexical scores of 4, 5 and 6 year olds on the naming task. Forced-choice (prompted) responses have been excluded. The wide range of performance at each of the age levels should be noted. Table 5.4 summarises the articulatory performance of 4, 5 and 6 year olds on the naming task, scoring for percentage of consonants correctly articulated. The children are approaching ceiling on this task by 5 years of age. Questions to Consider • Is there a discrepancy between a child’s lexical and articulatory score compared to typically developing children? • Can the child produce words from forced-choice prompts that he or she was unable to access spontaneously? • Does the child have word-finding difficulties, or speech difficulties, or both? Other Assessments Tapping Expressive Vocabulary and Access to Lexical Representations • • • •

Word Finding Vocabulary Test (4th edn; Renfrew, 1995a) Test of Word Finding (TWF; German, 1989) Expressive Vocabulary Test 2 (Williams, 2006) Expressive One-Word Picture Vocabulary Test (Brownell, 2000)

Table 5.4 Mean (standard deviation) for percentage of consonants correct (PCC) for speech production accuracy for Picture Naming Task 2 (data from the longitudinal study) Age group PCC

4 years (N 47)

5 years (N 47)

6 years (N 44)

Mean (s.d.) Range

88.90 (9.81) 53.85–100

94.28 (3.91) 83.56–100

93.94 (4.88) 80–100

110


Speech Production: Single Words – Word Repetition Task (from Vance, Stackhouse & Wells, 2005 and Nathan et al., 2004a) Aims • To examine the child’s ability to repeat words. • To allow comparison of performances across speech production tasks (naming, word repetition and non-word repetition) using the same stimuli. Description of the Task The child is asked to repeat a word after the tester. There are no pictures. Use of the Task Word repetition performance can demonstrate if a child has articulatory difficulties. However, by comparing it directly with naming and nonword repetition performance a more precise location of these difficulties can be established (see the introduction to this chapter for a discussion). This task includes a long and short version with high-frequency, familiar words, administered to children at 4 and 5 years of age. The short form of these, differentiated between those children in the longitudinal study whose speech difficulties did and did not resolve by 6 years of age (see Chapter 8). For the 5-year-old children task performance also differentiated between those who went on to be typical or delayed readers at 6 years of age. The task also includes a list of low-frequency, less familiar words that have a wider range of syllable lengths. These words are therefore more suitable for school-age children and for children with more subtle speech difficulties. Repetition of this low-frequency word list differentiated between 6-year-old children with persisting and resolving speech difficulties, as well as between 6-year-old children with typical vs delayed reading and spelling development in our longitudinal study. Psycholinguistic Analysis of the Task When asked to repeat a high-frequency word the child is likely to have an existing motor programme for that word (i.e. instructions for the pronunciation of that word) already stored within their lexical representations. The child might access this motor programme to produce a familiar word (see Figure 5.3). However, the child does not have to use an existing motor programme and can merely ‘parrot’ the word back to the tester, treating it like an unfamiliar or new word and recreating a motor


111

Figure 5.3 Speech processing route for word repetition

programme (see Figure 5.4). It is possible that repetition of low-frequency words is less likely to result in access of an existing motor programme. It is also possible that some children may not know one or more of the low-frequency words, in which case the task, for those stimuli, becomes a non-word repetition task. ‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – No. Level on the Speech Processing Profile • Level I

112


Figure 5.4 Speech processing route for non-word repetition. Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

Stimuli Full version The list of 60 core words devised for the cross-sectional study is used (see Table 2.2, Chapter 2). The words are presented in two sets, Set A and Set B. These are the same words used in the Picture Naming Task 1 (see above) and the Auditory Lexical Decision: Mispronunciation Detection Task (see Chapter 4). Short form The 24 single words used in the short form of the Mispronunciation Detection Task were presented for repetition in the longitudinal study (see Table 4.2, Chapter 4). They are divided into two equal-sized lists, each comprising eight words of one syllable, eight words of two syllables,


113

Table 5.5 Stimuli for the Word Repetition Task: Low-Frequency Words (from the longitudinal study) 1 syllable

2 syllables

3 syllables

4 syllables

5–6 syllables

squeak wasp bran crutch

cartridge drummer cricket scraper

acrobat anchovy gorilla grasshopper

satisfaction ballerina chrysanthemum librarian

encyclopaedia refrigerator electricity multiplication

and eight words of three or four syllables. These words can also be used for picture naming, and were used for the short form of the Auditory Lexical Discrimination: Mispronunciation Detection Task (see Chapter 4). Low-frequency words The stimuli comprise 20 low-frequency words, four words of one syllable, four words of two syllables, four words of three syllables, four words of four syllables and four words of five or six syllables (see Table 5.5 for the list of stimuli). The words are divided into two equalsized lists containing two items from each syllable length. Materials • Score sheets on the CD-Rom (Appendix E.5 for full version, Appendix E.6 for short form, Appendix E.7 for low-frequency words) Procedure The child is asked to repeat each word in turn. The stimulus can be presented again, once only, if the child fails to respond to a stimulus or if they request a repetition. Three practice items are presented at the start of the task (one practice item for the low-frequency word list). When explaining the task the procedure can be demonstrated using the child’s name and then the practice item presented. Corrective feedback can be given if the child fails to repeat any of these practice items, but only general encouragement can be given during the main part of the task. All responses should be audio-recorded using a good-quality recorder and microphone, as well as transcribed live in phonetic script. The transcription can then be checked after assessment from the audio recording. If some, or all, of the auditory tasks (Auditory Discrimination Task 3 or Auditory Lexical Discrimination Task 1: Mispronunciation Detection) that use the same or matched stimuli are going to be administered to the child with the full version or short form of the Word Repetition Task, and/or

114


other speech production tasks described in this chapter, then the order of presentation of tasks should be as shown in Table 2.3 in Chapter 2. If both the low-frequency word and the matched non-words are going to be used for repetition, then word list A and non-word list A should be presented in one session and word list B and non-word list B in a different session (or, if this is not possible, at a different point in one session). List A of these non-words does not contain any items derived from low-frequency words in list A (and similarly for list B). This procedure will, therefore, ensure that non-words will not be given consecutively with the words from which they are derived and, thus, avoid carry-over or practice effects. Scoring Full version The scoring procedure is the same as for the Picture Naming Task 1, outlined above. The production of the consonants in each response is examined. The word is counted as correct if the consonants within the word match an accepted adult realisation, taking into account any regional accent variations. (See Appendix E.3 for scoring notes for the full version and short form.) Any words in which consonants deviated from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. The number of words produced correctly is totalled. Short form The percentage of consonants correct (PCC) is calculated. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 77 consonants for this stimuli set. However, if the child fails to make a response, the percentage should be calculated on the number of consonants in words attempted. Low-frequency words This stimulus set is scored for PCC as for the short form (see above). There are a total of 92 consonants in this word list. Normative Data Full version Data was collected from typically developing children aged 3 to 7 years. The mean number of correct responses for each age group is given in Table 5.6. There was a significant difference between the performance of


115

Table 5.6 Mean (standard deviation) number of words repeated correctly on the full version of the Word Repetition Task (data from the cross-sectional study, Vance, Stackhouse & Wells, 2005) Age group 3 years (N 17) 1 syllable N 20 15.06 (2.91) Mean (s.d) 2 syllable N 20 13.19 (3.15) Mean (s.d) 3–4 syllable N 20 11.75 (4.71) Mean (s.d) Total N 60 Mean (s.d.)

40.00 (9.85)

4 years (N 20)

5 years (N 20)

6 years (N 20)

7 years (N 20)

16.30 (2.11)

17.80 (1.20)

18.65 (1.35)

18.30 (0.80)

16.30 (2.11)

17.50 (1.76)

17.75 (1.68)

18.30 (1.59)

14.90 (3.31)

16.95 (2.58)

17.55 (1.99)

18.70 (1.17)

47.50 (5.77)

52.25 (4.31)

53.95 (3.99)

53.30 (3.01)

3 and 4 year olds and between 4 and 5 year olds on this task. However, there was no difference between 5-, 6- and 7-year-old children’s performance. Mean scores are also given for the number of items correct at each word length. These show that the younger children’s speech production was less accurate as syllable length increased. The older children (6 and 7 years of age) are less sensitive to increasing word length in word repetition than on the Picture Naming Task 1 (see Table 5.6). Short form Data was collected on the short form of the task from 4- and 5-year-old children. The mean and standard deviation of the PCC are given in Table 5.7. The children perform well on this task, many of them are scoring at 100 % correct at 5 years of age. Table 5.7 Mean (standard deviation) and range of percentage of consonants correct (PCC) on the short form of the Word Repetition Task (data from the longitudinal study) Age group PCC

4 years (N 47)

5 years (N 47)

Mean (SD) Range

92.82 (5.48) 77.92–100

98.4 (2.32) 90.91–100

116


Table 5.8 Mean (standard deviation) and range of percentage of consonants correct (PCC) on the Word Repetition Task: Low-Frequency Words (data taken from the longitudinal study) Age group PCC

5 years (N 47)

6 years (N 44)

Mean (s.d.) Range

90.86 (6.38) 70.60–100

92.51 (5.77) 76.09–98.90

Low-frequency words Data was collected from 5- and 6-year-old typically developing children on the low-frequency word list. Mean, standard deviation and range for percentage of consonants correct are shown in Table 5.8. As can be observed, there is a greater range of performance of 5 year olds on these low-frequency words than on the high-frequency word list (see Table 5.7). Questions to Consider • How does performance on this task compare with the Picture Naming Task 1 and the Non-word Repetition Task (below) using the same stimuli sets? • How does performance compare with the child’s discrimination of the same items (see Auditory Lexical Discrimination: Mispronunciation Detection Task, Chapter 4)? Other Tasks • Word and Non-word Repetition Task (Roy & Chiat, 2004) for 2 to 4 year olds.

Speech Production: Single Words – Non-word Repetition Task (from Vance, Stackhouse & Wells, 2005 and Nathan et al., 2004a) Aims • To examine children’s ability to assemble new motor programmes. • To allow comparison of performances across speech production tasks (picture naming, word repetition and non-word repetition). Description of the Task The child is asked to repeat a series of non-words after the tester.


117

Use of the Task Performance on this task gives some insight into how a child processes novel verbal material. This is an important skill given the number of new words children are required to learn at all ages. Comparison of performance on this task with other speech production tasks can highlight areas of speech processing difficulty (see the introduction to this chapter). Within the psycholinguistic framework, this task is considered to be particularly taxing for motor programming skills, since the child has to assemble a new motor programme for speech production quickly. This requires holding the target non-word in a temporary store long enough to reflect on its components and plan the articulatory output. If a child produces a similar-sounding real word when asked to repeat a non-word, this indicates that he or she may have auditory discrimination difficulties or be over-reliant on top-down processing skills; both suggest that the child will be disadvantaged when dealing with new words. Difficulties with non-word repetition indicate that a child may need more exposure in learning new words in order for lexical representations and /or motor programmes to be developed; have subtle and residual speech problems not necessarily apparent on the real word task; or have associated spelling problems. The short form of this task, administered to children at 4 and 5 years of age, differentiated between children who did and did not resolve their speech difficulties by 6 years of age in the longitudinal study (see Chapter 8). For the 5-year-old children it also differentiated between children who were typical or delayed readers/spellers at 6 years of age. Psycholinguistic Analysis of the Task In contrast to the Word Repetition Task above, the items to be repeated are unfamiliar. Non-word repetition cannot make use of existing lexical representations as the child will not have heard or said these non-words before. This task, therefore, reflects a child’s ability to assemble and execute a new motor programme for a new word (i.e. non-word; Figure 5.4). However, it is likely that use is made of stored programmes for similar known words to assist this process. For example, Dollaghan, Biber and Campbell (1995) report that non-words in which the stressed syllable sounds the same as a known lexical item are repeated more accurately than those in which the stressed syllable does not sound like a known word. Similarly, non-words that have been rated more highly for ‘word likeness’ are repeated more accurately than those rated less highly (Gathercole et al., 1991). Performance on the Non-Word Repetition Task can be compared to performance on the Word Repetition Task (See above and Book 1, Stackhouse & Wells, 1997, p. 46 for further discussion on this point.)

118


‘What does this test really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – No. • Does the child have to access these lexical representations in order to complete the task? – Not applicable. Level on the Speech Processing Profile • Level J Stimuli Full version Sixty non-words were derived from the set of real words used in the crosssectional study (see Table 2.2 in Chapter 2). The vowels were changed to create the non-words, while the consonants were kept the same. The words from which these non-words are derived are used in the full versions of Picture Naming and Word Repetition Tasks described above. Short form The 24 non-words (see Table 5.9) derived from the 24 words used in the Word Repetition Task above. Again the items are divided into two equalsized lists, comprising eight one-syllable non-words, eight two-syllable non-words and eight three- to four-syllable non-words. The items have then been randomised in their order of presentation within each list. Non-words matched to low-frequency words Non-words were derived from the list of 20 low-frequency words that were selected for the low-frequency list for the Word Repetition Task Table 5.9 Stimuli for short form of Non-word Repetition Task (from longitudinal study) 1 syllable

2 syllable

3–4 syllable

/brʃ/ /spænd/ /glεv/ /snak/ /trɔn/ /vIn/ /pləυt/ /fεʃ/

/jælɔ/ /fluw/ /dlə/ /trεkt/ /slɔpə/ /sεsəd/ /gεtɔ/ /sped/

/ktəpælə/ /spgtə/ /ælfɒnt/ /æmbræl/ /pIdɔmz/ /pεrəʃit/ /bætəfləυ/ /ɒrəpləυn/


119

Table 5.10 Stimuli for the Non-word Repetition Task, non-words matched to low-frequency words (used in longitudinal study) 1 syllable

2 syllables

3 syllables

/skwuk/ /wεsp/ /brɒn/ /krεtʃ/

/kitrυd/ /drɒm/ /krɒkɒt/ /skr υpi/

/k bυt/ /υntʃevɑ/ /gærɒlu/ /grushεp/

4 syllables

5–6 syllables

/sɒtæsfεkʃn/ /blrɑn/ /krzɒnImεm/ /laυbr υ aυn/

/υns klaIpɑd/ / ɒfrædɒr υt/ /ɒlktræsυtɑ/ /mɑltplkaυʃn/

(see above). The non-words were derived by changing the vowels of the low-frequency words, keeping the consonants the same. There are, therefore, four non-words of one syllable, four non-words of two syllables, four non-words of three syllables, four non-words of four syllables and four non-words of five or six syllables (see Table 5.10 for the list of stimuli). The words are divided into two equal-sized lists containing two items from each syllable length. Materials • Score sheets on CD-Rom (Appendix E.8 for full version, Appendix E.9 for short form and Appendix E.10 for non-words matched to low-frequency words) Procedure The procedure is similar to that for the Word Repetition Task above, but for the Non-word Task the child is told that the investigator is going to say some funny words that they may not have heard before and is then asked to repeat the non-word that he or she hears. For younger children, a soft toy, such as a monkey, is used to demonstrate the task. The child is told that the monkey says ‘made-up, monkey words’ and that he or she would not know these words. The child is then asked to say each word like the monkey said it. Three practice items are presented at the start of the task (one practice item only for the non-words matched to the low-frequency real words). Corrective feedback can be given if the child fails to repeat any of these practice items. If the child lexicalises any of these words – that is, produces a real word – he or she should be encouraged to say the word ‘just like the monkey says it’. Only general encouragement can be given during the main part of the task. One further presentation of a test item can be given if the child fails to respond to a stimulus or if he or she requests a repetition. All responses should be audio-recorded using a good-quality recorder and microphone as well as transcribed

120


live in phonetic script. The transcription can be checked after the assessment from the audio recording. If some, or all, of the auditory tasks (Auditory Discrimination Task 3 or Auditory Lexical Discrimination Task 1: Mispronunciation Detection) that use the same or matched stimuli are going to be administered to the child with the full version or short form of the Non-word Repetition Task, and/or other speech production tasks described in this chapter, then the order of presentation of tasks should be as shown in Table 2.3 in Chapter 2. If both the low-frequency word and the matched non-words are going to be used for repetition, then word list A and non-word list A should be presented in one session and word list B and non-word list B in a different session (or, if this is not possible, at a different point in one session). List A of these non-words does not contain any items derived from low-frequency words in list A (and similarly for list B). This procedure will, therefore, ensure that non-words will not be given consecutively with the words from which they are derived and, thus, avoid carry-over or practice effects. Scoring Full version Each response is marked as right or wrong using the same scoring criteria as for the full version of the Word Repetition Task. For a non-word response to be counted as correct the realisation of the consonants within the word should be an accepted adult realisation, taking into account any regional accent variation (see more detailed guidelines in Appendix E.3). Any responses in which consonants deviated from accepted adult realisations are scored as incorrect, including responses in which consonants have been added or omitted by the child. Short form As for the short form of the Word Repetition Task above, the percentage of consonants correct (PCC) is calculated. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 77 consonants for this stimuli set. However, if the child fails to make a correct response, the percentage should be calculated on the number of consonants in words attempted. Non-words matched to low-frequency real words This stimulus set is scored for PCC, as in the short form (see above). There are a total of 92 consonants in this word list.


121

Table 5.11 Mean (standard deviation) number of non-words repeated correctly on the full version of the Non-Word Repetition Task (data from the cross-sectional study, Vance, Stackhouse & Wells, 2005) Age group

1 syllable N 20 Mean (s.d) 2 syllables N 20 Mean (s.d) 3–4 syllables N 20 Mean (s.d) Total N 60 Mean (s.d.)

3 years (N 17)

4 years (N 20)

5 years (N 20)

6 years (N 20)

7 years (N 20)

14.19 (2.79)

15.20 (1.94)

15.25 (1.68)

16.70 (1.22)

16.05 (1.88)

12.94 (3.42)

14.65 (2.08)

16.30 (1.34)

16.05 (1.23)

16.95 (1.90)

11.50 (3.01)

13.30 (2.45)

14.70 (2.41)

15.00 (2.70)

15.80 (2.33)

38.76 (7.79)

43.15 (4.67)

46.25 (3.55)

47.75 (4.22)

48.85 (4.66)

Normative Data Full version Data was collected from typically developing children aged 3 to 7 years. As on the real word repetition task, the children’s performance plateaued on this task at around 5 years of age. There was no significant difference between the performance of 5-, 6- and 7-year-old children on non-word repetition; see Table 5.11. There was a significant effect of word length, with more accurate production of one- and two-syllable words than three- or four-syllable words at all ages. The mean performance at each word length is also given in Table 5.11. Short form Data was collected from typically developing children aged 4 and 5 years. Examination of percentage of consonants correct reveals that children found the non-word repetition slightly more difficult than real word repetition, but even so children can perform well on this task, with some 5-yearold children performing at the level of 100 % correct (see Table 5.12). Non-words matched to low-frequency words Data was collected from the 5- and 6-year-old children in the longitudinal study; see Table 5.13. There is a wider range of performance in the 5-year-old children with this set of non-words compared to the short form of the Non-Word Repetition Task reported above (see Table 5.12)

122


Table 5.12 Mean (standard deviation) and range of percentage of consonants correct (PCC) on the short form Non-word Repetition Task (data taken from the longitudinal study) Age group PCC

4 years (N 47)

5 years (N 47)

Mean (s.d.) Range

90.39 (5.67) 76.6–100

95.41 (5.1) 80.5–100

Table 5.13 Mean (standard deviation) and range of percentage of consonants correct (PCC) on the Non-words Matched to Low-Frequency Words (data taken from the longitudinal study) Age group PCC

5 years (N 47)

6 years (N 44)

Mean (s.d.) Range

83.17 (8.66) 63.0–95.7

85.66 (8.69) 54.3–97.8

Questions to Consider • How does performance on non-word repetition compare with real word repetition using the same stimuli set? • How does performance compare with that on matched auditory discrimination tasks? Other Assessments of Non-word Repetition • Children’s Test of Non-word Repetition (CNRep; Gathercole & Baddeley, 1996) • Grammar and Phonology Screening Test (GAPS) Non-word Repetition subtest (Van der Lely et al., 2006) • Word and Non-word Repetition Task (Roy & Chiat, 2004) for 2 to 4 year olds

Mispronunciation Self-Correction Aims • To investigate children’s ability to correct their own speech errors at the time they are made. • To investigate a child’s self-correction strategies on specific targets. • To measure consistency of self-correction strategies.


123

• To establish a pre-intervention baseline of self-correction for comparison with post-intervention performance. Description of the Task The task is to observe and record how a child spontaneously corrects his or her own errors ‘on line’. The investigator notes if a child attempts spontaneously to self-correct his or her own speech errors and, if so, how successful this is. It is not the same as when the investigator feeds back the child’s errors to him or her to detect if they are said correctly or not; that task, Mispronunciation Detection for Individual Children (see Chapter 4), involves an inter-speaker (between speakers) comparison rather than intra-speaker (within same speaker) monitoring. Information about awareness of speech errors can also be gathered from school-age children by asking them directly about their speech errors and what they find difficult. Use of the Task A child’s self-monitoring is an important aspect of developing speech appropriately. A child who is not aware that his or her speech is different from others will not be able to progress with speech development. It is necessary to know when planning intervention if self-monitoring is a weakness or strength; that is, can it be utilised to support other weaknesses or is it a target for intervention in itself? This task allows a profile of self-correction strategies and consistency to be drawn up for a child to help plan appropriate intervention. This profile can be used to monitor a child’s progress. For example, a sheet can be completed on target items before and after a block of intervention and the proportions of responses in each category compared (see Appendix E.11). Thus, self-correction can be used as an outcome measure in intervention. Psycholinguistic Analysis of the Task If a child can detect his or her own speech errors this implies that he or she has listened to his or her own output and matched it to a phonological representation that is accurate, or at least more accurate than the speech output; thus, his or her intra-speech input skills are intact. If he or she can then change speech production to a correct pronunciation of the word or phrase, he or she has intact speech output skills. This is typical of all of us when we make ‘slips of the tongue’ in everyday conversations. However, a child with speech difficulties may detect when they have produced a word wrongly, but be unable to correct it because of an incorrectly stored motor programme and/or lower-level articulatory difficulties; or not detect his or her own speech errors in

124


the first place and then make no attempt to self-correct, thus perpetuating the speech errors. ‘What do tests really test’ questions • Is this an input or output task? – Input to detect own errors; output to correct them. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – Yes. Level on Speech Processing Profile • Level L Stimuli Observations of a child’s self-correction skills may not involve any specific stimuli. However, an investigator may want to observe a child’s self-correction strategies on specific targets, for example incorporating a particular sound or sounds, and can devise his or her own targets for this. Asking school-age children direct questions has also triggered insights into Level L, which can be tested further. The following questions can be used to do this (it is not necessary to ask all of them; sometimes only one is needed): • • • • • •

Why do you come for speech and language therapy? What is hard about talking? What do you do when someone doesn’t understand you? Why do you think listening is important? Even if we speak English, do we all sound the same? Have you finished learning to talk yet?

Materials • Score sheet on the CD-Rom (Appendix E.11). Procedure Observe the child in a general conversation/play situation; include pictures/materials that will trigger specific targets to be investigated if required. Audio record the session and complete the score sheet as far as possible at the time, checking this against the recording later. Where an investigation of consistency of self-correction strategies is specifically required, ensure that targets are used more than once in the session.


125

Note the following when the child is performing speech output tasks or when speaking spontaneously: 1. Corrects own speech error spontaneously. 2. Attempts to correct speech error and produces a response closer to the target, but not yet correct. 3. Attempts to correct speech error and produces a variable response, which may or may not be closer to the target. 4. Attempts to correct speech error, but response is same as original error. 5. Only attempts to correct speech error if listener does not understand. 6. Makes no attempt to correct speech error. If asking direct questions, as given in the description of stimuli above, do so informally and only if appropriate for the child. Record the response for psycholinguistic interpretation (see the example under scoring below). Scoring Analysis can be of items of interest that arise during spontaneous conversation or play, or materials presented to elicit particular words, perhaps that begin or end with a target sound. Record on the score sheet the word or phrase and tick the appropriate box to denote the child’s response to his or her own error (see 1–6 in the procedure above). Add up the number of times each category of self-correction attempt is used by the child. Percentages of responses in each category can also be calculated (see Appendix E.11). This will indicate the ability of the child to self-correct: the greater the proportion of responses in category 1 the better. Examine the responses to see if there are any commonalities between words that the child can and cannot self-correct: for example, words beginning with /k/ are not self-corrected but words beginning with /f/ are; words beginning with /s/ are not self-corrected but words ending in /s/ are. Hypotheses about these words or error patterns can then be tested out further. If a target occurs more than once in the sample, note any variation within self-correction attempts at the same target. If the child always produces the same self-correction attempt to his or her error, note that he or she was consistent in responding to that target. If he or she produces different self-correction attempts to the error, note that he or she was inconsistent. Calculate the number of targets for which the child was consistent and inconsistent. Interpreting responses to questions Answers to the direct questions, given in the procedure above, can be interpreted psycholinguistically. For example, in answer to the question

126


Have you finished learning to talk yet? Ben, aged 7;11 with persisting speech difficulties that included pronouncing [g] for /d/, and [k] for /t/, replied: well not yet, because I think I have a problem with ‘g’s and ‘g’s (he means ‘d’s and ‘g’s) .......... and maybe on ‘k’s as well because it seems like if I say ‘k’ (he means ‘t’) it seems like its going into a ‘k’. (Boase, 2006, p. 100)

Ben is therefore very aware of his speech errors, indicating that he has phonological representations in which these contrasting sounds are accurately stored, but cannot yet correct them in his speech production. His difficulties seem to be at an articulatory level. Further investigation using the procedures included in this volume to complete a speech processing profile for Ben would allow this hypothesis to be tested. Normative Data There is no normative data for this procedure. The guidelines are that typically developing children more often self-correct their ‘slips of the tongue’ than not and have a higher proportion of responses in categories 1 and 2 above (see Book 1, Stackhouse & Wells, 1997, p. 165). Questions to Ask • Is there spontaneous speech correction indicating not only selfmonitoring skills but also an ability to change speech production? • Are there spontaneous attempts at speech correction but these are not always successful because of difficulties elsewhere noted on the speech processing profile? • Is any change to speech output only in response to the listener misunderstanding the child’s speech output? • Can the child change their speech output only if directed to do so in therapy or a teaching situation? • Is there a difference between self-correction of single words and of connected speech? • Is there a mixture of responses? Is this related to the targets, for example certain sounds, position in words, lexical items or grammatical structures, and/or to the speaker-listener contexts, for instance child with therapist/parent/sibling/friends/a group? Other Tasks Reflecting Children’s Responses to Their Own Speech Errors See Stemberger (1989) and Tarplee (1996) for a description of speech errors in typically developing children, and Gardner (1997) and pp. 413–419


127

in Book 2 (Stackhouse & Wells, 2001) for a discussion of children’s selfcorrection in therapy contexts.

A Comparison of Performance Across Different Speech Output Tasks The normative data presented in this book is useful for interpreting speech production assessment findings in children with speech difficulties. It is possible to examine how a child’s profile of performance across tasks compares with that of children of a similar age and younger, so that processing difficulties can be more accurately pinpointed. For example, the finding that a child is worse at non-word than word repetition is not enough to suggest a specific deficit in motor programming per se, as this pattern is found in typically developing children. Rather, it is the amount of the discrepancy, and the age of the child, that will be of interest in interpreting test results. Two 6-year-old boys with severe speech difficulties, C and D, were assessed using the full versions of the Picture Naming, Word Repetition and Non-word Repetition Tasks described in this chapter. They were both performing less well than the typically developing children, as would be expected. Their raw scores on the three tasks were compared with the mean scores of the typically developing 6-year-olds, presented above. Examination of their profiles (see Figure 5.5) indicates that C is having most difficulty with non-word repetition, compared to the 6 year olds, suggesting that his speech processing deficit may be rooted in motor programming skills. However, D is having most difficulty in the naming condition, and to some extent word repetition, suggesting that his speech processing deficit may lie in the inaccuracy of his stored lexical representations (existing motor programmes). In this regard he is similar to the case presented by Bryan and Howard (1992) whose speech skills for new words were better than for familiar words. By comparing C and D’s performance with that of younger children, it can be seen that D is showing a similar profile of performance to that of younger children aged 3 years (see Figure 5.5). Comparing C and D to each other is also a reminder of the individual variation present in children with speech difficulties and the need for individually designed intervention programmes for children with persisting speech difficulties in particular (see Book 3 in this series, Pascoe, Stackhouse & Wells, 2006). It can also be informative to examine how word length affects a child’s speech production accuracy. Z scores were calculated (see Chapter 2 for how to do this) to compare C to the mean performance of typically developing 6-year-old children on words of different length in

128


Figure 5.5 Bar chart showing number of responses correct on tasks of naming, word and non-word repetition for C, D and for groups of typically developing 3- and 6-year-old children.

each task (see Table 5.14). Examination of these scores confirms difficulty with non-word repetition across all word lengths. The z score for three- to four-syllable non-words is less discrepant from the normative data than for shorter non-words; this does not reflect less difficulty for C in these

Table 5.14 Comparison, using z scores, between C’s performance and that of typically developing 6 year olds on naming, word repetition and non-word repetition, using stimuli at three word lengths. Task

Naming

Word repetition

NWR

Word length

1 syllable 2 syllables 3–4 syllables 1 syllable 2 syllables 3–4 syllables 1 syllable 2 syllables 3–4 syllables

6 year olds

C

Mean (s.d.)

Raw score

Z score

18.35 (1.31) 17.5 (1.5) 15.5 (3.07) 18.65 (1.35) 17.75 (1.68) 17.55 (1.99) 16.7 (1.22) 16.05 (1.23) 15.0 (2.7)

12 9 3 13 12 1 8 7 1

2.27 5.89 4.38 4.91 3.52 8.44 7.13 6.78 5.82


129

longest words, rather the greater variation in the typically developing group, as can be seen in the standard deviation. There is also a trend for poorer performance on word repetition of longer words (particularly for three- to four-syllable words) and for longer words on picture naming (both two- and three- to four-syllable words). If C has a difficulty in creating new, accurate motor programmes, it is likely that it will take longer to lay down the more complex motor programmes for longer words than for single-syllable words with a simpler phonological structure. However, there is a dilemma in using such normative data in psycholinguistic assessment. Any standard word list, such as these used in this book, aims to sample a wide range of phonology and phonotactic structures. Children with speech difficulties may have individual patterns of errors that are confined to a small number of phonemes or structures, so that set word lists may not be sensitive enough to uncover a child’s difficulties. The psycholinguistic approach advocates the use of word lists based on individual speech errors to examine a child’s speech processing system more accurately (for example, see Auditory Lexical Discrimination Task 3: Mispronunciation Detection for Individual Children in Chapter 4). If such individual word lists are used to collect data from children with speech difficulties, the patterns of development demonstrated in the normative data presented in this book may be useful as an indication of how to interpret task performance. Conclusions have been drawn about C and D’s speech processing difficulties on the basis of their performance on the single-word speech production tasks. More precise conclusions can be made, and a fuller speech processing profile developed, by also considering their performance on auditory discrimination tasks (see Chapters 3 and 4) and connected speech (see Chapter 6).

Summary • This chapter describes three different types of tasks measuring speech production of single words: picture naming, word repetition and non-word repetition. • Performance on these tasks can be compared to performance on auditory discrimination and auditory lexical discrimination tasks using the same stimuli. • Performance on picture naming is dependent on existing lexical representations and motor programmes to produce the word. • Repeating a familiar word is likely to be based on a stored motor programme for that word. • It is also possible to repeat a real word without going via stored representations, thus treating it like a non-word.

130


• Non-word repetition cannot access stored representations: a new motor programme needs to be assembled in order to produce the novel utterance. • Identifying why a child is having lexical and/or speech difficulties when naming is best achieved by comparing performance across speech tasks, as words can be mispronounced as a result of topdown deficits (e.g. inaccurate motor programmes) and/or bottomup deficits (e.g. oral structural abnormalities). • This chapter emphasises the importance of interpreting the task results in relation to performance on other tasks, and data from children with speech difficulties is used to illustrate this process. • Hypotheses to account for a child’s performance on speech production tasks include: - Real word repetition is better than naming: check lexical representations and retrieval. - Naming is better than real word repetition: check auditory discrimination. - Real word repetition is better than non-word repetition: check motor programme assembly. - Non-word repetition is better than real word repetition: check lexical updating. - Real word and non-word repetition are equally poor: check articulatory skills. • To test these hypotheses it is important to compare the profile of performance across speech output tasks with the profile of performance from typically developing children. • Children with speech difficulties can perform very differently from each other on speech output tasks and these individual differences should be taken into account when planning an intervention. • Observing if/how children attempt to self-correct their speech errors provides important information for both planning and evaluating intervention.

Chapter Six Speech Production: Connected Speech

As children or adults we seldom speak in single words. We usually produce a continuous stream of sounds in which the constituent words are not bounded by silence. Although adults routinely use longer sentences and speak more rapidly than young children, connected speech soon becomes the norm for children who have moved beyond the single-word stage, in typical development usually in the second year of life. Many children with speech difficulties are unintelligible when using longer strings of speech in everyday, spontaneous communication, even though they may be able to produce single words in isolation quite accurately. Despite this common observation, connected speech is not routinely assessed and often is not specifically addressed in intervention. This chapter focuses on the assessment of connected speech. Using connected speech places a greater load on the speech processing system than does the production of single words. However, the challenge of connected speech is about more than just extra processing load: connected speech is also qualitatively different from single words, in terms of its phonology and therefore its phonetics. Connected speech is more than merely a string of individual target segments joined together in series. Each segment in a connected speech string is liable to influence the segments that surround it. The precise form that these influences take is determined by the particular language in question, and so the phonology of connected speech is a part of the phonology of the language that the child has to master, just like its systems of vowels and consonants and its phonotactic structures. In this chapter we examine connected speech from these two different perspectives: processing load and connected speech phonology. In Task 1, children are required to repeat sentences containing specific single words (from the cross-sectional study, described in Chapter 2). This

131

132


is a task where the focus is on the extra load of connected speech and its effect on the accuracy of production of a word when compared to the word produced in isolation. Tasks 2 and 3 focus on the phonetic and phonological ‘glue’ that binds words together in everyday speech. This glue is, very broadly, of two kinds: prosodic and segmental.

Prosodic Features in the Psycholinguistic Framework Prosodic features can be subcategorised as lexical and supra-lexical. The main lexical prosodic feature in English is word stress. Within the psycholinguistic framework, this is treated as an aspect of the phonological representation and motor programme of a word (see Figure 1.9 in Chapter 1) and may be manipulated systematically in tests of single words. More relevant in the present context are supra-lexical prosodic features, in particular intonation. One of the key functions of intonation is to group words into cohesive utterances – an ability that in normal development starts to emerge at the onset of two-word utterances (Branigan, 1979; Corrin, Tarplee and Wells, 2001) and continues to be refined up to age 8 and beyond (Dankoviˇcová et al., 2004). This, along with other functions of intonation (e.g. to express affect, focus and interactional meanings), can be tested in a systematic way within a psycholinguistic framework, for instance using the PEPS-C procedure, a prosodic assessment battery devised within the psycholinguistic framework. In Chapter 11 of Book 2 of this series (Stackhouse & Wells, 2001) case studies of the intonation abilities of two children with speech and language impairments are presented. Group studies of typically developing children and children with speech and language impairments have also been carried out using this battery (Wells & Peppé, 2003; Wells, Peppé & Goulandris, 2004). A computerised version of this battery has also been developed (Peppé & McCann, 2003; and see http://sls.qmu.ac.uk/ssrc/ProsodyinASD/index.htm As PEPS-C is already in the public domain, in this chapter the focus is on segmental juncture features.

Segmental Juncture within a Psycholinguistic Framework Segmental juncture features are defined here as features that characterise segments across word boundaries. It has been found helpful as a first descriptive step to make a broad distinction between open juncture and close juncture (cf. Wells, 1994; Newton & Wells, 2002). When a speaker of English produces two words in sequence, there may be features that serve to keep the words distinct (Shockey, 2003), such as a silence; an audible release of the final stop in the coda of the first word; a glottal stop

Speech Production: Connected Speech

133

at the onset of the second word if it begins with a vowel. Adult speakers may deploy such open juncture for the purposes of emphasis or repair and it may be more frequent in certain connected speech activities, such as reading aloud. Close juncture, on the other hand, is characterised by phonetic features that serve to bind adjacent words together. How this can be done depends on the phonological structures that abut at the juncture. Connected speech processes are discussed in greater detail in this chapter as they relate to the specific tasks presented. Newton and Wells (2002) studied a subset of close junctures in the spontaneous speech of a normally developing boy (CW) learning Southern British English, between the ages of 2;4 and 3;4. One juncture type studied was -C C-, where -C is /d/ or /n/, and where C- is target velar or bilabial, for example RED BALLOON, ONE CALLED. These are sites for alveolar-to-bilabial and alveolar-to-velar assimilation in adult English. The earliest recordings showed examples of the close juncture (assimilated) forms, but no open juncture forms. As the year progressed, open junctures also appeared, but close junctures (assimilation) were always the majority. This result suggests that this child did not learn to join the two words together phonologically having first learnt to combine them grammatically, but that junctural phonology and grammar had emerged simultaneously. This has some important implications for our understanding of phonological development. It suggests that some children do not start out by learning words then learning how to combine them; instead, the child learns and reproduces holistic chunks that embody a single grammatical or interactional piece. See Chapter 7 in Book 3 of this series (Pascoe, Stackhouse & Wells, 2006) for a more detailed description of processes within connected speech. This chapter includes the following tasks: • Connected Speech Task 1: Word-in-Sentence Repetition (Vance, Stackhouse & Wells, 1995) • Connected Speech Task 2: Connected Speech Processes (CSP) Repetition (Newton, 1999) • Connected Speech Task 3: Final Consonant Juncture Repetition (Pascoe, Stackhouse & Wells, 2005)

Connected Speech Task 1: Word-in-Sentence Repetition (Vance, Stackhouse & Wells, 1995) Aims • To examine speech production of target words in connected speech. • To compare performance on this task with naming and repetition of the same words produced in isolation (see Chapter 5, full version of Picture Naming Task 1 and Word Repetition Task).

134


Description of the Task The child is shown a picture that contains a representation of the target stimuli and is asked to repeat a sentence about the picture spoken by the investigator. Use of the Task Children may be able to produce words in isolation better than in connected speech. For older children in particular it is possible to miss the presence of subtle speech difficulties that interfere with the child’s communication and learning (e.g. spelling) if assessment is only carried out at the single-word level (e.g. see the case of Richard in Book 1 of this series, Stackhouse & Wells, 1997, pp. 124–138). Further, children with speech difficulties can learn to produce the single-word stimuli targeted in therapy, but do not necessarily transfer their new skills to connected speech (Pascoe, Stackhouse & Wells, 2005). It is therefore important to add connected speech assessment as a routine part of the investigation of speech difficulties in children. This task allows comparison of the production of words in sentences with the production of the same stimuli in isolation (for example, see two cases of children with apraxia of speech in Stackhouse & Snowling, 1992). The pictures can also be used to elicit the target words in spontaneous speech for comparison with single-word production and production in sentence repetition, although no normative data is available for this measure. Psycholinguistic Analysis of the Task Although this task requires the child to repeat a longer stretch of speech than on the single-word repetition task, it does not necessarily require the child either to understand the sentence or to generate a grammatical construction. Processing is, therefore, at the same level on the speech processing profile as word repetition (Level I). If spontaneous production of sentences were being assessed, processing would be at a higher level, alongside naming (Level G), since the child would need to draw on previously stored grammatical knowledge. The sentence repetition task is more challenging than single word repetition in that the child has to remember a longer utterance and plan more articulatory output. However, grammatical knowledge, semantic context and the presence of a pictorial representation of the sentence will support performance on the task, through top-down processing. Sentence repetition is particularly difficult for children with specific language impairment, and may be a clinical marker for language impairment (Conti-Ramsden, Botting & Faragher, 2001). It is also used as a ‘sentence memory’ task to investigate the interaction between short-term


135

memory and existing language knowledge (Alloway & Gathercole, 2005). As sentence repetition is, therefore, vulnerable to language and short-term memory constraints, it is important to use short, simple sentences when the aim of the assessment is to measure speech production specifically. ‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – No. Level on Speech Processing Profile • Level I Stimuli A sentence was constructed for each of the 60 word stimuli used across the range of tasks in the cross-sectional study (see Table 2.2). The word was placed in one of the following sentence structures: Subject Verb Object; Subject Verb Complement; Subject Verb Adverb (see Table 6.1 for a full list of sentences). The target stimulus was located in non-final, non-tonic position in almost every case in order to differentiate this condition from single real word repetition, for example: • •

UMBRELLA: HIS UMBRELLA IS YELLOW LADDER: THE LADDER IS BY THE HOUSE

The tonic syllable is the final lexically stressed syllable of the sentence; that is, ‘yel—’ and ‘house’ in the examples above. This means that the target word, which is located earlier in the sentence, never carries the tonic. This provides a contrast to the single-word repetition task, where the tonic has to be on the target word. The target word is always followed by ‘is’, ‘has’ or ‘a’ to avoid any assimilation resulting from the consonant in the onset position of the following word influencing production of the coda of the target word. This is in contrast to the stimuli for the other two connected speech tasks described in this chapter. Each sentence is presented with an intonation pattern that is unmarked for declarative statements in English. Fifty-nine pictures represent sentences which include the 60 stimulus words used in the full versions of the high-frequency real word repetition tasks. Some sentences and pictures are designed to be amusing or absurd.

136


Table 6.1 The sentence stimuli used in Connected Speech Task 1: Words in Sentences, target word in capital letters (used in the cross-sectional study) Sentences Look, the MOUSE jumped out. The AEROPLANE is flying away. His PARACHUTE is green and blue. The COMPUTER is on. The BISCUIT has bites in it. The TELEPHONE is like a banana. His PYJAMAS are stripey. The LADDER is by the house. The ROOF is high. His UMBRELLA is yellow. Look, the SPIDER is hanging there. The TOILET is green. The TORCH is bright. The SAUSAGE is for the boy. The SPAGHETTI is for the boy. His PLATE is empty. The GLOVE is big. The KNIFE is for cutting. The JELLY is wobbly. The YELLOW hat has a bobble. His WATCH is on his leg. The MONEY is in his hand. The CAT is on the bed. The FISH is in the bowl. The TELEVISION is on. The HAIRDRESSER is cutting his hair. Her BRUSH is blue. The HELICOPTER is rescuing the clown. The TRAIN is going very fast. The KANGAROO is riding down the road. His SCOOTER is green. The TRACTOR is pulling very hard. The VAN is stuck in the mud. The ELEPHANT is too big. The CARAVAN is too small. The CROCODILE sings and plays. The GUITAR has a broken string. The DUCK has caught a fish. He’s FISHING with a twig. The SLIPPER is old. The SOCK has a hole. The DUSTBIN is very full. The PARROT is sitting in a tree. A LEAF falls off. A FEATHER is on the ground. The ROUNDABOUT has horses. The SEESAW is long.


137

Table 6.1 (Continued) The CATERPILLAR is sitting at one end. The BUTTERFLY is sitting at the other end. The HAMBURGER is tasty. The FLOWER is pretty. The CHAIR is broken. The BOOK has a good story. His THUMB has a bandage on. The HOSPITAL is near his house. The SNAKE is in the KITCHEN. The SANDWICH is ready to eat. The DINOSAUR is having a bath. The SPONGE is full of bubbles.

Materials • pictures (on CD-Rom). • Score sheet (Appendix F.1). Procedure The child is told that he or she will see some pictures and hear some sentences to go with the pictures. The child is asked to repeat the sentence associated with a picture while looking at it. One repetition of the sentence by the investigator is allowed if the child requests this or was not attending. All responses should be audio-recorded using a good-quality recorder and microphone as well as transcribed live in phonetic script. The transcription can be checked after the assessment from the audio recording. Scoring Only the production of the target word is scored; that is, not the entire sentence. The production of the consonant sounds in each response is examined. For each word to be scored as correct the production of the consonants within the word should be an accepted adult realisation, taking into account regional accent variation. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any target words in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Normative Data Data was collected from typically developing children aged 3 to 7 years. As with naming and real and non-word repetition, children’s accuracy of

138


Table 6.2 Mean (standard deviation) correct scores for the Connected Speech Task 1: Words in Sentences (data from the cross-sectional study) Age group

1 syllable N 20 Mean (s.d) 2 syllables N 20 Mean (s.d) 3–4 syllables N 20 Mean (s.d) Total words N 60 Mean (s.d)

3 years

4 years

5 years

6 years

7 years

(N 16)

(N 19)

(N 20)

(N 20)

(N 20)

15.53 (2.5)

16.16 (2.18)

18 (1.18)

18.55 (0.86)

18.8 (1.03)

13.12 (2.85)

15.89 (2.17)

17.25 (1.51)

17.35 (1.53)

18.3 (1.19)

8.59 (4.07)

12.58 (3.22)

15.75 (2.45)

15.85 (2.95)

17.6 (2.31)

37.24 (8.91)

44.90 (6.69)

50.5 (4.84)

51.25 (4.82)

54.2 (4.29)

repetition of words in sentences increased with age and correlated with performance on the other speech production tasks. Children’s performance was affected by the number of the syllables in the words, with less accurate production of longer words, particularly in the younger children (see Table 6.2). Questions to Consider • How does performance on this task compare to single-word repetition using the same stimuli (full version of repetition task in Chapter 5)? • How does performance on word repetition (see above) and the Connected Speech Task 1: Words in Sentences compare to performance on production of the stimuli on the naming task (Picture Naming Task 1: full version, Chapter 5)? • If data is available for production of these words when used spontaneously in connected speech, are they produced more or less accurately than in sentence repetition? Other Assessments that Include a Measure of Speech Production in Connected Speech • Diagnostic Evaluation of Articulation and Phonology (DEAP; Dodd et al., 2002) • Goldman-Fristoe Test of Articulation 2 (Goldman & Fristoe, 2000) • Phonological Process Analysis (Weiner, 1979)


139

Connected Speech Task 2: Connected Speech Processes (CSP) Repetition (Newton, 1999) Aims To assess use of the connected speech processes (CSPs) of • Assimilation: where word final alveolar /t, d, n/ assimilate in place of articulation to following word-initial bilabial or velar oral and nasal stop consonants. For example, RED BALLOON: [ εbblun]; ONE MISSING: [wmmIsIŋ]; RED CAR: [ εgkɑ]. • Elision: where /t/ and /d/ are omitted before a word that begins with a consonant. For example, LOST BERTIE: [lɒsb3ti]; FIND JACK: [fandæk]. • Liaison: where, in certain accents of English, glides /j, r, w, / rather than a glottal stop are inserted at the word boundary between two vowels. For example, TIDY UP: [tadi j p]; SAW A: [sɔ]; SHow uS: [ʃ υwS]. • The adult allomorphs of definite and indefinite articles. For example, AN ORANGE: [n ɒ nd] THE ORANGE: [ðj ɒ nd]

Description of the Task A set of 42 sentences is presented to the child. The child’s repetition of each sentence is recorded. No pictures are involved. Use of the Task An earlier version of the task was administered as part of a cross-sectional study on the use of connected speech processes by typically developing children, aged 3 to 7 years (Newton & Wells, 1999). There were 14 children aged 3 and 20 in each of the following age bands: 4, 5, 6 and 7 years. All children were from Hereford, in the west of England. As part of this study, the same connected speech processes were also analysed in a sample of spontaneous conversation from each child, and from a structured retelling of five short stories, using pictures that were designed to target the processes. There was some difference between three types of data elicitation: the highest proportion of CSPs was found in spontaneous speech and the lowest in story retelling, with sentence repetition performance lying between the two. For ease of administration and standardisation, it was therefore decided to use and recommend the sentence repetition task alone for assessment purposes. However, when interpreting scores on this task two points should be borne in mind. First, a score on the CSP Repetition Task may underestimate the occurrence of CSPs in the child’s spontaneous speech. This could

140


% adult-like CSPs produced

100

80

60

40

20

0 t#

n#

d#

#sh

Ct#C

Cd#C

j

w

r

process type 3-7 yr olds

Percy

Edward

James

Figure 6.1 Percy, Edward and James’ use of connected speech processes for specific phonemes when compared to typically developing children Key. t#, n#, d#, #sh – assimilation processes; Ct#C and Cd#C – elision processes; j,w,r – liaison processes (# word boundary)

be checked by recording a spontaneous speech sample and examining the child’s use of CSPs in target environments. Second, the CSP Repetition task does not tap into exactly the same set of psycholinguistic processes as spontaneous speech (see the psycholinguistic analysis of the task below). Newton (1999) used the present CSP Repetition task to investigate the use of CSPs by three children. Percy, Edward and James were all aged 11–12 years, had severe speech difficulties, and attended a special residential school. The present version is slightly simplified from the earlier version, in terms of length of sentence and syntactic complexity, however the phonological CSP targets remain the same. The performance of the three children with speech difficulties is compared to that of typically developing children aged 3 to 7 years, for specific CSPs, in Figure 6.1. From the data presented in Figure 6.1 it can be seen that each child displays some differences from the typically developing group. The typically developing children only produced adult-like open juncture (that is, emphatic speech, for example as for when reading aloud) or close juncture (for example assimilation), while Percy, Edward and James all produced atypical realisations too. One striking feature is the comparative dominance of open juncture, for example for liaison. This is particularly shown by the low scores for liaisons of ‘w’ and ‘r’ compared to the typically developing group. The three children may also differ considerably from one another in their realisation of a particular CSP, for instance the difference in the use of /j/ liaison by Percy and Edward compared to James. This illustrates how the CSP Repetition task can be


141

used to create a profile of an individual child’s use of CSPs, and to compare him or her to the typical population. In addition to the quantitative scores based on perceptual measures, Newton carried out more detailed phonetic analysis of the children’s productions of CSPs in the repetition task, using electropalatography and acoustic analysis. Such qualitative analysis can indicate in more detail the articulatory strategies and constraints of the individual child (cf. Howard, 2004).

Psycholinguistic Analysis of the Task This task particularly taps the motor planning stage of speech output, as it addresses the way in which a child is able to join up words to produce a connected utterance. If a child is able to repeat in isolation each of the two words involved in the juncture with accurate coda of the first word and onset of the second, for example RED [ εd] CAR: [kɑ], yet is unable to repeat accurately the juncture targeted in the task (the assimilated form RED CAR: [ εgkɑ]), this may indicate that the child has a problem at the motor planning stage (see Book 1, Stackhouse & Wells, 1997, pp. 226–228 for further discussion). As the task involves sentence repetition, the route through the speech processing model is as for ‘real word repetition’ (see Figure 5.3, Chapter 5). Like all repetition tasks, this task involves input processing, so failure to apply CSPs could be due to failure to process the input. One way of establishing whether the breakdown is at the input stage or at the motor planning stage would be to compare performance on this task (addressing Level I of the speech processing profile) with performance in spontaneous speech, or picture description (see the section on use of the task, above), both Level G tasks with no speech input processing. If the child’s realisation of CSPs is much greater in spontaneous speech than in Connected Speech Task 2: CSP Repetition, this may indicate a deficit on the input side (phonological recognition on the ‘box-and-arrow’ model). However, it could also indicate that the child is sensitive to the ‘test’ situation and is attempting to produce the required items with maximum clarity: typical speakers frequently suspend the application of CSPs in the interests of emphasis or clarity. Conversely, if the child’s realisation of CSPs is much greater on this repetition task than in spontaneous speech, this may indicate that, although the child can perceive the juncture (input processes intact) and execute the juncture, possibly bypassing the phonological representation (motor execution intact), the motor programmes for the words involved in the juncture do not contain the information needed to realise close juncture at the motor planning stage. For further research on children’s input processing of CSPs, in particular assimilation, see Loucas and Marslen-Wilson (1998).

142


‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – No. Level on Speech Processing Profile • Level I Stimuli The set of stimuli consists of 42 sentences that target four connected speech processes, specifically: • Assimilation to bilabial velar place of /t/, /d/, /n/ and assimilation to post-alveolar place of /s/, /z/ • Elision of /t/ and /d/ • Liaison with /w/, /j/ and /r/ • Prevocalic allomorphs of indefinite article /n/ and definite article /ði/ Each sentence is a maximum of seven words long and is presented with an intonation pattern that is unmarked for declarative statements in English. The tonic syllable is the final lexically stressed syllable of the sentence. This means that the target CSP site – that is, the two words containing the target juncture, located earlier in the sentence – never carries the tonic. This is to avoid the situation where the CSP site receives intonational emphasis, as this might be expected to lead the speaker to use open juncture rather than the CSP. The full list of sentences is included in Appendix F.2. Materials • Score sheet (Appendix F.2). Procedure The investigator says each sentence once only with the application of CSPs and with the prescribed intonation pattern. The child is asked to repeat each sentence just as it was spoken. This instruction is important: in a more formal or careful style, speakers may use fewer CSPs. Where possible, the stimuli should be presented in an accent that is close to that of the child. This is because there are some differences in CSPs according to regional accent: for example, South African English does not use the liaison form of the definite article. If the investigator is


143

not confident of being able to produce the stimuli accurately in real time, it is advised that the stimuli should be pre-recorded and presented from tape or disc. Scoring Only the production of the two-word CSP environment (underlined in the score sheet) is scored. The child’s production is transcribed phonetically. This is then coded according to the type of realisation produced: • c for production of adult-like close juncture • o for open juncture • n for a non-adult-like realisation of the juncture Investigators should refer to the descriptions of adult-like close and open juncture for the different process types in the section on segmental juncture at the beginning of this chapter and in the section on aims at the beginning of this task description. An important feature of normal adult connected speech to note when scoring a child’s repetitions is the common realisation of word-final /t/ as a glottal stop: for example FAT : [fæ], so that FAT PIG may be as likely produced [fæpIg] as [fæppIg]. Therefore a similar production of wordfinal /t/ as [] in assimilation environments should be coded as an adultlike close juncture. The number of adult-like junctures for each different connected speech process is recorded, and a percentage correct score for each of these is calculated. In addition, the number of open and non-adult-like junctures is noted, as qualitative analysis may provide a greater insight into an individual’s approach to these juncture types. There may be patterns in the realisation of certain juncture types; for example, Newton (1999) reports that Edward omitted entire word-final consonant clusters in the elision environment CC#C). (# word boundary). Normative Data Of the four processes investigated, only one showed a developmental trend: the number of correct allomorphs of the indefinite and definite articles (e.g. AN ORANGE/THE ORANGE) increased progressively through the age bands (see Figure 6.2). The incidence of each of the other processes (assimilation, elision and liaison) remained fairly consistent across the age range, at between 70 and 90% (see Table 6.3). Questions to Consider • How does performance on this task compare with spontaneous speech production?

144

Compendium of Auditory and Speech Tasks % adult-like articles produced

100 90 80 70 60 50 40 30 20 10 0

3-yr-olds

4-yr-olds

5-yr-olds

6-yr-olds

7-yr-olds

age group indefinite

definite

Figure 6.2 Percentage of adult-like articles produced by children aged 3 to 7 years

• Would further useful insights be provided by going beyond the task to carry out more detailed phonetic analysis, for example qualitative analysis of the child’s errors, possibly including the use of instrumental techniques such as electropalatography (EPG; e.g. see Newton, 1999; Howard, 2004)? Table 6.3 Percentage use of adult-like connected speech processes by typically developing children (# word boundary) Age group

Assimilation t# n# d# /#ʃ/ Elision Ct#C Cd#C Liaison j w r

3 years

4 years

5 years

6 years

7 years

(N 14)

(N 20)

(N 20)

(N 20)

(N 20)

93.19 76.85 46.77 52.38

91.08 75.18 41.79 78.03

91.62 75.44 29.78 81.67

91.57 77.48 38.1 74.16

92.40 80.43 43.18 83.83

94.21 76.33

84.14 71.72

79.23 61.76

84.54 59.83

86.94 72.63

68.24 87.04 88.81

89.13 91.51 90.26

81.17 92.80 87.67

88.44 93.47 88.36

91.49 95.35 86.15


145

• What does performance on this task suggest about the nature of the child’s phonological representations and his or her motor planning abilities? • How might the child’s ability to realise CSPs affect intelligibility and fluency? Other Assessments of Connected Speech Processes Procedures described in the following research papers can be adapted to investigate connected speech processes: Howard (2004); Howard, Wells & Local (2008); Loucas & Marslen-Wilson (1998).

Connected Speech Task 3: Final Consonant Juncture Repetition (Pascoe, Stackhouse & Wells, 2006) Aims • To assess production of word-final (coda) consonants in a connected speech environment. • To compare this with production of final consonants in words pronounced in isolation. Description of the Task The child repeats short, phonologically controlled phrases or sentences: for example, ‘The X in the picture’ where X represents the target word. The phrases are graded in terms of juncture complexity. The child also repeats the same list of words in isolation for comparison. Use of the Task This task was constructed for initial assessment and ongoing evaluation of intervention for Katy, aged 6;5 at the start of the study (Pascoe, Stackhouse & Wells, 2005). She had severe and persisting speech difficulties. The aim of the research was to determine if psycholinguistically based intervention could result in improvements in her speech production. One of the most predominant patterns of phonological simplification in Katy’s speech was final consonant deletion, occurring 96% of the time. Katy’s percentage of consonants correct for connected speech overall was 25%, similar to that of her production of single words. Her speech rate was typically slow, with many pauses between words and many single words emphasised with primary stress. Detailed psycholinguistic analysis suggested that there were deficits at a number of levels, including motor

146


planning. Motor planning is considered to involve phonetic aspects of speech production, moving beyond the abstract linguistic knowledge of the previous stages in the speech production process. It is at this level that coarticulation comes into play. While motor programming is conceived as being at a single-word level, motor planning involves the connection of words into strings of speech. Phases I and II of the intervention focused on developing accurate motor programmes for words with final consonants. However, children do not generally talk in single words, and the ultimate aim was for final consonants to be used in CVC words embedded in sentences, implicating motor planning, and this was addressed in Phase III of the intervention (see Chapter 8 in Book 3, Pascoe, Stackhouse & Wells, 2006). Katy’s connected speech did not improve until it was specifically addressed in therapy. It is likely that, as a result of the Phase III intervention, Katy’s attention shifted to the production of larger units of speech that involve motor planning, rather than being focused at the single-word level. Other children might also benefit from this type of specific intervention targeted at connected speech. Psycholinguistic Analysis of the Task As with the previous connected speech tasks, this task particularly taps the motor planning stage of speech output, as it addresses the way in which a child is able to use final consonants to produce a connected utterance. A child (such as Katy after the first two phases of therapy) may be able to repeat in isolation, with accurate coda (final consonant), the target words involved in the juncture; yet be unable to repeat them accurately in a connected speech frame. This indicates that the child has a problem at the motor planning stage. Another child may have the reverse pattern, being unable to pronounce the final consonant when repeating a word in isolation, but being able to do it before a following word beginning with a vowel: for example, a child may produce [pɑ] for PART in isolation, but be able to say ‘part of the puzzle’. This indicates that the final consonant is in the child’s phonological representation and in the motor programme for the word, but that he or she has a general phonotactic constraint on motor plans that blocks the realisation of a consonant unless it is followed by a vowel (i.e. the child has a CV utterance ‘template’). As for Connected Speech Tasks 1 and 2, the sentence stimuli are likely to be processed as for word repetition. Like all repetition tasks, input processing is involved and this could be implicated, if the child is failing on this task. In order to establish whether this is the case, it is possible to administer an auditory discrimination task using words from one of the lists: for example, a same or different task using sentences ‘The BOAT in the picture’ vs ‘The BOW in the picture’.


147

‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes. • Does the child have to access these lexical representations in order to complete the task? – No.

Level on the Speech Processing Profile • Level I

Stimuli The stimuli are based on 16 CVC or CCVC words. These were chosen to highlight the functional importance of final consonants, so that when the final consonant is removed, a CVC vs CV minimal pair is created (e.g. BOAT/BOW). The vowel nucleus is therefore either a long vowel or a diphthong. A graded hierarchy of phrases was devised using each target word, moving from a facilitatory context to a more demanding one. For example, in the case of the target word ROPE the facilitatory sentence used as a starting point (Level 1) was THIS ROPE PULLED THE CAR, where the onset consonant of the following word PULLED is the same as the coda consonant of the target word ROPE. Children using final consonant deletion may be able to produce the initial [p] in PULLED even if they omit the final [p] in ROPE. In order to achieve an acceptable realisation of this final consonant, the child merely has to lengthen the closure phase for the (single) consonant articulation. At the next level (neutral level, Level 2), the child is required to produce a sentence such as THERE’S ROPE ON THE ROAD, with the target ROPE being followed by a vowel. Most challenging for children are Level 3 sentences such as THIS ROPE GOT FRAYED where they must change the place of articulation (and voicing) between the final [p] in ROPE and the following consonant [g]. The Level 3 sentences varied in terms of the normal adult-like patterns of assimilation that would be expected. For example in THIS ROPE GOT FRAYED no assimilation between the hetero-organic consonants at the juncture (/p/, / /) was expected. However, in a sentence like THIS NOTE CAN’T BE READ one would expect assimilation of the final /t/ in NOTE, for example /ðs nəυk kɑm bi rεd/. A distinction is made in Level 3 between sentences where assimilation would and would not be expected. The full list of graded phrases is given in Table 6.4. They can be used to assess a child’s ability to produce final consonants in these four different juncture environments.

148


Table 6.4 Graded phrases used in Connected Speech Task 3: Final Consonant Juncture Repetition Test (Pascoe, Stackhouse & Wells, 2006) Target Word

Sentences

1

Note

2

Plane

3

Heart

4

Nail

5

Cage

6

Slide

7

Wheel

8

Rake

9

Stork

10

Leaf

11

Sauce

12

Ice

13

Soap

14

Pipe

15

Barn

16

Road

1: This NOTE teaches father Xmas 2: There’s a NOTE under the table 3ii: This NOTE can’t be read 1: The PLANE knocked it 2: There’s a PLANE in the sky 3ii: This PLANE must be loaded 1: This HEART tastes nice 2: There’s a HEART on my jumper 3ii: This HEART can break 1: This NAIL looks pretty 2: There’s a NAIL in the wood 3i: This NAIL got painted twice 1: The CAGE joined my class 2: There’s a CAGE on the bed 3ii: The CAGE got stolen 1: This SLIDE dumped me 2: The SLIDE in the park is nice 3iii: The SLIDE bounces him 1: The WHEEL looks broken 2: There’s a WHEEL on the bike 3i: This WHEEL got fixed 1: This RAKE cost ten pounds 2: There’s a RAKE on the ground 3i: This RAKE takes a bath 1: The STORK carries a baby 2: There’s a STORK on the log 3i: This STORK teaches swinging 1: The LEAF feels wet 2: The LEAF is in the air 3i: This LEAF got torn 1: The SAUCE seems nice 2: There’s SAUCE in the jar 3ii: The SAUCE shouldn’t burn 1: This ICE seems cold 2: The ICE is in the bucket 3ii: This ICE should melt 1: This SOAP pulled my hair 2: There’s SOAP on the towel 3i: This SOAP got soft 1: The PIPE pushes through the roof 2: There’s a PIPE on the wall 3i: This PIPE curls round the floor 1: The BARN needs painting 2: There’s a BARN on the farm 3ii: The BARN must be cleaned 1: The ROAD divides the hill 2: There’s a ROAD over the river 3ii: This ROAD brings us home

Key. 1 Facilitatory sentences; 2 Neutral sentences; 3i Challenging sentences, without assimilation; 3ii Challenging sentences with assimilation (marked **).


149

Materials • Score sheet (Appendix F.3). Procedure The child is asked to repeat each sentence just as it was spoken. There are two practice items in which feedback can be given to the child. Level 3ii items should be produced with the final consonants of the target words assimilated into the following consonant, as in the example given above: THIS NOTE CAN’T BE READ presented as /ðs n υk kɑm bi rεd/. Where possible, the stimuli should be presented in an accent that is close to that of the child’s. Each item is presented with an intonation pattern that is unmarked for declarative statements in English. The tonic syllable is the final lexically stressed syllable of the sentence or phrase. The tester may prefer not to present the stimuli in real time and to pre-record the stimuli on tape or disc. Following the sentence repetition task, the child’s ability to produce the same words in isolation is assessed, also through repetition. The child is asked to copy the investigator’s production of the single words.

Scoring The child’s production of the coda of the target words is scored. Two points are awarded for final consonants that are accurately realised; one point for using an inaccurate final consonant; and no points for omission of a final consonant. The scoring of Level 3 is slightly more complex. For items marked for assimilation i.e. 3ii, the target final consonant is scored as correct, and given two points, if the consonant is assimilated and realised as perceptibly long (as in the target), for example [ðs n υk kɑm bi rεd]. One point is awarded if the child realises the final consonant in unassimilated form, for instance [ðs n υt kɑm bi rεd]. Both these are possible forms in English, but the latter is not an accurate imitation of the target. If the child omits the final consonant (for example [ðs n υ kɑm bi rεd] with no perceptible lengthening of the [k], no points are awarded. Level 3i items without assimilation are scored as for the Level 1 and 2 items. Qualitative observation of how the child produces these items may offer more insights than the scores awarded. For example, if a child generally has a high percentage of final consonants correct, but omits final consonants on the ‘ assimilation’ items, it suggests that he or she is beginning to apply the adult connected speech process of assimilation. However, if a child generally has a low percentage of final consonants correct, the omission of final consonants on the ‘assimilation’ items is likely to be further evidence of final consonant deletion; that is, an immature process.

150


Normative Data There are currently no norms available specific to this task. However, general developmental norms (Grunwell, 1989; James, 2001) suggest that typically developing children are using final consonants consistently by 3;6–4;0 years. Data from Newton and Wells (1999), as described above, indicates that children can produce adult-like assimilations of final consonants from the age of 3 years. Questions to Consider • How does a child’s performance on the sentence and single-word parts of the task compare? • If the single words are used spontaneously, are they produced more or less accurately than in sentence repetition? If there is a discrepancy, then it may suggest that the child’s motor programmes have not been definitively updated to manage production of the final consonants in a connected speech environment. Children may go through an intermediate phase in which they have sufficient motor planning and motor programming ability to produce the final consonants in repetition tasks, but the skills are not yet automatic enough to cope with the demands of spontaneous speech production. • Would going beyond the task to carry out more detailed phonetic analysis, for instance using acoustic analysis (for example see Matthews, 2005), provide further useful data about covert contrasts in a child’s speech? Omission of final consonants by a child may not be an ‘all-or-nothing’ act – rather, it may be viewed on a continuum. Evidence of the final consonant may not be immediately perceptible to the ear but may become apparent through acoustic analysis, which may reveal lengthening of the vowel or appearance of a glottal stop.

Summary • Production of target words in connected speech can differ from production of the same word in isolation. • Typically developing children use connected speech processes early in development. • Children with speech difficulties may need help to ‘glue’ words together appropriately in connected speech. • This chapter presents three tasks of connected speech. – Connected Speech Task 1: Words-in-Sentence Repetition asks the child to repeat sentences containing the same single words


• •

•

•

•

•

151

as used in the Naming Task 1 and the Word Repetition Task from Chapter 5. – Connected Speech Task 2: Connected Speech Processes (CSP) focuses on the connected speech processes of assimilation, elision, liaison and production of articles, examining these processes using a sentence repetition format. – Connected Speech Task 3: Final Consonant Juncture Repetition focuses on production of final consonants in a graded hierarchy of connected speech environments. Each of these tasks focuses specifically on motor planning, requiring the child to join up words to produce a connected utterance. Sentence repetition tasks tap Level I of the speech processing profile, the same as for single-word repetition. However, sentence repetition is more challenging than single-word repetition in that the child has to remember a longer utterance and plan more lengthy articulatory output. In addition, connected speech tasks tap into the child’s knowledge of those aspects of the phonology of the language that relate specifically to connected speech processes such as assimilation and liaison. Comparisons of results from these connected speech tasks can be made with a child’s performance on single-word repetition tasks incorporating the same or similar items. Comparisons of results from these repetition tasks can be made with the child’s production of similar targets in spontaneous connected speech. Connected speech is often not routinely assessed in children with speech difficulties, but assessing it in a systematic way may offer important insights for intervention.

Chapter Seven Speech Accuracy, Rate and Consistency

Children with speech difficulties often have problems performing tasks that require them not only to produce speech accurately but also to produce it consistently on more than one occasion and/or at speed. Some children’s speech difficulties may arise from a structural abnormality (e.g. cleft lip and palate) or neurological dysfunction (e.g. as in cerebral palsy) that will directly affect the movement of the articulators. Others may have no obvious reason for limited or inconsistent speech production. One aim of a speech assessment is to decide if the child has sufficient movement to produce sounds and sequences of sounds as a foundation for sound production in spontaneous speech. Often a child is asked to repeat a sequence of syllables comprising the same sound, for example pe-pe-pe, or a sequence where the place of articulation varies, for example pe-te-ke. He or she may be asked to repeat the sequence as fast as possible to assess accuracy at speed. This comprises a commonly used assessment of children with speech difficulties – the investigation of diadochokinetic (DDK) rates. Performance on such tasks has been used to differentiate between types of speech difficulties such as dysarthria vs dyspraxia vs phonological impairment with varying degrees of success (Milloy & Morgan Barry, 1990; Maasen et al., 2004). However, in spite of frequent reference to the importance of such tasks in the diagnostic process, there has been little information about the design and scoring of such tasks or how typically developing children perform on them. Studies carried out by Williams and Stackhouse (2000) and Habgood (2000) explored the performance of typically developing children between the ages of 3 and 7 years on a range of DDK tasks comprising oral movements, repetition of syllable sequences, repetition of words and repetition of non-words. The tasks were therefore designed to tap different levels of speech output processing. The children repeated each stimulus and then

153

154


produced it again five times in succession. This allowed a measurement of accuracy of a single production of the stimulus in terms of number of consonants produced correctly; consistency of production by rating how different each response was from the original single response and from the other responses within the five productions; and rate of the child’s speech production, measured by the time taken in seconds to produce five repetitions regardless of accuracy. There was a striking amount of variability across the ages at which children could make oral movements and the rate of producing oral movements in sequence, and there was no increase in rate of movement with age. These ‘silent’ oral movements may, therefore, not be a reliable measure on which to base a diagnosis of speech difficulty or an intervention programme. In contrast, measures of accuracy and consistency of repetition of linguistic stimuli were developmentally sensitive, with performance increasing with age. These tasks may, therefore, be more reliable measures for comparison with children with speech difficulties than oral movement tasks. However, generally 3 year olds were much more variable in their performance than older children, and the tasks may, therefore, be most reliable for children of 4 years and over. In repeating sounds in isolation, there was a significant difference between 3 and 4 year olds, though not between 4 and 5 year olds, suggesting that by 4 years of age children can be expected to repeat most speech sounds in isolation. Words, non-words and syllable sequences were also presented, and when children repeated each stimulus only once there was no significant difference in accuracy of production across the stimuli types. This suggests that, in normal development, if a child can perform well with one stimulus type they will perform similarly well on the others. However, when asked to repeat each item five times, 3 and 4 year olds were less accurate in producing non-words than either real words or syllable sequences. By 5 years of age this discrepancy had disappeared. A developmental relationship between accuracy and consistency became apparent. At 3 years of age the children were more consistent than they were accurate in five repetitions of all stimuli types. For example, they may repeat BUTTERCUP as ‘buttertup’ all five times; that is, inaccurate compared to the adult model, but consistently so. By 4 years of age, accuracy and consistency scores were more in line, but the children were less accurate and less consistent with non-words than with real words. By 5 years of age the children’s accuracy and consistency scores were stable across all stimuli types. Thus, accuracy and consistency come together around 4 years of age, particularly on familiar words. Thus, overall a feature of typical development appears to be that young children are very consistent within their own speech system, even if inaccurate compared to an adult model. Children with delayed speech development may use simplifying processes but in a consistent way and be performing like younger children in terms of accuracy of production; this suggests they are in the

Speech Accuracy, Rate and Consistency

155

systematic simplification phase of the developmental phase model presented in Chapter 1. Inconsistent speech output is characteristic of the whole-word phase of this phase model and can be a sign of atypical or arrested development, perhaps implying a more pervasive input and output speech processing difficulty, particularly with motor programming. This distinction between consistency and inconsistency is a useful one to make in practice and has been used to classify children’s speech difficulties. For example, Barbara Dodd describes subgroups of children with speech difficulties as delayed phonological acquisition, consistent deviant disorder and inconsistent disorder (Dodd, 1995, 2005; Dodd et al., 2002) and makes predictions about their performance and educational outcome. Unlike measures of accuracy of production, there was no clear developmental progression in the children’s rates of responses in repeating test items five times; that is, children generally did not get quicker at the tasks as they got older. The mean rate of production for each age group increased as children got older, but there is no significant difference, except for three-syllable items produced significantly more rapidly by 5 year olds than by 3 year olds. Children aged 5 years and over were significantly slower to produce five repetitions of non-words than word stimuli, with rate of producing syllable sequences falling between that of words and non-words. Whatever their age, children approached the tasks in different ways. Some showed care in being very accurate in their production but with low rate scores. Others followed the opposite pattern. Many children became louder as they got faster and organised their repetitions rhythmically by beating the table in time to their speech output or by producing ‘sing-song’ rhythms. Walker and Archibald (2006) also found that articulation rate did not increase with age in their longitudinal study of 16 typically developing children between the ages of 4 and 6 years. These findings combined cast doubt on the worth of comparing children’s rate of production of sound sequences (e.g. pe-te-ke) against age norms in standardised tests. Rather, a child’s performance should be checked to see it falls within a normal range of time calculated across ages. By comparing children with speech difficulties with typically developing children, Williams and Stackhouse (1998) showed that, although the children with speech difficulties demonstrated a similar performance on a picture-naming task (articulation age between 3 and 4 years), they performed quite differently from each other on speech consistency and rate measures. The oldest girl had a chronological age of 8;07 and an articulation age of 4;05. She had difficulties with both the accuracy and consistency of her production. The youngest girl was aged 4;04 with an articulation age of 3;03. Although she was less accurate than typically developing children, her own productions were highly consistent. The third girl was aged 5;03, with an articulation age of less than 3;0. She had difficulties on all three speech output measures – accuracy, consistency and rate – reflecting the pervasive nature of her speech difficulties. Williams and

156


Stackhouse (1998) argued that the first child performed as might be expected of a child with ‘dyspraxia’, the second as a child with specific phonological delay and the third presented a mixture of both dysarthric and dyspraxic features. Thus, although they performed similarly on a naming accuracy task in terms of number of consonants correct, their accuracy/consistency/rate profile discriminated between them and helped to plan appropriate intervention. In order to investigate the nature of speech difficulties in adolescents with Down’s syndrome, Coffield (1994) administered naming, word and non-word tasks (as presented in Chapter 5) to four adolescents with Down’s syndrome in the age range of 13–19 years and to seven typically developing children aged 3–4 years. She extended the naming, word and non-word repetition task procedures by asking the participants to name or repeat the stimulus word/non-word once and then to repeat it again three times in succession. As a result, consistency as well as accuracy could be measured. The typically developing children performed equally well on naming and real word repetition, but made more errors on non-word repetition and as word length increased (also found by Vance, Stackhouse & Wells, 2005). They were most consistent across three productions in the naming condition. A qualitative analysis of errors produced by these typically developing children confirmed that they made similar errors to each other and that these errors would be described as ‘immature’, accounted for by normal simplifying processes. There was consistent production of these errors, with very little variability over the three repetitions of a target. In contrast, the adolescents with Down’s syndrome made a wider range of errors and were inconsistent in their productions, variable in both accuracy and consistency scores. As with the children with speech difficulties, described by Williams and Stackhouse (1998), and children diagnosed as having ‘verbal dyspraxia’ (Stackhouse & Snowling, 1992), the adolescents with Down’s syndrome performed differently from each other across the measures and stimuli types. This chapter includes the tasks used by Williams and Stackhouse (2000) and Coffield (1994). As with the tasks described in other chapters in this volume, it would not be appropriate to present all tasks and all stimuli sets to a single child. Rather, the investigator should select which measures and which stimuli set would be most appropriate for the child. The following tasks are described in this chapter: • Speech Accuracy, Rate and Consistency Task 1: Oral Movements and Silent DDK Rates (from Williams & Stackhouse, 2000) • Speech Accuracy, Rate and Consistency Task 2: Repetition and Spoken DDK Rates (from Williams & Stackhouse, 2000) • Speech Accuracy and Consistency Task 3 (from Coffield, 1994)


157

Speech Accuracy, Rate and Consistency Task 1: Oral Movements and Silent DDK Rates (from Williams & Stackhouse, 2000) Aims • To provide an investigation of lower-level articulatory skills and diadochokinetic rates (DDK). • To enable a comparison to be made between ability to perform silent oral movements and speech production on repetition tasks. Description of the Task The task comprises silent oro-motor tasks (from The Nuffield Centre Dyspraxia Programme, Connery, 1992; Williams & Stephens, 2004) and the repetition of oro-motor sequences. Use of the Task This task provides an investigation of lower-level articulatory skills and extends the use of repetition tasks. The oro-motor assessment is used to confirm if motor execution skills are intact and whether more peripheral articulatory work should be included in a child’s intervention programmes. Psycholinguistic Analysis of the Task The assessment of oro-motor skills does not involve lexical access but taps into the more peripheral motor execution aspects of the speech processing mechanism. ‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – No. • Does the child have to access these lexical representations in order to complete the task? – Not applicable. Level on the Speech Processing Profile • Level K Stimuli The stimuli for this oro-motor task are derived from the assessment in The Nuffield Centre Dyspraxia Programme (Connery, 1992; Williams &

158


Stephens, 2004). These appear in Appendix G.1, and include both single (e.g. lip spreading and tongue protrusion) and sequential lip and tongue movements (e.g. alternating rounded and spread lips; alternating tongue protrusion and retraction). Materials • Score sheet on the CD-Rom (Appendix G.1). • Not supplied: stopwatch timer. Procedure The child is asked to make some movements with his or her lips and tongue. Instructions are given verbally as well as demonstrated, for example ‘Can you stick out your tongue?’. A mirror can be used to provide visual feedback. Two or three of the movements are presented first for practice, with feedback to ensure the child understands what is required. All the movements are then re-presented one at a time and scored. The child is then asked to copy each movement sequence and repeat it five times as quickly as possible. The time taken for the child to complete the five repetitions of each silent DDK sequence is measured in seconds using a stopwatch. Scoring There is one measurement of accuracy and one measurement of rate of production for the oral movements and silent DDK tasks. Accuracy measure Each lip or tongue movement and movement sequence is scored as accurate or inaccurate. A clearly defined tongue tip shape is not essential to score as correct on protrusion or lateral movement tasks, but tongue elevation should only be scored as correct if a differentiated tongue tip was demonstrated (following Ozanne, 1992). The percentage of accurate movements can be calculated. Rate measure The time taken to produce five repetitions of each sequenced lip and tongue movements is measured in seconds. When the child is not able to achieve one or more of the component movements, a score of 0 is given. The mean time is then calculated for the set of lip movement sequences, for the set of tongue movement sequences, and for all movement sequences.


159

Table 7.1 Mean (standard deviation) and range of percentage scores for accuracy measure of oral movements (data from Williams & Stackhouse, 2000 and Habgood, 2000) Age group

% correct Lips Mean (s.d.) Range Tongue Mean (s.d.) Range Lip and tongue Mean (s.d.) Range

3 years (N 10)

4 years (N 10)

5 years (N 10)

6 years (N 10)

7 years (N 10)

92 (10.3) 80–100

94 (13.5) 60–100

98 (6.3) 80–100

98 (6.3) 80–100

98 (6.32) 80–100

77.7 (20.3) 33.3–100

93.3 (9.4) 77.7–100

87.7 (9.7) 77.7–100

89.9 (8.8) 77.7–100

91 (7.1) 77.7–100

82.8 (15.1) 50–100

93.5 (7.8) 78.5–100

91.4 (6.5) 77.7–100

92.8 (6.75) 85.7–100

93.5 (5.3) 85.7–100

Normative Data Table 7.1 summarises the performance of typically developing children aged 3 to 7 years for single movements. A significant proportion of the typically developing children in each age group could not perform the tongue elevation movements: seven out of ten 3 year olds; four out of ten 4 year olds and seven out of ten 5 year olds. This is, therefore, not a skill that necessarily improves with age. In all age groups, the children scored more poorly on the tongue movement tasks than on the lip movement tasks. Table 7.2 summarises the mean time taken, in seconds, for the typically developing children, aged 3 to 7 years, to complete the sequences of Table 7.2 Mean (standard deviation) and range for time taken, in seconds, for five repetitions of sequenced oral movements (data from Williams & Stackhouse, 2000) Age group Time taken in seconds Lips Mean (s.d.) Range Tongue Mean (s.d.) Range Lip and tongue Mean (s.d.) Range

3 years (N 10)

4 years (N 10)

5 years (N 10)

5.77 (1.58) 3.79–9.62

5.76 (2.22) 3.89–13.52

6.26 (1.33) 4.0–8.46

5.62 (2.67 1.6–11.89

5.88 (2.28 2.0–11.18

5.43 (1.83) 2.78–10.21

5.69 (2.22) 1.6–11.89

5.83 (2.23) 2.0–13.52

5.80 (1.66) 2.78–10.21

160


oral movements. Wide individual variation was noted, and there was no clear developmental progression with age, suggesting that, in this age range, rate may not be an important measure to take. Questions to Ask • Is there a lower-level articulatory difficulty that might need addressing in intervention? • What is the relationship between performance on Task 1 and Task 2 in this chapter? Other Assessments of Oro-motor Movement • Diagnostic Evaluation of Articulation and Phonology (Oro-motor Assessment, DEAP; Dodd et al., 2002) • The Nuffield Centre Dyspraxia Programme (Williams & Stephens, 2004) • Paediatric Oral Skills Package (POSP; Brindley et al., 1994)

Speech Accuracy, Rate and Consistency Task 2: Repetition and Spoken DDK Rates (from Williams & Stackhouse, 2000) Aims • To assess a child’s speech rate, accuracy and consistency at a single-word level. • To enable a comparison to be made between spoken DDK rates (Task 2) and silent DDK rates (Task 1). Description of the Task The child is asked to repeat words, non-words and syllable sequences five times each. A child’s ability to carry out these repetitions is evaluated in terms of speech rate (i.e. time taken to repeat the item five times) and the accuracy and consistency of productions. Use of the Task This task differs from the repetition tasks described in Chapter 5, in that the focus is on a detailed assessment of the accuracy, rate of production and consistency of production of the stimuli on more than one occasion. As a result, use of the tasks can differentiate children who may have inaccurate but consistent speech production from those with inconsistent speech output. This information can be used to plan appropriate


161

intervention (e.g. Crosbie, Holm & Dodd, 2005) and monitor a child’s progress through the normal phases of speech production (see the developmental phase model in Chapter 1). Psycholinguistic Analysis of the Task The word and non-word repetition tasks tap into the speech output processing system. The real word task taps a child’s ability to retrieve a motor programme of a familiar word and also the ability to trigger that programme repeatedly both accurately, consistently and speedily (see Figure 5.3 for processing involved in word repetition when the child makes use of existing motor programmes). The non-word task assesses a child’s ability to devise and trigger a new motor programme repeatedly (see Figure 5.4). The repetition of a sequence of speech output, whether words or non-words, and the speed and consistency with which the child can do this, also tax motor planning skills. ‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes for real words; no for non-words. • Does the child have to access these lexical representations in order to complete the task? – No. Level on the Speech Processing Profile • Level I (for real word stimuli) • Level J (for non-word stimuli) Stimuli Three types of stimuli are presented for spoken repetition: 16 two- and three-syllable words; 16 phonetically matched non-words; and syllable sequences that have the same consonants as the words but different vowels (see Table 7.3). The non-words contain different vowels in each syllable and the syllable sequences contain the same vowel in all syllables, for example word: SELLOTAPE; non-word: SOLITOPE; syllable sequence: SE-LE-TE, /sə-lə-tə/. There are also six practice items for each stimuli type. Materials • Pictures and score sheets on the CD-Rom (Appendix G.2 for word stimuli; Appendix G.3 for non-word stimuli; Appendix G.4 for syllable sequences). • Not supplied: stopwatch timer.

162


Table 7.3 Linguistic stimuli used in Speech Accuracy, Rate and Consistency Task 2: Repetition and Spoken DDK Rates (from Williams & Stackhouse, 2000) Real word stimuli 2 syllables 3 syllables

Non-word stimuli 2 syllables 3 syllables

Syllable sequence stimuli 2 syllables 3 syllables

paper seesaw cooker party fizzy yellow money digger coffee jelly

/pap/ /sɑsi/ /kεk/ /putə/ /fɒzə/ /jlɒ/ /mnə/ /d / /kfə/ /dlə/

/pəpə/ /səsə/ /kəkə/ /pətə/ /fəzə/ /jələ/ /mənə/ /də ə/ /kəfə/ /dələ/

letterbox patacake sellotape telephone buttercup cardigan

/ltbks/ /pɒtkəυk/ /sɒltɑυp/ /tɒləfan/ /btkεp/ /kud n• /

/lətəbə/ /pətəkə/ /sələtə/ /tələfə/ /bətəkə/ /kədə ə/

Procedure Timed real word repetition The investigator presents a picture of each word, says the word and asks the child to repeat it; this provides a baseline for comparing consistency of repeated responses. After the child has imitated the target word once, he or she is then asked to say it five times as quickly as possible. This procedure is followed for all items. A tick chart is completed as the child repeats the word to indicate to him or her how many times he or she has produced the item. The time taken to perform each set of five repetitions is recorded with a stopwatch. All responses should be audio-recorded using a good-quality recorder and microphone, and transcribed for scoring after the event. Six practice items are presented, with feedback, to ensure that the child understands the task. For example, if a child stops after three repetitions, he or she is encouraged to continue for longer; if the child speaks so fast that the word becomes unrecognisable, the child is reminded that the investigator needs to hear the word; if the child produces the words very slowly, he or she is encouraged to try again and go faster. The test items are then presented with no further help. Timed non-word repetition The investigator says a non-word for the child to copy. For young children, a toy, such as a monkey, can be introduced and the child asked to repeat some ‘monkey words’ after the toy. The investigator should explain to the child that the words are not real words that he or she will have heard before. The procedure is the same as for word repetition, above: the child is asked to say each non-word once and then to produce it five times as quickly as possible.


163

Timed syllable sequence repetition The child is asked to copy the investigator saying some sounds (the sequences are presented with equal stress on each syllable). After the target sequence has been imitated once, the child is asked to repeat it five times as quickly as possible. The procedure is the same as for word and non-word repetition, above. Scoring Two measurements are taken of accuracy, two of consistency and one of rate of production. Accuracy measures • Accuracy of first attempt: The child’s first attempt at repeating the target (real word, non-word and syllable sequence) is scored for accuracy in terms of consonants correct in comparison to a local adult model, allowing for regional accent variation. Responses with speech sound errors relating to consonants, for instance substitutions or omissions, are scored as incorrect. • Accuracy of five repetitions: Each set of five repetitions is scored as correct if the target is repeated correctly, according to the local adult model, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Consistency measures The child’s first imitated response is taken as the baseline for consistency measures, irrespective of whether it is accurate compared to the adult model. • Consistency of five repetitions: Each set of five repetitions is scored as correct if the target is repeated in the same way as the child’s first (baseline) response, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Essentially this consistency score is the same procedure as above for accuracy, but instead of an adult model, the child’s own first initial response (before the five repetitions) is used. • Second consistency measure: This compares each repetition with the first, baseline production from the child and across the other four repetitions of the same target. The following scale is used for each of the five repetitions: • Rating I: Repetition identical to child’s baseline response. • Rating II: Repetition different from child’s baseline response. • Rating III: Repetition different from baseline and one other previous repetition.

164

Compendium of Auditory and Speech Tasks • Rating IV: Repetition different from baseline and two other previous repetitions.

Example 1 For the stimulus BUTTERCUP, the child’s first, baseline response is [bkətp]. For the five repetitions, the child produces: 1. 2. 3. 4. 5.

[bkətp] Consistency rating is I as production is the same as the baseline. [bkətp] Consistency rating is I as production is the same as the baseline and as 1. [bkətp] Consistency rating is I as production is the same as the baseline and as 1 and 2. [bkətp] Consistency rating is I as production is the same as the baseline and as 1, 2 and 3. [bkətp] Consistency rating is I as production is the same as the baseline and as 1, 2, 3 and 4.

Final consistency rating for this item is, therefore, I. Example 2 For the stimulus BUTTERCUP, the child’s first, baseline response is [bkətp]. For the five repetitions, the child produces: 1. 2. 3. 4. 5.

[bkətp] Consistency rating is I as production is the same as the baseline. [pkətp] Consistency rating is II as production is different from the baseline and different from 1. [ptətp] Consistency rating is III as production is different from baseline and from 1 and 2. [ptətp] Consistency rating is III as production is different from baseline and from 1 and 2, but the same as 3. [ptətp] Consistency rating is III as production is different from baseline and from 1 and 2, but the same as 3 and 4.

Final consistency rating for this item is, therefore, III. Example 3 For the stimulus BUTTERCUP, again the child’s first, baseline response is [bkətp]. For the five repetitions the child produces: 1.

[pkətp] Consistency rating is II as production is different from the baseline.

Speech Accuracy, Rate and Consistency 2. 3. 4. 5.

165

[pkətp] Consistency rating is II as production is different from the baseline but the same 1. [ptətp] Consistency rating is III as production is different from baseline and from 1 and 2. [ptətp] Consistency rating is III as production is different from baseline and from 1 and 2, but the same as 3. [ptətt] Consistency rating is IV as production is different from baseline, from 1 and 2, and from 3 and 4.

Final consistency rating for this item is, therefore, IV. Rate measure The time taken to produce the five repetitions of each item is measured in seconds, and both accurate and inaccurate productions are accepted. However, if a child omits a syllable or stops before completing five repetitions, timing of this item is not recorded or included in the mean scores. The mean rate is calculated for all sets of five repetitions. Normative Data Table 7.4 summarises the mean accuracy scores of typically developing children on their first production of the words, non-words and syllable sequences, as compared to an adult model. It can be seen that the overall performance of the 3 year olds was significantly poorer than that of the 4 and 5 year olds. This is not surprising given the normal range of speech development that occurs at these ages (Grunwell, 1987).

Table 7.4 Mean (standard deviation) scores for accuracy of production of stimuli at first attempt (data from Williams & Stackhouse, 2000 and Habgood, 2000) Age group

Words (N 16) Mean (s.d) Non-words (N 16) Mean (s.d.) Syllable sequence (N 16) Mean (s.d.) Total (N 48) Mean (s.d.)

3 years (N 10)

4 years (N 10)

5 years (N 10)

6 years (N 10)

7 years (N 10)

14.2 (2.57)

15.8 (0.42)

15.9 (0.31)

15.6 (0.97)

15.7 (0.48)

13.4 (2.87)

15.3 (0.67)

15.2 (0.91)

15.4 (0.97)

15.9 (0.32)

14.1 (2.51)

15.5 (0.70)

15.4 (0.69)

15.3 (1.25)

15.3 (1.06)

41.7 (7.86)

46.6 (1.17)

16.5 (1.78)

46.3 (3.02)

46.9 (1.66)

166


Table 7.5 Mean (standard deviation) scores for accuracy of all five repetitions of each stimulus, compared to adult model production (data from Williams & Stackhouse, 2000 and Habgood, 2000) Age group

Words (N 16) Mean (s.d) Non-words (N 16) Mean (s.d.) Syllable sequence (N 16) Mean (s.d.) Total (N 48) Mean (s.d.)

3 years (N 10)

4 years (N 10)

5 years (N 10)

6 years (N 10)

7 years (N 10)

13.3 (2.9)

14.9 (1.2)

14.7 (1.2)

14.0 (1.56)

14.3 (1.64)

11.7 (2.2)

13.3 (1.6)

14.5 (1.1)

14.3 (1.42)

14.1 (1.97)

12.2 (2.2)

14.1 (1.4)

14.9 (1.0)

13.4 (1.43)

13.6 (1.78)

37.2 (6.6)

42.3 (2.7)

44.1 (2.3)

41.7 (3.97)

42.0 (4.9)

Table 7.5 presents the mean number of sets of five items repeated accurately for each stimuli type (where the child’s responses are compared to the adult model). Table 7.6 shows the mean number of sets of five items repeated consistently for each stimuli type (where the responses are compared to the child’s own baseline response as a model). Comparison of mean scores in Tables 7.5 and 7.6 shows that there is a significant difference between accuracy and consistency for the typically developing 3 year olds: their consistency scores are higher than the accuracy scores. This difference was not found for the older children. Accuracy is, therefore, not such a sensitive measure as consistency in 3 year olds. As children get older, accuracy becomes a more important measure. Table 7.6 Mean (standard deviation) scores for consistency of all five repetitions of each stimulus, compared to the child’s baseline production (data from Williams & Stackhouse, 2000 and Habgood, 2000) Age group 3 years (N 10)

4 years (N 10)

5 years (N 10)

6 years (N 10)

7 years (N 10)

Words (N 16) Mean (s.d) Non-words (N 16) Mean (s.d.)

15.0 (0.9)

14.9 (1.2)

14.9 (0.9)

14.5 (1.35)

14.4 (1.35)

13.8 (1.7)

13.6 (1.4)

14.5 (1.1)

14.5 (1.08)

14.1 (1.97)

Syllable sequence (N 16) Mean (s.d.) Total (N 48) Mean (s.d.)

13.3 (2.0)

14.3 (1.2)

14.9 (1.0)

13.9 (1.37)

13.9 (1.45)

42.1 (3.78)

42.8 (2.57)

44.3 (2.16)

42.9 (3.45)

42.4 (4.12)


167

Table 7.7 Performance of normally developing children on Task 2: percentage of items rated I, II, III and IV for the three age groups across all repetitions (data from Williams & Stackhouse, 2000) Age group

Rating I Mean (s.d) Range Rating II Mean (s.d.) Range Rating III Mean (s.d.) Range Rating IV Mean (s.d.) Range

3 years (N 10)

4 years (N 10)

5 years (N 10)

84.5 (8.5) 70.8–95.8

89.7 (5.3) 83.3–97.9

91.6 (5.9) 79.1–97.9

11.2 (5.4) 4.2–18.7

7.3 (4.5) 2.1–14.5

7.1 (4.3) 2.1–12.5

3.7 (4.3) 0–10.4

2.5 (1.6) 0–4.2

0.6 (1.0) 0–2.1

0.4 (0.9) 0–2.1

0.4 (0.9) 0–2.1

0.2 (1.0) 0–2.1

A further consistency measure is given using the rating scale I–IV (see details above) and the percentage score is given for the number of set of five repetitions that were scored at each rating (see Table 7.7). The results show that the predominant consistency rating for typically developing children of 3, 4 and 5 years of age is I; that is, most repetitions are the same as their first baseline response. If percentages of each rating are calculated for the child being investigated, it is possible to see if the consistency ratings follow the typical distribution shown in Table 7.7. The mean time taken in seconds to produce five repetitions of the two- and three-syllable items was calculated for the typically developing children. The means and standard deviations for each of the age groups are presented in Table 7.8. They do not change significantly over time, suggesting that rate is not a developmentally sensitive measure. However, the ranges of scores presented can be used as guidelines for what might be considered to be normal limits. Questions to Ask • Is there a difference in performance between accuracy of a single production of an item and accuracy/consistency of five repetitions of the item? • What is the relationship between performance on oro-motor movement tasks in Task 1 and these speech tasks? • Is there a typical profile of performance across the speech tasks? • How does the child’s profile of performance compare to typically developing children?

168


Table 7.8 Performance of typically developing children on Task 2: Time in seconds to produce five repetitions of two- and three-syllable stimuli (data from Williams & Stackhouse, 2000 and Habgood, 2000) Age group

2-syllable words Mean (s.d.) Range 3-syllable words Mean (s.d.) Range 2-syllable non-words Mean (s.d.) Range 3-syllable non-words Mean (s.d.) Range 2-syllable sequence Mean (s.d.) Range 3-syllable sequence Mean (s.d.) Range 2-syllable all items Mean (s.d.) Range 3-syllable all items Mean (s.d.) Range

3 years (N 10)

4 years (N 10)

5 years (N 10)

6 years (N 10)

7 years (N 10)

2.68 (0.77) 2.03–4.71

2.63 (0.63) 2.03–4.19

2.51 (0.38) 1.86–3.21

2.05 (0.33) 1.57–2.64

2.2 (0.34) 1.87–2.93

4.21 (0.76) 2.9–5.56

3.79 (0.59) 3.11–4.97

3.60 (0.65) 2.98–5.23

3.09 (0.41) 2.5–3.9

3.28 (0.49) 2.54–4.42

2.95 (0.71) 1.89–4.1

2.85 (0.52) 2.29–3.95

2.78 (0.39) 3.46–2.39

2.36 (0.45) 1.83–3.09

2.51 (0.51) 2.02–3.64

4.53 (0.85) 2.8–5.99

4.51 (0.93) 3.31–6.09

4.18 (0.64) 3.39–5.15

3.66 (0.74) 2.66–4.85

3.86 (0.35) 3.45–4.5

2.99 (0.61) 2.12–3.85

2.71 (0.38) 2.19–3.29

2.61 (0.52) 1.87–3.78

2.12 (0.31) 1.8–2.76

2.35 (0.55) 1.76–3.25

4.28 (0.77) 2.89–5.55

3.70 (0.44) 3.06–4.31

3.86 (0.66) 2.96–5.25

3.15 (0.53) 2.33–3.98

3.39 (0.47) 1.57–3.98

2.87 (0.69) 1.89–4.61

2.73 (0.51) 2.03–4.19

2.63 (0.43) 1.86–3.78

2.17 (0.47) 1.57–3.98

2.35 (0.52) 1.76–3.64

4.34 (0.78) 2.8–5.9

4.00 (0.76) 3.06–6.09

3.88 (0.67) 2.96–5.25

3.3 (0.57) 2.33–4.85

3.51 (0.53) 2.58–4.5

Other Assessments of Speech Rate and Consistency over Several Repetitions Rate • Fletcher Time-by-Count Test of Diadochokinetic Syllable Rate (DDK; Fletcher, 1978) Consistency • The Diagnostic Evaluation of Articulation and Phonology (Inconsistency Assessment, DEAP; Dodd et al., 2002)

Speech Accuracy and Consistency Task 3 (from Coffield, 1994) Aim To extend the naming and repetition tasks presented in Chapter 5 by providing a measure of consistency as well as accuracy.


169

Description of the Task The child is asked to name pictures and to repeat words and non-words three times each in order to examine consistency of responses as well as accuracy. Rate is not scored in this task. Use of the Task This task was originally used to investigate the speech processing profiles of adolescents with speech difficulties associated with Down’s syndrome (Coffield, 1994). The study clearly showed that the adolescents were different from each other in terms of their speech processing skills and their intervention needs. For example, one individual was similar in accuracy to younger, typically developing children with the same articulation age, but was different from them in terms of consistency across conditions. Another adolescent showed a significant difficulty with consistency on non-word repetition, but not on word repetition. The procedure allows examination of consistency as well as accuracy of speech production across the three conditions, which helps in understanding the nature of a child’s speech difficulties and in planning appropriate intervention. Psycholinguistic Analysis of the Task The naming and word and non-word repetition tasks tap into the speech output processing system. The real word tasks tap a child’s ability to retrieve a motor programme of a familiar word and also the ability to trigger that programme repeatedly both accurately and consistently (see Figure 5.1 for processing involved in picture naming and Figure 5.3 for processing involved in word repetition when the child makes use of existing motor programmes). The non-word task assesses a child’s ability to devise and trigger a new motor programme repeatedly (see Figure 5.4). The repetition of a sequence of speech output, whether words or non-words, and the consistency with which the child can do this, also tax motor planning skills ‘What do tests really test’ questions • Is this an input or output task? – Output. • Does the child have lexical representations for the stimuli used in the task? – Yes for real words and naming; no for non-words. • Does the child have to access these lexical representations in order to complete the task? – Yes for naming; no for repetition. Level on the Speech Processing Profile • Level G (for picture naming) • Level I (for word repetition) • Level J (for non-word repetition)

170


Stimuli This task uses the 60 words and matched non-words from the cross-sectional study (see Table 2.2 in Chapter 2). Materials • Pictures and score sheets on the CD-Rom (Appendix G.5) Procedure Each set of stimuli is presented separately. Non-words are presented before words, these two repetition tasks are presented before naming, and each set is constructed such that the same or matched stimuli are not contained in the same list. The instructions are as follows. Non-word repetition ‘I am going to say some silly words. They are not real words. I want you to say the silly words exactly as I say them. I want you to say them three times. Listen to a silly word MUB, now I will say it three times (demonstration is given with emphasis on a measured rate of repetition).’ For repetition tasks, the investigator’s mouth should be obscured (casually by hand) to prevent lip cueing. The child is asked to repeat each item three times. Word repetition ‘I am going to say some real words this time. They are words that you have heard before. I want you to say the words exactly as I say them. Listen to a word TABLE, now you say it three times.’ Naming ‘I am going to show you some pictures. I want you to tell me the names of the pictures. I want you to tell me the names of the pictures three times just as you did before.’ Each picture is presented and the child asked to say the name of the picture three times. If the child is unable to name the picture, semantic and gap-fill cues can be given. Children may need reminding that more than one production is required, and this can be cued by saying ‘again’ if the child stops before he or she has produced the item three times. It may also be necessary to remind the child not to rush the repetitions, as rate is not scored in this task. Scoring The child’s responses are measured for accuracy and for consistency.


171

Accuracy measure Each production of each item is scored as either correct or incorrect, according to the adult model. (See Appendix E.3 for scoring guidelines for this stimuli set.) As with all speech production tasks, local accent/variants need to be taken into account, as does the child’s own accent. As each task (naming, real word repetition and non-word repetition) consists of 60 words, each produced three times, there is a possible maximum of 180 correct. The accuracy score for each task is then converted to a percentage. Performance on words of different lengths can also be compared to normative data by calculating the number of correct responses for one-, two- and three- to four-syllable words separately across the tasks. Again the total scores at each word length are converted into a percentage. Consistency measure Consistency is measured by identifying the number of items for which the child produced the word or non-word three times with no change in pronunciation. The child’s productions might be correct or incorrect, as compared to an adult model. The percentage of items that were produced three times with the same pronunciation is calculated for each task. The percentage of items produced three times with the same pronunciation at each word length can also be calculated. Normative Data Data was collected from seven children with typical development and aged between 3 years 7 months and 4 years. The mean and standard deviation of the percentage of items repeated accurately and consistently within each task are reported in Table 7.9. The typically developing children produced their most consistent performance on naming and their least consistent performance on non-word repetition. Table 7.10 provides the mean and standard deviation of the percentage of items repeated accurately and consistently at each word length,

Table 7.9 The mean (standard deviation) and range of percentage scores for accuracy and for consistency across different tasks, in typically developing children, aged 3 years (N 7), on Speech Accuracy and Consistency Task 3 (data from Coffield, 1994)

Accuracy Consistency

Mean (s.d.) Range Mean (s.d.) Range

Naming

Real word repetition

Non-word repetition

63 (13.06) 48–83 74 (2.47) 61–90

64 (11.26) 47–83 66 (4.85) 45–83

42 (15.56) 21–71 63 (4.47) 43–80

172


Table 7.10 The mean (standard deviation) and range of percentage scores for accuracy and for consistency across different word lengths, in typically developing children aged 3 years (N 7) on Speech Accuracy and Consistency Task 3 (data from Coffield, 1994)

Accuracy Consistency

Mean (s.d.) Range Mean (s.d.) Range

1 syllable

2 syllables

3–4 syllables

69 (10.11) 56–85 76 (3.76) 60–88

56 (15.06) 39–82 70 (3.85) 53–88

44 (15.90) 23–67 58 (3.21) 43–76

across all three tasks. There is a steady decrease in both accuracy and consistency as syllable length increases. However, as discussed with reference to other tasks, there is a wide range of performance on speech tasks in children in this age group. Questions to Ask • Is there a difference in performance between accuracy of a single production of an item and consistency of three repetitions of the item? • Is there a difference in accuracy and/or consistency between word and real-word repetition? • Is there a difference in accuracy and/or consistency between word repetition and picture naming? • What effect does syllable length have on the child’s accuracy and consistency of speech production? Other Assessments of Accuracy and Consistency in Speech Production • Diagnostic Evaluation of Articulation and Phonology (Inconsistency Assessment, DEAP; Dodd et al., 2002) • The Nuffield Centre Dyspraxia Programme (Williams & Stephens, 2004)

Summary • This chapter includes two procedures for the assessment of children’s speech accuracy, rate and consistency. • The procedures included in this chapter are different to naming and repetition tasks (outlined in Chapter 5) in that the children are required to repeat their productions more than once to determine the stability of their productions.


173

• DDK and oro-motor tasks are frequently used in speech assessments, but there is limited normative data to help one determine if intervention is needed. This chapter has attempted to address this problem and provide some guidelines to enable interpretation of speech assessment findings. • Tasks 1 and 2, from Williams and Stackhouse (2000), involve nonlinguistic oro-motor tasks and linguistic tasks (repetitions of real words, non-words and syllable sequences). • Task 3, the consistency measure from Coffield (1994), uses the stimuli from the cross-sectional study and, like the tasks outlined in Chapter 5, requires children to name and repeat the stimuli. This task differs in that the children are required to provide three repetitions of each word so that their consistency can be assessed. • These tasks focus specifically on Levels I, J and K of the speech processing profile. Level K is specifically tapped in the oral skills task (from Task 1); Level J is tapped in the non-word repetition tasks (Tasks 2 and 3); and Level I is tapped in the real word repetition tasks (Tasks 2 and 3). Naming in Task 3 taps Level 3. • Together the tasks outlined in this chapter help to determine if a child with speech difficulties does or does not have problems with the movement of the articulators and if the child has sufficient movement to produce sounds and sequences of sounds in isolation as a foundation for sound production in spontaneous speech. • Young typically developing children may be inaccurate (compared to the adult model) in their productions, but are consistent, producing a word, or non-word, the same way each time they say it. • As children get older, accuracy becomes a more sensitive measure. • Children do not get faster on repetition tasks as they get older (in the age range 3–7 years).

Chapter Eight Predicting Persisting Speech Difficulties

‘Don’t worry, he’ll grow out of it.’ How do we know when this is the right thing to say to worried parents or nursery school staff? How do we know which young children to prioritise for intervention targeting their speech processing skills? How do we know which children are at risk for persisting speech difficulties and associated literacy problems? This chapter pulls together research findings from investigations using the procedures in Chapters 3–7 to try to address these important questions. It summarises the results of the longitudinal study, introduced in Chapter 2, which compared the speech processing skills and literacy development of children with and without speech difficulties between the ages of 3 and 7 years of age and draws on further analysis of this data by Nathan (2001). In particular, this chapter will focus on how the tasks presented in this volume can be used to predict which children will and will not resolve their speech difficulties by school age. It is vital that we improve our ability to assess which children in speech and language therapy are likely to have persisting speech problems in order to target resources effectively, particularly as children with persisting speech and language difficulties are at risk for developing literacy difficulties. One way to examine the nature of speech difficulties is by exploring how speech and language skills develop over time in children with and without speech difficulties. Longitudinal studies are essential for this (e.g. Leitao & Fletcher, 2004). Our own longitudinal study reported in this volume (Nathan et al., 2004a, b) had an unusually low attrition rate, which makes it particularly valuable for examining how outcomes might be predicted. No children were lost from the group with speech difficulties and only three control children were not traceable at the end of the study. The careful matching of each child with speech difficulties to a normally

175

176


developing control on age, gender, non-verbal IQ and, where possible, education means that any differences found between the group of children with speech difficulties and their controls cannot be explained by differences in general ability or education experience. The tracking of children over a four-year period between the ages of 3 and 7 years allows the cohort to be examined retrospectively for at risk characteristics in two ways: the children could be grouped according to whether they had or had not resolved their speech difficulties during this time, and by the end of the study it would be clear which children also had typical or delayed reading and spelling skills. These two outcomes are not unrelated. The ‘critical age hypothesis’ put forward by Bishop and Adams (1990) states that children who have not resolved their spoken language difficulties by the age of 5 years 6 months are likely to have associated literacy difficulties. The longitudinal study was ideal for examining this hypothesis further. We used speech input and output procedures presented in this volume plus additional language, phonological awareness and literacy tasks. Thus, the outcome measures taken for each child were: • • • • • •

speech output speech input language (expressive and receptive) phonological awareness literacy non-verbal

In addition, we collected background information on each child about: • • • • •

developmental history family data psychosocial data speech and language therapy education

This chapter will examine what tasks can help to predict speech and literacy outcomes by comparing the performance of the children with speech difficulties and their controls, and the children who persisted in having speech difficulties and those who had resolved them by the end of the study. The background information is presented in Chapter 9.

Children with and without Speech Difficulties As a group the children with speech difficulties performed significantly less well than their matched controls without speech difficulties on all of

Predicting Persisting Speech Difficulties

177

the speech, language, auditory, phonological awareness and alphabetic tasks at 4 years of age (T1, mean age 4;06). These differences persisted at age 5 (T2, mean age 5;08) when a significant difference between the groups also emerged on spelling skills. At age 6 (T3, mean age 6;08) there were group differences on both reading and spelling measures. At the end of the study when the children were around 7 years of age they took national tests at school (Standard Attainment Tests or SATs). As a group the children with speech difficulties performed less well on these tests than their IQ matched controls (Nathan et al., 2004b). Although there were significant differences between the group of children with speech difficulties and their controls, there were also clear differences between the children within the speech-disordered group based on the severity of the speech difficulty, the presence or absence of a language difficulty, and the pervasiveness of the speech processing difficulties, in terms of whether both speech input and output were affected rather than speech output alone. Each of these was examined separately and fed into the final analysis aimed at identifying predictors of persisting speech difficulties and associated literacy problems.

Severity of Speech Difficulty At age 4;06, the children in the speech-disordered group had significant speech difficulties and were scoring below 3 standard deviations (s.d.s) compared to the control group on a composite score of speech output tasks: picture naming (Snowling, van Wagtendonk & Stafford, 1988, Chapter 5), word repetition (Nathan et al., 2004a, Chapter 5) and non-word repetition (Nathan et al., 2004a, Chapter 5). However, the group of children with speech difficulties as a whole made progress with their speech over the course of the study. At T2 (CA 5;08) 45% of the children with speech difficulties were scoring above 3 s.d.s and 15% of them were scoring within normal limits (i.e. within 1 s.d. of the mean). By the end of the study at T3 (CA 6;08) 28% of the children had resolved their speech difficulties and 79% of the remainder were scoring within 3 s.d.s of the norm. However, the majority of the children’s speech at 6 years of age still fell below the level expected for this age group. Thus, the children identified with significant speech difficulties at age 4 were still performing differently compared to the normally developing control group at age 6 on word repetition, non-word repetition and naming tasks.

Presence of an Additional Language Difficulty There were group differences between the children with speech difficulties and their controls on both expressive and receptive language measures,

178

Compendium of Auditory and Speech Tasks Speech only

23%

Speech & Language (Expressive) Speech & Language (Expressive & Receptive)

60% 17%

Figure 8.1 Proportion and type of language difficulty in the speech difficulties group at T1

except for on the Information Score of the Renfrew Action Picture Test (RAPT; Renfrew, 1988) at 5 and 6 years of age. Poor speech intelligibility might explain the reduced expressive language performance in the speech difficulties group, with grammatical markers omitted or sentence length reduced in order to make the message more salient, but not poorer performance on receptive language measures or on lexical knowledge tasks. However, it was apparent that not all of the children in the speech difficulties group had associated language difficulties. To establish how many of the children with speech difficulties also had a language problem the criteria for this were set as follows: a child was regarded as having a significant language difficulty if he or she gained a score at or below the 10th centile on two or more language measures (which could be receptive and/or expressive). Based on this criteria, 19 of the 47 children in the speech difficulties group had additional language difficulties, leaving 28 (60%) with specific speech difficulties in the absence of language delay at the age of 4. Eleven (23%) of the children with language impairment had deficits in both expressive and receptive language, while 8 (17%) had expressive language problems alone (see Figure 8.1).

Speech Input Processing Skills At age 4, the speech difficulties group performed less well than the controls on auditory discrimination (i.e. Auditory Discrimination Task 1: Same/Different Test, Bridgeman & Snowling, 1988, Chapter 3) and auditory lexical discrimination (Auditory Lexical Discrimination Task 1: Mispronunciation Detection, Nathan et al. 2004a, Chapter 4). At ages 5 and 6, the two groups differed on auditory discrimination: ABX (Auditory Discrimination Task 3, Nathan et al. 2004a, Chapter 3). At each of the testing phases (T1, T2 and T3) around 20% of the children


179

Table 8.1 Relationship between speech severity score on the Edinburgh Articulation Tests (EAT) for children with and without language difficulties EAT standard score (ss 86 significant speech delay) Off scale–55 56–70 70–85 86

Number of children with speech difficulties only 2 9 15 1

Number of children with speech and language difficulties 9 8 2 1

with speech difficulties had poor speech input skills, with difficulty discriminating differences between similar-sounding pairs of words and non-words, or difficulty making judgements of whether a word was said accurately or not. The level of speech input skill at T1, T2 and T3 correlated strongly with speech output skills for both the speech difficulties group and the control group. Literacy skills at the end of the study (T3) also correlated with speech input performance for the speech difficulties group at T1 and T2, and for the control group at T1, T2 and T3.

Relationship between Severity of Speech Difficulties, Language Development and Speech Input Skills Examining the profiles of the individual children with speech difficulties revealed that the children with language problems tended to have more severe speech difficulties. Table 8.1 shows the relationship between language performance and articulation scores on the Edinburgh Articulation Test (Anthony et al., 1971). Further, children who had both speech output difficulties and language problems were more likely to have speech input problems as well. To examine if this relationship might determine eventual speech outcome at T3 (CA 6;08), the children with speech difficulties were subdivided into those who had and had not resolve their speech difficulties by the end of the study.

Children with Resolved vs Persisting Speech Difficulties Through analysing children’s performance on the test battery at T1, T2 and T3 we were able to explore the factors relating to typical and atypical speech development not only by describing the performance of the children with speech difficulties compared to their matched controls, as

180


in the section above, but also by examining retrospectively the children within the speech difficulties group who did and did not resolve their speech difficulties by the end of the study. Thus, the 47 children in the speech-disordered group were divided into two subgroups: those whose speech difficulties had resolved by the end of the longitudinal study, and those whose speech difficulties persisted. Defining ‘Resolved’ First, in order to examine which speech processing and language skills differentiate these two subgroups, it was necessary to define ‘resolved’ in order to allocate children to the appropriate subgroup. Children with speech difficulties were subdivided at T3 (CA 6;08) according to their speech output abilities as follows: if a child scored below 1 s.d. on a speech production composite z score (using the control groups’ average scores and standard deviations) of articulatory naming, word repetition and non-word repetition, he or she was classified as having ‘persisting’ speech difficulties; if a child scored above 1 s.d., he or she was classified as having ‘resolved’ speech difficulties; that is, performing like the control group. This resulted in 28% (N 13) of children in the speech difficulties group being classified as ‘resolved’ and 72% (N 34) as having persisting speech difficulties. Thus, the majority of children recruited into the speech difficulties group at the beginning of the study continued to have speech difficulties of some kind. Including performance on non-word repetition tasks in our criteria ensured that ‘hidden’ persisting speech difficulties – that is, those not apparent when only real/familiar words are produced – were not missed. Eleven of the 34 children with persisting speech difficulties performed poorly on non-word speech output tasks only; five of these performed as well as controls on the articulatory naming pictures tasks; and six performed as well as controls on the real word repetition task. Thus, if only real words had been used in the investigation, these 11 children would have appeared to have resolved their speech difficulties. Further, a feature of the children with persisting speech difficulties as a group was that there was an increasing discrepancy between their performance on real and non-word tasks between 4 and 5 years old, and to a greater extent between 5 and 6 years old, which was not characteristic of their controls or of children who had truly resolved their speech difficulties. Children who have difficulties processing non-words (input or output) are likely not only to have speech difficulties but also to have problems with new word learning, since all words are non-words until their meaning and phonology are learnt. Repeating a word more than once is also a typical way in which children hold on to it while segmentation for spelling occurs. Thus, children with repetition difficulties are vulnerable when it comes to storing new vocabulary, developing phonological


181

awareness skills and spelling. They are specifically disadvantaged in a classroom situation when exposed to the vocabulary of projects or topics in the curriculum such as in maths, science or geography. Non-word repetition tasks, such as those included in this volume, therefore make an important contribution to the identification and prediction of persisting speech, lexical and literacy difficulties. The subgroup of children who resolved their speech difficulties mainly did so by T2 (CA 5;08), suggesting that they have caught up on a ‘developmental delay’. This resolved subgroup represents what Shriberg, Kwiatkowski and Gruber (1994) term ‘short-term normalisation’. This term defines a child whose speech resolves within approximately two years of the identification of the speech problem. Shriberg, Kwiatkowski and Gruber identified the average age of referral in their study as about 4 years 3 months, thus normalisation would occur by age 6. The resolved group in our longitudinal study is the smaller of the two outcome subgroups (28%) and is slightly larger than Shriberg, Kwiatkowski and Gruber’s equivalent group of 18.5%, suggesting that the majority of children presenting with significant speech difficulties in the pre-school years may have more complex difficulties that will not necessarily resolve before the first year of school (around 5 years of age in the UK), thus putting their literacy development in jeopardy. Which Tasks Differentiated the Resolved vs Persisting Speech Difficulties Subgroups? If we now look back at the tasks administered in the longitudinal study we can identify which ones, if any, differentiated between children who did and did not resolve their speech difficulties. Table 8.2 summarises the differences found between the children with resolved vs persisting speech difficulties on the speech, language, phonological awareness and literacy tasks. This shows that even though at T1 all of the children had significant speech difficulties (at least 1 s.d. below the mean on the Edinburgh Articulation Test, Anthony et al., 1971), it is possible at this early stage to differentiate between children who did and did not go on to have persisting speech difficulties: those whose speech difficulties persisted performed less well on speech, auditory and expressive language tasks from the beginning of the study. Let’s examine these areas in more detail. Speech Input and Output Children whose speech problems persisted over the course of the study were noted to have more severe speech difficulties than children with resolved speech difficulties on the speech output tasks at T1 (CA 4;06) and performed less well on the auditory discrimination task. At T2 (CA

182


Table 8.2 Tasks where children with resolved speech difficulties performed significantly better than children with persisting speech difficulties T1 CA 4;6

Speech output and input processing

Language skills T2 CA 5;8

Speech output and input processing:

Language skills Phonological awareness T3 CA 6;8

Speech output and input processing

Language skills

Literacy skills

Word repetition: short form (Task 3, Chapter 5) Non-word repetition: short form (Task 4, Chapter 5) Articulatory naming (Task 2, Chapter 5) Auditory discrimination: ABX short form (Task 3, Chapter 3) Renfrew Bus Story (Renfrew, 1995b) Word repetition: short form (Task 3, Chapter 5) Non-word repetition: short form (Task 4, Chapter 5) Articulatory naming (Task 2, Chapter 5) Word repetition: low-frequency words (Task 3, Chapter 5) Non-word repetition: matched to low frequency words (Task 4, Chapter 5) Auditory discrimination: ABX short form (Task 3, Chapter 3) Auditory discrimination: same/different: shortest form (Task 1, Chapter 3) RAPT (Renfrew, 1988) Phoneme deletion (e.g. say ‘wind’ without the ‘w’) Articulatory naming (Task 2, Chapter 5) Word repetition: low-frequency words (Task 3, Chapter 5) Non-word repetition: matched to low frequency words (Task 4, Chapter 5) Auditory discrimination: ABX short form (Task 3, Chapter 3) RAPT (grammar score; Renfrew, 1988) Renfrew Bus Story (MLU; Renfrew, 1995b) BPVS (Dunn et al., 1997) TROG (Bishop, 1989) BAS Word Reading (Elliot, Murray & Pearson, 1983) Neale Comprehension (1997) Non-word spelling Spelling from pictures

5;08), the resolved speech subgroup was performing like controls on all speech input and output tasks. In contrast, the children with persisting speech difficulties not only performed less well on the speech tasks but also on two of the auditory discrimination tasks: the ABX task and also the same/different task involving cluster sequence discrimination, for example LOTS~ LOST/ VOST~VOTS. At T3, a similar pattern was found, with


183

the persisting speech group performing less well on speech output measures and on the ABX auditory discrimination task. Thus, children with persisting speech difficulties are more likely to have both input and output speech processing difficulties. Four of the children with persisting speech difficulties appeared to develop speech input problems during the course of the study. They did not show any difficulties on the input tasks at T1 (CA 4;06) but had significant difficulties on these by T3 (CA 6;08). They were therefore falling behind their peer group. This suggests that some children who appear to have no associated difficulties in the pre-school years can develop them when older as they fail to develop in line with or at the same rate as their peers. When working with the school-age child, it is therefore important to check input skills on a regular basis and not be complacent because a child had no difficulties with these in the past. Language Skills The children with resolved speech difficulties vs those with persisting speech difficulties also differed on their language development. At T1, the two outcome subgroups differed on both the information score and the mean length of utterance on the Renfrew Bus Story (Renfrew, 1995b). At T2, a significant difference was also noted on the Information and Grammar scores of the Renfrew Action Picture Test (Renfrew, 1988) and by T3 there were also differences on receptive language measures (BPVS, Dunn et al., 1997; TROG, Bishop, 1989) as well as on expressive grammar (RAPT, grammar score; and the Renfrew Bus Story MLU score). Thus, the children with persisting speech difficulties did not resolve their expressive language problems either. Further, these language problems extended into the children’s verbal comprehension as they got older; they fell behind their peers in their receptive vocabulary development and both receptive and expressive grammar at T3. However, out of the 34 children with persisting speech difficulties, there were 13 children with speech-only difficulties; that is, no associated language problems at T1. Only one of these 13 children developed language difficulties by T3 (CA 6;08), suggesting that language skills were generally more stable over time than speech input skills, given that four children with speech-only difficulties at T1 had developed speech input problems by T3 (see above section). Thus, by the end of the study there were only eight children with specific speech difficulties in the absence of any other obvious difficulties. Non-verbal Skills At the beginning of the study each child with speech difficulties had been matched on non-verbal IQ to a child without speech difficulties, so

184


that there was no significant difference between the two groups on nonverbal IQ (see Table 2.4 in Chapter 2). At T3 when the children were on average 6 years and 8 months old, the children in the persisting speech difficulties subgroup performed less well than their original matched controls on the picture completion task. In contrast, there was no difference between the resolved subgroup and their original matched controls. This suggests that non-verbal skills also need to be monitored in children with persisting speech difficulties. Indeed, there is evidence from other longitudinal studies of children with specific language impairment (SLI) of a decrease in their non-verbal performance over time (Stothard et al., 1998; Johnson et al., 1999; Conti-Ramsden et al., 2001). Chiat (2001) suggests that speech/language difficulties stem from problems with mapping sound and meaning, and as such language and cognitive difficulties can develop from an initial phonological deficit. Phonological Awareness and Literacy Measures Although a description of the phonological awareness and literacy measures could not be included in this volume, they are an important part of investigating children’s speech difficulties because of the likelihood of associated literacy problems in children whose spoken language difficulties persist beyond the age of 5;06. The children with persisting speech difficulties performed less well than the resolved subgroup at T2 (CA 5;08) on phonological awareness measures, in particular those involving segments rather than rhymes, for example phoneme deletion tasks (see Chapter 3 in Book 1 of this series, Stackhouse & Wells, 1997, for a discussion of a range of phonological awareness tasks). They also performed less well than the resolved group on literacy measures at T3 (CA 6;08), for example word reading, reading comprehension, non-word spelling and spelling names of animals from pictures. The longitudinal study supported the ‘critical age hypothesis’ referred to above (Bishop & Adams, 1990), in that the majority of children who had not resolved their spoken language difficulties by around 5 and a half years of age were showing associated literacy difficulties.

Examples of Children Who Have/Have Not Resolved Their Speech Difficulties Two children presented in Book 2 of this series (Stackhouse & Wells, 2001, pp. 17–21) illustrate the different outcome subgroups of children with speech difficulties discussed in this chapter. Zara’s profile at CA 4;03 revealed that she had no input processing or language difficulties. However, she had significant speech output difficulties on naming, word


185

and non-word repetition. By CA 5;03 she had no difficulty with any of these speech tasks (including non-word repetition) and her speech processing profile was no different from controls on either input or output tasks. This suggested that she had no residual underlying speech processing problems that would affect her reading and spelling development. Indeed, she progressed well with her phonological awareness and literacy development without any additional help and performed as well as controls on all reading and spelling measures at CA 6;03. In contrast, Tom had both speech and language difficulties at the beginning of the study. His speech processing profile at CA 4;0 showed difficulties with speech input as well as with speech output: he performed poorly on non-word auditory discrimination tasks and auditory lexical discrimination tasks, for example mispronunciation detection. This suggested that he had impaired ability to discriminate between unfamiliar words and that his phonological representations were inaccurately or incompletely stored. Compared to Zara, he had a more pervasive speech processing difficulty, additional language problems and his speech output problems were more severe. Whereas Zara’s speech problems had resolved by the time she was CA 5;03, Tom’s speech difficulties were still apparent when he was CA 5;02. At this age, he also had persisting problems with speech input and scored poorly on phonological awareness tasks that focused on the phoneme, for example alliteration fluency and phoneme completion. His letter knowledge was 2 s.d.s below the mean. Thus, Tom’s speech processing skills were a shaky foundation on which to build his literacy skills. At CA 6;02 his performance on single-word reading and spelling tasks was significantly below that of his normally developing peers.

Which Tasks Predict Speech Outcome? There is no one magic task that will identify children at risk for persisting speech and associated literacy difficulties. Rather, it is a collection of tasks tapping speech input and output processing, as presented in this volume, which is the key to identifying the at-risk child. Examining a child’s speech input and output skills, as well as their receptive and expressive language skills, enables an at-risk profile to be built up in the pre-school years. At this time phonological awareness tasks, though still important to include in the nursery curriculum or intervention programmes, are not yet reliable predictors of literacy outcome. As a child reaches school age (around 5 years in the UK), however, phonological awareness measures, particularly those dealing with segments (e.g. ‘sounds’ or clusters) rather than rhyme, become important at risk markers. By 6 years of age, poor performance on letter

186


Table 8.3 Areas that differentiated at T1 (CA 4;6), T2 (CA 5;8) and T3 (CA 6;8) the children with resolved vs persistent speech difficulties at T3 T1 CA 4;6

T2 CA 5;8

T3 CA 6;8

Auditory discrimination Severity of speech output Expressive language Auditory discrimination Speech output Expressive language Phoneme deletion Auditory discrimination Speech output Expressive language Receptive language Phoneme deletion and completion Letter naming Reading and spelling

knowledge and reading and spelling tasks features in an at-risk profile. Table 8.3 summarises the collection of areas at each age point that constitutes a developmental at-risk profile. Our longitudinal study is unusual in deliberately recruiting young children with primary speech difficulties in order to establish how early children at risk for persisting speech and literacy difficulties might be identified. The findings suggest that identification of children at risk for persisting speech difficulties is possible by the time the child is 4 years old based on the severity and pervasiveness of the speech and language difficulty itself. Those children who had persisting speech difficulties generally had more severe speech difficulties: speech input processing problems and delayed language skills. Including auditory discrimination and auditory lexical tasks in routine screening and assessments of young children with speech difficulties is an important part of this predictive process. Further, children whose speech difficulties persist are also at risk for literacy problems; knowing which children may not resolve their speech difficulties by around 5 years of age enables speech and language therapists, teachers and others to target their resources and also to monitor this vulnerable group in terms of literacy outcome. The following summarises the at-risk signs for associated literacy difficulties: • • • • •

Persisting speech difficulties between the ages of 5 and 6. Persisting language difficulties. Speech input difficulties at age 5. Poor phoneme awareness at age 5. Poor letter knowledge at age 5.


187

Speech vs Language Skills as Predictors of Literacy Outcome Although studies agree that both speech and language skills are important for literacy development, there have been conflicting results about whether it is the speech or the language skills that contribute most to literacy development. Some report that performance on tasks of syntactic knowledge is a particularly good predictor of literacy outcome (e.g. Bishop & Adams, 1990; Magnusson & Naucler, 1990), while others have emphasised aspects of speech production as being the strongest predictor (e.g. Bird, Bishop & Freeman, 1995; Webster & Plante, 1992; Larrivee & Catts, 1999), including the ‘type’ of speech difficulties. Leitao and Fletcher (2004), for example, followed up at ages 12–13 years 14 children from their original study (Leitao, Hogben & Fletcher, 1997) and found that those children who had had ‘non-developmental’ speech errors around the age of 5 years performed less well on phonological awareness and literacy measures at follow-up than children whose errors had been more typical of ‘developmental delay’. Our longitudinal study goes some way to explaining the above conflict. In our study, both speech and language skills predicted the literacy performance of children identified with specific speech difficulties, but each was more significant at different times. First, language ability at T1 (CA 4;06) was a main predictor of literacy development at T3 (CA 6:08). Both speech (input and output) and language ability were predictors at T2 (CA 5;08). However, by T3 (CA 6;08) speech output ability had taken over as the best concurrent predictor of literacy performance. This suggest that language skills are an important foundation for getting started with literacy development, but that intact speech processing skills are necessary to move a child through the early stages of literacy development, which involve cracking the alphabetic code. Here, letter knowledge and phonological awareness are essential. This accounts for why children who do not resolve their speech difficulties are most at risk. Children with persisting speech difficulties after the age of around 5;6 often have poor phonological awareness and cannot take advantage of literacy instruction at school that involves alphabetic skills. This affects spelling in particular. These children may also carry with them an additional language problem, which compounds other aspects of their literacy development, such as comprehension. However, even the children who had resolved their speech difficulties before taking Statutory Assessment Tests (SATs) around the age of 7 years performed less well than their IQ matched controls on spelling. This implies that having had an earlier deficit in speech output is still sufficient to cause a problem with the development of phonological awareness and literacy skill (Nathan et al., 2004b). In summary, the different predictors of literacy

188


performance at different ages suggest that literacy development may require different resources at different stages. At an early age, language skill is an important foundation for literacy development: children learn to read words more quickly if they are familiar with their meanings and spoken forms (Laing & Hulme, 1999) and children who have better syntactic skills will be better placed to use context to bootstrap inefficient decoding (Nation & Snowling, 1998). However, by the age of six years, phoneme awareness is an important precursor of the ability to decode and thereby to generalise reading and spelling strategies to unfamiliar words. At this stage, it is children’s speech skills that have a more direct influence on reading and spelling; thus, children who have persisting speech difficulties will have difficulty with these. From this developmental perspective, the findings highlight the interactive nature of speech and language development, with changing relationships between speech, language and literacy skills over time.

Conclusion Speech and language skills are the foundation on which literacy develops. Difficulties in the development of these skills renders a child not ready to take advantage of literacy instruction offered at school. Children with persisting speech difficulties may not be laying down clear representations of words in their lexical storage system and may be missing important information in the classroom, for example when new words are introduced. Fuzzy phonological representations and faulty motor programmes not only affect speech production skills but also phonological awareness development and spelling in particular. Generally, children resolve their speech difficulties, if they are going to, during their fifth year. However, the majority of children with significant speech difficulties in the early years continue to have speech difficulties at 6 and 7 years of age; sometimes these are hidden and are only evident on non-word tasks. Children whose speech difficulties are still present at 5 and a half years are most at risk for not only persisting speech difficulties, but also associated literacy difficulties. Most children whose speech difficulties persist have associated difficulties, such as language delay and/or speech input tasks. Children with pervasive speech processing difficulties are most likely to go on to have persisting speech and literacy difficulties. A smaller proportion of children have persisting speech difficulties in the absence of any other obvious speech processing difficulties (e.g. no associated language or speech input difficulties). It is more difficult to predict if these children will also have literacy problems, but the severity of their speech output difficulties and persisting difficulties with processing non-words may be important at-risk markers when pervasiveness does not apply.


189

Summary This chapter outlined findings from investigations of predictors of speech and literacy outcomes, using some of the tasks included in this volume. The following points are highlighted: • Individual variation exists in the speech and language profiles of children with significant speech difficulties at the age of 4 based on severity of their speech difficulties; presence/absence of language problems; and speech input as well as output difficulties. • The majority of children with speech difficulties at age 4 had not resolved them by age 5;06. • The more severe and pervasive a child’s speech processing difficulties, the more likely he or she is to go on to have persisting speech difficulties and associated literacy problems. • Children with speech output difficulties alone may well have associated literacy problems, but it is more difficult to predict outcome in this group. • There is no one task that will predict a child’s speech and literacy outcome; prediction is based on a collation of information. • It is possible to predict which children will/will not resolve their speech difficulties by collecting together information about a child’s speech input and output skills and their language development. • Speech input (e.g. auditory discrimination tasks) and non-word tasks (e.g. non-word repetition) are necessary additions to routine assessments of children with speech difficulties. • Children have only resolved their speech difficulties if they can perform as well as controls on non-word/new word tasks as well as on real word tasks that are familiar to them.

Chapter Nine Assessing Risk Factors Through Questionnaires

Although the speech processing tasks presented in this volume are a useful source of information about the nature of speech difficulties in children, the psycholinguistic approach itself is only one perspective on a child’s development. Another source of information comes from people involved with the child, and where possible from the children themselves (McLeod, 2004; and see Appendix 8 in Book 3 in this series, Pascoe, Stackhouse & Wells, 2006). Chapter 8 focused on identifying children at risk of persisting speech and literacy problems by administering specific speech and language processing tasks. Chapter 9 looks at how this information can be supplemented by more qualitative background data and examines characteristics of children with and without speech difficulties. Children with speech difficulties do not form a homogeneous group and there have been various attempts to examine associated risk factors. The ones most frequently reported are a positive family history; hearing impairment, such as ‘glue ear’; pre- and perinatal problems, including prematurity and low birth weight; and psychosocial factors (Fox, Dodd & Howard, 2002). Speech language and literacy problems often co-occur and run in families. Studies of identical and fraternal twins have shown a higher concordance rate in identical twins; that is, if one twin has speech and language difficulties there is a greater chance that the other identical twin will also have similar difficulties if the twins are monozygotic (Bishop, North & Donlan, 1995). Further, phonological coding difficulties, in particular, are highly heritable (Lewis et al., 2006). Lewis, Ekelman and Aram (1989) compared the performance of 20 children with ‘severe speech disorder’ and their siblings with normal controls on measures of phonological skill, word repetition, oro-motor skill, gross

191

192


and fine motor skills, and language and reading development, as well as on case history information. The siblings of children with ‘speech disorder’ performed significantly less well than the controls on measures taken and there was higher incidence of speech, language and dyslexic difficulties in their families. Shriberg (1997) reports studies quoting a range of between 20 and 80% of affected children having at least one affected nuclear family member (e.g. Felsenfeld, McGue & Broen, 1995; Lahey & Edwards, 1995). In a study of 84 children with speech delay, Shriberg and Kwiatkowski (1994) found that 56% of the children had one or more family members with ‘the same speech problem’, as reported by the carers/parents. More recent studies have tracked specific families with persisting speech and language difficulties. For example, VarghaKhadem et al. (2005) describe the discovery of a mutation in a gene known as FOXP2 that is thought to give rise to apraxia of speech in those with the affected gene. Their investigations of the speech and language skills of the KE family, half of whose members have significant speech difficulties, have suggested that this mutated gene gives rise to the cluster of deficits associated with apraxia. This type of study is extremely useful in shedding light on the neural mechanisms involved in speech. However, while it informs our knowledge about why some children have conditions such as apraxia, it does not tell us why these difficulties persist or how they might be addressed. The hypothesis that children with a family history of speech difficulties might form a specific subgroup within the population of children with speech difficulties has not yet been supported empirically (Lewis & Freebairn, 1997; Fox, Dodd & Howard, 2002). Thus, there is a great deal of further work to be done in considering how genes and environment interact. Although it is important that every child with speech difficulties has a hearing assessment, it is not the case that all children with speech difficulties do present with a hearing loss. The reverse is also true. It is not uncommon for pre-school children to experience bouts of otitis media or otitis media with effusion (OME), and yet not have any speech difficulties whatsoever. Where hearing problems are detected, these do not necessarily cause permanent deficits in a child’s spoken language development (Bishop & Edmundson, 1986), though if they are chronic rather than transitory there is more likelihood of associated speech and language problems: Many children referred to our speech clinic for suspected speech delay have had nearly continuous OME during some period of the first 3 years of life and beyond. (Shriberg, 1997, p. 115)

Thus, a fluctuating hearing loss in itself may not be the cause of a child’s speech difficulties. However, if it is persistent and co-occurring with other risk factors, then its significance increases (see Bishop, 1997, pp. 45–46 for further discussion).

Assessing Risk Factors Through Questionnaires

193

Similarly, there is little evidence of a one-to-one relationship between isolated examples of pre- and perinatal difficulties and the occurrence of specific speech difficulties, though these may contribute to a child’s vulnerability. Psychosocial factors have also been discussed as both a cause (Shriberg, 1994) and consequence (Nash, 2006; Nash et al., 2002) of children’s speech difficulties. They include being teased or bullied, fear or anxiety associated with communication, and low self-esteem. Shriberg (1994) classified children with speech difficulties on the basis of such risk factors and proposes five subtypes: • Speech delay: Occurs in around 60% of pre-school children with speech difficulties of unknown origin. • Speech delay – otitis media with effusion (OME): Includes children with speech delay who have had at least six episodes of OME in the first three years of life. This accounts for approximately 30% of the pre-school population with speech delay of unknown origin. • Speech delay – developmental apraxia of speech: Accounts for 3–5% of children with speech delay. Focus on prosodic features, in particular ‘excessive-equal stress’, which could be a genetically transmitted diagnostic marker for at least a subgroup of these children. • Speech delay – developmental psychosocial involvement: Based on clinical observations and reports that some children with speech delay have associated psychosocial needs. Occurs in around 7% of children with speech delay. • Speech delay – special populations: In contrast to above where the speech/language delay is the primary need, this category acknowledges that speech delay can occur in children with other health or educational needs, for example where there is a brain injury or learning difficulties or cleft lip and palate. Although this is a useful classification in terms of thinking about risk factors and their occurrence in children with primary speech difficulties, Fox, Dodd and Howard (2002) were not able to fit all of their children into these categories. Through parental questionnaires, they examined the risk factors in 65 German-speaking children with ‘functional speech disorders’ in the age range of 2;7 to 7;2 and compared them to 48 normally speaking controls in the age range of 3;4 to 6;1. There was a higher incidence of pre and perinatal problems, ear, nose and throat problems, sucking habits, and positive family history in the children with speech difficulties compared to the controls. However, audiological factors were not a feature of the group of children with speech difficulties. The study also examined if there were any risk factors associated with subgroups of children with speech difficulties based on surface speech

194


errors. The children were classified according to Dodd’s (1995/2005) subgroups: articulation disorder; delayed phonological development; deviant consistent phonological disorder; and deviant inconsistent phonological disorder. The one significant finding was that there was a high (56%) incidence of perinatal problems (e.g. foetal distress) in children in the deviant inconsistent subgroup. We examined the background information of the children with and without speech difficulties in our longitudinal study introduced in Chapter 2 and described further in Chapter 8 of this volume. Two questionnaires were completed by parents of all the children involved in the final year of the study when the children were aged between 6 and 7 years. The first questionnaire focused on medical history and development, and the second on family history of speech, language and literacy problems. At the same time, each child’s class teacher completed two questionnaires: one questionnaire covered educational progress, and the other psychosocial development. For the children in the speech-disordered group only, each child’s speech and language therapist completed a questionnaire three times during the study (at T1 mean CA 4;06; T2 mean CA 5;08; and T3 mean CA 6;08) in order to record the type and amount of therapy received. There were exceptionally high return rates on these questionnaires and the sections in this chapter summarise the findings from the following: • • • • •

Questionnaire Questionnaire Questionnaire Questionnaire Questionnaire

1: Developmental History 2: Family Information 3: Psychosocial 4: Educational Progress 5: Speech and Language Therapy

Copies of Questionnaires 1, 2, 4 and 5 above are included in Appendices H.1–H.4. Questionnaire 3 is the Strengths and Difficulties Questionnaire (SDQ) published in Goodman, 1997.

Questionnaire 1: Developmental History At T3 (CA 6;08), a questionnaire was sent out to all parents of the children in the longitudinal study to gather information about the children’s development (see Appendix H.1). The questionnaire covered the following areas. Birth and general health details • Child’s birth position in relation to siblings. • Whether child was born prematurely. • Complications associated with the birth.


195

• Feeding difficulties. • Allergies/fits/asthma/frequent coughs and colds/ear infections/ catarrh. Physical development • • • • •

Age child started to walk. Parental concern over child’s physical development. Has help/treatment been sought? Has help/treatment been received? Current problems.

Hearing • • • •

Parental concern over child’s hearing. Has help/treatment been sought? Has help/treatment been received? Current problems.

Speech and Language • • • • •

Age child began talking. Parental concern over child’s speech or language development. Has help/treatment been sought? Has help/treatment been received? Current problems.

Vision • • • •

Parental concern over child’s vision. Has help/treatment been sought? Has help/treatment been received? Current problems.

Reading • • • •

Parental concern over child’s reading. Has help/treatment been sought? Has help/treatment been received? Current problems.

Spelling • • • •

Parental concern over child’s spelling. Has help/treatment been sought? Has help/treatment been received? Current problems.

196


Additional information • Handedness. Response rates were high: 80.9% of forms were returned. The response was slightly better in the group of children with speech difficulties (87.2%, N 41) than in the control group (74.5%, N 35), which included three no-responses from children whom it was not possible to retrace at T3. Table 9.1 summarises the responses for the two groups. There were some significant differences between these two groups. Compared to controls, children with speech difficulties: • were reported to be on average around 8 months later than controls in saying their first words • were later born children in families, i.e. there were more first-born children in the non-speech disordered group • had a higher frequency of coughs and colds • had more visual difficulties There were no significant differences between the two groups on: • • • • • •

birth history health (apart from coughs and colds) hearing feeding physical development handedness

Nathan (2001) examined the group of children with speech difficulties further by dividing the children with speech difficulties into two subgroups: children whose speech difficulties had resolved by the end of the study (see Chapter 8 for definition of ‘resolved’ and for further information about these two subgroups), and those with persisting speech difficulties. Although there had been a significant difference between the speech-disordered group and their controls on the age first words were spoken, this measure did not differentiate children who had and had not resolved their speech difficulties. The only significant difference between children with persisting and resolved speech difficulties was a higher incidence of asthma in children who had resolved their speech difficulties by around 6 years of age. Children in the resolved subgroup also had a higher incidence of reported coughs and colds (50% compared to 22.6% of the persisting group), though this did not reach statistical significance. These findings might suggest that children with resolved speech difficulties had been more susceptible to upper respiratory health problems at an early age than those with persisting difficulties. It could be hypothesised that their poor speech development is related to difficulties of a

Table 9.1 Developmental information for the group of children with speech difficulties and their controls (from Stackhouse et al., 2002)

Measure Birth and general health details Premature Birth complications Feeding difficulties Allergies Fits Asthma Frequent coughs and colds Ear infections Catarrh Physical development Age when first walkeda Concerns over physical development If yes, was: Treatment sought If yes, was: Treatment given Current problems Hearing Concerns over hearing If yes, was: Treatment sought If yes, was: Treatment given Current problems Speech and language Age of first wordsa Concerns over speech or language If yes, was: Treatment sought If yes, was: Treatment given Current problems Vision Concerns over vision If yes, was: Treatment sought If yes, was: Treatment given Current problems Reading Concerns over reading If yes, was: Treatment sought If yes, was: Treatment given Current problems Spelling Concerns over spelling If yes, was: Treatment sought If yes, was: Treatment given Current problems Other Right handed a

Children with speech difficulties (N 41) % yes responses

Controls (N 35) % yes responses

17.1 10 19.5 15 0 22.5 29.3 24.4 7.3

5.7 17.1 5.7 6.1 0 17.1 8.8 20.0 2.9

13.03 (3.33) 14.6

12.35 (2.48) 5.7

100 83.3 50 39

50 100 100 22.9

93.3

100

93.3 6.7

100 33.3

21.64 (9.68) 97.6

13.02 (5.71) 2.9

97.6

0

92.7 45.0

0 0

22.0

5.7

100

100

88.9 77.8

50 50

53.7

20

76.2

62.5

78.9 100

83.3 50

45

8.8

71.4

75

66.7 85.7

75 66.7

87.2

79.5

In months (s.d.)

197

198


more peripheral nature and that these health problems are associated with poor speech development. In particular, they are associated with those children whose speech problems are transitory. However, these results need to be interpreted cautiously because of the small numbers.

Questionnaire 2: Family Information At T3 (CA 6;08), a questionnaire was also sent out to both parents of each child in the study (see Appendix H.2): 80% of the mothers’ questionnaires were returned (85% of speech-disordered group; 74% of controls) and 66% of the fathers’ questionnaires were returned (70.2% of speech-disordered group; 61.7% of controls). The lower response rate from fathers was mainly due to not being able to trace some fathers who were not living with the child in the study. The questionnaire covered the following areas: • • • •

Mother’s/father’s qualifications Mother’s/father’s occupation Family history of reading/spelling difficulties Family history of speech difficulties

There were no differences between children with speech difficulties and their controls on level of parents’ education or social class (see Table 2.5 in Chapter 2). Table 9.2 summarises the results of family history of speech, reading and spelling difficulties. There were no significant differences between the two groups on reported incidence of family history of speech and literacy difficulties. However, there was a trend towards a slightly higher incidence of reading difficulties on the mother’s side in the speech-disordered group, and 33% of the speech-disordered group compared to 14% of the control group reported having a family member with reading difficulties. There was a significant difference between the groups on the number of children who had a sibling or siblings with speech and/or literacy difficulties. In the group of children with speech difficulties, 44.7% of them had siblings with spoken/written language difficulties compared to only 11.1% of children in the control group. However, it was also the case that children with speech difficulties tended to have more siblings than controls. The children who had speech difficulties were more often second born, while more of the controls were first born. The findings of no other significant differences in family history could be due to a number of reasons, for example the sample is relatively small and heterogeneous; and parents with literacy difficulties might be less likely to return questionnaires (this could be reflected in a relatively low return rate of questionnaires from fathers).


199

Table 9.2 Family information about children with speech difficulties and their controls

Measure

Mother’s questionnaire Children with speech difficulties (N 40) Controls (N 35) % of yes responses % of yes responses

Reading difficulties (self) Reading difficulties (parents) Reading difficulties (grandparents) Reading difficulties (siblings) Reading difficulties (any family member) Speech difficulties (self) Speech difficulties (parents) Speech difficulties (grandparents) Speech difficulties (siblings) Speech difficulties (any family member) Attendance at speech therapy clinic (self) Hearing loss (self)

Measure

10 17.5 2.5 17.5 32.5 7.5 7.5 2.5 10 17.5 7.5 10

11.4 2.9 2.9 8.6 14.3 2.9 0 0 8.6 11.4 2.9 14.3

Father’s questionnaire Children with speech difficulties (N 33) Controls (N 28) % of yes responses % of yes responses

Reading difficulties (self) Reading difficulties (parents) Reading difficulties (grandparents) Reading difficulties (siblings) Reading difficulties (any family member) Speech difficulties (self) Speech difficulties (parents) Speech difficulties (grandparents) Speech difficulties (siblings) Speech difficulties (any family member) Attendance at speech therapy clinic (self) Hearing loss (self)

Measure Siblings Child has sibling with speech problem Birth order 1st born 2nd born 3rd born 4th born No. of siblings

6.1 12.5 3.1 15.6 25 9.1 0 0 3.2 6.5 9.1 9.4

10.7 3.6 7.1 3.6 17.9 0 0 0 7.1 7.1 0 10.3

Children with speech difficulties (N 47) Controls* (N 32) % of yes responses % of yes responses 48.8

11.1

15 59 17 4 1.83 (.97)

51 23 23 1 1.29 (.86)

Note: 1. 4 of the controls not included here were either only children or had one infant sibling.

200


Questionnaire 3: Psychosocial Strengths and Difficulties Questionnaire (Goodman, 1997) At T3 (CA 6;08) the teachers of the children with speech difficulties and their controls were requested to fill in a short questionnaire on each child’s psychosocial development: the Strengths and Difficulties Questionnaire (SDQ; Goodman, 1997). This provided information about the children’s emotional status, conduct, hyperactivity, peer relations and pro-social behaviour. Again, there was a high return rate of questionnaires: 98.9%. The children were categorised into three bands (normal, borderline and abnormal). Following Goodman (1997), it would be expected that around 80% of children would fall into the ‘normal’ category, 10% into the ‘borderline’ band and 10% into the ‘abnormal’ band. This was the case here, with the exception of ‘hyperactivity’ where more than 10% of children in both groups were considered to have difficulties (15.6% of the children with speech difficulties and 18.2% of the controls). Table 9.3 summarises these results for both the children with speech difficulties and their controls. Two significant findings emerged: • Overall, the children with speech difficulties did not score more poorly than the controls on any of the psychosocial areas. • There was a small significant difference between the two groups on the conduct score, with a higher incidence of difficulties in the control group. This study investigated children with speech and language difficulties at a younger age than most other studies of a similar nature. Therefore, we cannot say that the children will not develop psychosocial problems as they get older, particularly as a typical time for developing these is from around 7 years of age, the time at which our longitudinal study

Table 9.3 Performance of the children with speech difficulties and their controls on the Strengths and Difficulties Questionnaire (Goodman, 1997). Children with speech difficulties Controls (N 45) (N 44) % % % % % % Normal Borderline Difficulties Normal Borderline Difficulties Emotional Conduct Hyperactivity Peer relationships Prosocial

77.8 91.1 71.1 84.4 75.6

13.3 4.4 13.3 11.1 17.8

8.9 4.4 15.6 4.4 6.7

88.6 90.9 75 93.2 70.5

2.3 0 6.8 2.3 13.6

9.1 9.1 18.2 4.5 15.9

Total

71.1

17.8

11.1

77.3

15.9

6.8


201

stopped. Further, when Nathan (2001) examined this data from the perspective of speech outcome subgroups, children with persisting speech difficulties scored higher on the hyperactivity measure than those who had resolved their speech difficulties. This needs careful interpretation. We cannot say that hyperactivity is a cause of children’s speech difficulties. Rather, it may be a correlate (i.e. a co-occurrence) or a consequence of difficulties a child experiences in the classroom (e.g with communication or with literacy) that is manifested in poor attention skills and/or distracting behaviour. However, this does seem to be an important at-risk marker as there were no other significant differences between these two subgroups on the psychosocial measures taken: emotional symptoms, conduct problems, peer problems or pro-social behaviours.

Questionnaire 4: Educational Information At T3 (CA 6;08), the teachers were also requested to fill in a short questionnaire on each child’s educational development (see Appendix H.3). There was a high return rate (98.9%) and a summary of the educational information collected about the children with speech difficulties and their controls is presented in Table 9.4. Three significant differences emerged between children with speech difficulties and their controls: • The teachers’ views of the children’s literacy attainment specifically (there was no significant difference for other subjects). • Children in the speech-disordered group were receiving more support at school than children in the control group. • More children in the speech-disordered group had statements of special educational needs. In addition to this questionnaire, information was also collected about the children’s performance on Standard Attainment Tests (SATs; Nathan et al., 2004b). As a group the children with speech difficulties did less well than IQ-matched controls. However, there was variation within this group. When the group was divided into those who had and had not resolved their speech difficulties, the resolved group did as well as the controls on all aspects of SATs apart from on spelling, where they significantly underachieved compared to their controls. The children with persisting speech difficulties did less well overall on SATs than their IQ-matched controls. The combined results of the above show that children with speech difficulties are at risk for underachieving at school, particularly with literacy. An important finding is that even children who have resolved their speech difficulties are vulnerable when it comes to spelling development.

202


Table 9.4 Educational information about the children with speech difficulties and their controls % Children with speech difficulties % Controls Year group Year 1 Year 2 Year 3 Use of reading scheme1 Yes No Not specified Literacy performance Difficulties reported No difficulties reported Other school performance Difficulties reported No difficulties reported Extra help received2 Extra help No extra help Stage of Code of Practice for Special Educational Need Statemented Stage 1 (school action) Stage 2 (individual educational plan) Stage 3 (school action plus) Stage 4 (request for statutory assessment)

26.1 60.9 13.0

22.7 63.6 13.6

65.2 21.7 13.0

63.6 25.0 11.4

63.0 37.0

31.8 68.2

46.7 53.3

34.1 65.9

50.0 50.0

25.0 75.0

15.9 6.52 10.9 10.9 6.5

0 6.8 6.8 0 0

Notes 1 Reading schemes used: Oxford Reading Tree, Ginn 360, Moon, New Way, Sunshine Spiral, I Can Read Alone, Owls, Bangers and Mash, Marie Clay approach, combinations of reading schemes. 2

Help was given mainly in groups, occasionally individually by a classroom teacher, support teacher, classroom assistant, welfare assistant, learning support assistant, SEN teacher, SEN coordinator, parent, volunteer, home tutor. Equal numbers received help in and out of the classroom.

Questionnaire 5: Speech and Language Therapy The longitudinal study described in this volume was not designed to measure efficacy of intervention. However, knowing whether the children with speech difficulties were receiving ongoing help, and what form this took, was important information when interpreting the results and examining risk factors. Each child’s speech and language therapist was therefore asked to complete a questionnaire (see Appendix H.4) covering the following areas: • Age of referral to speech and language therapy (SLT). • Number of individual sessions of therapy.


203

• Number of group therapy sessions. • Total number of minutes of therapy. • *Type of therapy: phonological/oro-motor skills/articulatory/ phonological awareness/expressive language/receptive language/ play skills/listening skills/social skills/parent workshop/parent– child interaction/other. • Therapy setting: clinic/school/language unit/hospital/combination of settings/specialist centre. • Frequency of liaison: yearly/termly/occasionally/none/ongoing. • ∗Management: regular for therapy/on review/on waiting list/planned discharge/discharged. • Involvement of other agencies. This questionnaire was completed at three points in time: • At T1 (mean CA 4;6), covering the period when the child was first referred to speech and language therapy up until the T1 assessments were carried out. • From T1 up until the T2 (mean CA 5;8) assessments. • From T2 up until the T3 (mean CA 6;8) assessments. There was a high return rate of questionnaires: 97.8% at T1; 95.7% at T1–T2; and 89% at T2–T3. All of the children had been referred to speech and language therapy at the age of about 3 years because of concern about their speech development, with a mean age of 2.96 (standard deviation 0.91). Table 9.5 summarises where and what type of speech and language therapy was received. This varied within the group: 80% of the children attended community clinics for their therapy following initial referral. This declined to 53% of the sample when the children were around 6 years of age. A significant number (15/41, 36.6%) were placed on a waiting list and/or review list at some point during T1 (CA around 4;06), following which this approach declined. Eight of these children were discharged by T2, leaving seven still in need of therapy. Overall, a range of therapy was delivered. Working with parents was more common when the children were younger and a number of therapists included phonological awareness activities alongside speech work, particularly at T1–T2 (CA 4;06–5;08 years). Therapy was either individual or group or a mixture of both; individual therapy was more common in community clinics and group therapy was mainly delivered at specialist centres. The amount of therapy received generally decreased with age (see Table 9.6). On average children received equivalent of either 13 individual or 14 group sessions per year around the age of 3–4 years and either 15 individual or 11 Note: ∗items are not mutually exclusive categories.

204


Table 9.5 Where and what type of speech and language therapy was received Up to T1 (N 45) % of children *Type of therapy Phonological Oro-motor skills Articulatory Phonological awareness Expressive language Receptive language Play skills Listening skills Social skills Parent workshop Parent–child interaction Other Therapy setting Clinic School Language unit Hospital Combination of settings Specialist centre Frequency of liaison Yearly Termly Occasionally None Ongoing ∗Management Regular for therapy On review On waiting list Planned discharge Discharged Involvement of other agencies

T1 to T2 (N 44) % of children

T2 to T3 (N 41) % of children

66.7 37.8 17.8 48.9 20 15.6 4.4 37.8 6.7 20 11.1 6.7

57.1 23.8 40.5 51.2 11.9 9.5 0 21.4 7.1 0 4.8 11.9

34.1 9.8 34.1 36.6 24.4 7.3 0 12.2 4.9 0 0 2.4

80 0 2.5 2.5 15 0

69.8 2.3 7 0 6.9 0

53.7 2.4 7.3 0 0 2.4

2.1 5.1 33.3 30.8 7.7

4.9 22 34.1 29.3 9.8

2.4 24.4 14.6 51.2 7.3

78 36.6 36.6 0 0 33.3

75.6 24.4 7.3 0 26.2 19

53.7 29.3 2.4 0 56.4 15.4

Note: ∗not mutually exclusive categories.

group sessions per year around the age of 4–5 years. However, the number of sessions dropped to around either seven individual or five group sessions per year for those children with specific and persisting speech difficulties at the age of 5–6 years. Although it may well be the case that ‘six hours of therapy is not enough’ (Law & Conti-Ramsden, 2000, p. 908), increasing the amount and intensity of therapy does not necessarily guarantee normalised speech difficulties. Amount of therapy received did not significantly affect the outcome of the speech-disordered group. Children with severe


205

Table 9.6 Summary of amount of therapy received by the group of children with speech difficulties between the ages of 3;06 and 6;06 years

Mean no. of individual sessions Mean no. of group sessions Mean length of a session (in minutes)

Up to T1 (N 45)

T1 to T2 (N 44)

T2 to T3 (N 41)

13.93 14.27 54.27

15.07 11.40 44.85

7.27 5.07 42.88

and persisting speech difficulties were generally receiving more therapy than those with resolving difficulties, even at an early stage. Speech and language therapists were appropriately allocating resources to those children with more pervasive speech and language difficulties. However, not all of the children’s speech difficulties had resolved by T3 (6;08), although this is more to do with the nature of persisting speech difficulties in children rather than any reflection on the quality of the therapy received. The contact time they had may well have allowed the subgroup of children with persisting speech difficulties to make progress that otherwise would not have occurred. Three of the children with severe speech difficulties attended language units and received intensive speech and language therapy. One of these made significant progress on the Nuffield Centre Dyspraxia Programme (Williams & Stephens, 2004). This was against predictions given the severity of her speech difficulties. Although she did not resolve her speech difficulties completely by T3, she did transfer to mainstream school, indicating that for some children intensive and structured intervention early enough can really help. The other two children made less improvement and still had severe speech difficulties at T3. They both had pervasive speech and language problems and this response to therapy is likely to be indicative of the intractable nature of their speech difficulties. In summary, the findings of this questionnaire suggest that children with persisting and more severe speech problems were receiving more therapy than those who resolved their difficulties. Therapy resources were therefore being targeted appropriately. However, older children with persisting speech difficulties received a limited amount of therapy. Some received as little as four individual sessions or one group session a year at around 6 years, the age when they are particularly at risk for literacy difficulties. This suggests that their long-term needs were not being met (see Chapters 13 and 14 in Book 3 of this series, Pascoe, Stackhouse & Wells, 2006, for further discussion of service delivery for children with persisting speech difficulties).

206


Conclusion Adding the qualitative data presented in this chapter allows a greater understanding of the nature of speech difficulties and adds further information to what risk factors might contribute to persisting speech and literacy difficulties. Limited opportunity for speech and language therapy and support at school is part of these risk factors. The questionnaires are included in this volume to aid the gathering of supplementary data for use in practice and research with children with normal and atypical speech and language difficulties.

Summary Data from a longitudinal study comparing children with and without speech difficulties between the ages of 3 and 7 years revealed the following: • Information collected from people living and working with a child provides useful information for understanding his or her difficulties and predicting possible outcomes. • Questionnaires are one way of doing this. • Hearing difficulties did not feature as a correlate in children with speech difficulties in children aged 3–7 years who had been recruited to a study on the basis of having passed a routine hearing screen at the age of 3 years. • Children with speech difficulties had more visual problems than children without speech difficulties. • Children with speech difficulties were less likely to be first born in a family and had more siblings than children without speech difficulties. • Birth history, feeding history, physical development or handedness did not differentiate children with and without speech difficulties. • Children with speech difficulties spoke later than children without speech difficulties; children who persisted with their speech difficulties did not speak significantly later than those who resolved their speech difficulties. • Children whose speech difficulties resolved had more coughs/colds and asthma than children with persisting speech difficulties. • Presence of speech difficulties in children was not related to their parents occupation or level of education. • There is a tendency for specific speech difficulties, particularly of an apraxic nature, to run in families, though for individual children this information may not always be clear or known. • There was a tendency for children with speech difficulties to have a family history of literacy problems.


207

• There were no psychosocial differences between children with and without speech difficulties at around C.A. 6;08. • Children with persisting speech difficulties had more of a tendency to be hyperactive that children whose speech difficulties resolved. • Children with speech difficulties are more likely to have literacy difficulties and be receiving support at school than children without speech difficulties. • There is variation in the amount, type and frequency of speech and language therapy received by children with speech difficulties between the ages of 3 and 7 years. • Children with more severe speech difficulties tend to get more therapy but the amount of therapy declines with age. • The younger the child the more likely that therapy is delivered at a community clinic and that parents are involved in the intervention programme.

Chapter Ten Using Auditory and Speech Tasks in Practice

This volume has described auditory and speech tasks used in investigations of children with speech difficulties, and what psycholinguistic and environmental factors are characteristic of children with persisting speech difficulties and associated literacy difficulties. Use of psycholinguistic tasks in practice can help to ensure a better understanding of speech difficulties in children and more accurate identification of children in need of support for both their spoken and current or future literacy development. Baker et al. (2001) note: Another clinically important implication lies in the potential of psycholinguistic approaches to provide clinicians with new ways of conceptualizing the causes and correlates of speech impairment in children. (Baker et al., 2001, p. 686)

A further aim of using the tasks included in this volume is to plan appropriate intervention and management for children with speech difficulties. Examples of how to do this are a main focus of Books 2 and 3 in this series (Stackhouse & Wells, 2001; Pascoe, Stackhouse & Wells, 2006). This chapter will examine what contribution auditory and speech tasks can make to developing assessment procedures; interpreting assessment findings; planning intervention; predicting outcomes; defining ‘resolved’ speech difficulties; specifying discharge criteria; and training others. It includes practical activities and summarises the steps involved in adopting a psycholinguistic approach to assessment for planning intervention. The importance of moving on from prescribed procedures to internalising the approach for creative practice and research is emphasised.

209

210


Developing Assessment Procedures Routine assessments of children’s speech difficulties have not necessarily included an investigation of children’s auditory skills other than a basic hearing test or simple CVC discrimination (see Chapter 2). Using the auditory tasks presented in this volume has shown the importance of having a range of simple to complex stimuli (see Chapter 3). For example, children’s subtle auditory discrimination problems with cluster sequences (e.g. LOST/LOTS) or with syllable sequences within words (e.g. IBIKUS/IKIBUS) may be missed if only simple auditory discrimination tasks are used (e.g. CVC same/different tasks – PIN/BIN). Including auditory tasks that tap different levels of speech input processing (Levels B–E on the Speech Processing Profile in Appendix A.2) allows an investigation of a child’s lexical storage skills as well as lower-level auditory discrimination. Children who cannot detect when another speaker is pronouncing a target word in a picture correctly may well have stored a ‘fuzzy’ phonological representation of that word that will have a negative impact on their speech production and spelling skills (see Chapter 4). Comparing performance on speech output tasks reveals if a child has difficulties with articulation or lexical retrieval or both (see Chapter 5). The inclusion of complex words and non-words is particularly important. Some children appear to have resolved their difficulties but persist with subtle speech difficulties that are only evident in longer words or when novel material is introduced. Examining a child’s ability to discriminate and produce non-words indicates how well a child is able to deal with new words, a necessary skill for survival at school, for example when learning the vocabulary for specific projects and in the curriculum generally. Including a procedure to investigate children’s self-correction skills (see the Mispronunciation Self-Correction Task in Chapter 5 for Level L on the Speech Processing Profile) ensures that important information is collected about children’s self-monitoring skills and how these can be used or targeted in intervention. By using matched stimuli across speech and auditory tasks, it is possible to examine the relationship between a child’s input and output skills more closely. Constable, Stackhouse & Wells (1997) put this principle to good use when investigating the word-finding skills of school-age children. A close relationship was found between the children’s discrimination and retrieval of words: the words they could not discriminate well were also the ones they could not retrieve (see Chapter 10 in Book 2, Stackhouse & Wells, 2001). This sort of intra-child comparison – that is, comparing a child’s performance across different types of tasks – is useful for identifying processing strengths and weaknesses and how these might be used in intervention (see case studies in Book 3, Pascoe, Stackhouse & Wells, 2006). However, an inter-child comparison with normally developing

Using Auditory and Speech Tasks in Practice

211

children is also important in order to establish if the child has delayed or atypical development, and if this development is changing over time to be more in keeping with age-matched controls. Using the auditory and speech tasks with typically developing children has been invaluable for informing what to include in our assessment procedures and when. For example, there now seems little diagnostic value in asking a child to elevate his or her tongue tip given that many typically developing young children cannot do this; or to rely on age norms for 3–5 year olds for rate of articulation, as the evidence suggests that this is not a developmentally sensitive measure in this age range. However, comparing a child’s accuracy and consistency of speech productions with typically developing children will identify if he or she has a delayed or atypical speech development (see Chapter 7). Similarly, examining how young children develop connected speech processes draws attention to what to look for in children with speech difficulties (see Chapter 6) and how that information might be used in intervention (see Book 3, Pascoe, Stackhouse & Wells, 2006, Chapters 7 and 8).

Interpreting Assessment Data Assessment tasks are only useful if the data collected can be interpreted appropriately. Adopting the psycholinguistic framework presented in Book 1 of this series (Stackhouse & Wells, 1997) is a means of doing this. A psycholinguistic training activity devised by Rachel Rees at University College London demonstrates not only the importance of including both auditory and speech tasks when investigating the nature of speech difficulties in children, but also how to interpret data from such psycholinguistic procedures (cf. Rees, 2001). The activity presents four children (George, Ben, Harry and Cleo) in the same age group (5–6 years) who all realise /k/ as [t] when naming pictures, so that KEY, CAP, CAT, CAR are all pronounced by the children as “tea, tap, tat, tar”. However, we know from using the psycholinguistic approach in the past that children with the same surface speech difficulties (e.g. /k/ : /t/) may not necessarily be the same ‘underneath’: they can have very different speech processing strengths and weaknesses that will influence what is included in therapy (see Book 1. Stackhouse & Wells, 1997). By presenting a range of auditory and speech tasks, the extent of these differences can become apparent. Tasks devised specifically to investigate further the tk contrast based on hypotheses about how children’s speech difficulties may arise can differentiate between difficulties arising in speech input or output or both; this hypothesis testing approach is an essential aspect of a psycholinguistic investigation (see Chapter 1). Stimuli for such tasks are presented in Table 10.1 and a description of the tasks follows.

212

Compendium of Auditory and Speech Tasks Table 10.1 Stimuli used for investigating tk contrast in children Target word

Target non-word

key cap cat car tea tap two ten

/ k/ / kɔp / / kIt / / ku / / t / / tɔp / / teI / / tɒn /

Speech Tasks Naming (Level G) Two pictures for each of the words in the table were presented one by one on a computer screen. Presentation was randomised so that a picture was never followed by an identical picture. The child was asked to name each picture on two different occasions, to check consistency, and when he or she had done so the next picture was presented. The child’s productions were transcribed at the time and audio/video-recorded so that the transcription could be checked. Word repetition task (Level I) Two recorded audio presentations of each of the words in the table were played one by one and were accompanied by a white speech bubble in a blue square that was presented on a computer screen. Presentation was randomised so that a spoken word was never followed by an identical spoken word. The child was asked to repeat each word on two different occasions, to check consistency, and when he or she had done so, the next speech bubble and word were presented. The child’s productions were transcribed at the time and audio/video-recorded so that the transcription could be checked. Non-word repetition task (Level J) Non-words were designed to match the words in the table by altering the vowels. Two recorded audio presentations of each of the nonwords in the table were played one by one and were accompanied by a white speech bubble in a blue square that was presented on a computer screen. Presentation was randomised so that a spoken nonword was never followed by an identical spoken non-word. The child


213

was asked to repeat each word on two different occasions, to check consistency, and when he or she had done so the next speech bubble and non-word were presented. The child’s productions were transcribed at the time and audio/video-recorded so that the transcription could be checked. Auditory Tasks (no pictures) Non-word same/different task (Level B) Three recorded audio presentations of the following minimal pairs were made and played in a random order: / k t/, /t k/, /k k/, /t t/; /kɔp tɔp/, /tɔp kɔp/, /kɔp kɔp/, /tɔp tɔp/. Each pair was accompanied by two white speech bubbles in blue squares that appeared on the left and then on the right of a computer screen as each word was played. After each pair of presentations the child was asked to press a ‘same’ or ‘different’ key on the keyboard. Word same/different task (Level D) Three recorded audio presentations of each of the following minimal pairs were made and played in a random order: KEY/TEA, TEA/KEY, KEY/KEY, TEA/TEA, CAP/TAP, TAP/CAP, CAP/CAP, TAP/TAP. Each pair was accompanied by two white speech bubbles in blue squares that appeared on the left and then on the right of a computer screen as each word was played. After each pair of presentations the child was asked to press a ‘same or ‘different’ key on the keyboard.

Lexical discrimination task: mispronunciation detection (Level E) A target picture of a CAR was presented at the top of the computer screen. Underneath this picture ten white speech bubbles in blue squares appeared one by one and were accompanied by ten audio presentations. Five of these presentations were the same as the target and five were different. The order of the same or different presentations was randomised. For example, a picture of a CAP was presented at the top of a computer screen. Underneath this picture the ten white speech bubbles in blue squares appeared one by one and were accompanied by ten audio presentations. Five of these presentations were CAP and five were TAP. The order of the presentations was randomised. After each spoken stimulus and bubble were presented under the picture, the child was asked to press the ‘right’ or ‘wrong’ key on the keyboard, depending on whether he or she thought the stimulus was the right or wrong pronunciation of the target word.

214


Responses on the Auditory and Speech Tasks The following are possible profiles of performance on the auditory and speech tasks. Assume that children aged 5–6 years should be able to perform these speech and auditory tasks perfectly well, and that scores on the auditory tasks of 23 or 22/24 are acceptable. George Table 10.2 summarises George’s responses on the speech tasks. On no occasion does George produce /k/ for any target in either naming or repetition conditions. However, he scores near ceiling on the auditory tasks and shows typical processing of real words and non-words making his performance acceptable for his age: • Non-word same/different task: 22/24. • Word same/different task: 23/24. • Lexical discrimination task: 23/24. Ben Table 10.3 summarises Ben’s responses on the speech tasks. Ben produced /k/ initially in words across each condition but not consistently. This inconsistency may be related to his difficulties with the auditory tasks: • Non-word same/different task: 12/24. • Word same/different task: 10/24. • Lexical discrimination task: 16/24. Harry Table 10.4 summarises Harry’s responses on the speech output tasks. Although he can produce /k/ initially in words on some occasions, these Table 10.2 George’s two responses for each target on the speech tasks

Target word

Naming

Word repetition

Target non-word

Non-word repetition


[ti] [ti] [tæp] [tæp] [tæt] [tæt] [tɑ] [ti] [ti] [tæp] [tæp] [tu] [tu] [tεn] [tεn]

[ti] [ti] [tæp] [tæp] [tæt] [tæt] [tɑ] [tɑ] [ti] [ti] [tæp] [tæp] [tu] [tu] [tεn] [tεn]

/k/ /kɔp/ /kt/ /ku/ /t/ /tɔp/ /teI/ /tɒn/

[t] [t] [tɔp] [tɔp] [tIt] [tIt] [tu] [tu] [t] [t] [tɔp] [tɔp] [teI] [teI] [tɒn] [tɒn]


215

Table 10.3 Ben’s two responses for each target on the speech tasks

Target word

Naming

Word repetition

Target non-word

Non-word repetition


[ti] [ti] [tæp] [tæp] [tæʔ] [tæʔ] [tɑ] [kɑ] [ki] [ti] [tæp] [kæp] [tu] [ku] [tεn] [tεn]

[ki] [ki] [tæp] [tæp] [kæʔ] [tæʔ] [tɑ] [tɑ] [ki] [ti] [kæp] [tæp] [tu] [tu] [tεn] [kεn]

/ k/ / kɔp / / kIt / / ku / / t/ / tɔp / / teI / / tɒn /

[t] [k] [tɔp] [tɔp] [kIʔ] [tIʔ] [ku] [ku] [t] [k] [tɔp] [tɔp] [keI] [teI] [kɒn] [kɒn]

are only when he is repeating words or non-words. The absence of /k/ when he is naming pictures suggests that his stored motor programmes are faulty, even though in the repetition tasks he has demonstrated the motor execution ability to articulate [k] accurately. His poor performance on the auditory lexical decision task (see below) also suggests that he may have inaccurate phonological representations of stored items. • Non-word same/different task: 22/24. • Word same/different task: 23/24. • Lexical discrimination task: 13/24. His performance on non-word auditory discrimination, however, is encouraging, suggesting that he can now deal with new items. The finding that his repetition of non-words is more accurate than his spontaneous production of familiar words may indicate that his lexicon contains ‘frozen forms’ of earlier learned words, but that new words may be more accurately stored. In this sense, Harry is similar to the boy with speech difficulties described by Bryan and Howard (1992).

Table 10.4 Harry’s two responses for each target on the speech tasks

Target word

Naming

Word repetition

Target non-word

Non-word repetition


[ti] [ti] [tæp] [tæp] [tæʔ] [tæʔ] [tɑ] [tɑ] [ti] [ti] [tæp] [tæp] [tu] [tu] [tεn] [tεn]

[ti] [ki] [tæp] [tæp] [tæʔ] [kæʔ] [kɑ] [kɑ] [ti] [ti] [tæp] [tæp] [tu] [tu] [tεn] [tεn]

/ k/ / kɔp / / kt / / ku / / t/ / tɔp / / te/ / tɒn /

[k] [k] [kɔp] [kɔp] [tʔ] [kʔ] [ku] [ku] [t] [t] [tɔp] [tɔp] [te] [te] [tɒn] [tɒn]

216


Table 10.5 Cleo’s two responses for each target on the speech tasks

Target word

Naming

Word repetition

Target non-word

Non-word repetition


[ti] [ti] [tæp] [tæp] [tæ?] [tæ?] [tɑ] [tɑ] [ti] [ti] [tæp] [tæp] [tu] [tu] [tεn] [tεn]

[ti] [ti] [tæp] [tæp] [tæ?] [tæ?] [tɑ] [tɑ] [ti] [ti] [tæp] [tæp] [tu] [tu] [tεn] [tεn]

/ k/ / kɔp / / kt / / ku / / t/ / tɔp / / te/ / tɒn /

[t] [t] [tɔp] [tɔp] [t?] [t?] [tu] [tu] [t] [t] [tɔp] [tɔp] [te] [te] [tɒn] [tɒn]

Cleo Table 10.5 summarises Cleo’s responses on the speech output tasks. Cleo does not produce /k/ accurately in any of the target words or non-words when naming or repeating words. In this sense she is the same as George. However, the two children differ on their performance on the auditory tasks. Unlike George, Cleo has some difficulty with auditory discrimination of real words. Her strong performance on auditory discrimination of non-words and her poor performance on the lexical decision task suggest that she has fuzzy representations for familiar words rather than an auditory discrimination problem per se. • Non-word same/different task: 23/24. • Word same/different task: 20/24. • Lexical discrimination task: 10/24. Having looked at the children’s auditory and speech data, now try the following activity.

ACTIVITY 10.1 Aims: • To summarise the above children’s speech processing profiles. • To compare and contrast performances on the auditory and speech tasks. With reference to the above auditory and speech data, complete the grid below on each of the children by answering questions 1–5 with a yes/no answer. Place a ‘Y’ for ‘Yes’ and an ‘N’ for ‘No’ in each square of the grid.

Using Auditory and Speech Tasks in Practice George

Ben

Harry

217 Cleo

1. Can we hypothesise that the child’s auditory discrimination of /k/ and /t/ in initial position in the nonwords tested is accurate? 2. Can we hypothesise that the child’s auditory discrimination of /k/ and /t/ in initial position in the words tested is accurate? 3. Can we hypothesise that the child’s phonological representations of words beginning with /k/ (that were tested) are accurate? 4. Can we hypothesise that the child’s motor programmes for the words beginning with /k/ (that were tested) are accurate? 5. Can we hypothesise that the child has adequate motor execution skills to produce /k/ before V and VC?

Here are some guidelines to help you answer questions 1–5 above: Q1: See number correct out of possible 24 on the non-word auditory discrimination. Q2: See number correct out of possible 24 on the word auditory discrimination. Q3: See number correct out of possible 24 on the auditory lexical discrimination task. Q4: If stored motor programmes are accurate the child will produce /k/ consistently at the beginning of each target word beginning with /k/. Q5: Is there evidence that the child can produce /k/ at all, i.e. anywhere in the words, even if not consistently?

218


When you have completed the grid compare your responses to those in the Key to Activity 10.1 at the end of this chapter. Look down the columns to see each child’s profile and look across the rows to see what areas of strength or weakness they all have in common. Looking down the columns of the completed grid shows that each child has a slightly different profile. George is the only child that has specific speech output difficulties in the absence of input difficulties. Harry and Cleo both have stored fuzzy representations of familiar words, but do have some auditory discrimination skill to draw on. Unlike Harry, however, Cleo also has problems with her articulation. In this respect she is like George, but does not have his advantage of precisely stored phonological representations. Like Harry, Ben does not have an articulatory difficulty since he can produce /k/ in words, albeit inconsistently, but he does have significant speech input problems. Looking across the rows, the only area they have in common is that none of them has stored accurate motor programmes for producing words with /k/ in initial position. Putting it another way, the only thing that makes them a ‘group’ is that they all do /t/ for /k/. However, in terms of why this might be the case is quite different for each individual. This has implications for their intervention needs.

Planning Intervention A main motivation for investigating children’s speech difficulties is to plan appropriate intervention for them. Approaches drawn from different perspectives are well illustrated in the case of ‘Jarrod’, a 7-year-old boy with highly unintelligible speech who is the subject of a special issue of the journal Advances in Speech-Language Pathology (Vol. 8, No. 3, Sept. 2006). This volume comprises a collection of papers, each focusing on a particular intervention approach for Jarrod, for example Cycles (Hodson, 2006); Core Vocabulary (Dodd et al., 2006) Working through Parents (Bowen & Cupples, 2006); Prompts for Restructuring Oral Muscular Phonetic Targets – PROMPT (Hayden, 2006) and Psycholinguistic (Stackhouse, Pascoe & Gardner, 2006). As argued in Chapter 9, a psycholinguistic approach should never be used in isolation and other perspectives need to be taken into account, for example medical, educational, psychosocial, educational and linguistic. A linguistic perspective allows description of a child’s speech difficulties and includes a phonetic and phonological analysis of a child’s speech. This will identify what to target in intervention, for example a particular sound, group of sounds or features, clusters, connected speech. The psycholinguistic perspective will inform how to do this. For example, speech


219

input tasks might be incorporated as a strength to support speech output (e.g. see ‘Alan’, presented by Waters in Chapter 6 of Book 2, Stackhouse & Wells, 2001); or because it is a weakness that needs work (see ‘Anna’, presented by Corrin in Chapters 4 and 5 of Book 2, Stackhouse & Wells, 2001). Direct speech work may or may not be required; when it is, some children can benefit from instrumental techniques such as electropalatography (see cases presented by Dent in Chapter 7 of Book 2, Stackhouse & Wells, 2001). Whatever approach is taken, intervention tasks require the same careful analysis of what processing they are tapping, as we have done with the assessment tasks presented in this volume (see Chapter 3 by Rees in Book 2, Stackhouse & Wells, 2001). To illustrate how auditory and speech tasks can contribute to intervention planning let’s return to George, Ben, Harry and Cleo. All produce /t/ for /k/. How might the speech processing information we have about them influence their intervention programmes? Try the following activity.

ACTIVITY 10.2 Aim: • To plan therapy on the basis of speech processing strengths and weaknesses. Consider how each child’s profile discussed above and summarised on the grid in the Key to Activity 10.1 would influence intervention planning. Decide where you would start and what strengths could be used to support any weaknesses. Jot these down and then read on.

Therapy for George George needs help through articulatory work to produce /k/, preferably with a following vowel. However, before starting therapy, you would need to ensure that there were no neurological or anatomical abnormalities that are causing his articulatory difficulties. Ruling these out, and once George has developed some articulatory ease for producing /k/ in initial position in syllables, he would be a prime candidate for phonological therapy, since his strong input skills allow him to monitor differences between minimal pairs presented. Further, the therapist can use his phonological representations to ‘play back’ to him the right/wrong way of saying target words with /k/ initial and encourage him to self-monitor. Until he can articulate /k/ in syllables, however, the phonological phase of therapy that targets the distribution of segments will not be successful and would only increase his frustration.

220


Therapy for Ben It would be important to start with auditory tasks, for as long as auditory discrimination remains poor, speech output tasks will be confusing. When Ben can discriminate between simple non-words, minimal pair real words may be successful in helping Ben to realise the significance of the change from one sound to another. Unlike with George above, work on motor execution directly should not be necessary. In fact, his articulatory skill can be used to help him with auditory tasks. Therapy for Harry Harry’s strength in discriminating /k/ /t/ at the beginning of words could be utilized to update his phonological representations of words beginning with /k/. A character/toy who sometimes gets his words ‘right’ or ‘wrong’ can be used in games with Harry. By asking Harry to detect when the toy gets a word right or wrong, the therapist is working on his phonological representations. This activity often stimulates the child to update their own motor programmes as well because they want to tell the toy how to say a word correctly. Stabilising the phonological representations first will facilitate the updating of the motor programmes. Harry’s ability to imitate /k/ before a vowel (plus/minus a consonant) could be used to practise new motor programmes for known or new words.

Therapy for Cleo Cleo’s strength at discriminating between /t/ /k/ at the beginning of words could be utilised to update her phonological representations of words beginning with /k/. Unlike Harry, however she is unable to produce /k/ in words and therefore games targeting the phonological representations should not yet include output work. In parallel, Cleo can be helped to produce /k/ through articulatory tricks and practice; checking first, as with George, that there are no neurological or physical reasons for her inability to produce /k/ so far. The above therapy suggestions are based on the following principles: • Children with the same speech error can have different speech processing strengths and weaknesses. • A child’s speech processing profile can help to plan appropriate therapy. • Results from the auditory and speech tasks presented in this volume (Chapters 3–7) are key to drawing up a child’s speech processing profile


221

• Individually tailored tasks can be devised based on psycholinguistic principles. Another theme of this book has been the identification of children at risk for persisting speech and associated literacy difficulties (see Chapter 8).

Predicting Speech and Literacy Outcomes It is now well established that children who do not resolve their spoken language difficulties by CA 5:06 are likely to have associated literacy problems (see Chapter 8). The advantage of the longitudinal study (Nathan et al., 2004a) presented in this volume is that we could track children’s speech and emerging literacy development between the ages of 4 and 7 years. By the end of the study it was apparent which children had and had not resolved their speech difficulties, and which were developing associated literacy problems. The results indicated that children with both speech and language difficulties not only had more severe speech output difficulties, but also more pervasive speech processing problems, which included speech input difficulties. These children were more likely to have persisting speech output difficulties beyond the age of 5;06 and to develop associated literacy problems than children with speech difficulties alone. With this in mind, try the following activity.

ACTIVITY 10.3 Aim: • To speculate on which children from the four cases presented (George, Ben, Harry and Cleo) are most at risk for literacy problems. Examine the children’s profiles and decide which, if any, are at risk for associated literacy problems and why. What other information would you need to know about them in order to be more confident in your decision? Refer to Chapters 8 and 9 to help you with this activity. Jot down your thoughts and then read on. All of the children are above 5 years of age and would therefore be considered at risk since they have persisting speech difficulties beyond the ‘critical age’ of around 5;06 (Bishop & Adams, 1990). However, research

222


findings suggest that it is the children who have more pervasive speech processing problems that are the most at risk for both persisting speech difficulties and associated literacy problems. This puts Ben at the top of the list for concern, closely followed by Cleo, as they both have speech input and output problems. Harry and George are lowest on the list but are still of concern. For all of them a key factor will be how they respond to intervention in speech and language therapy and at school, and whether they can make links between their spoken and written language. In order to be more confident about who is most at risk, further information is required about the following areas: receptive and expressive language skills, because children who have both speech and language problems are more at risk for literacy problems than children with speech problems alone; phonological awareness, because it is phonological awareness development that moves spoken language skills into literacy (see Book 1, Stackhouse & Wells, 1997, particularly Chapter 3); and letter knowledge, because phonological awareness alone is not sufficient to result in successful literacy development unless it is coupled with a developing awareness of letter sounds and names. These are the three key areas in which children with speech difficulties can be weak. They are not necessarily delayed in the visual processing aspects of developing literacy, though some can be. Additional background information is also helpful, for example from the parents/carers and teachers. Children with a family history of literacy problems may be particularly at risk, for example, or those who are reported to be poor at listening and attention at school (see Chapter 9, this volume).

Defining Resolved Speech Difficulties Ideally, children need to start school with intact speech and language skills in order to progress with their literacy development. However, not all children will resolve their speech difficulties by the time they start school (around 5 years of age in the UK). In the longitudinal study presented in this volume (Nathan et al., 2004a), there was a tendency for children to have resolved their speech difficulties during their fifth year if they were going to at all. In other words, if they had not resolved their speech difficulties by T2 (mean CA 5;08) they were not likely to have done so by T3 (mean CA 6;08). Further, the discrepancy between real and non-word production increased between T2 and T3 in the children with persisting speech difficulties, indicating an underlying speech processing difficulty with motor programming and new word learning. Some children’s persisting speech difficulties are quite subtle and not obvious to the people around them, such as parents or teachers, who believe that earlier speech difficulties have resolved because the child is intelligible in conversation. Luke, described by Nathan and Simpson


223

(2001), is a good example of this. He was referred to speech and language therapy at 3 years of age because of speech difficulties and was first assessed on a selection of the auditory and speech tasks described in this volume when he was 4 years of age. At CA 4;04 he had no difficulties discriminating similar-sounding real words and non-words (see tasks in Chapter 3), or on the mispronunciation detection task in Chapter 4. Luke’s speech was intelligible and his family and teacher were no longer concerned about his speech. However, his naming, word repetition and non-word repetition skills were below the level expected for his age using the high-frequency stimuli presented in Chapter 5, and speech errors were noted on voice/voiceless contrast, clusters and affricates. A year later (CA 5;05) he performed as well as controls on these highfrequency speech input and output tasks, but when asked to produce low-frequency complex word stimuli he performed poorly, suggesting that underlying speech processing difficulties remained and that he may be at risk for literacy problems. In spite of this, his ability to communicate with others was not impaired, or of concern at home or at school. By CA 6;05 his performance on non-word repetition tasks was below that of typically developing children and he was struggling to develop phonological awareness skills. He now had obvious literacy problems and needed specialist help to support his progress at school (see Chapter 8 by Nathan and Simpson in Book 2, Stackhouse & Wells, 2001, for further details of his assessments, his speech processing profile and the subsequent intervention programme designed for him). Luke’s case is not only a good reminder to include more challenging stimuli in assessment procedures for school-age children, but also that persisting speech problems are not necessarily obvious or problematic for everyday communication. Sometimes a child with obvious difficulties is in a better position to get help compared to children like Luke who can go unnoticed until they are failing at school because of literacy or behaviour problems that prevent them accessing the curriculum. Luke’s case also raises the issue of what criteria should be used to discharge children from speech and language therapy.

Specifying Discharge Criteria We calculated that 28% of the children with speech difficulties in our longitudinal study had resolved their problems, leaving the remainder with significant speech problems on one or more of the speech output measures (naming, and real and non-word repetition, see Chapter 5). However, over half of the group had been discharged from speech and language therapy (56.4%) and over half of the parents (53%) felt their children did not have problems with their speech. These figures may not be unrelated, since a mutual lack of concern about a child’s speech development on

224


the part of the therapist and the parent is a common reason for discharging a child from speech and language therapy. Failure to attend appointments can also be a reason for discharge, especially if this nonattendance is the result of parents no longer being worried about their child’s speech development. Whatever the reason, in the longitudinal study the situation was that a significant number of children who had not truly resolved their speech difficulties were being discharged from the care of speech and language therapy and, like Luke, were struggling at school with new word learning, phonological awareness and literacy development in particular (see Chapter 8). Indeed, data from the questionnaires presented in Chapter 9 suggests that, although the children in the ‘persisting’ speech difficulties group may have had poorer performance on these speech tasks, many of them were not considered to have functional speech difficulties any more. It is understandable that many speech and language therapy services use intelligibility as a main criterion for discharging clients. However, this may be too superficial if it is the only measure used. Consideration also needs to be given to a child’s speech processing skills and a check done before discharge on whether these are in line with age, for example by using a non-word repetition task. If not, these subtle and more hidden difficulties will have a negative impact on the child’s learning at school. Follow-up studies have confirmed that children with a history of speech difficulties do less well than IQ-matched controls in both primary school (Nathan et al., 2004b) and secondary school assessments (Stothard et al., 1998). Thus, this is a long-term issue that needs to be addressed as early as possible. Follow-up or monitoring of these children through and with the education services is also a wise precaution to establish if the earlier speech and language difficulties have truly resolved. Teachers in particular need support to be able to identify subtle spoken language difficulties and the relationship between these and literacy development. A consequence of inclusive education policies is that teachers are being confronted more than ever before by children with speech difficulties in their classrooms (Lindsay et al., 2002) and they are concerned about their level of training in this area (Marshall, Stojanovik & Ralph, 2002; Sadler, 2005; Mroz, 2006).

Training Others The psycholinguistic framework presented in this series grew out of practice with children with speech and literacy difficulties and out of training speech and language therapy students and others. It is a training tool for the practitioner as much as for the practitioner’s clients; it is


225

meant to facilitate the user’s thinking about speech and literacy difficulties in children, particular if the user also has the role of training others to understand or use it. It can be adapted to present to a range of audiences, including teachers and parents who find the simple speech processing model depicted in Figure 1.1 (see Chapter 1) helpful, particularly when talking about the relationship between speech difficulties, vocabulary development and phonological awareness skills. Training professionals and carers to identify the risk signs presented in Chapters 8 and 9 is an important aspect of using the psycholinguistic framework in practice, as well as developing appropriate teaching and learning strategies to support the children at home and school. Training professionals and carers working in the early years – that is, pre-school – is particularly important if a solid spoken language foundation is to be developed ready for starting school when more formal literacy instruction is given (Wood, Wright & Stackhouse, 2001; Wright & Wood, 2006). Wood, Wright and Stackhouse’s (2001) programme of training sessions for early-years workers on the development of spoken and written language development and difficulties in children was positively evaluated by participants from a range of disciplines from education, health and social services. Other useful resources for supporting teachers, in particular, to work with children’s speech and language development at school include Locke (2006) and Speake (2003). Research is an important aspect of the psycholinguistic framework (see Book 3, Pascoe, Stackhouse & Wells, 2006). A number of theoretically driven procedures, such as the tasks presented in this volume, have been developed as part of research projects, including undergraduate and postgraduate dissertations. This is why some of the tasks have long versions that may not be practical in everyday practice, and where possible shorter forms are included. Training in research based on this approach may involve using the procedures to investigate typically developing children; in itself this can be a valuable experience for students hoping to work with children with special needs. The collation of results from such studies, and from case studies of children with speech and literacy difficulties, provides a basis for further development and application of the components of this approach, namely the Speech Processing Profile, the box-and-arrow speech processing model and the developmental phase model presented in Book 1 (Stackhouse & Wells, 1997; see Chapter 1, this volume, for a summary).

A Psycholinguistic Approach Most of what is presented in a psycholinguistic approach is not new in terms of procedures to follow in assessment or intervention. Rather, it is

226


an active questioning about how and why we do what we do. There are no prescribed ways of working within a psycholinguistic approach other than to continually base it on hypothesis testing, interpretation of data and evaluation of outcomes. This might involve administering published procedures like the ones included in this volume, but it may not. We stated in Book 1 (Stackhouse & Wells, 1997) that the psycholinguistic approach is a particular way of thinking about children’s speech and literacy difficulties rather than a new assessment battery or intervention programme as such. As it is internalised in the ‘head of the user’, it can be carried out at any moment, for example in a conversation with a child or observing a play session; it is often in these informal sessions that key observations are made and hypotheses formulated and tested. For example, in the Mispronunciation Self-correction Task (see Chapter 5), which addresses Level L on the Speech Processing Profile, we included questions to ask children directly. In answer to the question Have you finished learning to talk yet?, Ben, aged 7;11 with persisting speech difficulties that included pronouncing [ ] for /d/, and [k] for /t/, replied: well not yet, because I think I have a problem with ‘g’s and ‘g’s (he means ‘d’s and ‘g’s) . . . and maybe on ‘k’s as well because it seems like if I say ‘k’ (he means ‘t’) it seems like its going into a ‘k’. (from Boase, 2006, p. 100)

Without testing Ben in any formal way, we can we can hypothesise from this one response that Ben has separate phonological representations for these contrasting sounds but cannot yet correct them. His difficulties are likely to be with speech output rather than with speech input, and we would want to check out his articulatory skills. We could carry out procedures included in this volume to fill in a speech processing profile and test out this hypothesis, and/or we could devise specific tasks to investigate gd and tk contrasts. Daniel (CA 7;07), another child with speech difficulties, revealed his problems with processing speech input in his response to the question: Do you ever get teased at school?, when he replied: Yeah. And we have dinner here. (from Boase, 2006, p. 70)

He had misinterpreted what he had heard as Do you ever get teas at school. Again, without administering any formal procedures we can hypothesise that he was experiencing learning difficulties at school, since this level of input difficulty would militate against processing instructions and learning new material. This was indeed the case. Further investigation of his auditory skills on a one-to-one basis could include discrimination of single words on the Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences (from Bridgeman &


227

Snowling, 1988) in Chapter 3, or in connected speech on the Auditory Lexical Discrimination Task 4: Words in Sentences (Cassidy, 1994). This, together with further investigation of these skills in the classroom situation, would help to plan appropriate support for him at school. In summary, the following steps are presented as guidelines for what is included in a psycholinguistic approach to assessment for children in need of intervention. These guidelines need to be interpreted flexibly, as users are working in different contexts and each child’s needs are different. Not all steps have to be covered; what is important is a cogent interpretation of data collected and an appropriate and evaluated intervention programme.

Steps in a Psycholinguistic Approach for Children in Need of Intervention 1. Formulate hypotheses about the nature of a child’s speech difficulties from first impressions and rate intelligibility (see Chapter 11 in Book 3, Pascoe, Stackhouse & Wells, 2006) 2. Examine background data from different perspectives and decide who else should be involved in the intervention programme. 3. Analyse and describe speech output phonetically and phonologically, and formulate hypotheses about underlying ‘causes’. 4. Select appropriate speech targets for intervention from this linguistic perspective, paying particular attention to stimuli design (see Chapters 3, 5 and 7 in Book 3, Pascoe, Stackhouse & Wells, 2006). 5. Administer appropriate psycholinguistic tasks to test out hypotheses, incorporating the child’s own speech errors into the stimuli design where possible (see Chapter 11 in Book 1, Stackhouse & Wells, 1997). 6. Collate results onto a Speech Processing Profile with reference to normative data where possible (see Appendix A.2). 7. Interpret results further with reference to box-and-arrow model (see Appendix A.3) and/or developmental phase model (see Figure 1.2) as appropriate. 8. Carry out a programme of intervention bringing together linguistic, psycholinguistic, education, psychosocial and medical perspectives as needed. 9. Monitor progress with reference to each of these perspectives, and evaluate outcomes at regular intervals. 10. Specify discharge criteria to take into account underlying speech processing skills as well as intelligibility.

228


Summary This chapter summarises how the auditory and speech tasks presented in this volume can be used to understand the nature of speech difficulties in children and to plan appropriate intervention for them, including training others involved. These tasks have been devised for investigating children’s speech and literacy difficulties from a psycholinguistic perspective. It is not expected that all of the tasks will be administered in any one case. Far from it: that would be completely unnecessary. Rather, the tasks add to our hypothesis-testing resources and are there to be selected as appropriate to investigate children’s needs. They can be plotted onto the Speech Processing Profile (see Appendix A.2) in order to facilitate interpretation. However, they are not standardised ‘tests’ but assessment procedures, with some normative data for guidelines. Practitioners and researchers are encouraged to develop their own, more tailor-made procedures based on psycholinguistic principles to investigate individual differences in children with speech difficulties. Such a procedure was included in this chapter to investigate the t k contrast in four children. These children also illustrate why an intervention programme based on underlying ‘causes’ of children’s speech difficulties is often more effective than one based on ‘symptoms’. The focus in this volume has been on the performance of typically developing children on the auditory and speech tasks, for comparison with children presenting with spoken and written language difficulties. This normative data leads us to question some assumptions often made about children’s speech and auditory skills. Further, data from the longitudinal study of children with speech difficulties helps to identify which children might not resolve their speech difficulties before starting school. It raises the issue of what ‘resolved speech difficulties’ really means, and the criteria for discharging children from speech and language therapy. Certainly, professionals should not be complacent about children who have or have had speech and language difficulties; even when these appear to have resolved, since there can be a lasting impact on their literacy development. This volume has aimed to share the auditory and speech tasks we have found useful and to disseminate the insights gained from using them. It is hoped that this will facilitate working within a common framework that may add to our understanding of children’s speech and literacy difficulties and how best to support them.


229

Key to Activity 10.1 George

Ben

Harry

1. Can we hypothesise that the child’s auditory discrimination of /k/ and /t/ in initial position in the nonwords tested is accurate?

Y

N

Y

2. Can we hypothesise that the child’s auditory discrimination of /k/ and /t/ in initial position in the words tested is accurate?

Y

N

Y

N

N

N

3. Can we hypothesise that the child’s phonological representations of words beginning with /k/ (that were tested) are accurate? 4. Can we hypothesise that the child’s motor programmes for the words beginning with /k/ (that were tested) are accurate? 5. Can we hypothesise that the child has adequate motor execution skills to produce /k/ before V and VC?

Y

N

N

N

N

Y

Cleo

Y

N

N

Y

N

References

Alloway, T. P. & Gathercole, S. E. (2005). Working memory and short-term sentence recall in young children. European Journal of Cognitive Psychology, 17: 207–220. Anthony, A., Bogle, D., Ingram, T. & McIsaac, M. (1971). Edinburgh Articulation Test. Edinburgh: Churchill Livingstone. Armstrong, S. & Ainley, M. (1988). South Tyneside Assessment of Phonology (STAP). Ponteland: STASS Publications. Baker, E., Croot, K., McLeod, S. & Paul, R. (2001). Psycholinguistic models of speech development and their application to clinical practice. Journal of Speech, Language and Hearing Research, 44: 685–702. Beving, B. & Eblen, R. E. (1973). ‘Same’ and ‘different‘ concepts and children’s performance on speech sound discrimination. Journal of Speech and Hearing Research, 16: 513–517. Bird, J. & Bishop, D. (1992). Perception and awareness of phonemes in phonologically impaired children. European Journal of Disorders of Communication, 27: 289–311. Bird, J., Bishop, D. & Freeman, N. (1995). Phonological awareness and literacy development in children with expressive phonological impairments. Journal of Speech and Hearing Research, 38(2): 446–462. Bishop, D. (1989). Test for Reception of Grammar (2nd edn). University of Manchester: Age and Cognitive Performance Research Centre. Bishop, D. (1997). Uncommon Understanding. London: Psychology Press. Bishop, D. & Adams, C. (1990). A prospective study of the relationship between specific language impairment, phonological disorders and reading retardation. Journal of Child Psychology and Psychiatry, 31: 1027–1050. Bishop, D. & Edmundson, A. (1986). Is otitis-media a major cause of specific developmental language disorders? British Journal of Disorders of Communication, 21: 321–338.

231

232

References

Bishop, D. & Edmundson, A. (1987). Language-impaired 4-year olds: Distinguishing transient from persistent impairment. Journal of Speech and Hearing Disorders, 52: 156–173. Bishop, D., North, T. & Donlan, C. (1995). Genetic basis of specific language impairment: Evidence from a twin study. Developmental Medicine and Child Neurology, 37: 56–71. Boase, A. (2006). Talking about Talking: Children’s Views on Speech. Unpublished Master’s thesis. Sheffield: University of Sheffield. Bortolini, U., Arfe, B., Caselli, C., Degasperi, L., Deevy, P. & Leonard, L. (2006). Clinical markers for specific language impairment in Italian: The contribution of clitics and non-word repetition. International Journal of Language and Communication Disorders, 41(6): 695–712. Botting, N. & Conti-Ramsden, G. (2000). Educational outcomes for children with speech and language impairment. In Bishop, D. & Leonard, L. (eds), Speech and Language Impairments in Children: Causes, Characteristics, Intervention and Outcome. Hove: Psychology Press. Bowen, C. & Cupples, L. (2006). PACT: Parents and children together in phonological therapy. Advances in Speech-Language Pathology, 8(3): 282–292. Bradford, A. & Dodd, B. (1994). The motor planning abilities of phonologically disordered children. European Journal of Disorders of Communication, 29: 349–369. Branigan, G. (1979). Some reasons why successive single word utterances are not. Journal of Child Language, 6: 411–421. Bridgeman, E. & Snowling, M. (1988). The perception of phoneme sequence: A comparison of dyspraxic and normal children. British Journal of Disorders of Communication, 23(3): 245–252. Brindley, C., Cave, D., Lees, J., Crane, S. & Maffat, V. (1994). Paediatric Oral Skills Package (POSP). Chichester: John Wiley & Sons Ltd. Broadbent, P. (2000). An Investigation of the Speech Processing Skills of Children Attending a Mainstream Primary School. Unpublished MSc project, London: Department of Human Communication Science, University College London. Broomfield, J. & Dodd, B. (2004). The nature of referred subtypes of primary speech disability. Child Language Teaching and Therapy, 20(2): 135–151. Brownell, R. (2000). Expressive One-Word Picture Vocabulary Test (EOWPVT2000). London: Pearson Assessments. Bryan, A. & Howard, D. (1992). Frozen phonology thawed: The analysis and remediation of a developmental disorder of real word phonology. European Journal of Disorders of Communication, 27: 343–365. Cassidy, B. (1994). An Investigation into the Auditory Discrimination Skills of Normally Developing Children Using Single Word and Sentence Stimuli. Unpublished MSc project, Department of Human Communication Science, University College London. Chiat, S. (2000). Understanding Children with Language Problems. Cambridge: Cambridge University Press. Chiat, S. (2001). Mapping theories of developmental language impairment: Premises, predictions and evidence. Language and Cognitive Processes, 16(2/3): 113–142. Chiat, S. & Hunt, J. (1993). Connections between phonology and semantics: An exploration of lexical processing in a language impaired child. Child Language Teaching and Therapy, 9: 200–213. Claessen, M., Heath, S., Fletcher, J., Hogben, J. & Leita ˜ o, S. (submitted) Quality of phonological representations: A window into the lexicon?

References

233

Coffield, C. (1994). An Investigation into the Speech Output of Four Down’s Syndrome Adolescents. MSc Thesis, London: National Hospital’s College of Speech Sciences/Institute of Neurology. Connery, V. (1992). The Nuffield Centre Dyspraxia Programme. London: The Nuffield Hearing and Speech Centre. Constable, A. (2001). A psycholinguistic approach to word-finding difficulties. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Constable, A., Stackhouse, J. & Wells, B. (1997). Developmental word-finding difficulties and phonological processing: The case of the missing handcuffs. Applied Psycholinguistics, 18: 507–536. Conti-Ramsden, G., Botting, N. & Faragher, B. (2001). Psycholinguistic markers for specific language impairment. Journal of Child Psychology and Psychiatry, 42(6), 741–748. Conti-Ramsden, G., Botting, N., Simkin, Z. & Knox, E. (2001). Follow-up of children attending infant language units: Outcomes at 11 years of age. International Journal of Language and Communication Disorders, 36(2): 207–219. Corrin, J. (2001a). From profile to programme: Steps 1–2. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Corrin, J. (2001b). From profile to programme: Steps 3–6. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Corrin, J., Tarplee, C. & Wells, B. (2001). Interactional linguistics and language development: A conversation analytic perspective on emergent syntax. In Selting, M. & Couper-Kuhlen, E. (eds.), Studies in Interactional Linguistics, Amsterdam: Benjamins. Crosbie, S. & Dodd, B. (2001). Training auditory discrimination: A single case study. Child Language Teaching and Therapy, 17(3), 173–194. Crosbie, S., Holm, A. & Dodd, B. (2005). Intervention for children with severe speech disorder: A comparison of two approaches. International Journal of Language and Communication Disorders, 40: 467–491. Cruttenden, A. (1994). Gimson’s Pronunciation of English (5th edn). London: Edward Arnold. Dankovicová, J., Pigott, K., Peppé, S. & Wells, B. (2004) Temporal markers of prosodic boundaries in children’s speech production. Journal of the International Phonetic Association, 34: 17–36. Darili, E. (1994) An Investigation into the Lexical-phonological Representations of Normal Nursery Children. Unpublished BSc project. London: Department of Human Communication Science, University College London. Dean, E., Howell, J., Hill, A. & Waters, D. (2000). Metaphon Resource Pack. Windsor: NFER-Nelson. Dent, H. (2001). Electropalatography: A tool for psycholinguistic therapy. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Dodd, B. (ed.). (1995). Differential Diagnosis and Treatment of Children with Speech Disorder. London: Whurr Publishers. Dodd, B. (ed.). (2005). Differential Diagnosis and Treatment of Children with Speech Disorder (2nd edn). London: Whurr Publishers.

234

References

Dodd, B. & Bradford, A. (2000). A comparison of three therapy methods for children with different types of developmental phonological disorder. International Journal of Language and Communication Disorders, 35: 189–209. Dodd, B., Holm, A., Crosbie, S. & McIntosh, B. (2006). A core vocabulary approach for management of inconsistent speech disorder. Advances in SpeechLanguage Pathology, 8(3): 220–230. Dodd, B., Hua, Z., Crosbie, S., Holm, A. & Ozanne, A. (2002). Diagnostic Evaluation of Articulation and Phonology (DEAP), New York: Psychological Corporation. Dollaghan, C., Biber, M. and Campbell, T. (1995). Lexical influences on non-word repetition. Applied Psycholinguistics, 16: 211–222. Dry, S. (1997) An Investigation into the Auditory Processing and Literacy Skills of a Boy with Landau-Kleffner Syndrome. Unpublished MSc project. London: Department of Human Communication Science, University College London. Dunn, L., Dunn, L., Whetton, C. & Burley, J. (1997). British Picture Vocabulary Scale (2nd edn). Windsor: NFER-Nelson. Ebbels, S. (2000). Psycholinguistic profiling of a hearing-impaired child. Child Language Teaching and Therapy, 16(1): 3–22. Elliot, C., Murray, D. & Pearson, L. (1983). British Ability Scales. Windsor: NFER-Nelson. Enderby, P. & Phillips, R. (1986). Speech and language handicap: Towards knowing the size of the problem. British Journal of Disorders of Communication, 21: 151–165. Felsenfeld, S., McGue, M. & Broen, P. (1995). Familial aggregation of phonological disorders: Results from a 28-year follow-up. Journal of Speech and Hearing Research, 38, 1091–1107. Fletcher, S. (1978). The Fletcher Time-by-Count Test of Diadochokinetic Syllable Rate. Tigard, OR: C.C. Publications. Fox, A. (2002). PLAKSS – Psycholinguistische Analyse kindlicher Sprechstörungen. Frankfurt: Swets-Test Services. Fox, A. (2004). Kindliche Aussprache-Störungen. Idstein: Schulz-Kirchner Verlag. Fox, A., Dodd, B. & Howard, D. (2002). Risk factors for speech disorders in children. International Journal of Language and Communication Disorders, 37(2): 117–133. Gardner, H. (1997). Are your minimal pairs too neat? The dangers of phonemicisation in phonology therapy. European Journal of Disorders of Communication, 32(2): 167–175. Gathercole, S. & Baddeley, A. (1996). The Children’s Test of Nonword Repetition. London: Psychological Corporation. Gathercole, S., Willis, C., Emslie, H. & Baddeley, A. (1991). The influences of syllables and wordlikeness on children’s repetition of nonwords. Applied Psycholinguistics, 12: 349–367. German, D. J. (1989). Test of Word Finding (TWF), Allen, TX: DLM Teaching Resources. German, D. J. (1998a). Test of Word Finding (TWF) (2nd edn), Allen, TX: DLM Teaching Resources.

References

235

German, D. J. & Newman, R. S. (2004). The impact of lexical factors on children’s word finding errors. Journal of Speech, Language and Hearing Research, 47(3): 624–636. Gillon, G. (2004). Phonological Awareness: From Research to Practice. New York: The Guilford Press. Goldman, R. & Fristoe, M. (2000). The Goldman-Fristoe Test of Articulation (2nd edn, GFTA-2). London: Pearson Assessments. Goldman, R., Fristoe, M. & Woodcock, W. (1972). Goldman-Fristoe-Woodcock Tests of Articulation, Circle Pines, MN: American Guidance Service. Goldman, R., Fristoe, M. & Woodcock, W. (1975). Goldman-Fristoe-Woodcock Auditory Skills Battery. Circle Pines, MN: American Guidance Service. Goodman, R. (1997). The Strengths and Difficulties Questionnaire: A research note. Journal of Child Psychology, Psychiatry, and Allied Disciplines, 38: 581–586. Gourlay, S., Joffe, V. & Levin, A. (2005). Speech impairment in children: A psycholinguistic approach. Bulletin of the Royal College of Speech and Language Therapists, 638: 12–17. Grunwell, P. (1985). PACS – Phonological Assessment of Child Speech. Windsor: NFER-Nelson. Grunwell, P. (1987). Clinical Phonology (2nd edn). London: Croom Helm. Grunwell, P. (1989). Developmental phonological disorders and normal speech development: A review and illustration. Child Language Teaching and Therapy, 5: 304–319. Grunwell, P. & Harding, A. (1995). PACS TOYS Developmental Assessment. Windsor: NFER-Nelson. Habgood, M. (2000). An Investigation into the Performance of Normally Developing Children aged 6–7 years on Oral Movements and Silent and Spoken Diadochokinetic Tasks. Unpublished BSc project, London: Department of Human Communication Science, University College London. Hayden, D. (2006). The PROMPT model: Use and application for children with mixed phonological-motor impairment. Advances in Speech-Language Pathology, 8(3): 265–281. Henry, C. (1990). The development of oral diadochokinesis and non-linguistic rhythmic skills in normal and speech-disordered young children. Clinical Linguistics and Phonetics, 4(2): 121–138. Hewlett, N., Gibbon, F. & Cohen-Mackenzie, W. (1998). When is a velar an alveolar? Evidence supporting a revised psycholinguistic model of speech production in children. International Journal of Language and Communication Disorders, 33: 161–176. HMSO (1995). Standard Occupational Classification (Employment Department Group, Office of Population Censuses and Surveys. 2nd edn, Vol. 2 Coding Index). London: HMSO (Great Britain Office of Population Censuses and Surveys). Hodson, B. (2004). The Hodson Assessment of Phonological Patterns (3rd edn, HAPP-3). Austin, TX: Pro-Ed. Hodson, B. (2006). Identifying phonological patterns and projecting remediation cycles: Expediting intelligibility gains of a 7 year old Australian child. Advances in Speech-Language Pathology, 8(3): 257–264.

236

References

Howard, S. (2004). Connected speech processes in developmental speech impairment: Observations from an electropalatographic perspective. Clinical Linguistics and Phonetics, 18: 405–417. Howard, S., Wells, B. & Local, J. (in press 2008). Connected speech. In Ball, M., Perkins, M., Mueller, N. & Howard, S. (eds). The Handbook of Clinical Linguistics. Oxford: Blackwell. James, D. (2001). The use of phonological processes in Australian children aged 2 to 7:11 years. Advances in Speech-Language Pathology, 3: 109–128. Johnson, C., Beitchman, J. H., Young, A. R., Escobar, M., Atkinson, L., Wilson, B., Brownlie, E. B., Douglas, L., Taback, N., Lam, I. & Wang, M. (1999). Fourteenyear follow-up of children with and without speech/language impairments: Speech/language stability and outcomes. Journal of Speech, Language and Hearing Research, 42: 744–760. Kamil, M. L. & Rudegeair, R. (1972). Methodological improvements in the assessment of phonological discrimination in children. Child Development, 43: 1087–1091. Knox, E. & Conti-Ramsden, G. (2003). Bullying risks of 11-year-old children with specific language impairment (SLI): Does school placement matter? International Journal of Language & Communication Disorders, 38(1): 1–12. Lahey, M. & Edwards, J. (1995). Specific language impairment: Preliminary investigation of factors associated with family history and patterns of language performance. Journal of Speech and Hearing Research, 38: 643–657. Laing, E. & Hulme, C. (1999). Phonological and semantic processes influencing beginning readers ability to learn to read words. Journal of Experimental Child Psychology, 73: 183–207. Larrivee, L. & Catts, H. (1999). Early reading achievement in children with expressive phonological disorders. American Journal of Speech-Language Pathology, 8: 118–128. Law, J. & Conti-Ramsden, G. (2000). Treating children with speech and language impairments. British Medical Journal, 321: 908–909. Law, J. & Sivyer, S. (2003). Promoting the communication skills of primary school children excluded from school or at risk of exclusion: An intervention study. Child Language Teaching and Therapy, 19(1): 1–25. Law, J., Boyle, J., Harris, F., Harkness, A. & Nye, C. (2000) Prevalence and natural history of primary speech and language delay: Findings from a systematic review of the literature. International Journal of Language and Communication Disorders, 35: 165–188. Leitao, S. & Fletcher, J. (2004). Literacy outcomes for students with speech impairment: Long term follow-up. International Journal of Language and Communication Disorders, 39: 245–256. Leitao, S., Hogben, J. & Fletcher, J. (1997). Phonological processing skills in speech and language impaired children. European Journal of Disorders of Communication, 32(2): 91–113. Lewis, B. & Freebairn, L. (1997). Speech production skills of nuclear family members of children with phonology disorders. Language and Speech, 41: 45–61. Lewis, B., Ekelman, B. & Aram, D. (1989). A familial study of severe phonological disorders. Journal of Speech, Language and Hearing Research, 32: 713–724. Lewis, B., Shribog, L., Freebairn, L., Hansen, A., Stein, C., Taylor, H.G., Iyengar, S.K., (2006) The genetic bases of speech sound disorders: Evidence

References

237

from spoken and written language. Journal of Speech, Language, and Hearing Research 49(6) 1294–1310 Lindsay, G., Soloff, N., Law, J., Band, S., Peacey, N., Gascoigne, M. & Radford, J. (2002). Speech and language therapy services to education in England and Wales. International Journal of Language and Communication Disorders, 37(3): 273–288. Locke, A. (2006). One Step at a Time. London: Network Continuum Education. Locke, A., Ginsborg, J. & Peers, I. (2002). Development and disadvantage: Implications for the early years and beyond. International Journal of Language and Communication Disorders, 37(1): 3–15. Locke, J. (1980a). The inference of speech perception in the phonologically disordered child. Part I: A rationale, some criteria, the conventional tests. Journal of Speech and Hearing Disorders, 45: 445–468. Locke, J. (1980b). The inference of speech perception in the phonologically disordered child. Part II: Some clinically novel procedures, their use, some findings. Journal of Speech and Hearing Disorders, 45: 445–468. Loucas, T. & Marslen-Wilson, W. D. (1998). The development of spoken word recognition: Experimental and computational studies. In Proceedings of the Twentieth Annual Conference of the Cognitive Science Society. Mahwah, NJ: Lawrence Erlbaum Associates. Maassen, B., Hulstijn, W., Kent, R., Peters, H. & van Lieshout, P. H. M. M. (eds) (2004). Speech Motor Control in Normal and Disordered Speech. Oxford: Oxford University Press. Magnusson, E. & Naucler, K. (1990). Reading and spelling in language disordered children – linguistic and metalinguistic pre-requisites: Report on a longitudinal study. Clinical Linguistics and Phonetics, 4(1): 49–61. Marshall, J., Stojanovik, V. & Ralph, S. (2002). ‘I never gave it second thought’: PGCE students’ attitudes towards the inclusion of children with speech and language impairments. International Journal of Language and Communication Disorders, 37(4): 475–489. Martin, N. & Brownell, R. (2005). Test of Auditory Processing Skills (3rd edn, TAPS-3). Novato, CA: Academic Therapy Publications. Matthews, K. (2005). Psycholinguistic Assessment and Acoustic Analysis of a Child with Persisting Speech Difficulties: A Follow-up Study. Unpublished thesis. Department of Human Communication Sciences, University of Sheffield. McGregor, K. & Leonard, L. (1989). Facilitating word-finding skills of languageimpaired children. Journal of Speech and Hearing Disorders, 54: 141–147. McLeod, S. (2004). Speech pathologists’ application of the ICF to children with speech impairment. Advances in Speech-Language Pathology, 6: 75–81. Milloy, N. & Morgan Barry, R. (1990) Developmental neurological disorders. In Grunwell, P. (ed.), Developmental Speech Disorders. London: Whurr Publishers. Morgan Barry, R. (1988). Auditory Discrimination and Attention Test. Windsor: NFER Nelson. Morgan Barry, R. (1994). Observing and assessing auditory skills in children. In Wirz, S. (ed.) Perceptual Approaches to Communication Disorders. London: Whurr Publishers.

238

References

Mroz, M. (2006). Providing training in speech and language for education professionals: Challenges, support and the view from the ground. Child Language, Teaching and Therapy, 22(2): 155–176. Muter, V., Hulme, C. & Snowling, M. (1997). Phonological Abilities Test (PAT). London: Psychological Corporation. Nash, P. (2006). Assessment and management of psychosocial aspects of reading and language impairments. In Snowling, M. & Stackhouse, J. (eds), Dyslexia Speech and Language: A Practitioner’s Handbook (2nd edn). London: Whurr Publishers. Nash, P., Stengelhofen, J., Toombs, L. & Brown, J. (2002). Improving Children’s Communication: Managing Persistent Communication Difficulties. London: Whurr Publishers. Nathan, L. (2001). The Development of Speech Processing Skills in Children with and without Speech Difficulties. Unpublished PhD thesis. London: Department of Human Communication Science, University College London. Nathan, L. & Simpson, S. (2001). designing a literacy programme for a child with a history of speech difficulties. In Stackhouse, J. & Wells, B. (eds., Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Nathan, L., Stackhouse, J., Goulandris, N. & Snowling, M. (2004a). The development of early literacy skills among children with speech difficulties: A test of the ‘critical age hypothesis’. Journal of Speech, Language and Hearing Research, 47: 377–391. Nathan, L., Stackhouse, J., Goulandris, N. & Snowling, M. (2004b). Educational consequences of developmental speech disorder: Key Stage I National Curriculum assessment results in English and mathematics. British Journal of Educational Psychology, 74: 173–186. Nation, K. & Snowling, M. (1998). Individual differences in contextual facilitation: Evidence from dyslexia and poor reading comprehension. Child Development, 69: 996–1011. Neale, M. (1997). Neale Analysis of Reading Ability: Second Revised British Edition. Windsor: NFER Nelson. Newcomer, P. & Hammill, D. (1997). Test of Language Development: Primary (3rd edn). Austin, TX: Pro-Ed. Newton, C. (1999). Connected Speech Processes in Phonological Development: Word Glue and Other Sticky Situations. PhD dissertation. London: University College London. Newton, C. & Wells, B. (1999). The development of between-word processes in the connected speech of children aged between three and seven. In Maassen, N. & Groenen, P. (eds), Pathologies of Speech and Language: Advances in Clinical Phonetics and Linguistics. London: Whurr Publishers. Newton, C. & Wells, B. (2002). Between word junctures in early multiword speech. Journal of Child Language. 29: 275–299. Newton, M. & Thomson, M. (1982). Revised Aston Index. Cambridge: Learning Development Aids.

References

239

Ozanne, A. (1992). Normative data for sequence oral movements and movements in context for children aged three to five years. Australian Journal of Human Communication Disorders, 20: 47–63. Pascoe, M., Stackhouse, J. & Wells, B. (2005). Phonological therapy within a psycholinguistic framework: Promoting change in a child with persisting speech difficulties. International Journal of Language and Communication Disorders, 39: 1–32. Pascoe, M., Stackhouse, J. & Wells, B. (2006). Persisting Speech Difficulties in Children. Children’s Speech and Literacy Difficulties 3. Chichester: John Wiley & Sons Ltd. Peppé, S. & McCann, J. (2003). Assessing intonation and prosody in children with atypical language development: The PEPS-C test and the revised version. Clinical Linguistics and Phonetics, 17(4/5): 345–354. Popple, J. & Wellington, W. (2001). Working together: The psycholinguistic approach within a school setting. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Rees, R. (2001). Principles of psycholinguistic intervention. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Renfrew, C. (1988). Renfrew Action Picture Test (3rd edn). Oxford: Winslow Press. Renfrew, C. (1995a). Word Finding Vocabulary Test (4th edn). Oxford: Winslow Press. Renfrew, C. (1995b). The Renfrew Bus Story (3rd edn). Oxford: Winslow Press. Rvachew, S., Ohberg, A., Grawburg, M. & Heyding, J. (2003). Phonological awareness and phonemic perception in 4-year old children with delayed expressive phonology skills. American Journal of Speech-Language Pathology, 12: 463–471. Roy, P. & Chiat, S. (2004). A prosodically controlled word and nonword repetition task for 2-4 year olds: Evidence from typically developing children. Journal of Speech, Language and Hearing Research, 47(1): 223–234. Sadler, J. (2005). Knowledge, attitudes and beliefs of the mainstream teachers of children with a preschool diagnosis of speech/language impairment. Child Language Teaching and Therapy, 21: 147–163. Schissel, R. (1980). The role of selected auditory skills in the misarticulation of /s/, /r/ and /t/ by third grade children. British Journal of Disorders of Communication, 15(2): 129–139. Shockey, L. (2003). Sound Patterns of Spoken English. Oxford: Blackwell. Shriberg, L. (1994). Five subtypes of developmental phonological disorders. Clinics in Communication Disorders, 4: 38–53. Shriberg, L. (1997). Developmental phonological disorders: One or many? In Hodson, B. & Edwards, M. (eds), Perspectives in Applied Phonology. Frederick, MD: Aspen Publications. Shriberg, L. & Kwiatkowski, J. (1994). Developmental phonological disorders I: A clinical profile. Journal of Speech and Hearing Research, 37: 1100–1126. Shriberg, L., Kwiatkowski, J. & Gruber, F. (1994). Developmental phonological disorders II: Short-term speech sound normalisation. Journal of Speech and Hearing Research, 37: 1127–1150.

240

References

Shriberg, L., Tomblin, J. & McSweeney, J. (1999). Prevalence of speech delay in 6-year-old children and comorbidity with language impairment. Journal of Speech, Language and Hearing Research, 42(6): 1461–1481. Simkin, Z. & Conti-Ramdsen, G. (2006). Evidence of reading difficulty in subgroups of children with specific language impairment. Child Language Teaching and Therapy, 22(3): 315–331. Snowling, M. (2000). Language and literacy skills: Who is at risk and why? In Bishop, D. V. M. & Leonard, L. (eds), Speech and Language Impairments in Children: Causes, Characteristics, Intervention and Outcome. Hove: Psychology Press. Snowling, M. (2006). Language skills and learning to read: The dyslexia spectrum. In Snowling, M. & Stackhouse, J. (eds). Dyslexia Speech and Language: A Practitioner’s Handbook (2nd edn). London: Whurr Publishers. Snowling, M., Chiat, S. & Hulme, C. (1991) Words, non-words, and phonological processes: Some comments on Gathercole, Willis, Emslie, and Baddeley. Applied Psycholinguistics, 12: 369–373. Snowling, M., Stothard, S. E. & McLean, J. (1996). The Graded Nonword Reading Test. Reading: Thames Valley Test Company. Snowling, M., van Wagtendonk, B. & Stafford, C. (1988). Object-naming deficits in developmental dyslexia. Journal of Research in Reading, 11(2): 67–85. Speake, J. (2003). How to Identify and Support Children with Speech and Language Difficulties. Wisbech: LDA. Speech Assessment and Interactive Learning System (SAILS). Software development and Information Technology Utilization Group. http://sdg.cllrnet.ca/ SAILS/about.htm. Stackhouse, J. (1989). Phonological Dyslexia in Children with Developmental Verbal Dyspraxia. Unpublished PhD thesis. London: Psychology Department, University College London. Stackhouse, J. (2006). Speech and spelling difficulties: What to look for. In Snowling, M. & Stackhouse, J. (eds). Dyslexia Speech and Language: A Practitioner’s Handbook. (2nd edn). London: Whurr Publishers. Stackhouse, J. & Snowling, M. (1992). Developmental verbal dyspraxia II: A developmental perspective on two case studies. European Journal of Disorders of Communication, 27(1): 35–54. Stackhouse, J. & Wells, B. (1993). Psycholinguistic assessment of developmental speech disorders. European Journal of Disorders of Communication, 28: 331–348. Stackhouse, J. & Wells, B. (1996). Developmental supermodels. Bulletin of the Royal College of Speech and Language Therapists. 527: 9–10. Stackhouse, J., & Wells, B. (1997). Children’s Speech and Literacy Difficulties 1: A Psycholinguistic Framework. London: Whurr Publishers. Stackhouse, J. & Wells, B. (2001). Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers. Stackhouse, J., Nathan, L., Goulandris, N. & Snowling, M. (2002). The Relationship Between Speech Disorders and Literacy Problems: Identification of the At Risk Child. Report on a 4 year Longitudinal Study to Research and Development, London: North Thames Regional Health Authority. Stackhouse, J., Pascoe, M. & Gardner, H. (2006). Intervention for a child with persisting speech and literacy difficulties: A psycholinguistic approach. Advances in Speech–Language Pathology, 8(3): 231–244.

References

241

Stemberger, J. (1989). Speech errors in early child language production. Journal of Memory and Language, 28: 164–188. Stothard, S., Snowling, M., Bishop, D., Chipchase, B. & Kaplan, C. (1998). Language-impaired preschoolers: A follow-up into adolescence. Journal of Speech, Language, and Hearing Research, 41: 407–418. Tarplee, C. (1996). Working on young children’s utterances: Prosodic aspects of repetition during picture labelling. In Couper-Kuhlen, E. & Selting, M. (eds). Prosody in Conversation: Interactional Studies. Cambridge: Cambridge University Press. Thompson, I. (2000). Børns taleog skriftsprologue varskeligher. Herning: SpecialPaedagogisk Forlag. Thoonen, G., Maassen, B., Gabreels, F., Schreuder, R. & de Swart, B. (1997). Towards a standardised assessment procedure for developmental apraxia of speech. European Journal of Disorders of Communication, 32(1): 37–60. Tomblin, J., Smith, E. & Zhang, X. (1997). Epidemiology of specific language impairment: Prenatal and perinatal risk factors. Journal of Communication Disorders, 30: 325–343. Vance, M. (1995). Investigating speech processing skills in young children. In Caring to Communicate, Proceedings of the Golden Jubilee Conference. London: Royal College of Speech and Language Therapists. 287–293. Vance, M. (1996). Portuguese translation of ‘Assessing speech processing skills in children: A task analysis’. In Snowling, M. & Stackhouse, J. (eds), Dislexia, Fala e Linguagem: um Manual do Profissional. Sao Pablo: Artmet. Vance, M. (2001). Speech Processing and Short-term Memory in Children with Normal and Atypical Speech and Language Development. Unpublished PhD thesis, London: University College London. Vance, M., Rosen, S. & Coleman, M. (in preparation). Speech Input Processing in Children (SIPc). Vance, M., Stackhouse, J. & Wells, B. (1994). ‘Sock the Wock the Pit-Pat-Pock’: Children’s responses to measures of rhyming ability. Work in Progress, National Hospitals College of Speech Sciences, 4: 171–185. Vance, M., Stackhouse, J. & Wells, B. (1995). The relationship between naming and word-repetition skills in children aged 3 to 7 years. Work in Progress, UCL Department of Human Communication Sciences, 5: 127–133. Vance, M., Stackhouse, J. & Wells, B. (2005). Speech production skills in children aged 3–7 Years. International Journal of Language and Communication Disorders. 40(1): 29–48. Van der Lely, H., Gardner, H., Froud, K. & McClelland, A. (2006). Grammar and Phonology Screening (GAPS). London: DLDCN.com. Vargha-Khadem, F., Gadian, D., Copp, A. & Mishkin, M. (2005). FOXP2 and the neuroanatomy of speech and language. Nature Reviews Neuroscience, 6: 131–138. Walker, J. & Archibald, L. (2006). Articulation rate in preschool children: A 3 year longitudinal study. International Journal of Language and Communication Disorders, 41(5): 541–565. Waters, D. (2001). Using input processing strengths to overcome speech output difficulties. In Stackhouse, J. & Wells, B. (eds), Children’s Speech and Literacy Difficulties 2: Identification and Intervention. London: Whurr Publishers.

242

References

Watts, N. (2004). Assessment of vowels summary. ACQuiring Knowledge in Speech, Language and Hearing (ACQ), 6(1): 22–25. http://www.speechlanguage-therapy.com/vowelassessmentACQ2004.pdf. Webster, P. & Plante, A. (1992). Effects of phonological impairment on word, syllable and phoneme segmentation and reading. Language, Speech and Hearing Services in Schools, 23: 176–182. Webster, P., Plante, A. & Couvillion, L. (1997). Phonologic impairment and prereading: Update on a longitudinal study. Journal of Learning Disabilities, 30(4): 365–375. Wechsler, D. (1990). Wechsler Preschool and Primary Scale of Intelligence (WPSSI). London: Psychological Corporation. Weiner, F. (1979). Phonological Process Analysis (PPA). Baltimore, MD: University Park Press. Weiss, C., Gordon, M. & Lillywhite, H. (1987). Clinical Management of Articulatory and Phonological Disorders. Baltimore, MD: Williams and Wilkins. Wells, B. (1994). Junction in developmental speech disorder: A case study. Clinical Linguistics and Phonetics, 8(1): 1–25. Wells, B. & Peppé, S. (2003) Intonation abilities of children with speech and language impairments. Journal of Speech Language and Hearing Research, 46: 5–20. Wells, B., Peppé, S. & Goulandris, N. (2004). Intonation development from five to thirteen. Journal of Child Language, 31: 749–778. Wells, B., Stackhouse, J. & Vance, M. (1999). La conscience phonologique dans le cadre d’une evaluation psycholinguistique de l’enfant. Re-education Orthophonique, 197: 3–12. Wepman, J. & Reynolds, W. (1987). Wepman’s Auditory Discrimination Test (2nd edn). Los Angeles, CA: Western Psychological Services. Wiig, E., Secord, W. & Semel, E. (2000). Clinical Evaluation of Language Fundamentals (3rd edn). London: Psychological Corporation. Williams, K. (2006). Expressive Vocabulary Test (2nd edn, EVT-2). London: Pearson Assessments. Williams, N. & Chiat, S. (1993). Processing deficits in children with phonological disorder and delay: A comparison of responses to a series of output tasks. Clinical Linguistics and Phonetics, 7: 145–160. Williams, P. & Stackhouse, J. (1998). Diadochokinetic skills: Normal and atypical performance in children aged 3–5 years. International Journal of Language and Communication Disorders, 33(Suppl): 481–486. Williams, P. & Stackhouse, J. (2000). Rate, accuracy and consistency: Diadochokinetic performance of young, normally developing children. Clinical Linguistics and Phonetics, 14(4): 267–293. Williams, P. & Stephens, H. (2004). Nuffield Centre Dyspraxia Programme (3rd edn). Windsor: Miracle Factory. Wood, J., Wright, J. & Stackhouse, J. (2001). Language and Literacy: Joining Together. Reading: British Dyslexia Association. Wright, J. & Wood, J. (2006). Supporting language and literacy in the early years: Interdisciplinary training. In Snowling, M. & Stackhouse, J. (eds). Dyslexia Speech and Language: A Practitioner’s Handbook (2nd edn). London: Whurr Publishers.

Appendix A.1 Tasks for Each Question in the Psycholinguistic Assessment Framework

(Presented in top to bottom order for both input and output sides of the model. The tasks included here are all in this volume and focus on speech and auditory discrimination. This list should therefore be viewed in conjunction with the list that appeared in Book 1 (Stackhouse & Wells, 1997, Appendix 2) and Book 3 (Pascoe, Stackhouse & Wells, 2006, Appendix 2).

INPUT F. Is the child aware of the internal structure of phonological representations? • See Appendix 2 in Book 1 or Book 3 (Stackhouse & Wells, 1997; Pascoe, Stackhouse & Wells, 2006) E. Are the child’s phonological representations accurate? • Auditory Lexical Discrimination Task 1: Mispronunciation Detection (Vance, 1995), Chapter 4. • Auditory Lexical Discrimination Task 2: With and Without Pictures (Constable, Stackhouse & Wells, 1997), Chapter 4. • Auditory Lexical Discrimination Task 3: Mispronunciation Detection for Individual Children (Locke, 1980a, b), Chapter 4. • Auditory Lexical Discrimination Task 4: Words in Sentences (Cassidy, 1994), Chapter 4.

243

244

Appendix A.1

D. Can the child discriminate between real words? • Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words (Bridgeman & Snowling, 1988), Chapter 3. C. Does the child have language-specific representations of word structures? • Auditory Discrimination Task 4: Legal vs Illegal Non-words (from Dry, 1997 and Darili, 1994, Chapter 3). B. Can the child discriminate speech sounds without reference to lexical representations? • Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Non-Words (Bridgeman & Snowling, 1988), Chapter 3. • Auditory Discrimination Task 2: Complex Non-words (Stackhouse, 1989), Chapter 3. • Auditory Discrimination Task 3: ABX Task (Vance, 1995), Chapter 3. A. Does the child have adequate auditory perception? • See Appendix 2 in Book 1 or Book 3 (Stackhouse & Wells, 1997; Pascoe, Stackhouse & Wells, 2006).

OUTPUT G. Can the child access accurate motor programmes? • Picture Naming Task 1 (Vance, Stackhouse & Wells, 2005), Chapter 5. • Picture Naming Task 2 (Snowling, van Wagtendonk & Stafford, 1988), Chapter 5. • Speech Accuracy and Consistency Task 3, Picture Naming (Coffield, 1994), Chapter 7. H. Can the child manipulate phonological units? • See Appendix 2 in Book 1 or Book 3 (Stackhouse & Wells, 1997; Pascoe, Stackhouse & Wells, 2006).

Appendix A.1

245

I. Can the child articulate real words accurately? • Word Repetition Task (Vance, Stackhouse & Wells, 2005; Nathan et al., 2004a), Chapter 5. • Words-in-Sentences Task (Vance, Stackhouse & Wells, 1995), Chapter 6. • Connected Speech Processes Repetition Task (Newton, 1999), Chapter 6. • Final Consonant Juncture Repetition Task (Pascoe, Stackhouse & Wells, 2006), Chapter 6. • Speech Accuracy, Rate and Consistency Task 2, Words (Williams & Stackhouse, 2000), Chapter 7. • Speech Accuracy and Consistency Task 3, Word Repetition (Coffield, 1994), Chapter 7. J. Can the child articulate speech without reference to lexical representations? • Non-word Repetition Task (Vance, Stackhouse & Wells, 2005; Nathan et al., 2004a), Chapter 5. • Speech Accuracy, Rate and Consistency Task 2: Non-Words and Syllable Sequences (Williams & Stackhouse, 2000), Chapter 7. • Speech Accuracy and Consistency Task 3: Non-Word Repetition (Coffield, 1994), Chapter 7. K. Does the child have adequate sound production skills? • Speech Accuracy, Rate and Consistency Task 1: Oral Movements and Silent DDK Rates (Williams & Stackhouse, 2000), Chapter 7. L. Does the child reject his/her own erroneous forms? • Mispronunciation Self-Correction, Chapter 5. • Tester observes if child attempts to correct his or her output spontaneously.

Appendix A.2 Speech Processing Profile Name: Age:

Comments: d.o.b:

Date: Profiles:

Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

247

Appendix A.3 Speech Processing Model

Source: Stackhouse & Wells (1997). Reproduced by permission of John Wiley & Sons, Ltd.

249

Appendix B Standardised Assessments used in the Longitudinal Study

British Ability Scales (BAS; Elliott, Murray & Pearson, 1983). British Picture Vocabulary Scale (BPVS, 2nd edn; Dunn et al., 1997). Clinical Evaluation of Language Fundamentals (CELF-3 (UK)), Recalling Sentences (Wiig, Secord & Semel, 2000). Edinburgh Articulation Test (Anthony et al., 1971). Graded Non-word Reading Test (Snowling, Stothard & McLean, 1996). Neale Analysis of Reading Ability (2nd revd British edn; Neale, 1997). Phonological Abilities Test (PAT; Muter, Hulme & Snowling, 1997). Renfrew Action Picture Test (3rd edn; Renfrew, 1988). Test for Reception of Grammar (TROG, 2nd edn; Bishop, 1989). Renfrew Bus Story (3rd edn, Renfrew, 1995b). Wechsler Pre-School and Primary Scale of Intelligence (WPSSI-R), Block Design and Picture Completion (Wechsler, 1990).

251

Appendix C.1 Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words and Non-words – Reduced Version (from Bridgeman & Snowling, 1988) Name:___________ Date:_________ Age:______ Investigator:_________ Instructions: The task is demonstrated using the child’s name, e.g. if the child’s name is BEN, the child is asked if ‘BEN/BEN’ are the same or different; then if ‘BEN/OLIVER’ are the same or different, i.e. deliberately choosing a name that sounds quite different, then if ‘BEN/PEN’ are the same or different, i.e. choosing a name or an object that sounds similar. Intersperse ‘BEN/BEN’ with these practice items, feedback is given to ensure the child understands the task. The 4 practice items can then be presented, followed by the test items. No feedback is given for the test items. For the non-words tell the child that they will hear some ‘silly, made-up words’. Before presenting the real words, tell the child that ‘now the words will be words that you know’. All items can be presented in one session. One repetition is permitted for the test items if the child requests it or is not attending. The investigator should cover his/her mouth to avoid use of visual cues. Scoring: Record the child’s response, by circling S or D. BOLD CAPITALS indicate correct responses. Add up correct responses to obtain child’s scores, calculating totals for each stimulus type and total for all items. Normative data for this task can be found in Table 3.4.

253

254

Appendix C.1

REDUCED VERSION: NON-WORDS STIMULI

RESPONSES Feature change s

Sequence change

P1

/vɒs/ - /vɒt/

P2

/fεst/ - /fεts/

s

d

P3

/vɒst/ - /vɒts/

s

d

P4

/tεt/ - /tεt/

1

/kεst/ - /kεts/

s

D

s

D

s

d

d

2

/bles/ - /blet/

s

3

/zεt/ - /zεt/

S

d

4

/fɒt/ - /fɒs/

s

D

5

/kεs/ - /kεt/

s

D

6

/dts/ - /dst/

7

/vt/ - /vs/

s

D

8

/pəυts/ - /pəυts/

S

d

9

/zεts/ - /zεts/

S

d

10

/fɒts/ - /fɒst/

s

D

11

/vts/ - /vst/

s

D

12

/bs/ - /bs/

13

/jets/ - /jest/

s

D

14

/dt/ - /ds/

s

15

/pəυt/ - /pəυt/

S

d

16

/jes/ - / jet/

s

D

17

/bst/ - /bst/

S

d

18

/blest/ - /blets/

s

D

/3

/6

S

/3 TOTAL SCORE

D

d D

/6 / 18

Appendix C.1

255

REDUCED VERSION: REAL-WORDS STIMULI

RESPONSES Feature change Same

Different

S

d

Sequence change Same

Different

s

D

S

d

s

D D

19

kit / kit

20

hits / hissed

21

messed / messed

22

guess / get

s

D

23

race / rate

s

D

24

mitts / missed

25

plate / place

26

guessed / gets

s

27

kissed / kissed

S

d

28

rates / raced

s

D

29

tots / tossed

s

D

30

tot / toss

s

D

31

miss / mitt

s

D

32

hit / hiss

s

D

33

goats / goats

S

d

34

met / met

S

d

35

placed / plates

s

D

36

goat / goat

S

d

/3

/6

/3

/6

s

D

TOTAL SCORE

/ 18

SUMMARY TABLE TOTAL FEATURE OR CLUSTER WORDS NON-WORDS FEATURE CHANGE

/9

/9

/18

CLUSTER SEQUENCE

/9

/9

/18

TOTAL WORD TYPE

/18

/18

TOTAL ALL

/36

Appendix C.2 Auditory Discrimination Task 1: Same/Different, S-Cluster Sequences, Words and Non-words – Shortest Form (from Bridgeman & Snowling, 1988) Name:___________ Date:________ Age:______ Investigator:__________ Instructions: The task is demonstrated using the child’s name, e.g. if the child’s name is BEN, the child is asked if ‘BEN/BEN’ are the same or different; then if ‘BEN/OLIVER’ are the same or different, i.e. deliberately choosing a name that sounds quite different, then if ‘BEN/PEN’ are the same or different, i.e. choosing a name or an object that sounds similar. Intersperse ‘BEN/BEN’ with these practice items, feedback is given to ensure the child understands the task. The 4 practice items can then be presented, followed by the test items. No feedback is given for the test items. For the non-words tell the child that they will hear some ‘silly, made-up words’. Before presenting the real words, tell the child that ‘now the words will be words that you know’. All items can be presented in one session. One repetition is permitted for the test items if the child requests it or is not attending. The investigator should cover his/her mouth to avoid use of visual cues. Scoring: Record the child’s response, by circling S or D. BOLD CAPITALS indicate correct responses. Add up correct responses to obtain child’s total score for all items. Normative data for this task can be found in Table 3.5.

257

258

Appendix C.2

SHORTEST FORM: NON-WORDS STIMULI

RESPONSES Feature change s

Sequence change

P1

/vɒs/ - /vɒt/

d

P2

/fεst/ - /fεts/

s

d

P3

/vɒst/ - /vɒts/

s

d

P4

/tεt/ - /tεt/

s

d

1

/bles/ - /blet/

s

D

2

/zεts/ - /zεts/

S

d

3

/jets/ - /jest/

s

D

4

/pəυt/ - /pəυt/

S

d

5

/jes/ - / jet/

s

D

6

/blest/ - /blets/

s

D

7

/dts/ - /dst/

s

D

8

/zεt/ - /zεt/

9

/pəυts/ - /pəυts/

S

d

10

/dt/ - /ds/

/2

/3

S

d

s

D

/2

/3

NON-WORD SCORE

/10

SHORTEST FORM: REAL WORDS STIMULI

RESPONSES Feature

Sequence

Same

Different

Same

Different

s

D

1

race / rate

s

D

2

met / met

S

d

3

placed / plates

4

goat / goat

S

d

5

miss / mitt

s

D

6

rates / raced

s

D

7

messed / messed

S

d

8

goats / goats

S

d

9

plate / place

10

mitts / missed

s

D

/2

/3

s

/2

D

/3

TOTAL WORD SCORE

/10

TOTAL SCORE

/20

Appendix C.3 Auditory Discrimination Task 2: Complex Non-words (from Stackhouse, 1989)

Name:__________ Date:________ Age:_____ Investigator:____________ Instructions: ‘I’m going to say some words and I want you to tell me if the two words sound the same or different. The words are silly, made-up words. If they are the same, then you must say “same” (or yes). If they sound different, then you must say ‘different’ (or no, or not the same). First we’ll practise some (practise using child’s own name, e.g. PETE/BEAT). Let’s do some more practice words.’ Administer P1–P4 below and then the test items. Feedback can be given on the practice items, but not for the test items. One repetition of each test item is permitted if the child requests it or is not attending. The investigator should casually cover his or her mouth to avoid use of visual cues. Scoring: Record the child’s response, by circling S or D; BOLD CAPITALS indicate correct responses. Add up correct responses to obtain child’s score. Circle incorrect responses in final column to indicate contrasts of difficulty. Add number of same and number of different responses correct and total responses correct. Calculate the percentage correct for all items, and for same responses and for different responses. Normative data for this task can be found in Table 3.8.

259

260

Appendix C.3

PRACTICE SET

RESPONSES

P1

/nst/ /nst/

s

P2

/gil/ /dil/

s

d

P3

/skɑ/ /stɑ/

s

d

P4

/kɔt/ /kɔt/

s

d

TEST ITEMS SET A

RESPONSES

CIRCLE IF INCORRECT

d

CIRCLE IF INCORRECT

1

/wεsp/ /wεps/

s

D

Cluster sequence

2

/ snmɒn/ / snmɒn/

S

d

(Same)

3

/wb/ /jb/

s

D

Place of articulation

4

/ ləυthas/ / ləυthaf/

s

D


5

/d ks/ /d ks/

S

d

(Same)

6

/ skkre/ / sttre/

s

D


7

/ ə tɔ/ /tə ɔ/

s

D

Metathesis

8

/ pnsəl/ / pnsəl/

S

d

(Same)

9

/ bkəs/ / kbəs/

s

D

Metathesis

10

/ bεskət/ / bεksət/

s

D

Cluster Sequence

S

d

(Same)

TEST ITEMS SET B 11

/stεmp/ /stεmp/

12

/bet/ /pet/

s

D

Voicing

13

/ rεkət/ / rεtək/

s

D

Metathesis

14

/smak/ /smak/

S

d

(Same)

15

/ r liskəυts/ / l riskəυts/

s

D

Metathesis

16

/ drgən/ / drgən/

S

d

(Same)

17

/ bkt/ / btk/

s

D

Metathesis

18

/ krvn/ / krvm/

s

D


19

/ rnd / / r /

s

D


20

/ spəυdə/ / spəυdə/

S

d

(Same)

TEST ITEMS SET C 21

/bg/ /bg/

S

d

(Same)

22

/ slεpə slεtə

s

D


23

/ pεpi/ / tεpi/

s

D


24

/ ks l/ / ksn /

s

D

Manner of articulation

25

/t nt/ /tnt/

s

D

Vowel

26

/tʃps/ /tʃps/

S

d

(Same)

27

/ b skɔts/ / p skɔts/

s

D

Voicing

28

/d l/ /d l/

S

d

(Same)

29

/d sk/ /d ks/

s

D

Cluster sequence

30

/ tʃkləυt/ / tʃkləυt/

S

d

(Same)

Appendix C.3 TEST ITEMS SET D

261

RESPONSES

CIRCLE IF INCORRECT

31

/sti/ /ski/

s

D


32

/be/ /be/

S

d

(Same)

33

/spəυb/ /spəυd/

s

D


34

/ tr ɑ/ / trðɑ/

s

D


35

/krεb/ /krb/

s

D

Vowel

36

/ bɑgli/ / bɑdli/

s

D


37

/tʃsp/ / tʃps/

s

D

Cluster sequence

38

/tɒlɒl v n / /tɒlɒ v n /

S

d

(Same)

39

/spəυd/ /spəυd/

S

d

(Same)

40

/ mtbəυk/ / mkbəυt/

s

D

Metathesis

TOTAL SAME / DIFFERENT % SAME / DIFFERENT TOTAL SCORE

/14

/26

%

%

/40 % Total correct:

Appendix C.4 Auditory Discrimination Task 3: ABX Task – Full Version (Vance, 1996)

Name:__________ Date:________ Age:_____ Investigator:____________ Instructions: Two soft toys are placed in front of the child (e.g. two different monkeys) and practice items are presented by pointing to one toy and saying ‘He says /səυf/’ and pointing to the second toy and saying ‘He says / təυf /’ and then asking ‘Who said /səυf /?’. The child is encouraged to point to one of the toys. Corrective feedback is given as required for the practice items. During the practice items the cue phrases are phased out, so that just the non-word stimuli are spoken as each toy is pointed to, e.g. /grb/ /sb/, and then the child is asked ‘Who said X?’. Finally all the cue phrases can be omitted, so that the X stimulus is also presented as a single non-word /spəυd/ /pəυd/; /spəυd/. The task items are presented with general encouragement as required but no feedback. One repetition is allowed on each item if requested by the child or if he or she is not attending. The investigator should cover his or her mouth to avoid use of visual cues. Scoring: Record the child’s response, by circling A or B. BOLD CAPITALS indicate correct responses. Add up correct responses to obtain child’s score. Circle incorrect responses in final column to indicate type of difficulty and look for any patterns. Normative data for this task can be found in Table 3.11.

263

264

Appendix C.4

FULL VERSION: LIST A A

B

X

RESPONSE

P1 /səυf/

/təυf/

/səυf/

a

b

P2 /spin/

/stin/

/stin/

a

b

CIRCLE IF INCORRECT

P3 /glzəs/

/klzəs/

/klzəs/

a

b

P4 /grb/

/sb/

/grb/

a

b

P5 /baυg/

/baυf/

/baυf/

a

b

P6 /spəυd/

/pəυd/

/spəυd/

a

b

1

/pəυt/

/pləυt/

/pləυt/

a

B

2

/sɔk/

/zɔk/

/zɔk/

a

B

Voicing

3

/vn/

/zn/

/vn/

A

b


4

/ɹɔf/

/ɹɔs/

/ɹɔs/

a

B


5

/nak/

/snak/

/nak/

A

b

Cluster reduction

6

/tɔn/

/tɹɔn/

/tɔn/

A

b

Cluster reduction

7

/ðɒm/

/θɒm/

/θɒm/

a

B

Voicing

8

/vεʃ/

/fεʃ/

/fεʃ/

a

B

Voicing

9

Cluster reduction

/wυtʃ/

/ɹυtʃ/

/wυtʃ/

A

b


10 /dutʃ/

/tutʃ/

/tutʃ/

a

B

Voicing

11 /bɹs/

/bɹʃ/

/bɹʃ/

a

B


12 /mɔt/

/mɔs/

/mɔt/

A

b


13 /tεt/

/kεt/

/kεt/

a

B


14 /sp nd /

/sp ndz/

/sp ndz/

a

B


15 /məυf/

/nəυf/

/məυf/

A

b


16 /lɔf/

/jɔf/

/lɔf/

A

b


17 /mɔk/

/bɔk/

/bɔk/

a

B


18 /d k/

/g k/

/d k/

A

b


19 /tʃi/

/tsi/

/tʃi/

A

b


20 /glεb/

/glεv/

/glεb/

A

b

Place and manner

21 / pɹt/

/ bɹt/

/ bɹt/

a

B

Voicing

22 / sped/

/ steb/

/ steb/

a

B

Cluster

23 / bɒkset/

/ bɒsket/

/ bɒsket/

a

B

Cluster

24 / delət/

/ telət/

/ delət/

A

b

Voicing

25 / d sbn /

/ b sdn /

/ b sdn /

a

B


26 / dzlə/

/ d lə/

/ dzlə/

A

b


27 / sɑsi/

/ sɑti/

/ sɑti/

a

B


28 / dɹεkt/

/ tɹεkt/

/ tɹεkt/

a

B

Voicing

29 / fluw/

/ sluw/

/ sluw/

a

B


30 / ʃɒfŋ/

/ fɒʃŋ/

/ ʃɒfŋ/

A

b

Metathesis

LIST A SCORE

/30

Appendix C.4

265

FULL VERSION: LIST B A

B

X

CIRCLE IF RESPONSES INCORRECT

1

/ smwəd /

/ fmwəd /

/ smwəd /

A

b


2

/ kɒtʃn /

/ gɒtʃn /

/ kɒtʃn /

A

b

Voicing

3

/ f ð/

/ ð f/

/ f ð/

A

b

Metathesis

4

/ nεmə/

/ mεnə/

/ mεnə/

a

B

Metathesis

5

/ jεd/

/ lεd/

/ jεd/

A

b


6

/tε ɔ/

/ ε tɔ/

/ ε tɔ/

a

B

Metathesis

7

/ stikə/

/ skitə/

/ skitə/

a

B

Metathesis

8

/ l jɔ/

/ j lɔ/

/ l jɔ/

A

b

Metathesis

9

/ sεd əs/

/ sεsəd /

/ sεd əs/

A

b

Metathesis

10 / slɔpə/

/ lɔpə/

/ slɔpə/

A

b

Cluster reduction

11 / b təfləυ/

/ b təfəυ/

/ b təfləυ/

A

b

Cluster reduction

12 / denzɑ/

/ densɑ/

/ densɑ/

a

B

Voicing

13 / hɔdɹ s/

/ hɔdɹ t/

/ hɔdɹ t/

a

B


14 /sp gtə/

/g sptə/

/sp gtə/

A

b

Metathesis

15 / pεɹəsit/

/ pεɹəʃit/


/ pεɹəʃit/

a

B

16 / kɹdəkaυl/ / kɹkədaυl/

/ kɹdəkaυl/

A

b

Metathesis

17 / flɒnt/

/ lfɒnt/

/ lfɒnt/

a

B

Metathesis

18 / ɒɹəpɹəυn/

/ ɒɹəpləυn/

/ ɒɹəpɹəυn/

A

b


19 / p d ɔmz/ / p d ɔbz/

/ p d ɔmz/

A

b


20 / dɒləfan/

/ tɒləfan/

/ dɒləfan/

A

b

Voicing

21 / hmbɑ /

/ hmbɑd/

/ hmbɑd/

a

B


22 / ɹəυndəbat/ / wəυndəbat/ / ɹəυndəbat/ A

b


23 / kvɒɹn /

/ kɹɒvn /

24 / ktəp lə / / kpət lə /

/ kɹvn /

a

B

Metathesis

/ kpət lə /

a

B

Metathesis

25 / gεm pjaυt/ / kεm pjaυt/

/ kεm pjaυt/

a

B

Voicing

26 / hlət pkə/ / hlək ptə/

/ hlət pkə/

A

b

Metathesis

27 / h stpl/

/ h sptl/

/ h sptl/

a

B

Metathesis

28 / m bɹ l/

/ m bl ɹ/

/ m bl ɹ/

a

B

Metathesis

29 / kɒŋgɹɑ/

/ gɒŋkɹɑ/

/ kɒŋgɹɑ/

A

b

Metathesis

/ tυlɒv

n /

a

B

Metathesis

30 / tυvɒl

n / / tυlɒv

n / LIST B SCORE

/30

TOTAL SCORE

/60

Appendix C.5 Auditory Discrimination Task 3: ABX Task – Short Form (Nathan et al., 2004)

Name:__________ Date:________ Age:_____ Investigator:____________ Instructions: Two soft toys are placed in front of the child (e.g. two different monkeys) and practice items are presented by pointing to one toy and saying ‘He says /səυf/’ and pointing to the second toy and saying ‘He says / təυf /’ and then asking ‘Who said /səυf /?’. The child is encouraged to point to one of the toys. Corrective feedback is given as required for the practice items. During the practice items the cue phrases are phased out, so that just the non-word stimuli are spoken as each toy is pointed to, e.g. /grb/ /səυb/, and then the child is asked ‘Who said X?’. Finally all the cue phrases can be omitted, so that the X stimulus is also presented as a single non-word, /spəυd/ /pəυd/; /spəυd/. The task items are presented with general encouragement as required but no feedback. One repetition is allowed on each item if requested by the child or if he or she is not attending. The investigator should cover his or her mouth to avoid use of visual cues. Items marked * are used for the reduced version. Scoring: Note the child’s response, by circling A or B. BOLD CAPITALS indicate correct responses. Add up correct responses to obtain child’s score. Circle incorrect responses in final column to indicate type of difficulty and look for any patterns. Normative data for this task can be found in Table 3.12.

267

268

Appendix C.5

LIST A A

B

X

RESPONSE CIRCLE IF INCORRECT

P1

/səυf/

/təυf/

/səυf/

a

P2

/spin/

/stin/

/stin/

a

b

P3

/glzəs/

/klzəs/

/klzəs/

a

b

P4

/grb/

/sb/

/grb/

a

b

1*

/ dɹεkt/

/ tɹεkt/

/ tɹεkt/

a

B

Voicing

2

/bɹs/

/bɹʃ/

/bɹʃ/

a

B


3*

/ sped/

/ steb/

/ steb/

a

B

Metathesis

4*

/glεb/

/glεv/

/glεb/

A

b


5

/sp nd /

/sp ndz/

/sp ndz/

a

B


6

/ dzlə/

/ d lə/

/ dzlə/

A

b


7*

/ fluw/

/ sluw/

/ sluw/

a

B


8*

/tɔn/

/tɹɔn/

/tɔn/

A

b

Cluster reduction

9*

/ ktəp lə / / kpət lə / / kpət lə / a

B

Metathesis

10

/nak/

/snak/

/nak/

A

b

Cluster reduction

11

/ b təfləυ/

/ b təfəυ/

/ b təfləυ/ A

b

Cluster reduction

12

/ slɔpə/

/ lɔpə/

/ slɔpə/

A

b

Cluster reduction

/g sptə/

/sp gtə/

A

b

Metathesis

b

LIST B 1*

/sp gtə/

2

/ p d ɔmz/ / p d ɔbz/ / p d ɔmz/ A

b


3*

/ ɒɹəpɹəυn/ / ɒɹəpləυn/ / ɒɹəpɹəυn/ A

b


4*

/ l jɔ/

/ j lɔ/

/ l jɔ/

A

b

Metathesis

5*

/ pεɹəsit/

/ pεɹəʃit/

/ pεɹəʃit/

a

B


6

/ flɒnt/

/ lfɒnt/

/ flɒnt/

A

b

Metathesis

7

/tε ɔ/

/ ε tɔ/

/ ε tɔ/

a

B

Metathesis

8

/ m bɹ l/

/ m bl ɹ/ / m bl ɹ/ a

9*

/vεʃ/

/fεʃ/

10

/pəυt/

B

Metathesis

/fεʃ/

a

B

Voicing Cluster reduction

/pləυt/

/pləυt/

a

B

11* /vn/

/zn/

/vn/

A

b


12

/ sεsəd /

/ sεd əs/

A

b

Metathesis

/ sεd əs/

TOTAL SCORE

/24

Appendix C.6 Auditory Discrimination Task 4: Legal versus Illegal Non-words

Name:__________ Date:________ Age:_____ Investigator:____________ Instructions: An example of a procedure is as follows. Two puppets are introduced to the child, one called ‘Tommy’ (a boy) and one called ‘Kivos’ /xvɒs/ (a robot). ‘These are my two puppets that can talk but they talk differently. Listen to them talk. Tommy is talking like all the boys and girls, he’s saying: Hello little boy/girl. My name is Tommy and I am very happy that you are going to play this game with us. Listen to Kivos now (produce a non-word string that uses nonEnglish sounds and clusters). They talk differently, don’t they? I’m going to say some words, some of which you will know and some of which you don’t know, and you point to who you think said the word. Was it Tommy or was it Kivos?’ On score sheet write in stimuli selected and indicate if non-words are legal or illegal, and if they are illegal due to non-English segments or non-English clusters. Scoring: Record whether the child correctly identified each non-word as being legal or illegal. Scores can be recorded for legal non-words and for illegal non-words with non-English segments and with non-English clusters. No normative data is available but children recognise illegal productions by around 4 gears of age.

269

270

STIMULI

Appendix C.6

LEGAL NON-WORD

ILLEGAL NON-WORDS Non-English segment Non-English cluster

NO CORRECT RESPONSES TOTAL CORRECT ILLEGAL NON-WORDS

/

/

/

/

Appendix D.1 Auditory Lexical Discrimination Task 1: Mispronunciation Detection – Full Version (from Vance, 1995) Name:_________Date:________Age:_________Investigator:__________ Instructions: The child should be asked to name each of the pictures first. This provides a vocabulary check and allows the child’s speech production to be recorded as for a naming task (see Chapter 5 and Appendix E.1) and compared with his or her performance on this auditory task, if required. If the child doesn’t know the name of the picture it can be supplied. The child is asked to look at each picture in turn and to decide if the name of the picture has been said correctly or not. A soft toy monkey can be used to explain the task. The child is told that the monkey is going to say some words, sometimes he is ‘clever’, and sometimes ‘silly’, or sometimes the monkey says the words ‘right’ and sometimes ‘wrong’ (the investigator can decide which best suits the child). There are two practice items during which corrective feedback can be given. Only general encouragement is given during the main part of the task. One repetition of a test item is allowed if the child fails to respond to a stimulus or he or she requests a repetition. The investigator should cover his or her mouth to avoid use of visual cues. Scoring: Record the child’s response, by circling Y or N. BOLD CAPITALS indicate correct responses. Add up the correct responses to obtain the child’s score. Remember that items marked with * and appearing in brackets are administered but not scored. Circle incorrect responses in final column to look for difficulty with specific contrasts. Total number of items correct at each word length and number correct overall. Normative data for this task can be found in Table 4.3.

271

272

Appendix D.1

FULL VERSION: LIST A

P1

PICTURE

STIMULUS

table

table

y

n

table

y

n

/slebl /

y

n

/teʃl /

y

n

table

y

n

P2

1

2

3

4

5

house

brush

sponge

glove

duck

leaf

1 SYLL

2 SYLL

3–4 SYLL

CIRCLE INCORRECT

/spaυs/

y

n

house

y

n

/faυs/

y

n

/haυf/

y

n

house

y

n

brush

Y

n

*(brush)

y

n

/brs/

y

N


/spndz/

y

N


sponge

Y

n

*(sponge)

y

n

/glb/

y

N

glove

Y

n

/gk/

y

N

duck

Y

n

*(/gk/)

y

n

/jif/

y

N

leaf

Y

n




Appendix D.1

6

7

8

9

10

11

12

13

14

STIMULUS

1 SYLL

sock

sock

Y

n

/zɒk/

y

N

*(sock)

y

n

/tæt/

y

N

cat

Y

n

*(/tæt/)

y

n

book

Y

n

/mυk/

y

N


/dɔtʃ/

y

N

Voicing

torch

Y

n

/maυt/

y

N

mouse

Y

n

knife

Y

n

/maf/

v

N

*(/maf/)

y

n

snake

Y

n

/nek/

y

N

Cluster reduction

/ten /

y

N

Cluster reduction

train

Y

n

/zæn/

y

N

van

Y

n

*(van)

y

n

book

torch

mouse

knife

snake

train

van

3–4 SYLL

CIRCLE INCORRECT

PICTURE

cat

2 SYLL

273

Voicing






274

Appendix D.1

(Continued)

15

16

17

18

19

20

21

22

23

24

25

STIMULUS

1 SYLL

watch

/rɒtʃ/

y

N

watch

Y

n

/pet/

y

N

*(/pet/)

y

n

plate

Y

n

roof

Y

n

*(roof)

y

n

/rus/

y

N


/vʃ/

y

N

Voicing

fish

Y

n

chair

Y

n

/tseə/

y

N

/ðm/

y

N

thumb

Y

n

plate

roof

fish

chair

thumb

sandwich

toilet

money

feather

yellow

2 SYLL

3–4 SYLL

CIRCLE INCORRECT

PICTURE

Metathesis

Cluster reduction


/ fæmwd /

y

N


sandwich

Y

n

/ dɔlət/

y

N

toilet

Y

n

/ nm/

y

N

money

Y

n

feather

Y

n

/ ðεfə/

y

N

Metathesis

/ lεləυ/

y

N


yellow

Y

n

Voicing

Metathesis

Appendix D.1

PICTURE

26

27

28

29

30

kitchen

ladder

flower

dustbin

jelly

STIMULUS

1 SYLL

275

2 SYLL

3–4 SYLL

CIRCLE INCORRECT

kitchen

Y

n

/ gtʃn /

y

N

ladder

Y

n

/ jædə/

y

N


/ slaυwə/

y

N


flower

Y

n

*(/ slaυwə/)

y

n

dustbin

Y

n

/ bsdn/

y

N

Metathesis

/ dzεl/

y

N


jelly

Y

n

TOTAL CORRECT LIST A

/40

/20

1 SYLL

2 SYLL

Voicing

FULL VERSION: LIST B

1

2

3

STIMULUS

tractor

/ dræktə/

y

N

tractor

Y

n

fishing

Y

n

*(fishing)

y

n

/ ʃfŋ/

y

N

Metathesis

/ bkst/

y

N

Metathesis

*(/ bkst/)

y

n

biscuit

Y

n

fishing

biscuit

3–4 SYLL

CIRCLE INCORRECT

PICTURE

Voicing

276

Appendix D.1

(Continued)

4

5

6

7

8

9

10

11

12

13

PICTURE

STIMULUS

scooter

/ stukə/

y

N

scooter

Y

n

*(/ stukə/)

y

n

/ bærət/

y

N

parrot

Y

n

/ stɔ/

y

N

seesaw

Y

n

*(seesaw)

y

n

/ lpə/

y

N

slipper

Y

n

sausage

Y

n

/ sɒd s/

y

N

*(/ sɒd s/)

y

n

/tgɑ/

y

N

guitar

Y

n

spider

Y

n

/ stapə/

y

N

*(spider)

y

n

parrot

seesaw

slipper

sausage

guitar

spider

caterpillar

spaghetti

elephant

1 SYLL

2 SYLL

3–4 SYLL

CIRCLE INCORRECT Metathesis

Voicing


Cluster reduction

Metathesis

Metathesis

Metathesis

caterpillar

Y

n

/ kæpətlə/

y

N

*(caterpillar)

y

n

/ gəspεt/

y

N

spaghetti

Y

n

*(spaghetti)

y

n

/ εflənt/

y

N

elephant

Y

n

Metathesis

Metathesis

Metathesis

Appendix D.1

3–4 SYLL

CIRCLE INCORRECT

/ kævəræn/

y

N

Metathesis

caravan

Y

n

*(/ kævəræn/)

y

n

/ krɒdəkal/

y

N

crocodile

Y

n

umbrella

Y

n

/ mblεrə/

y

N

helicopter

Y

n

*(helicopter)

y

n

/ hεltɒpkə/

y

N

Metathesis

/ gæŋkəru/

y

N

Metathesis

kangaroo

Y

n

/ tεvəl n /

y

N

*(/ tεvəl n /)

y

n

television

Y

n

/ hɒstpl /

y

N

hospital

Y

n

telephone

Y

n

/dεlfəυn/

y

N

*(/ dεlfəυn/)

y

n

/ pærəsut/

y

N

parachute

Y

n

/ btəfa/

y

N

butterfly

Y

n

*(butterfly)

y

n

PICTURE STIMULUS 14

15

16

17

18

19

20

21

22

23

caravan

crocodile

umbrella

helicopter

kangaroo

television

hospital

telephone

parachute

butterfly

277

1 SYLL

2 SYLL

Metathesis

Metathesis

Metathesis

Metathesis

Voicing


Cluster reduction

278

Appendix D.1

(Continued)

24

25

26

27

28

29

30

PICTURE

STIMULUS

computer

computer

Y

n

/ gəmpjutə/

y

N

*(computer)

y

n

/ waυndəbaυt/

y

N

roundabout

Y

n

hairdresser

Y

n

/ hεədrεtə/

y

N

/ εərəpren/

y

N

aeroplane

Y

n

pyjamas

Y

n

/pəd ɑbəz/

y

N

*(/pəd ɑbəz/)

y

n

/ hæmb də/

y

N

Hamburger

Y

n

/ danəzɔ/

y

N

*(/ danəzɔ/)

y

n

dinosaur

Y

n

roundabout

hairdresser

aeroplane

pyjamas

hamburger

dinosaur

1 SYLL

LIST B TOTALS WORD LENGTH TOTALS OVERALL TOTAL

/40

2 SYLL

3–4 SYLL

/20

/40

/40

/40

/120

CIRCLE INCORRECT

Voicing





Voicing

Appendix D.2 Auditory Lexical Discrimination Task 1: Mispronunciation Detection – Short Form (from Nathan et al., 2004) Name:_________Date:________Age:_________Investigator:__________ Instructions: The child should be asked to name each of the pictures first. This provides a vocabulary check and allows the child’s speech production to be recorded as for a naming task (see Chapter 5) and compared with his or her performance on this auditory task, if required. If the child doesn’t know the name of the picture it can be supplied. The child is asked to look at each picture in turn and to decide if the name of the picture has been said correctly or not. A soft toy monkey can be used to explain the task. The child is told that the monkey is going to say some words, sometimes he is ‘clever’, and sometimes ‘silly’, or sometimes the monkey says the words ‘right’ and sometimes ‘wrong’ (the investigator can decide which best suits the child). There are two practice items during which corrective feedback can be given. Only general encouragement is given during the main part of the task. One repetition of an item is allowed if the child fails to respond to a stimulus or requests a repetition. The investigator should cover his or her mouth to avoid use of visual cues. Scoring: Note the child’s response, by circling Y or N. BOLD CAPITALS indicate correct responses. Add up the correct responses to obtain the child’s score. Remember that items marked with * and appearing in brackets are administered but not scored. Circle incorrect responses in final column to look for difficulty with specific contrasts. Total number of items correct at each word length and number correct overall. Normative data for this task can be found in Table 4.4.

279

280

Appendix D.2

SHORT FORM: LIST A

P1

PICTURE

STIMULI

table

table

y

n

/slebl /

y

n

/teʃl /

y

n

table

y

n

house

y

n

/faυs/

y

n

/haυf/

y

n

house

y

n

/spndz/

y

N

sponge

Y

n

*(sponge)

y

n

/ slaυwə/

y

N

flower

Y

n

*(/ slaυwə/)

y

n

/ten /

y

N

train

Y

n

pyjamas

Y

n

/pəd ɑbəz/

y

N

*(/pəd ɑbəz/)

y

n

/ gəspεt/

y

N

spaghetti

Y

n

caterpillar

Y

n

/ kæpətlə/

y

N

* (caterpillar)

y

n

P2

1

2

3

4

5

6

house

sponge

flower

train

pyjamas

spaghetti

caterpillar

RESPONSES

CIRCLE IF INCORRECT

Place of Articulation


Cluster reduction


Metathesis

Metathesis

Appendix D.2

7

8

9

10

11

12

281

PICTURE

STIMULI

aeroplane

/ εərəpren/

y

N

aeroplane

Y

n

/ dræktə/

y

N

tractor

Y

n

brush

y

n

*(brush)

Y

n

/brs/

y

N

spider

Y

n

/ stapə/

y

N

*(spider)

y

n

/glb/

y

N

glove

Y

n

jelly

Y

n

/ dzel/

Y

N

tractor

brush

spider

glove

jelly

SCORE LIST A

RESPONSES

CIRCLE IF INCORRECT Place of articulation

Voicing


Metathesis



/24

SHORT FORM LIST B

1

2

3

STIMULUS

snake

snake

Y

n

/nek/

y

N

Cluster reduction

/ pærəsut/

y

N


parachute

Y

n

sausage

Y

n

/ sɒd s/

y

N

*(/ sɒd s/)

y

n

parachute

sausage

RESPONSES

CIRCLE IF INCORRECT

PICTURE

Metathesis

282

Appendix D.2

(Continued)

4

5

6

7

8

9

10

11

12

STIMULUS

fish

/vʃ/

y

N

fish

Y

n

*(fish)

y

n

/ εflənt/

y

N

elephant

Y

n

umbrella

Y

n

/ mblrə/

y

N

Metathesis

/ btəfa/

y

N

Cluster reduction

butterfly

Y

n

*(butterfly)

y

n

/t gɑ/

y

N

guitar

Y

n

yellow

Y

n

/ lεləυ/

y

N

*(yellow)

y

n

/ lpə/

y

N

slipper

Y

n

/pet/

y

N

*(/pet/)

y

n

plate

Y

n

/zæn/

y

N

van

Y

n

*(van)

y

n

elephant

umbrella

butterfly

guitar

yellow

slipper

plate

van

RESPONSES

CIRCLE IF INCORRECT

PICTURE

SCORE LIST B

/24

TOTAL SCORE (LIST A & B)

/48

Voicing

Metathesis

Metathesis


Cluster reduction

Cluster reduction


Appendix D.3 Auditory Lexical Discrimination Task 2: Without Pictures (from Constable, Stackhouse & Wells, 1997) Name:__________Date:________Age:_______Investigator:__________ Instructions: The child is given instructions along the following lines: ‘I’m going to say some words, some of the words you will know, but some you won’t: they will sound strange or silly. When you hear a word that you think is said right, you can say “yes”, but if you don’t know the word, you can say “no”. OK, let’s practise that. What if you heard the word “butterfly”?. . . . . . . . / desənɔ/? . . . . . . . . / sεnttid/?. The investigator CQ1 should cover his or her mouth to avoid use of visual cues. Scoring: Note the child’s response, by circling Y or N. BOLD CAPITALS indicate correct responses. Add up correct responses for each type of stimulus, to obtain child’s scores and calculate the percentage correct. Distractor items marked with * and appearing in brackets are administered but not scored. Normative data for this task can be found in Table 4.6.

283

284

Appendix D.3

WITHOUT PICTURES

STIMULUS

REAL WORD

1

elephant

Y

n

2

*(eskimo)

y

n

3

/ hεlkɒpkə/

4

*(president)

5

/ kæpətlə/

6

*(competition)

y

n

7

crocodile

Y

n

8

/εstəlekə/

9

/ εllənt/

NON-WORD TYPE A (PERSEVERATION)

y y

N

n

y

*(porcupine)

y

n

11

helicopter

Y

n

12

/ mafrəkəυn/

13

*(calculator)

14

/ tεll n /

y

N

15

/ɒʔtətəs/

y

N

16

hospital

17

/ εflənt/

18

/bnɒkjunəz/

y

N

19

/ krɒkəkal/

y

N

20

*(radiator)

y

n

21

*(rhinocerous)

y

n

22

*(octagon)

y

n

23

/ hɒstpl /

24

/ makrəkəυn/

25

caterpillar

26

/εskəlekə/

Y

N

y

N

y

N

y

N

y

N

n

n

Y

y

N

10

y

NON-WORD TYPE B (SEQUENCE)

y

N

y

N

n

Appendix D.3

STIMULUS

REAL WORD

27

binoculars

Y

n

28

*(alligator)

y

n

29

/ hεltɒpkə/

30

octopus

31

/ tεvl n /

32

/ hɒsppl /

33

microphone

34

/ krɒdəkal/

35

escalator

Y

n

36

*(telephone)

y

n

37

/ ktətlə/

38

/ ɒʔpətəs/

39

television

40

/ blɒkjunəz/

Y

% SCORES



y

N

y

N

y

N

y

N

y

N

n

y

SCORES

285

Y

N

n

y

Y

N

n

/10

/10

/10

%

%

%

Appendix D.4 Auditory Lexical Discrimination Task 2: With Pictures (from Constable, Stackhouse & Wells, 1997) Name:_____________Date:_______Age:________ Investigator:________ Instructions: ‘Now you are going to hear some more words. I will show you a picture and say a word. I want you to tell me if you think the word I say goes with the picture, and you can say “Yes”. If you think I say it wrong, or it sounds silly, then you can say “No”.’ The investigator should casually cover his or her mouth to avoid use of visual cues. Scoring: Note the child’s response, by circling Y or N. BOLD CAPITALS indicate correct responses. Add up the correct responses for each type of stimulus to obtain the child’s scores, and calculate the percentage. Items marked with * and appearing in brackets are administered but not scored. Normative data for this task can be found in Table 4.6.

287

288

Appendix D.4

WITH PICTURES

STIMULUS

REAL WORD

1

elephant

Y

n

2

*(eskimo)

y

n

3

/ hεlkɒpkə/

4

*(president)

5

/ kpətlə/

6

*(competition)

7

crocodile

8

/εstəlekə/

9

/ εllənt/


y y


N

n y

y Y

n n

y

*(porcupine)

y

n

11

helicopter

Y

n

12

/ mafɹəkəυn/

13

*(calculator)

14

/ tεll n /

y

N

15

/ ɒʔtətəs/

y

N

16

hospital

17

/ εflənt/

18

/bnɒkjunəz/

y

N

19

/ kɹɒkəkal/

y

N

20

*(radiator)

y

n

21

*(rhinocerous)

y

n

22

*(octagon)

y

n

23

/ hɒstpl /

24

/ makɹəkəυn/

25

caterpillar

Y

N

y

N

y

N

y

N

n

n

y Y

y N

10

y

N

n

N

Appendix D.4 26

/ εskəlekə/

y

27

binoculars

Y

n

28

*(alligator)

y

n

29

/ hεltɒpkə/

30

octopus

31

/ tεvl n /

32

/ hɒsppl /

33

microphone

34

/ kɹɒdəkal/

35

escalator

Y

n

36

*(telephone)

y

n

37

/ ktətlə/

38

/ ɒʔpətəs/

39

television

40

/ blɒkjunəz/

Y

289 N

y

N

y

N

n

y

Y

N

n

y

Y

y

N

y

N

y

N

N

n

SCORES

/10

/10

/10

% SCORES

%

%

%

Appendix D.5 Auditory Lexical Discrimination Task 3: Mispronunciation Detection for Individual Children (after Locke, 1980) Name:___________Date:________Age:______Investigator:___________ Instructions: The child names selected pictures, and his or her inaccurate productions are recorded. For each word to be examined, the target, e.g. SOCK; the child’s error, e.g. TOCK; and a control mispronunciation (i.e. not an error that the child makes), e.g. SHOCK (/ʃɒk/), are noted at the top of the score sheet (see example score sheet below). The child is presented with the picture and asked ‘Is this right?’ for each production: SHOCK? (control), TOCK? (error) or SOCK? (target). Six presentations of each of the three stimuli are given randomly. The process can be repeated for several words/pictures. To the child: ‘I am going to say some words. I’ll show you a picture as I say the words. I want you tell me if you think the word that I say goes with the picture. If you think I say the word wrong, or if it sounds silly, then say “Wrong” or “No”. OK, let’s have a practice.’ The tester should casually cover his or her mouth to avoid use of visual cues. Scoring: Note the child’s response, by circling Y or N. BOLD CAPITALS indicate correct responses. Add up the correct responses for each type of production. Normative data: There is no normative data, as the task is specific to an individual child’s errors.

291

292

Appendix D.5

EXAMPLE SCORE SHEET: Picture:

SOCK

Target production /sɒk/

Child’s production: Error production /tɒk/

[tɒk] Control production /ʃɒk/

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

/6

/6

Picture:

/6

Child’s production:

Target production

Error production

Control production

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

/6

/6

Picture:

/6


Target production

Error production

Control production

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

/6

/6

/6

Appendix D.5

Picture:

293


Target production

Error production

Control production

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

/6

/6

Picture:

/6


Target production

Error production

Control production

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

Y

n

y

N

y

N

/6

/6

/6

Appendix D.6 Auditory Lexical Discrimination Task 4: Words in Sentences (from Cassidy, 1994) Name:__________Date:___________Age:_______ Investigator:________ Part 1: Single Word Naming Instructions: Present a pair of pictures and ask the child to name each of them. Transcribe the response. If the child is unable to name a picture, say the word and ask him or her to point to the one named, indicate this on the score sheet. Scoring: Circle 1 if the picture is named correctly and 0 if incorrectly, and calculate total number of accurate responses. If child does not name the picture, put a tick or a cross in the ‘Pointed to Picture’ column to indicate if he or she correctly identified the picture. SCORE SHEET 1: SINGLE WORD NAMING PICTURE NAMED POINTED TO PICTURE STIMULUS TRANSCRIPTION CORRECTLY PICTURE coat

0

1

goat

0

1

lock

0

1

log

0

1

tea

0

1

key

0

1

mouse

0

1

mouth

0

1

bat

0

1

295

296

Appendix D.6

(Continued) PICTURE STIMULUS TRANSCRIPTION

PICTURE NAMED POINTED TO CORRECTLY PICTURE

mat

0

1

head

0

1

hen

0

1

glass

0

1

grass

0

1

clown

0

1

crown

0

1

SCORE

/16

Part 2: Single Word Discrimination Task Instructions: Present a pair of pictures to the child and ask him or her to point to the one you name. The order of presentation is given on the score sheet. One repetition can be given for each item if the child requests this, or if he or she fails to respond. The investigator should casually cover his or her mouth to avoid use of visual cues. Scoring: Circle the appropriate score for each item: 0 for an incorrect or no response, 1 point for a correct response following one repetition and 2 points for a correct response. Add the total points for each pair, and enter these on the summary sheet. Normative data for this task is found in Table 4.9.

SCORE SHEET 2: SINGLE WORD DISCRIMINATION PRACTICE ITEMS pear

bear

bear

pear

pear

bear

012

012

012

012

012

012

tin

tin

bin

tin

bin

bin

012

012

012

012

012

012

TEST ITEMS key

tea

key

tea

tea

key

012

012

012

012

012

012

key–tea

TOTAL SCORE

/12

Appendix D.6

297

mouse

mouse

mouse

mouth

mouth

mouth

012

012

012

012

012

012

mouse–mouth

TOTAL SCORE

/12

bat

bat

mat

mat

mat

bat

012

012

012

012

012

012

bat–mat

TOTAL SCORE

/12

head

hen

head

head

hen

hen

012

012

012

012

012

012

head–hen

TOTAL SCORE

/12

glass

glass

grass

grass

glass

grass

012

012

012

012

012

012

glass–grass

TOTAL SCORE

/12

clown

crown

crown

clown

clown

crown

012

012

012

012

012

012

clown–crown

TOTAL SCORE

/12

coat

coat

goat

coat

goat

goat

012

012

012

012

012

012

coat–goat

TOTAL SCORE

/12

log

lock

lock

lock

log

log

012

012

012

012

012

012

log-lock

TOTAL SCORE

/12

Part 3: Words in sentences Instructions: Present a pair of pictures. Tell the child, ‘You will hear the name of one of these pictures in a sentence: point to the picture that you hear. Listen to the whole sentence before pointing to one of the pictures. Some of the sentences may sound a bit funny, but listen carefully and point to the picture you hear.’ The order of presentation of sentences is given on the score sheet. One repetition is allowed for each test item, if necessary. The investigator should cover his or her mouth to avoid use of visual cues. 3-year-old children should be presented with the neutral sentences only. Scoring: Circle the appropriate score for each item: 0 for an incorrect or no response, 1 point for a correct response following one repetition and 2 points for a correct response. Add the total points for each pair for each sentence type, then enter these on the summary sheet. Normative data for this task can be found in Table 4.9.

298

Appendix D.6

SCORE SHEET 3: WORDS IN SENTENCES PRACTICE SENTENCES pear/ bear

The girl put the I put the

PEAR

in her bag

on my bed

The boy ate the

PEAR

for lunch

The boy ate the

BEAR

for lunch

The girl put the

BEAR

in her bag

I put the tin/bin

BEAR

The

TIN

PEAR

on my bed

is empty

We keep the

TIN

I ate the

of fruit

The

BIN

BIN

outside the door

is empty

We keep the

BIN

I ate the

of fruit

TIN

outside the door

TEST SENTENCES coat/ goat

NEUTRAL

I think your

COAT

012

GOAT

in the

012

Mum put her cupboard

COAT

in the

012

COAT

I think your

goat / coat

with long fur

GOAT


tea/key

NONSENSE


I liked the

tea/key

is lovely

BIASED

is lovely

COAT

012 012

in the

Set 1 score

012 /4

/2

/6

NEUTRAL

BIASED

NONSENSE

Mum put the

TEA

on the table.

012

I dropped the biscuits.

TEA

on the

012


TEA

on the

012

Mum put the

KEY

on the table.

012

The girl used the the door.

TEA

to open

012


KEY

to open

012

Set 1 score

/4

/4

/4

Appendix D.6 NEUTRAL mouse/ mouth

My teacher drew a my book. The boy’s food.

MOUSE

My teacher drew a my book.

MOUTH

The woman used the light the fire.

The woman used the light the fire. I put the

LOG

LOG

LOG

I liked the

GOAT

GOAT

Mum put her cupboard. I think your I liked the

Mum put her cupboard.

012 012

beside

012

012 012

with long fur.

Set 2 score

/4

012

in the

is lovely.

COAT

/4

012

with long fur.

GOAT

COAT

012

to

is lovely.

/2

012

/4

GOAT

/4

012

beside

Set 1 score I think your

012

to

in the cupboard.

The man put the the key.

goat/coat

LOCK

LOCK

012

/6

in the cupboard.


goat/ coat

in

Set 1 score

LOCK

012

012

The cat chased the around the house

I put the

lock/log

in

NONSENSE

012

was full of

MOUSE

mouse/mouth lock/log

MOUTH

BIASED

012

was full of

MOUSE

My teacher drew a my book. The boy’s food.

in

MOUSE

299

012

in the

012 /4

/4

/4

300

Appendix D.6

(Continued) NEUTRAL mat/bat

I threw the stairs.

BAT

down the

The boy used the the ball. I put the

MAT

BAT

I put the

BAT

MAT

to hit

BAT

key/tea

Mum put the

KEY

on the table.


KEY

on the

Mum put the

on the table.

TEA

The girl used the the door. I dropped the biscuits.

I liked the

I liked the

KEY

COAT

COAT

I put the

012 012 012

GOAT

in the

is lovely.

LOG

012 012 012

with long fur.

012 /4

in the cupboard.


LOG

beside

/4

012

with long fur.

COAT

/4

012

with long fur.

coat/goat Set 3 score log/lock

012

to open

is lovely

/6

012

on the

GOAT

GOAT

I think your

TEA

/4

012

/4

Mum put her cupboard. I liked the

012

to open

KEY

I think your

012

/2


coat/ goat

012

to hit

Set 1 score

Set 2 score

012

to hit

mat/bat

key/tea

012

outside the door.

The boy used the the ball.

NONSENSE

012

outside the door.


BIASED

/6

012 012

/2

Appendix D.6 NEUTRAL The man put the the key.

beside

LOCK


LOCK


LOG

I put the log/lock

Set 2 score

key/tea


012

to

012 012

on the

/4

/4

012


KEY

to open

012


TEA

to open

012

Mum put the

TEA

on the table.

012

Mum put the

KEY

on the table.

012


TEA

on the

key/tea

Set 3 score

hen/ head

The farmer’s away.

HEN


HEAD

The girl saw the picture.

HEN

in the

in the

HEN

012 012 012

on

012

has run

hen/head Set 1 score

/4 012

has run

HEAD

HEN

/4

has run



012 /4

The boy rested his the pillow.

crown/ clown

NONSENSE

012

/4 KEY

BIASED

to

in the cupboard.

LOG

301

012 /4

We looked at the CROWN in the picture book.

012


012

/2

/6

302

Appendix D.6

(Continued) NEUTRAL The king put the his head.

CLOWN

on 012

The children watched the CROWN on television.

crown/clown glass/ grass

CROWN

012 /6

GLASS

on

/2

in

012

The man sat on the the garden.

GLASS

in

012

GLASS

The girl filled the lemonade. glass/grass

with

GRASS

GRASS

012 012

with

Set 1 score

012 /4

My teacher drew a in my book.

MOUSE

012

My teacher drew a in my book.

MOUTH

012

The boy’s of food.

MOUTH

was full

My teacher drew a in my book. The boy’s of food.

MOUSE

I put the

in the cupboard.

/4

012 012 012

MOUTH

Set 2 score LOCK

/4

012

was full

The cat chased the around the house. mouse/mouth

MOUSE

/4

012

GRASS

The girl dropped the on the ground.

lock/

on


The girl filled the lemonade.

mouse/ mouth

012

Set 1 score

The girl dropped the the ground.

NONSENSE 012


The king put the his head.

BIASED

/6 012

/4

/2

Appendix D.6 NEUTRAL log


LOCK


LOG

beside


LOCK

mat/ bat

I threw the stairs. I put the

I threw the stairs. I put the

down the

down the

BAT

MAT

Set 2 score

mouse/ mouth

The cat chased the around the house. The boy’s of food.

MOUTH

The cat chased the around the house. The boy’s of food.

012

MOUTH

012 012 012 /6

/4

012 012

MOUTH

was full

/2 012

MOUSE

was full

/4

012

to hit

down the

mat/bat

/4

012

outside the door.

The boy used the the ball. I threw the stairs.

012

outside the door.

MAT

MAT

012

/4

MAT

BAT

012

to

I put the LOCK in the cupboard.

NONSENSE

012

to

LOG

Set 3 score

BIASED 012

beside


lock/log

303

012

The cat chased the around the house.

MOUSE

012

The cat chased the around the house.

MOUTH

012

mouse/mouth

Set 3 score

/4

/8

304

Appendix D.6

(Continued) NEUTRAL hen/ head

The girl saw the the picture.

HEAD

The girl saw the the picture

HEN

The boy rested his the pillow. The girl saw the the picture.

hen/head mat/ bat

clown/ crown

012 on

HEAD

HEN

in on


HEAD

012

on

Set 2 score MAT

012 012

HEN

I threw the stairs.

NONSENSE

012

in


012 /6

down the

/6

012

I put the door.

MAT

outside the

012

I put the door.

BAT

outside the

012


BAT

to hit

012


MAT

to hit

012

I threw the stairs. mat/bat

in

BIASED

BAT

down the

Set 3 score The king put the his head.

/4 CLOWN

The children watched the CROWN on television. CROWN

/4

on

We looked at the CLOWN in the picture book.


012

012 012 012

on

The children watched the CLOWN on television.

/4

012 012

Appendix D.6 NEUTRAL The king put the his head. clown/crown glass/ grass

CLOWN



012 012 012

in

012

with

012


GLASS

with

012

Set 2 score



HEAD

HEN


grass/ glass

/6

GRASS


head/hen

on

/4


glass/grass head/ hen

GRASS

on

in

GRASS

NONSENSE 012

/2

GLASS

GLASS


BIASED

on

Set 2 score


305

/4

has run

HEN

on

012 012

in the

012

on


HEAD


has run

HEAD

012 012

Set 3 score

/2

The girl dropped the the ground. The man sat on the the garden.

GRASS

GRASS

GLASS


/4 012

has run

HEAD


/4

on

012

with on

/6

012

in

GLASS

/4

012 012

306

Appendix D.6

(Continued) NEUTRAL The man sat on the the garden. The girl filled the lemonade. grass/glass clown/ crown

GLASS

GRASS

in

CLOWN

012 /4

in the

NONSENSE

012

with

Set 3 score

We looked at the picture book.

BIASED

/4

/4

012

The children watched the on television.

012

CROWN


CROWN

on

012

The children watched the CLOWN on television. We looked at the picture book.

CLOWN

in the

012 012

The children watched the CLOWN on television. clown/crown

Set 3 score

012 /4

/6

/2

Appendix D.6

307

SUMMARY SHEET: AUDITORY LEXICAL DISCRIMINATION TASK (WORDS IN SENTENCES) SINGLE WORDS

NEUTRAL CONTEXT

BIASED CONTEXT

NONSENSE CONTEXT

Set 1 Set 2 Set 3 Set 1 Set 2 Set 3 Set 1 Set 2 Set 3 COAT/GOAT

/12

LOCK/LOG

/12

TEA/KEY

/12

MOUSE/ MOUTH

/12

BAT/MAT

/12

HEAD/HEN

/12

GLASS/ GRASS

/12

CLOWN/ CROWN

/12 /34

TOTAL SCORES

/96

/36 /96

/26

/26

/34 /96

/36

/36

/26 /96

/34

Appendix E.1 Picture Naming Task 1 – Full Version (from Vance, Stackhouse & Wells, 2005) Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Present the pictures and ask the child to name them. If a child fails to name the picture or responds with a different lexical item, use cues such as semantic, gap fill or first sound to prompt him or her. If a cue is used then, after a brief pause, the child should be asked to name the picture again and the second response recorded and scored. No further help should be given. Scoring: Transcribe the child’s response phonetically. Circle 1 if child’s production is correct, 0 if incorrect. The production of the consonant sounds in each response is examined. For each word to be scored as correct the production of the consonants within the word should be an accepted adult realisation. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any words in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Add total number correct at each word length, and calculate total overall. Normative data for this task can be found in Table 5.1.

309

310

Appendix E.1

FULL VERSION: SET A PICTURE STIMULI

CHILD’S RESPONSE

1 SYLLABLE

2 SYLLABLE

PRACTICE ITEMS light sofa stickerbook TEST ITEMS brush sponge glove duck leaf sock cat book torch mouse knife snake train van watch plate roof fish chair thumb sandwich toilet money feather yellow kitchen ladder flower dustbin jelly ONE SYLLABLE SCORE TWO SYLLABLE SCORE – LIST A

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1

/20 /10

Appendix E.1

311

FULL VERSION: SET B PICTURE STIMULI

CHILD’S RESPONSE

tractor fishing biscuit scooter parrot seesaw slipper sausage guitar spider caterpillar spaghetti elephant caravan crocodile umbrella helicopter kangaroo television hospital telephone parachute butterfly computer roundabout hairdresser aeroplane pyjamas hamburger dinosaur TWO SYLLABLE SCORE – LIST B

2 SYLLABLE 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

/10

THREE / FOUR SYLLABLE SCORE

/20

SUMMARY SHEET: FULL VERSION 1-SYLLABLE WORDS (FROM LIST A) 2-SYLLABLE WORDS (FROM LISTS A & B) 3–4-SYLLABLE WORDS (FROM LIST B) TOTAL SCORE

3–4 SYLLABLE

/20 /20 /20 /60

Appendix E.2 Picture Naming Task 1 – Short Form

Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Present the pictures and ask the child to name them. If a child fails to name the picture or responds with a different lexical item, use cues such as semantic, gap fill or first sound to prompt him or her. If a cue is used then, after a brief pause, the child should be asked to say the word again and the second response recorded and scored. No further help should be given. Scoring: Transcribe the child’s response phonetically. Circle 1 if child’s production is correct, 0 if incorrect. The production of the consonant sounds in each response is examined. For each word to be scored as correct the production of the consonants within the word should be an accepted adult realisation. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any words in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Normative data is not available for this set of stimuli on this task but qualitative comparisons can be made across tasks which use the same stimuli.

313

314

Appendix E.2

SHORT FORM PICTURE STIMULI

CHILD’S RESPONSE 1 SYLLABLE 2 SYLLABLE 3–4 SYLLABLE

PRACTICE ITEMS light sofa stickerbook TEST ITEMS brush sponge glove snake train van plate fish yellow flower jelly tractor slipper sausage guitar spider caterpillar spaghetti elephant umbrella pyjamas parachute butterfly aeroplane 1-SYLLABLE SCORE 2-SYLLABLE SCORE

0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0

/8 /8

3–4-SYLLABLE SCORE TOTAL SCORE

1 1 1 1 1 1 1 1

/8 /24

Appendix E.3 Guidelines for Scoring Full Versions and Short Forms of Speech Production Tasks The production of the consonant sounds in each response is examined. To be counted as correct all consonants within a word should be an accepted adult realisation. The following variations were considered to be acceptable for speech data collected in the South-East of England: (a) /υ/ for syllable final /l/ in e.g. HOSPITAL, CROCODILE. (b) labio-dental ‘r’ as /υ/ in e.g. BRUSH (but scored incorrect if the phoneme articulated was /w/). (c) interdental /s/ as /s/ (but scored incorrect if the phoneme articulated was //. (d) /ʔ/ for phonemes in syllable final position, e.g. /tɹʔtə/ for TRACTOR. (e) PYJAMAS with an unaspirated /p/. (f) /tʃɹ/ for /tɹ/ in, e.g. TRAIN, TRACTOR. (g) /εəplen/ for AEROPLANE. (h) HAIRDRESSERS or HAIRDRESSER both accepted. In the cross-sectional study the responses of the 7-year-old children from the sample on naming and real-word repetition were used to establish an appropriate pattern. For example, as the majority of 7-year-old children produced dental fricative // in words, then the common dialectal variation of /f/ used for // was scored as an error. Within other dialectal contexts the relevant adult model should be considered in deciding whether a child’s realisation is appropriate or not.

315

Appendix E.4 Speech Production: Single Words – Picture Naming Task 2 (from Snowling, van Wagtendonk & Stafford, 1988) Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Present each picture in turn and ask the child to name it. All responses should be audio-recorded using a good-quality recorder and microphone as well as transcribed live in phonetic script on the lexical score sheet. If the child is unable to name a picture or uses a different word from the target, three choices are given to the child, e.g. ‘Is it a TIGER, a HIPPOPOTAMUS or an ELEPHANT?’; see Forced Choice Prompts on the score sheet. The choice the child makes can be circled on the score sheet, to indicate that this was not a spontaneous naming response, and the child’s actual production transcribed. Scoring: Lexical Score: Spontaneous naming responses recognisable as the target word are scored as correct, by circling 1 on the score sheet, even if production is inaccurate. Circle 0 if response is incorrect. Forced choice (prompted) responses are noted, as these indicate that the child knew the word, but was unable to access it immediately. However, prompted responses are not included in the total correct score. Speech Score: Responses transcribed on sheet 1 can be transferred to the Speech Score (Sheet 2). This does include responses given to a forced choice prompt. Each response is scored for the number of consonants that are correct. The percentage of consonants produced correctly is calculated. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 86 consonants for this stimuli set (but see notes below). However, if the child fails to make a correct response, the percentage should be calculated on the number of consonants in words attempted.

317

318

Appendix E.4

Normative data for this task can be found in Tables 5.3 and 5.4. SHEET ONE: LEXICAL SCORE PICTURE

CHILD’S RESPONSE

FORCED CHOICE PROMPTS

hippopotamus

0

1

tiger/hippopotamus/elephant

screwdriver

0

1

screwdriver/hammer/saw

acorn

0

1

hazelnut/acorn/pinecone

strawberry

0

1

strawberry/grape/gooseberry

microscope

0

1

telescope/microscope/microphone

ballerina

0

1

gymnast/ice-skater/ballerina

elephant

0

1

rhino/elephant/giraffe

helicopter

0

1

helicopter/glider/aeroplane

umbrella

0

1

umbrella/sunshade/parasol

giraffe

0

1

giraffe/elephant/antelope

seahorse

0

1

seahorse/chameleon/shrimp

puppet

0

1

doll/puppet/mannequin

hammock

0

1

bunk-bed/cot/hammock

stethoscope

0

1

stethoscope/syringe/telescope

binoculars

0

1

spectacles/binoculars/monocle

pyramid

0

1

pyramid/sphinx/triangle

octopus

0

1

conger eel/crab/octopus

rhinoceros

0

1

rhinoceros/crocodile/elephant

saddle

0

1

saddle/reins/stirrup

unicorn

0

1

centaur/mermaid/unicorn

TOTAL CORRECT SPONTANEOUSLY

/20

Appendix E.4

319

SHEET TWO: SPEECH SCORE PICTURE

CHILD’S RESPONSE (FROM SHEET 1)

TARGET NO. OF CONSONANTS*

hippopotamus*

6 (or 2)

screwdriver

6

acorn

2

strawberry

5

microscope

6

ballerina

4

elephant

4

helicopter

5

umbrella

4

giraffe

3

seahorse

3

puppet

3

hammock

3

stethoscope

6

binoculars

6

pyramid

4

octopus

4

rhinoceros*

CONSONANTS CORRECT

5 (or 2)

saddle

3

unicorn

4

TOTAL NUMBER OF CONSONANTS CORRECT

/86

PERCENTAGE CONSONANTS CORRECT

%

*HIPPOPOTAMUS: either ‘hippopotamus’ or ‘hippo’ can be accepted as correct. If the child says ‘hippo’, the percentage correct is calculated on four fewer consonants in the total possible. RHINOCEROUS: either ‘rhinoceros’ or ‘rhino’ can be accepted as correct. If the child says ‘rhino’, the percentage correct is calculated on three fewer consonants in the total possible.

Appendix E.5 Word Repetition Task – Full Version (from Vance, Stackhouse & Wells, 2005) Name:__________ Date:________ Age:______ Investigator:___________ Instructions: The child is asked to repeat the word that he or she hears. Three practice items are presented at the start of the task. Corrective feedback can be given if the child fails to repeat any of these practice items, but only general encouragement can be given during the main part of the task. Each test item can be presented again, once only, if the child does not respond or asks for the word to be repeated. Scoring: Transcribe the child’s response phonetically. Circle 1 if child’s production is correct, 0 if incorrect. Production of the consonant sounds in each response is examined. For each word to be scored as correct the production of the consonants within the word should be an accepted adult realisation, taking into account regional accent variation. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any words in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Add total number correct at each word length, and calculate total overall. Normative data for this task can be found in Table 5.6.

321

322

Appendix E.5

FULL VERSION: LIST A STIMULI

CHILD’S RESPONSE

PRACTICE ITEMS light sofa stickerbook TEST ITEMS brush sponge glove duck leaf sock cat book torch mouse knife snake train van watch plate roof fish chair thumb sandwich toilet money feather yellow kitchen ladder flower dustbin jelly 1-SYLLABLE SCORE 2-SYLLABLE SCORE–LIST A

1 SYLLABLE

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

2 SYLLABLE

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1

/20 /10

Appendix E.5

323

FULL VERSION: LIST B SCORE STIMULI

CHILD’S RESPONSE

TEST ITEMS tractor fishing biscuit scooter parrot seesaw slipper sausage guitar spider caterpillar spaghetti elephant caravan crocodile umbrella helicopter kangaroo television hospital telephone parachute butterfly computer roundabout hairdresser aeroplane pyjamas hamburger dinosaur 2-SYLLABLE SCORE–LIST B 3–4-SYLLABLE SCORE

2 SYLLABLE 3–4 SYLLABLES

0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

/10 /20

324

Appendix E.5

SUMMARY SHEET: WORD REPETITION, FULL VERSION 1-SYLLABLE WORDS (FROM LIST A)

/20

2-SYLLABLE WORDS (FROM LIST A B)

/20

3–4-SYLLABLE WORDS (FROM LIST B)

/20

TOTAL SCORE

/60

Appendix E.6 Word Repetition Task – Short Form (from Nathan et al., 2004a)

Name:__________ Date:________ Age:______ Investigator:___________ Instructions: The child is asked to repeat the word that he or she hears. Three practice items are presented at the start of the task. Corrective feedback can be given if the child fails to repeat any of these practice items, but only general encouragement can be given during the main part of the task. Each item can be presented again, once only, if the child fails to respond to a stimulus or requests a repetition. Scoring: Transcribe each item, and record the number of consonants correct for each word. Add up the number of consonants correct for words at each syllable length for qualitative comparison. Calculate the percentage of consonants correct for all the items for comparison with normative data. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 77 consonants for this stimuli set. However, if the child fails to make a correct response, the percentage should be calculated on the number of consonants in words attempted. Normative data for this task can be found in Table 5.7.

325

326

Appendix E.6

SHORT FORM LIST A

STIMULUS

CHILD’S RESPONSE

PRACTICE ITEMS light sofa stickerbook TEST ITEMS snake guitar van plate sausage slipper elephant umbrella butterfly parachute fish yellow

ENTER NUMBER OF CONSONANTS CORRECT – TOTAL NO OFCONSONANTS IN WORD GIVEN IN BRACKETS. 1 SYLL

2 SYLL

3–4 SYLL

(3) (2) (2) (3) (3) (3) (4) (4) (4) (4) (2) (2)

SHORT FORM – LIST B 1 SYLL train aeroplane tractor brush jelly caterpillar spider pyjamas sponge flower spaghetti glove NO. OF CONSONANTS CORRECT AT EACH SYLLABLE LENGTH TOTAL CONSONANTS CORRECT PERCENTAGE CONSONANTS CORRECT

2 SYLL

3–4 SYLL

(3) (4) (4) (3) (2) (4) (3) (4) (4) (3) (4) (3) /23

/22 /77 %

/32

Appendix: E.7 Word Repetition Task: Low-Frequency Words (from Nathan et al., 2004a)

Name:__________ Date:________ Age:______ Investigator:___________ Instructions: The child is asked to repeat the word that he or she hears. One practice item is presented at the start of the task. Each item can be presented again, once only, if the child fails to respond to a stimulus or requests a repetition. Scoring: Transcribe each item, and score the number of consonants correct for each word. Add up the number of consonants correct for words at each syllable length for qualitative comparison. Calculate the percentage of consonants correct (PCC) for all the items for comparison with normative data. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 92 consonants for this stimuli set. However, if the child fails to make a correct response, the percentage should be calculated on the number of consonants in words attempted. Normative data for this task can be found in Table 5.8.

327

328

Appendix E.7

LIST A

STIMULUS

CHILD’S RESPONSE

ENTER NUMBER OF CONSONANTS CORRECT – TOTAL NO. OF CONSONANTS IN WORD IS GIVEN IN BRACKETS. 1 SYLL

PRACTICE hippopotamus TEST ITEMS satisfaction encyclopaedia acrobat squeak (4) ballerina cartridge drummer refrigerator gorilla wasp (3) CONSONANTS CORRECT /7 BY WORD LENGTH LIST A TOTAL CONSONANTS CORRECT

2 SYLL

3 SYLL

4 SYLL 5–6 SYLL

(7) (6) (4) (4) (4) (3) (6) (3) /7

/7

/44

/11

/12

Appendix E.7

329

LIST B

STIMULUS

ENTER NUMBER OF CONSONANTS CHILD’S CORRECT – TOTAL NO. OF CONSONANTS RESPONSE IN WORD IS GIVEN IN BRACKETS. 1 SYLL

electricity bran cricket grasshopper anchovy chrysanthemum scraper crutch librarian multiplication CONSONANTS CORRECT BY WORD LENGTH

2 SYLL

3 SYLL

4 SYLL 5–6 SYLL (6)

(3) (4) (5) (3) (7) (4) (3) (5) (8) /6

/8

LIST B TOTAL CONSONANTS CORRECT

/8

/12

/14

/48

SUMMARY SHEET: LOW-FREQUENCY WORD REPETITION NUMBER OF CONSONANTS CORRECT FROM LIST A 1-SYLLABLE WORDS 2-SYLLABLE WORDS 3-SYLLABLE WORDS 4-SYLLABLE WORDS 5–6-SYLLABLE WORDS TOTAL CONSONANTS CORRECT

/7 /7 /7 /11 /12

PERCENTAGE CONSONANTS CORRECT (PCC)

FROM LIST B

TOTAL

/6 /8 /8 /12 /14

/13 /15 /15 /23 /26 /92 %

Appendix E.8 Non-Word Repetition Task – Full Version (from Vance, Stackhouse & Wells, 2005) Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Pronounce the non-words with the same stress pattern as the matched real word. The child is told that he or she is going to say some funny words that he or she won’t have heard before, and is asked to repeat the non-word that he or she hears. For younger children, a soft toy, e.g. a monkey, is used to demonstrate the task. The child is told that the monkey says made up, monkey words’ and that he or she will not know these words. He or she is then asked to say each word like the monkey said it. Three practice items are presented at the start of the task. Corrective feedback can be given if the child fails to repeat any of these practice items. If the child lexicalises any of these words, i.e. produces a real word, he or she should be encouraged to say the word ‘just like the monkey says it’. Only general encouragement can be given during the main part of the task. A further presentation of an item can be given if the child fails to respond to a stimulus or requests a repetition. Scoring: Transcribe the child’s response phonetically. Circle 1 if child’s production is correct, 0 if incorrect. The production of the consonant sounds in each response is examined. For each word to be scored as correct the production of the consonants within the word should be an accepted adult realisation. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any words in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Add total number correct at each word length, and total overall. Normative data for this task can be found in Table 5.11.

331

332

Appendix E.8

FULL VERSION: SET A SCORE STIMULI

CHILD’S RESPONSE

PRACTICE ITEMS /sap/ / ɔɹεd / / nazpəυp/ TEST ITEMS /bɹʃ/ /spnd / /glεv/ /dk/ /lɔf/ /sɔk/ /kεt/ /bɔk/ /tutʃ/ /mɔs/ /nəυf/ /snak/ /tɹɔn/ /vn/ /wυtʃ/ /pləυt/ /ɹɔf/ /fεʃ/ /tʃi/ /ɒm/ / smwəd / / telət/ / mεnə/ / fð/ / jlɔ/ / kɒtʃən/ / lεd/ / fluw/ / dsbən/ / dzlə/ 1-SYLLABLE SCORE – LIST A 2-SYLLABLE SCORE – LIST A

1 SYLL

0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

2 SYLL

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1

/20 /10

Appendix E.8

333

FULL VERSION: SET B SCORE STIMULI

CHILD’S RESPONSE

PRACTICE ITEMS /sap/ / ɔɹεd / / nazpəυp/ TEST ITEMS / tɹεkt/ / fɒʃŋ/ / bɒskεt/ / skitə/ / pɹt/ / sɑsi/ / slɔpə/ / sεsəd / /gε tɔ/ / sped/ / ktəplə / / sp gtə/ / lfɒnt/ / kɹɒvn / / kɹkədaυl/ / m bɹl/ / hləkptə/ / kɒŋgɹɑ/ / tυləv ən/ / hspətl/ / tɒləfan/ / pεɹəʃit/ / btəfləυ/ / kεm pjaυt/ / ɹəυndəbat/ / hɔdɹs/ / ɒɹəpləυn/ / p d ɔmz/ / hmbɑ/ / densɑ/ 1-SYLLABLE SCORE (LIST A) 2-SYLLABLE SCORE (LIST A B) 3–4-SYLLABLE SCORE (LIST B) TOTAL SCORE

2 SYLL

0 0 0 0 0 0 0 0 0 0

3–4 SYLL

1 1 1 1 1 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

/20 /20 /20 /60

Appendix E.9 Non-Word Repetition Task – Short Form (from Nathan et al., 2004)

Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Pronounce the non-words with the same stress pattern as the matched real word. The child is told that he or she is going to say some funny words that he or she won’t have heard before, and is asked to repeat the non-word that he or she hears. For younger children, a soft toy, e.g. a monkey, can be used to demonstrate the task. The child is told that the monkey says made up, monkey words’ and that he or she will not know these words. He or she is then asked to say each word like the monkey said it. Three practice items are presented at the start of the task. Corrective feedback can be given if the child fails to repeat any of these practice items. If the child lexicalises any of these words, i.e. produces a real word, he or she should be encouraged to say the word ‘just like the monkey says it’. Only general encouragement is given during the test items. A further presentation of an item can be given if the child fails to respond to a stimulus or requests a repetition. Scoring: Transcribe each item, and record the number of consonants correct for each word. Add up the number of consonants correct for words at each syllable length for qualitative comparison. Calculate the percentage of consonants correct (PCC) for all the items for comparison with normative data. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 77 consonants for this stimuli set. However, if the child fails to make a correct response, the percentage should be calculated on the number of consonants in words attempted. Normative data for this task can be found in Table 5.12.

335

336

Appendix E.9

SHORT FORM LIST A

STIMULUS

ENTER NUMBER OF CONSONANTS CORRECT – TOTAL NO. OF CONSONANTS IN WORD GIVEN CHILD’S RESPONSE IN BRACKETS.

PRACTICE ITEMS /sap/ / ɔɹεd / / nazpəυp/ TEST ITEMS / sped/ /spnd / / fluw/ /tɹɔn/ / p d ɔmz/ /sp gtə/ / ktəplə / / ɒɹəpləυn/ / tɹεkt/ /bɹʃ/ / dzlə/ /glεv/

1 SYLL

2 SYLL

3–4 SYLL

(3) (4) (3) (3) (4) (4) (4) (4) (4) (3) (2) (3)

SHORT FORM – LIST B 1 SYLL /snak/ / sεsəd / /vn/ /gε tɔ/ / slɔpə/ / m bɹl/ / lfɒnt/ / btəfləυ/ / jlɔ/ / pεɹəʃit/ /fεʃ/ /pləυt/ NO. CONSONANTS CORRECT AT EACH LENGTH TOTAL CONSONANTS CORRECT PERCENTAGE CONSONANTS CORRECT (PCC)

2 SYLL

3–4 SYLL

(3) (3) (2) (2) (3) (4) (4) (4) (2) (4) (2) (3) /23

/22

/32

/77 %

Appendix E.10 Non-word Repetition Task: Matched to Low-Frequency Words (from Nathan et al., 2004a) Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Pronounce the non-words with the same stress pattern as the matched real word. The child is asked to repeat the word that he or she hears. The child is warned that the words are silly, made-up words. There is one practice item. All responses should be audio-recorded using a good-quality recorder and microphone as well as transcribed live in phonetic script. Scoring: Transcribe each item, and record the number of consonants correct for each word. Add up the number of consonants correct for words at each syllable length for qualitative comparison. Calculate the percentage of consonants correct (PCC) for all the items for comparison with normative data. If an extra consonant is produced in a word, one point is taken off the total possible score for that word. There is a maximum total of 92 consonants for this stimuli set. However, if the child fails to make a correct response, the percentage should be calculated on the number of consonants in words attempted. Normative data for this task can be found in Table 5.13.

337

338

Appendix E.10

LIST A

STIMULI

ENTER NUMBER OF CONSONANTS CHILD’S CORRECT – TOTAL NO. OF CONSONANTS RESPONSE IN WORD GIVEN IN BRACKETS. 1 SYLL

PRACTICE ITEM / l sɒsεtɑ/ TEST ITEM /ɒlk trsυtɑ/ /bɹɒn/ /laυ brəυraυn/ / krɒkɒt/ / skrəυpi/ /kr zɒnθmεm/ / rushεp/ /krεtʃ/ / υntʃevɑ/ / mɑltplkaυʃn / CONSONANTS CORRECT BY WORD LENGTH LIST A TOTAL CONSONANTS CORRECT

2 SYLL

3 SYLL 4 SYLL 5–6 SYLL

(6) (3) (5) (4) (4) (7) (4) (3) (4) (8) /6

/8

/8

/12

/14

/48

LIST B STIMULUS

CHILD’S RESPONSE 1 SYLL 2 SYLL 3 SYLL 4 SYLL 5–6 SYLL

PRACTICE ITEM / l sɒsεtɑ/ TEST ITEMS /g rɒlu/ /rɒ frd ɒrəυt/ /bl rɑn/ / sɒtsfεkʃn / /υn səklapɑd / / kitrυd / /skwuk/ / krbυt/ / drɒm/ /wεsp/ CONSONANTS CORRECT BY WORD LENGTH LIST B TOTAL CONSONANTS CORRECT

(3) (6) (4) (7) (6) (4) (4) (4) (3) (3) /7

/7

/7

/44

/11

/12

Appendix E.10

339

SUMMARY SHEET: NON-WORDS MATCHED FREQUENCY WORDS NUMBER OF CONSONANTS CORRECT FROM LIST A 1-SYLLABLE WORDS 2-SYLLABLE WORDS 3-SYLLABLE WORDS 4-SYLLABLE WORDS 5–6-SYLLABLE WORDS TOTAL SCORE PERCENTAGE CONSONANTS

/6 /8 /8 /12 /14 /48 CORRECT

WITH

LOW-

FROM LIST B

TOTAL

/7 /7 /7 /11 /12 /44

/13 /15 /15 /23 /26 /92 %

Appendix E.11 Mispronunciation Self-Correction

Name:__________ Date:________ Age:______ Investigator:___________ Instructions: Observe the child in a general conversation/play situation; include pictures/materials that will trigger specific targets to be investigated, if required. Audio-record the session and complete the score sheet as far as possible at the time, checking this against the audio-recording later. Where an investigation of consistency of self-correction strategies is specifically required, ensure that targets are attempted more than once in the session. Record on the sheet below items on which the child attempts self-correction, and /or targets about which you want to record whether a self-correction attempt has been made, or not. Write the targets on the sheet below in advance of the session if focusing on specific sounds/contrasts. Scoring: For each self-correction attempt recorded, tick the box to mark type of self-correction response: 1. Corrects own speech error spontaneously. 2. Attempts to correct speech error and produces a response closer to the target but not yet correct. 3. Attempts to correct speech error and produces a variable response that may or may not be closer to the target. 4. Attempts to correct speech error but response is same as original error. 5. Only attempts to correct speech error if listener does not understand. 6. Makes no attempt to correct speech error.

341

342

Appendix E.11

To obtain an overall view of the child’s self-correction attempts, total the number of ticks in each category for all targets and enter the total in the row at the bottom of the sheet. Calculate the percentage of selfcorrection attempts in each category as follows (the total number of target attempts is the total number of ticks for that target): Number in category 100 Total number of target attempts If the same target is attempted more than once, note any variation in self-correction. Read across the row for each target and see if each attempt is recorded in the same box, showing consistency of self-correction, or if more than one box is ticked. If ticks for a target are all in same box, circle C for consistent; if ticks are in more than one box, circle I for inconsistent. Total the number of consistent and inconsistent responses at the bottom of the consistency column, and calculate the percentage of consistent and inconsistent self-corrections.

Total Percentage

/

/ %

/ %

/ %

6

/ %

Consistency

5

no selfcorrection

4

responds to listener

3

same error

2

variable

1

closer to target

Child’s Production

selfcorrects

Target

/ %

%

C

I

C

I

C

I

C

I

C

I

C

I

C

I

C

I

C

I

C

I

C

I

/

/ %

%

Appendix E.11 CHILD’S VIEWPOINT Note the answer the child gives to any of the following questions: 1. 2. 3. 4. 5. 6.

Why do you come for speech and language therapy? What is hard about talking? What do you do when someone doesn’t understand you? Why do you think listening is important? Even if we speak English, do we all sound the same? Have you finished learning to talk yet?

343

Appendix F.1 Connected Speech Task 1: Words in Sentences (from Vance, Stackhouse & Wells, 1995) Name:________Date:__________ Age:_______Investigator:___________ Instructions: Show the child the relevant picture. Read each sentence to the child in turn, asking the child to repeat the sentence that he or she hears. All responses should be audio-recorded using a good-quality recorder and microphone. There is one practice item on which help can be given. One repetition only can be given on test items if the child requests this or if he or she was not attending. Scoring: Transcribe the child’s production of the target words (in capitals). Circle 0 for an incorrect production of the target word, and 1 for an accurate production. The production of the consonant sounds in each response is examined. For each word to be scored as correct, the production of the consonants within the word should be an accepted adult realisation, taking into account regional accent variation. (See Appendix E.3 for guidelines on appropriate realisations for this stimulus set.) Any responses in which consonants deviate from accepted adult realisations are scored as incorrect, including words in which consonants have been added or omitted by the child. Normative data for this task can be found in Table 6.2.

345

346

Appendix F.1

SENTENCE STIMULUS

P1 The

MONKEY

CHILD’S RESPONSE TO TARGET WORD

1 SYLL

2 SYLL

3–4 SYLL

is on the wall jumped

0 1

1.

Look, the out.

2.

The AEROPLANE is flying away.

0 1

3.

His PARACHUTE is green and blue.

0 1

4.

The

0 1

5

The BISCUIT has bites in it.

6.

The TELEPHONE is like a banana.

0 1

7.

His

0 1

8.

The LADDER is by the house.

9.

The

10. His

MOUSE

COMPUTER

PYJAMAS

ROOF

is on. 0 1

are stripey.

is high.

UMBRELLA

0 1 0 1

is yellow.

0 1

11. Look, the SPIDER is hanging there.

0 1

12. The TOILET is green.

0 1

13. The

TORCH

is bright.

0 1

14. The SAUSAGE is for the boy.

0 1

15. The SPAGHETTI is for the boy. is empty.

0 1

16. His

PLATE

17. The

GLOVE

is big.

0 1

18. The

KNIFE

is for cutting.

0 1

19. The

JELLY

is wobbly.

0 1

20. The YELLOW hat has a bobble.

0 1

21. His WATCH is on his leg.

0 1

0 1

Appendix F.1

SENTENCE STIMULUS 22. The

MONEY

23. The

CAT

24. The

FISH

25. The

TELEVISION


347

1 SYLL

is in his hand.

is on the bed. is in the bowl.

2 SYLL 0 1

0 1 0 1

is on.

0 1

26. The HAIRDRESSER is cutting his hair. 27. Her

BRUSH

is blue.

0 1 0 1

28. The HELICOPTER is rescuing the clown. 29. The TRAIN is going very fast.

0 1 0 1

30. The KANGAROO is riding down the road. 31. His

0 1

is green.

SCOOTER

0 1

32. The TRACTOR is pulling very hard. 33. The VAN is stuck in the mud.

0 1 0 1

is too big.

0 1

is too small.

0 1

34. The

ELEPHANT

35. The

CARAVAN

36. The CROCODILE sings and plays.

0 1

37. The GUITAR has a broken string. 38. The DUCK has caught a fish. 39. He’s

FISHING

40. The

SLIPPER

41. The

SOCK

42. The

DUSTBIN

3–4 SYLL

0 1 0 1

with a twig.

0 1

is old.

0 1

has a hole. is very full.

43. The PARROT is sitting in a tree.

0 1 0 1 0 1

348

Appendix F.1

(Continued) SENTENCE STIMULUS 44. A

LEAF


falls off.

1 SYLL

2 SYLL

0 1

45. A FEATHER is on the ground.

0 1

46. The ROUNDABOUT has horses. 47. The

SEESAW

3–4 SYLL

0 1

is long.

0 1

48. The CATERPILLAR is sitting at one end.

0 1

49. The BUTTERFLY is sitting at the other end.

0 1

50. The

HAMBURGER

0 1

51. The

FLOWER

52. The

CHAIR

is tasty.

is pretty.

0 1

is broken.

0 1

53. The BOOK has a good story.

0 1

54. His THUMB has a bandage on.

0 1

55. The HOSPITAL is near his house. 56/ The SNAKE is in the 57. KITCHEN.

0 1 0 1 0 1

58. The SANDWICH is ready to eat.

0 1

59. The DINOSAUR is having a bath. 60. The SPONGE is full of bubbles. WORDS CORRECT AT EACH SYLLABLE LENGTH TOTAL SCORE

0 1 0 1 /20

/20 /60

/20

Appendix F.2 Connected Speech Task 2: Connected Speech Processes (CSP) Repetition (from Newton, 1999) Name:_________Date:_______Age:________Investigator:___________ Instructions: The child is asked to repeat a sentence that he or she hears, just as it was spoken. Responses should be audio-recorded using a good-quality recorder and microphone. Read the sentences to the child with the prescribed intonation pattern, with stress on the syllable in bold, and using the appropriate CSP. The investigator should produce each sentence adhering to the following guidelines: where possible, the stimuli should be presented in an accent that is close to that of the child; each item is presented with an intonation pattern that is unmarked for declarative statements in English; the tonic syllable is the final lexically stressed syllable of the sentence or phrase, marked below in bold. The targeted juncture, underlined below, should be produced with the appropriate process, transcribed below in phonetic transcription. The child’s production of the target words should be transcribed on the score sheet and can then be checked from the recording. Scoring: Code the child’s responses according to juncture type produced: c for production of adult-like close juncture; o for open juncture (i.e. no assimilation, elision or liaison); and n for non-adult like realisation of the juncture. Circle the appropriate letter on the score sheet. On the summary sheet, for each c (i.e. adult-like) response, score through one of the tallies (1) for that particular CSP. Total the number correct for each different CSP and calculate the percentage correct. Numbers of (o) open junctures and (n) non-adult like realisations of the junctures can also be recorded. An important feature of normal adult connected speech to note when scoring a child’s repetitions is the common realisation of word-final /t/ as a glottal stop, e.g. FAT : [fʔ], so that FAT PIG may be as likely produced [fʔpI] as [fppI]. Therefore a similar production of word-final /t/ as [ʔ] in assimilation environments should be coded as adult-like close juncture. Normative data for this task can be found in Table 6.3. 349

1.

You must clean your TEETH. [msklin]

2.

I gave the elephant a baNAna [ðIjεləfənt]

3.

Mary’s shoes are CLEAN [mεəɹi ʃuz]

4.

Claire ate all her LUNCH. [klεəɹεt]

r con

5.

My uncle is a FARmer. [majnkl]

j con

6.

The red car went aWAY. [ɹε kɑ]

7.

They robbed the bank YEsterday. [ɹɒbðə]

8.

This shape is a SQUARE. [ðʃʃep]

#sh con

9.

The brown bear eats FISH. [bɹaυmmbεə]

n# con

DEFINITE ARTICLE

Ct#C con con #sh con

d# con Cd#C con

10. I live near a big WOOD. [nə ɹ ə]

r con

11. We saw an elephant at the ZOO. [ən’εləfənt] 12. John collects STAMPS. [d ɒŋkəlεks]

INDEFINITE ARTICLE

LIAISON

CHILD’S PRODUCTION TARGET SENTENCE OF JUNCTURE

ELISION

Appendix F.2

ASSIMILATION

350

con n# con

13. Sam loved to DANCE. [lvtə]

Cd#C con

14. She wrapped the PARcel. [ɹpðə]

Ct#C con

15. You can read my BOOK. [ɹib ma]

d# con

16. Good girls are NICE. [υlz]

d# con

con

18. I washed my hair last NIGHT. [ wɒʃmɑ]

Cd#C con

19. He judged the compe TItion. [ d d ðə]

Cd#C con

20. I wore a JUMper. [ wɔ ɹə]

r con t# con

22. The gold box was HEAvy. [gəυlbɒks] 23. John played TEnnis. [ d ɒmpled]

Cd#C con n# con

24. He gave me a baNAna. [ mijə]

j con

25. We saw the tent by the RIver. [ tεmba]

con

26. She picked some FLOwers. [ pksəm]

Cd#C con

Ct#C

27. Sam ate an orange very SLOWly. [ənɒɹnd ]

con w

28. Some smoke blew out of the CHIMney. [ bluwaυt]

con

29. He sneezed very LOUDly. [ snizveɹ]

Cd#C con

30. We found presents under the TREE. [ faυmpɹεzənʔs]

Cd#C con

31. Tom hit Claire very HARD. [ hkklεə]

t# con

DEFINITE ARTICLE

INDEFINITE ARTICLE

Cd#C

17. My mum hugged me when I was SAD. [hgmi]

21. You eat pudding with a SPOON. [ ippυdŋ]

LIAISON

351

ELISION


ASSIMILATION

Appendix F.2

352

Appendix F.2

32. You must stir in the SUgar. [ stɹn]

r con

33. The toy elephant was BROken. [ tɔjεləfənt]

j con

34. The yellow aeroplane CRASHED. [ jεləυwεəɹəplen]

w con

35. She cut my HAIR. [ kpma]

DEFINITE ARTICLE

t# con

36. He watched television all DAY. [wɒtʃtεlv ən] 37. Jane made some SOUP. [ d emmed]

INDEFINITE ARTICLE

LIAISON

ELISION


ASSIMILATION

(Continued)

Cd#C con n# con

38. She gave the orange to SAM. [ðjɒɹnd ]

con

39. My left leg HURTS. [ lεflεg]

Ct#C con

40. They argued all DAY. [ ðejɑjud]

j con

41. Alice put gloves on her HANDS. [ pυʔglvz]

t# con

42. We had bacon for LUNCH. [ hæbbekən]

d# con

Appendix F.2

353

SUMMARY SHEET

Adult-like close (c ) junctures

No. of No. of Number % open (o) non-adult correct correct junctures (n) junctures

Assimilation t#

1111

/4

%

n#

1111

/4

%

d#

1111

/4

%

#sh

11

/2

%

Ct#C 1 1 1 1

/4

%

Cd#C 1 1 1 1 1 1 1 1 1 1

/ 10

%

/j/

1111

/4

%

/w/

11

/2

%

/ɹ/

1111

/4

%

Indefinite article

11

/2

%

Definite article

11

/2

%

Elision

Liaison

Appendix F.3 Connected Speech Task 3: Final Consonant Juncture Repetition (from Pascoe, Stackhouse & Wells, 2006) Name:__________Date:_________Age:_________Investigator:________ Instructions: Read each sentence to the child. The child is asked to repeat the sentence that he or she hears. All responses should be audiorecorded using a good-quality recorder and microphone. On completion of the sentence repetition, the child is asked to repeat all single words. One repetition of each stimulus is allowed if the child requests it or was not attending. Where possible, the stimuli should be presented in an accent that is close to that of the child’s. The investigator should produce each sentence adhering to the following guidelines: each item is presented with an intonation pattern that is unmarked for declarative statements in English; the tonic syllable is the final lexically stressed syllable of the sentence or phrase, items marked with ** are presented with assimilation. The final consonant(s) of the target word is assimilated into the following consonant, e.g. THIS NOTE CAN’T BE READ produced as /ðs nəυk kɑm bi rεd/. The assimilation is marked below with an underline for these items. The child’s production of the target word (in capitals) should be transcribed on the score sheet, and can be checked with the recording. Scoring: The child’s production of the final consonant (in bold) is scored. Circle the appropriate number: 0 no final consonant, 1 a final consonant but not correct, 2 final consonant is correct. For items in which assimilation is modelled (marked ** below) the target final consonant is scored as follows: 2 consonant is assimilated and realised as perceptibly long (as in the target), e.g. [ðs nəυk kɑm bi rεd]. 1 child realises the final consonant in unassimilated form, e.g. [ðs nəυtkɑm bi rεd], or produces an incorrect final consonant. 0 final consonant omitted (e.g. [ðs nəυ kɑm bi rεd] with no perceptible lengthening of the [k]). No normative data is available for this task. 355

356

Appendix F.3

Child’s Level 1 Level 2 Level 3 production Facilita Neutral without of target tory assimilation word (including final consonant)

Sentences PRACTICE 1: The DOG ate the bone.

Level 3 with assimilation

0 1 2

PRACTICE 2: There was a BIG bang! 1.

This NOTE teaches father xmas

2.

There’s a HEART on my jumper

3.

The PLANE knocked it

4.

**

5.

The WHEEL looks broken

6.

**

The SLIDE bounces him

7.

There’s a the sky

8.

This NAIL got painted twice

9.

This nice

0 1 2 0 1 2

This NOTE can’t be read

PLANE

HEART

0 1 2

**

0 1 2 0 1 2

**

0 1 2

in

0 1 2 0 1 2

tastes

0 1 2

10. The CAGE joined my class

0 1 2

11. There’s a RAKE on the ground 12.

**

**

This

HEART

0 1 2

can

break 13. This NAIL looks pretty

0 1 2 0 1 2

Appendix F.3

Child’s production of target word (including final consonant)

Sentences

Level 1 Level 2 Level 3 Facilita Neutral without ory assimilation

14. This STORK teaches swinging 0 1 2

16. This WHEEL got fixed

0 1 2

17. There’s a CAGE on the bed

0 1 2

18. There’s a ROAD over the river

0 1 2

ICE

seems

0 1 2

20. The SLIDE in the park is nice 21.

**

0 1 2 **

This

0 1 2

must be loaded PLANE

22. The LEAF is in the air

0 1 2

23. There’s a WHEEL on the bike

0 1 2

24. The CAGE got stolen

0 1 2 0 1 2

25. There’s a NOTE under the table 26. This RAKE cost ten pounds 27. There’s a NAIL in the wood 28. This soft

SOAP

got


0 1 2

15. This SLIDE dumped me

19. This cold

357

0 1 2 0 1 2 0 1 2

358

Appendix F.3

(Continued) Child’s production of target word (including final consonant)

Sentences

29. The STORK carries a baby


0 1 2

30. The BARN needs painting

0 1 2

31. The SAUCE seems nice

0 1 2

32. The wet 33.

feels

0 1 2

**

The BARN must be cleaned

34. This torn 35.

LEAF

LEAF

**

0 1 2

got

0 1 2

**

**

This should melt

0 1 2

ICE

36. There’s the jar

SAUCE


in

0 1 2

37. There’s SOAP on the towel

0 1 2

38. There’s a STORK on the log

0 1 2

39. The ICE is in the bucket

0 1 2

40. There’s a PIPE on the wall

0 1 2

41. This SOAP pulled my hair

0 1 2

42. The ROAD divides the hill

0 1 2

43. The PIPE pushes through the roof

0 1 2

Appendix F.3

44.

Sentences

Child’s production of target word (including final consonant)

**

**

The SAUCE shouldn’t burn

45. There’s a the farm

BARN

359


0 1 2

on

0 1 2

46. This PIPE curls round the floor

0 1 2

47. This RAKE takes a bath

0 1 2

48.

**

This brings us home


**

0 1 2

ROAD

TOTAL FINAL CONSONANTS SCORE % FINAL CONSONANTS SCORE (connected speech)

/32

/32

/16

/16

%

%

%

%

360

Appendix F.3

Score: 0 no final consonant; 1 a final consonant but not correct; 2 final consonant is correct Single words

Child’s production

Score

note

0 1 2

plane

0 1 2

heart

0 1 2

nail

0 1 2

cage

0 1 2

slide

0 1 2

wheel

0 1 2

rake

0 1 2

stork

0 1 2

leaf

0 1 2

sauce

0 1 2

ice

0 1 2

soap

0 1 2

pipe

0 1 2

barn

0 1 2

road

0 1 2

TOTAL FINAL CONSONANTS SCORE

/32

% FINAL CONSONANTS (single words score)

%

Appendix G.1 Speech Accuracy, Rate and Consistency Task 1: Oral Movements and Silent DDK Rates (from Williams & Stackhouse, 2000) Name:____________Date:_________Age:_____Investigator:_________ Instructions: Instructions are given verbally as well as demonstrated, e.g. ‘Can you stick out your tongue?’. A mirror can be used to provide visual feedback. Two or three of the movements are presented first for practice, with feedback to ensure the child understands what is required. All the movements are then re-presented one at a time and scored. The child is then asked to copy each movement sequence and repeat it five times as quickly as possible. The time taken for the child to complete the five repetitions of each silent DDK sequence is measured in seconds using a stopwatch. Scoring Accuracy Measure: Circle 0 or 1 depending on the child’s accuracy in making each movement, and each sequence of movements. For protrusion or lateral movement a clearly defined tongue tip shape is not essential for a correct score; for tongue elevation a differentiated tongue tip should be demonstrated to be scored as correct. Percentage scores are calculated for each set of movements and total for both lip and tongue movement. Rate Measure: Rate is measured as the time taken in seconds to produce five repetitions of each sequenced lip and tongue movement. The mean rate for producing five repetitions is then calculated for each set of movements and for both lip and tongue movements together. When the child is not able to achieve one or more of the component movements, a score of 0 is given. Normative data for this task can be found in Tables 7.1 and 7.2.

361

362

Appendix G.1

LIP MOVEMENTS

ACCURACY

TIME FOR 5 REPETITIONS IN SECONDS

Single (isolated) Rounded

0

1

Spread

0

1

Open

0

1

Alternating round and spread

0

1

Alternating round and open

0

1

Sequenced (5 repetitions)

TOTAL NO. ACCURATE LIP MOVEMENTS

/5

% ACCURATE

%

MEAN RATE OF 5 REPETITIONS

secs

TONGUE MOVEMENTS Single (isolated) Protrusion

0

1

Retraction

0

1

Lateral (right)

0

1

Lateral (left)

0

1

Elevation and depression

0

1

0

1

Alternating lateral movement

0

1

Alternating elevation and depression

0

1

Sequenced (5 repetitions) Alternating protrusion and retraction

TOTAL NO. ACCURATE TONGUE MOVEMENTS % ACCURATE

/8 %

MEAN RATE OF 5 REPETITIONS

secs

TOTAL NO. ACCURATE LIP AND TONGUE MOVEMENTS

/13

% ACCURATE

%

MEAN RATE OF 5 REPETITIONS FOR LIP & TONGUE MOVEMENTS

secs

Appendix G.2 Speech Accuracy, Rate and Consistency Task 2a: Repetition and Spoken DDK Rates: Words (from Williams & Stackhouse, 2000) Name:_________Date:___________Age:____ Investigator:__________ Instructions: The investigator presents the picture, says the word and asks the child to repeat it. The child’s initial production of the target word is transcribed to provide a baseline production. After the child has imitated the target word once, he or she is asked to say it five times, as quickly as possible. The tick chart can be used to indicate to the child how many times he or she has produced the item so far, by ticking in each column (1–5) as he or she says each repetition. The time taken to perform each set of five repetitions is recorded with a stopwatch and entered on the form. The child’s responses are audio-recorded with a good-quality recorder and microphone for later transcription of the five repetitions. Six practice items are presented, with feedback, to ensure that the child understands the task. For example, if a child stops after three repetitions, he or she is encouraged to continue for longer; if the child speaks so fast that the word becomes unrecognisable, the child is reminded that the tester needs to hear the word; if the child produces the words very slowly, he or she is encouraged to try again and go faster. The test items are then presented with no further help. Scoring (see bottom of score sheet for a completed example) Accuracy Measures: 1. Accuracy of first attempt: The child’s first attempt at repeating the target is scored for accuracy in terms of consonants correct, in comparison to a local adult model, allowing for regional accent variation. Responses with speech sound errors relating to consonants, e.g. substitutions or omissions, are scored as incorrect.

363

364

Appendix G.2

2. Accuracy of five repetitions: Each set of five repetitions is scored as correct if the target is repeated correctly, according to the local adult model, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Consistency Measures: The child’s first imitated response is taken as the baseline for consistency measures, irrespective of whether it is accurate compared to the adult model. 1. Consistency of five repetitions: Each set of five repetitions is scored as correct if the target is repeated in the same way as the child’s first (baseline) response, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Essentially this consistency score is the same procedure as above for accuracy, but instead of comparing to an adult model, the child’s own initial response (before the five repetitions) is used. 2. A second consistency measure compares each repetition with the first, baseline production from the child and across the other four repetitions of the same target. The following scale is used for each of the five repetitions. Rating I: Repetition identical to child’s baseline response. Rating II: Repetition different from child’s baseline response. Rating III: Repetition different from child’s baseline and from one other previous repetition. Rating IV: Repetition different from child’s baseline and from two other previous repetitions. Scoring examples are given at the end of this score sheet. Rate Measure: The time taken to produce the five repetitions of each item is measured in seconds; both accurate and inaccurate productions are accepted. However, if a child omits a syllable or stops before completing five repetitions, timing of this item is not recorded or included in the mean scores. The mean rate is calculated for all sets of five repetitions. Normative data for this task can be found in Tables 7.4–7.8.

Appendix G.2

365

PRACTICE ITEMS: BABY

LOLLY

POTTY

TIGER

TOMATO

MOTORBIKE

TICK CHART Child’s first production PRACTICE

baby

ITEMS

lolly potty tiger tomato motorbike

TEST ITEMS party seesaw cardigan fizzy yellow pat-a-cake money paper letter-box telephone jelly digger buttercup cooker coffee sellotape

1

2

3

4

5

366

Appendix G.2

TEST ITEMS: Stimulus:

1

PARTY

Child’s production of single word Sets of 5

Accuracy measure 1: first attempt 1

2

3

4

0

1

5

Child’s production Accuracy measure 2 (adult model)

0

1

0

1

0

1

0

1

0

1

Consistency measure 1 (child’s baseline)

0

1

0

1

0

1

0

1

0

1

Consistency Measure 2: rating

I II III IV

I II III IV

I II III IV

I II III IV

I II III IV

Accuracy measure 2 (adult model)

0

1


0

1

Consistency measure 2: rating

I II III IV

Time taken in seconds Stimulus:

2

SEESAW



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV

Appendix G.2

367

(Continued) Accuracy measure 2 (adult model) Consistency measure 1 (child’s baseline) Consistency measure 2: rating

0

1

0

1

I II III IV

Time taken in seconds

Stimulus:

3

CARDIGAN



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1

Consistency measure 2: rating Time taken in seconds

I II III IV

368

Stimulus:

Appendix G.2

4

FIZZY



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

5

YELLOW



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.2

369

(Continued) Consistency Measure 2: rating

I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

6

PAT-ACAKE



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1

Consistency measure 1 (child’s base-line)

0

1


I II III IV

370

Stimulus:

Appendix G.2

7

MONEY



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

8

PAPER



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.2

371

(Continued) Consistency MEASURE 2: rating

I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

9

LETTERBOX



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1

Consistency measure 1 (child’s base-line)

0

1


I II III IV

372

Stimulus:

Appendix G.2

10

TELEPHONE



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

11

JELLY



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.2

373


I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

Time taken

Stimulus:

12

secs

DIGGER


I II III IV


2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

374

Stimulus:

Appendix G.2

13

BUTTERCUP



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

14

COOKER



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.2

375


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

15

COFFEE



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

376

Stimulus:

Appendix G.2

16

SELLOTAPE



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

Appendix G.2

377

SUMMARY SHEET

Stimulus

Accuracy first attempt

Accuracy Consistency of 5 of repetitions 5 repetitions

Consistency rating

Time taken, in seconds 2 syll 3 syll

party

0

1

0

1

0

1

I II III IV

xxx

seesaw

0

1

0

1

0

1

I II III IV

xxx

cardigan

0

1

0

1

0

1

I II III IV

fizzy

0

1

0

1

0

1

I II III IV

xxx

yellow

0

1

0

1

0

1

I II III IV

xxx

pat-acake

0

1

0

1

0

1

I II III IV

money

0

1

0

1

0

1

I II III IV

xxx

paper

0

1

0

1

0

1

I II III IV

xxx

letter-box

0

1

0

1

0

1

I II III IV

xxx

telephone 0

1

0

1

0

1

I II III IV

xxx

jelly

0

1

0

1

0

1

I II III IV

xxx

digger

0

1

0

1

0

1

I II III IV

xxx

buttercup 0

1

0

1

0

1

I II III IV

cooker

0

1

0

1

0

1

I II III IV

xxx

coffee

0

1

0

1

0

1

I II III IV

xxx

sellotape

0

1

0

1

0

1

I II III IV

Mean scores

/16

/16

/16

No. of items at each consistency rating (out of 16)

xxx

Percentage items at each consistency rating

xxx

xxx

xxx

xxxs

xxx

xxx

378

Appendix G.2

EXAMPLES OF SCORE SHEET COMPLETION Stimulus:

12

BUTTERCUP

Child’s production of single word Sets of 5 Child’s production

[butekup] 1

2

[butekup]

Accuracy first attempt 3

[butekup]

4

[buketup]

0

1

5

[buketup]

[butetup]


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

BUTTERCUP

Child’s production of single word

[buketup]

Sets of 5 Child’s production

1

2

[puketup]

I II III IV

Accuracy first attempt 3

[puketup]

3.45

4

[putetup]

0

1

5

[putetup]

[putetup]


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III I II III IV IV

Appendix G.2

379

(Continued) Accuracy measure 2 (adult model)

0

1


0

1


I II III IV 4.65

Appendix G.3 Speech Accuracy, Rate and Consistency Task 2b: Repetition and Spoken DDK Rates: Non-Words (from Williams & Stackhouse, 2000) Name:_____________Date:_________Age:____Investigator:__________ Instructions: The investigator says a non-word and asks the child to repeat it. The child’s initial production of the target non-word is transcribed to provide a baseline production. After the child has imitated the target once, he or she is asked to say it five times, as quickly as possible. The tick chart can be used to indicate to the child how many times he or she has produced the item so far, by ticking in each column (1–5) as he or she says each repetition. The time taken to perform each set of five repetitions is recorded with a stopwatch and entered on the form. The child’s responses are audio-recorded with a good-quality recorder and microphone for later transcription of the five repetitions. Six practice items are presented, with feedback, to ensure that the child understands the task. For example, if a child stops after three repetitions, he or she is encouraged to continue for longer; if the child speaks so fast that the word becomes unrecognisable, the child is reminded that the tester needs to hear the word; if the child produces the words very slowly, he or she is encouraged to try again and go faster. The test items are then presented with no further help. Scoring (see bottom of Appendix G.2 for a completed example): Accuracy Measures: 1. Accuracy of first attempt: The child’s first attempt at repeating the target is scored for accuracy in terms of consonants correct, in com-

381

382

Appendix G.3

parison to a local adult model, allowing for regional accent variation. Responses with speech sound errors relating to consonants, e.g. substitutions or omissions, are scored as incorrect. 2. Accuracy of five repetitions: Each set of five repetitions is scored as correct if the target is repeated correctly, according to the local adult model, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Consistency Measures: The child’s first imitated response is taken as the baseline for consistency measures, irrespective of whether it is accurate compared to the adult model. 1. Consistency of five repetitions: Each set of five repetitions is scored as correct if the target is repeated in the same way as the child’s first (baseline) response, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Essentially this consistency score is the same procedure as above for accuracy, but instead of comparing to an adult model, the child’s own initial response (before the five repetitions) is used. 2. A second consistency measure compares each repetition with the first, baseline production from the child and across the other four repetitions of the same target. The following scale is used for each of the five repetitions. Rating I: Repetition identical to child’s baseline response. Rating II: Repetition different from child’s baseline response. Rating III: Repetition different from child’s baseline and from one other previous repetition. Rating IV: Repetition different from child’s baseline and from two other previous repetitions. Scoring examples are given at the end of this score sheet. Rate Measure: The time taken to produce the five repetitions of each item is measured in seconds; both accurate and inaccurate productions are accepted. However, if a child omits a syllable or stops before completing five repetitions, timing of this item is not recorded or included in the mean scores. The mean rate is calculated for all sets of five repetitions. Normative data for this task can be found in Tables 7.4–7.8.

Appendix G.3

383

PRACTICE ITEMS: / babə/

/ lælə/

/ petə/

/ tɒ/

/ təmu:təυ/

/ maυtbek/

TICK CHART Child’s first production PRACTICE ITEMS

/ babə/ / lælə/ / petə/ / tɒ/ / təmu:təυ/ / maυtbek/

TEST ITEMS /pu: tə/ /sɑsi/ / kudæn/ /fɒzə/ /jælɒ/ / pɒtkəυk/ /mnə/ /pap / ltbæks/ / tɒləfan/ /dlə/ /dæ/ / bætkεp/ /kεk/ /kfə/ / sɒltaυp/

1

2

3

4

5

384

Appendix G.3

TEST ITEMS:

Stimulus:

1

/putə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

2

/sɑsi/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.3

385


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

3

/ kudæn/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Consistency MEASURE 2: rating

I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

386

Stimulus:

Appendix G.3

4

/fɒzə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

5

/jælɒ/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.3

387


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

6

/ pɒtkəυk/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

388

Stimulus:

Appendix G.3

7

/mnə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

8

/pap/



2

3

0

4

1 5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.3

389


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

9

/ ltbæks/



2

3

0

4

1 5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

Stimulus:

10

/ tɒləfan/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0


0


1 1

I II III IV


Stimulus:

11

/dlə/



2

3

0

4

1 5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

Appendix G.3

Stimulus:

12

/dæ/


391


2

3

0

4

1 5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

13

/ bætkεp/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

392

Appendix G.3


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

14

/kεk/



2

3

0

4

1 5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

Appendix G.3

Stimulus:

15

/kfə/


393


2

3

0

4

1 5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

16

/ sɒltɑυp/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

394

Appendix G.3


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


SUMMARY SHEET

Stimulus

Accuracy Accuracy Consistency first of 5 of Consistency attempt repetitions 5 repetitions rating


/ pu:tə/

0

1

0

1

0

1

I II III IV

xxx

/ sɑsi/

0

1

0

1

0

1

I II III IV

xxx

/ kudæn/

0

1

0

1

0

1

I II III IV xxx

/fɒzə/

0

1

0

1

0

1

I II III IV

xxx

/jælɒ/

0

1

0

1

0

1

I II III IV

xxx

/ pɒtkəυk/

0

1

0

1

0

1

I II III IV xxx

/mnə/

0

1

0

1

0

1

I II III IV

xxx

/pap/

0

1

0

1

0

1

I II III IV

xxx

/ ttbæks/

0

1

0

1

0

1

I II III IV xxx

/ tɒləfan/

0

1

0

1

0

1

I II III IV xxx

/dlə/

0

1

0

1

0

1

I II III IV

xxx

/dæ/

0

1

0

1

0

1

I II III IV

xxx

/ bætkεp/

0

1

0

1

0

1

I II III IV xxx

/kεk/

0

1

0

1

0

1

I II III IV

xxx

/kfə/

0

1

0

1

0

1

I II III IV

xxx

/ sɒltɑυp/

0

1

0

1

0

1

I II III IV xxx

Appendix G.3

395

(Continued)

Stimulus Mean scores

Accuracy Accuracy Consistency first of 5 of Consistency attempt repetitions 5 repetitions rating /16

/16

/16


xxx


xxx

xxx

Time taken, in seconds

Appendix G.4 Speech Accuracy, Rate and Consistency Task 2c: Repetition and Spoken DDK Rates: Syllable Sequences (from Williams & Stackhouse, 2000) Name:_________Date:________Age:_________ Investigator:_________ Instructions: The investigator says a syllable sequence and asks the child to repeat it. The sequence is produced with equal stress on each syllable. The child’s initial production of the target sequence is transcribed to provide a baseline production. After the child has imitated the target once, he or she is asked to say it five times, as quickly as possible. The tick chart can be used to indicate to the child how many times he or she has produced the item so far, by ticking in each column (1–5) as he or she says each repetition. The time taken to perform each set of five repetitions is recorded with a stopwatch and entered on the form. The child’s responses are audio-recorded with a good-quality recorder and microphone for later transcription of the five repetitions. Six practice items are presented, with feedback, to ensure that the child understands the task. For example, if a child stops after three repetitions, he or she is encouraged to continue for longer; if the child speaks so fast that the word becomes unrecognisable, the child is reminded that the tester needs to hear the word; if the child produces the words very slowly, he or she is encouraged to try again and go faster. The test items are then presented with no further help. Scoring (see bottom of Appendix G.2 for a completed example): Accuracy Measures: 1. Accuracy of first attempt: The child’s first attempt at repeating the target is scored for accuracy in terms of consonants correct, in com-

397

398

Appendix G.4

parison to a local adult model, allowing for regional accent variation. Responses with speech sound errors relating to consonants, e.g. substitutions or omissions, are scored as incorrect. 2. Accuracy of five repetitions: Each set of five repetitions is scored as correct if the target is repeated correctly, according to the local adult model, all five times. If the child does not produce all five repetitions, this is scored as incorrect. Consistency Measures: The child’s first imitated response is taken as the baseline for consistency measures, irrespective of whether it is accurate compared to the adult model. 1. Consistency of five repetitions: Each set of five repetitions is scored as correct if the target is repeated in the same way as the child’s first (baseline) response all five times. If the child does not produce all five repetitions, this is scored as incorrect. Essentially this consistency score is the same procedure as above for accuracy, but instead of comparing to an adult model, the child’s own initial response (before the five repetitions) is used. 2. A second consistency measure compares each repetition with the first, baseline production from the child and across the other four repetitions of the same target. The following scale is used for each of the five repetitions. Rating I: Repetition identical to child’s baseline response. Rating II: Repetition different from child’s baseline response. Rating III: Repetition different from child’s baseline and from one other previous repetition. Rating IV: Repetition different from child’s baseline and from two other previous repetitions. Rate Measure: The time taken to produce the five repetitions of each item is measured in seconds; both accurate and inaccurate productions are accepted. However, if a child omits a syllable or stops before completing five repetitions, timing of this item is not recorded or included in the mean scores. The mean rate is calculated for all sets of five repetitions. Normative data for this task can be found in Tables 7.4–7.8.

Appendix G.4

399

PRACTICE ITEMS: / bəbə/

/ lələ/

/ pətə/

/ təə/

/ təmətə/

/ mətəbəkə/

TICK CHART

Child’s first production PRACTICE ITEMS

/ bəbə/ / lələ/ / pətə/ / təə/ / təmətə/ / mətəbəkə/

TEST ITEMS /pətə/ /səsə/ /kədəə/ /fəzə/ /jələ/ /pətəkə/ /mənə/ /pəpə/ /lətəbə/ /tələfə/ /d ələ/ /dəə/ /bətəkə/ /kəkə/ /kəfə/ /sələtə/

1

2

3

4

5

400

Appendix G.4

TEST ITEMS: Stimulus:

1

/pətə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

2

/səsə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.4

401


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

3

/kədəə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

402

Stimulus:

Appendix G.4

4

/fəzə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

5

/jələ/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.4

403


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

6

/pətəkə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

404

Stimulus:

Appendix G.4

7

/mənə/



2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

8

/pəpə/


Accuracy measure 1: first attempt

Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.4

405


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

9

/lətəbə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

406

Stimulus:

Appendix G.4

10

/tələfə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

11

/dələ/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1

Appendix G.4

407


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

12

/dəə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

408

Stimulus:

Appendix G.4

13

/bətəkə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV


Stimulus:

14

/kəkə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV

Appendix G.4

409

(Continued) Accuracy measure 2 (adult model)

0

1


0

1


I II III IV


Stimulus:

15

/kəfə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

410

Stimulus:

Appendix G.4

16

/sələtə/



Sets of 5

1

2

3

4

0

1

5


0

1

0

1

0

1

0

1

0

1


0

1

0

1

0

1

0

1

0

1


I II III IV

I II III IV

I II III IV

I II III IV

I II III IV


0

1


0

1


I II III IV

Appendix G.4

411

SUMMARY SHEET

Stimulus

Accuracy first attempt

Accuracy of five repetitions

Consistency of five repetitions

Consistency rating


/pətə/

0

1

0

1

0

1

I II III IV

xxx

/səsə/

0

1

0

1

0

1

I II III IV

xxx

/kədəə/

0

1

0

1

0

1

I II III IV

/fəzə/

0

1

0

1

0

1

I II III IV

xxx

/jələ/

0

1

0

1

0

1

I II III IV

xxx

/pətəkə/

0

1

0

1

0

1

I II III IV

/mənə//

0

1

0

1

0

1

I II III IV

xxx

/pəpə/

0

1

0

1

0

1

I II III IV

xxx

/lətəbə/

0

1

0

1

0

1

I II III IV

xxx

/tələfə/

0

1

0

1

0

1

I II III IV

xxx

/d ələ/

0

1

0

1

0

1

I II III IV

xxx

/dəə/

0

1

0

1

0

1

I II III IV

xxx

/bətəkə/

0

1

0

1

0

1

I II III IV

/kəkə/

0

1

0

1

0

1

I II III IV

xxx

/kəfə/

0

1

0

1

0

1

I II III IV

xxx

/sələtə/

0

1

0

1

0

1

I II III IV

Mean scores

/16

/16

/16

xxx


xxx


xxx

xxx

xxx

xxx

xxx

Appendix G.5 Speech Accuracy and Consistency Task 3 (from Coffield, 1994)

Name:_____________Date:________Age:_____Investigator:__________ Instructions: The instructions are as follows. Non-word repetition: ‘I am going to say some silly words. They are not real words. I want you to say the silly words exactly as I say them. I want you to say them three times. Listen to a silly word MUB, now I will say it three times (demonstration is given with emphasis on a measured rate of repetition). Now you try. For repetition tasks, the investigator’s mouth should be obscured (casually by hand) to prevent lip cueing. Word repetition: ‘I am going to say some real words this time. They are words that you have heard before. I want you to say the words exactly as I say them. Listen to a word TABLE , now you say it three times.’ Naming: ‘I am going to show you some pictures. I want you to tell me the names of the pictures. I want you to tell me the names of the pictures three times just as you did before.’ Each picture is presented and the child is asked to say the name of the picture three times. If the child is unable to name the picture, semantic and gap-fill cues can be given. Children may need reminding that more than one production is required, and this can be cued by saying ‘again’ if the child stops before he or she has produced the item three times. It may also be necessary to remind the child not to rush the repetitions, as rate is not scored in this task. Scoring Accuracy Measure: Each production is scored as correct or incorrect by comparing the child’s realisation of the consonants with the adult model, and circling 0 for an

413

414

Appendix G.5

incorrect and 1 for a correct response. (See Appendix G.3 for scoring guidelines for this stimuli set.) Local accent/variants should be taken into account. Total accuracy scores for each condition and for each word length are collated on the summary sheet. The accuracy score for each condition and for each word length is then converted to a percentage. Consistency Measure: For each set, a score of 1 is given if the item is repeated with no change in pronunciation for all three repetitions. The child’s productions can be correct or incorrect as compared to an adult model. Total consistency scores for each condition and for each word length are collated on the summary sheet. The consistency score for each condition and for each word length is then converted to a percentage. Normative data for this task can be found in Tables 7.9 and 7.10.

LIST A: REPETITION OF 1-SYLLABLE NON-WORDS; REPETITION OF 2-SYLLABLE WORDS; NAMING OF PICTURES OF 3–4-SYLLABLE WORDS. SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

2 syll

CONSISTENCY

3–4 1 syll syll

NON-WORD REPETITION ITEMS /brʃ/

/spænd/

/glεv/

/dæk/

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

2 syll

3–4 syll

Appendix G.5

415

(Continued ) SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

/lɔf/

CONSISTENCY

2 syll 3–4 1 syll syll

0 1 0 1 0 1

/sɔk/

0 1

0 1 0 1 0 1

/kεt/

0 1

0 1 0 1 0 1

/bɔk/

0 1

0 1 0 1 0 1

/tutʃ/

0 1

0 1 0 1 0 1

/mɔs/

0 1

0 1 0 1 0 1

/nəυf/

0 1

0 1 0 1 0 1

/snak/

0 1

0 1 0 1 0 1

0 1

2 syll 3–4 syll

416

Appendix G.5


CHILD’S RESPONSE

ACCURACY 1 syll

/trɔn/

CONSISTENCY


0 1 0 1 0 1

/vn/

0 1

0 1 0 1 0 1

/wυtʃ/

0 1

0 1 0 1 0 1

/pləυt/

0 1

0 1 0 1 0 1

/rɔf/

0 1

0 1 0 1 0 1

/fεʃ/

0 1

0 1 0 1 0 1

/tʃi/

0 1

0 1 0 1 0 1

/θɒm/

0 1

0 1 0 1 0 1

0 1

2 syll 3–4 syll

Appendix G.5

417

REAL WORD REPETITION SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

sandwich

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 0 1

toilet

0 1

0 1 0 1 0 1

money

0 1

0 1 0 1 0 1

feather

0 1

0 1 0 1 0 1

yellow

0 1

0 1 0 1 0 1

kitchen

0 1

0 1 0 1 0 1

ladder

0 1

0 1 0 1 0 1

flower

0 1

0 1 0 1 0 1

dustbin

0 1

0 1

418

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 jelly

0 1

0 1 0 1 0 1

tractor

0 1

0 1 0 1 0 1

fishing

0 1

0 1 0 1 0 1

biscuit

0 1

0 1 0 1 0 1

scooter

0 1

0 1 0 1 0 1

parrot

0 1

0 1 0 1 0 1

seesaw

0 1

0 1 0 1 0 1

slipper

0 1

0 1

Appendix G.5

419

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 sausage

0 1

0 1 0 1 0 1

guitar

0 1

0 1 0 1 0 1

spider

0 1

0 1 0 1 0 1

0 1

NAMING STIMULI caterpillar

0 1 0 1 0 1

spaghetti

0 1

0 1 0 1 0 1

elephant

0 1

0 1 0 1 0 1

caravan

0 1

0 1 0 1 0 1

0 1

420

Appendix G.5


CHILD’S RESPONSE

ACCURACY 1 syll

crocodile

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 0 1

umbrella

0 1

0 1 0 1 0 1

helicopter

0 1

0 1 0 1 0 1

kangaroo

0 1

0 1 0 1 0 1

television

0 1

0 1 0 1 0 1

hospital

0 1

0 1 0 1 0 1

telephone

0 1

0 1 0 1 0 1

parachute

0 1

0 1 0 1 0 1

butterfly

0 1

0 1

Appendix G.5

421

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 computer

0 1

0 1 0 1 0 1

roundabout

0 1

0 1 0 1 0 1

hairdresser

0 1

0 1 0 1 0 1

aeroplane

0 1

0 1 0 1 0 1

pyjamas

0 1

0 1 0 1 0 1

hamburger

0 1

0 1 0 1 0 1

dinosaur

0 1

0 1 0 1 0 1

TOTAL ACCURACY

/60

TOTAL CONSISTENCY /20

0 1 XXX

XXX XXX

422

Appendix G.5

LIST B: NON-WORD REPETITION, 2 SYLLABLES; WORD REPETITION, 3–4 SYLLABLES; NAMING, 1 SYLLABLE SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

NON-WORD REPETITION ITEMS / smwəd/

0 1 0 1 0 1

/ telət/

0 1

0 1 0 1 0 1

/ mεnə/

0 1

0 1 0 1 0 1

/ fæð/

0 1

0 1 0 1 0 1

/ jælɔ/

0 1

0 1 0 1 0 1

/ kɒtʃ n/

0 1

0 1 0 1 0 1

/ lεd/

0 1

0 1 0 1 0 1

/ fluw/

0 1

0 1

Appendix G.5

423

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 / dæsbn/

0 1

0 1 0 1 0 1

/ dzlə/

0 1

0 1 0 1 0 1

/ trεkt/

0 1

0 1 0 1 0 1

/ fɒʃŋ/

0 1

0 1 0 1 0 1

/ bɒskεt/

0 1

0 1 0 1 0 1

/ skitə/

0 1

0 1 0 1 0 1

/ prt/

0 1

0 1 0 1 0 1

/ sɑsi/

0 1

0 1

424

Appendix G.5


CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 / slɔpə/

0 1

0 1 0 1 0 1

/sεsəd/

0 1

0 1 0 1 0 1

/εtɔ/

0 1

0 1 0 1 0 1

/ sped/

0 1

0 1 0 1 0 1

0 1

REAL WORD REPETITION caterpillar

0 1 0 1 0 1

spaghetti

0 1

0 1 0 1 0 1

elephant

0 1

0 1 0 1 0 1

caravan

0 1

0 1

Appendix G.5

425

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 crocodile

0 1

0 1 0 1 0 1

umbrella

0 1

0 1 0 1 0 1

helicopter

0 1

0 1 0 1 0 1

kangaroo

0 1

0 1 0 1 0 1

television

0 1

0 1 0 1 0 1

hospital

0 1

0 1 0 1 0 1

telephone

0 1

0 1 0 1 0 1

parachute

0 1 0 1

0 1

426

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY

2 syll 3–4 1 syll syll 0 1

butterfly

2 syll 3–4 syll 0 1

0 1 0 1 0 1

computer

0 1

0 1 0 1 0 1

roundabout

0 1

0 1 0 1 0 1

hairdresser

0 1

0 1 0 1 0 1

aeroplane

0 1

0 1 0 1 0 1

pyjamas

0 1

0 1 0 1 0 1

hamburger

0 1

0 1 0 1 0 1

dinosaur

0 1 0 1

0 1

Appendix G.5

427

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


0 1

NAMING STIMULI brush

0 1 0 1 0 1

sponge

0 1

0 1 0 1 0 1

glove

0 1

0 1 0 1 0 1

duck

0 1

0 1 0 1 0 1

leaf

0 1

0 1 0 1 0 1

sock

0 1

0 1 0 1 0 1

cat

0 1

0 1 0 1 0 1

book

0 1

2 syll 3–4 syll

0 1

428

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

torch

mouse

knife

snake

train

van

watch

plate

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

CONSISTENCY


0

1

0

1

0

1

0

1

0

1

0

1

0

0

1

1

2 syll 3–4 syll

Appendix G.5

429 SCORE

STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

roof

fish

chair

thumb

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0

1

0 TOTAL ACCURACY

CONSISTENCY


1

/60

0

1

0

1

0

1

0

1

0

1

XXX

2 syll 3–4 syll

XXX XXX


LIST C: NON-WORD REPETITION, 3–4 SYLLABLES; WORD REPETITION, 1 SYLLABLE; NAMING 2 SYLLABLES SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll


NON-WORD REPETITION ITEMS / ktəpælə /

CONSISTENCY

0 1 0 1

2 syll 3–4 syll

430

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll


/spgtə/

CONSISTENCY 2 syll 3–4 syll 0 1

0 1 0 1 0 1

/ ælfɒnt/

0 1

0 1 0 1 0 1

/ krɒvn /

0 1

0 1 0 1 0 1

/ krkədaυl/

0 1

0 1 0 1 0 1

/ æmbræl/

0 1

0 1 0 1 0 1

/ hləkæptə/

0 1

0 1 0 1 0 1

/ kɒŋgrɑ/

0 1

0 1 0 1 0 1

/ tυləvæn/

0 1 0 1

0 1

Appendix G.5

431

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


/ hæspətl/


0 1 0 1 0 1

/ tɒləfan/

0 1

0 1 0 1 0 1

/ pεrəʃit/

0 1

0 1 0 1 0 1

/ bætəfləυ/

0 1

0 1 0 1 0 1

/ kεmpjaυt/

0 1

0 1 0 1 0 1

/ rəυndəbat/

0 1

0 1 0 1 0 1

/ hɔdræs/

0 1

0 1 0 1 0 1

/ ɒrəpləυn/

0 1 0 1

0 1

432

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


/ pdɔmz/


0 1 0 1 0 1

/ hmbɑ/

0 1

0 1 0 1 0 1

/ densɑ/

0 1

0 1 0 1 0 1

0 1

REAL WORD REPETITION brush

0 1 0 1 0 1

sponge

0 1

0 1 0 1 0 1

glove

0 1

0 1 0 1 0 1

duck

0 1

0 1 0 1 0 1

leaf

0 1 0 1

0 1

Appendix G.5

433

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll 0 1

sock

CONSISTENCY


0 1 0 1 0 1

cat

0 1

0 1 0 1 0 1

book

0 1

0 1 0 1 0 1

torch

0 1

0 1 0 1 0 1

mouse

0 1

0 1 0 1 0 1

knife

0 1

0 1 0 1 0 1

snake

0 1

0 1 0 1 0 1

train

0 1 0 1

0 1

2 syll 3–4 syll

434

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll


0 1 van

CONSISTENCY

0 1

0 1 0 1 0 1

watch

0 1

0 1 0 1 0 1

plate

0 1

0 1 0 1 0 1

roof

0 1

0 1 0 1 0 1

fish

0 1

0 1 0 1 0 1

chair

0 1

0 1 0 1 0 1

thumb

0 1

0 1 0 1 0 1

0 1

NAMING STIMULI sandwich

0 1

2 syll 3–4 syll

Appendix G.5

435

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 toilet

0 1

0 1 0 1 0 1

money

0 1

0 1 0 1 0 1

feather

0 1

0 1 0 1 0 1

yellow

0 1

0 1 0 1 0 1

kitchen

0 1

0 1 0 1 0 1

ladder

0 1

0 1 0 1 0 1

flower

0 1

0 1 0 1 0 1

dustbin

0 1

0 1

436

Appendix G.5

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

CONSISTENCY


2 syll 3–4 syll

0 1 0 1 jelly

0 1

0 1 0 1 0 1

tractor

0 1

0 1 0 1 0 1

fishing

0 1

0 1 0 1 0 1

biscuit

0 1

0 1 0 1 0 1

scooter

0 1

0 1 0 1 0 1

parrot

0 1

0 1 0 1 0 1

seesaw

0 1

0 1 0 1 0 1

slipper

0 1 0 1

0 1

Appendix G.5

437

SCORE STIMULI

CHILD’S RESPONSE

ACCURACY 1 syll

sausage

CONSISTENCY


2 syll 3–4 syll

0 1

0 1

0 1 0 1 0 1

guitar

0 1

0 1 0 1 0 1

spider

0 1

0 1 0 1 0 1

TOTAL ACCURACY

/60


0 1 XXX

XXX XXX

438

Appendix G.5

SUMMARY SHEET 1 syll

2 syll

3–4 syll

TOTAL %

ACCURACY: Non-word Repetition

/60

/60

/60

/180

Word Repetition

/60

/60

/60

/180

Naming

/60

/60

/60

/180

TOTAL

/180

/180

/180

/20

/20

/20

/60

Word Repetition

/20

/20

/20

/60

Naming

/20

/20

/20

/60

TOTAL

/60

/60

/60

% CONSISTENCY: Non-word Repetition

%

Appendix H.1 Questionnaire 1: Developmental History

I. BIRTH AND GENERAL HEALTH DETAILS 1. Child’s name: 2. Child’s date of birth: 3. Please list your other children (if any) by age and date of birth: Children

Age

Date of birth

4. Was your child born prematurely? 5. Were there any complications associated with the birth? 6. Did your child have any feeding difficulties? 7. Does your child suffer from any of the following: Allergies Fits Asthma Frequent coughs and colds Ear infections Catarrh

YES/NO YES/NO YES/NO YES/NO YES/NO YES/NO

439

YES/NO YES/NO YES/NO

440

Appendix H.1

II. DEVELOPMENT SINCE BIRTH Physical development 8. Approximately how old was your child when s/he began to walk? 9. Have you ever been worried about your child’s physical development? YES/NO If Yes: 9a. What caused this concern?

9b. Was any help/treatment sought for this? If Yes: 9c. Was help/treatment given? 9d. Are there still problems?


Hearing 10. Have you ever been worried about your child’s hearing?YES/NO If Yes: 10a. What caused this concern?

10b. Was any help/treatment sought for this? If Yes: 10c. Was help/treatment given? 10d. Are there still problems?


Speech and Language 11. Approximately how old was your child when s/he began to talk? 12. Have you ever been worried about your child’s speech or language development? If Yes: 12a. What caused this concern?

If Yes:

12b. Was any help/treatment sought for this? 12c. Was help/treatment given? 12d. Are there still problems?

YES/NO


Appendix H.1

441

Vision 13. Have you ever been worried about your child’s vision? YES/NO If Yes: 13a. What caused this concern?

13b. Was any help sought for this? If Yes: 13c. Was help/treatment given? 13d. Are there still problems?


Reading and Spelling 14. Have you been worried about your child’s reading? If Yes: 14a. What caused this concern?

YES/NO

14b. Was any help sought for this? If Yes: 14c. Was help given? 14d. Are there still problems? 15. Have you been worried about your child’s spelling? If Yes: 15a. What caused this concern?

YES/NO YES/NO YES/NO YES/NO

15b. Was any help sought for this? If Yes: 15c. Was help/treatment given? 15d. Are there still problems?


442

Appendix H.1

III. FINAL SECTION Please use this space if there is any more information you would like us to have.

Name:___________________________ Signed:__________________________ Date:____________________________

THANK YOU VERY MUCH FOR COMPLETING THIS QUESTIONNAIRE

Appendix H.2 Questionnaire 2: Family Information

PARENTAL QUESTIONNAIRE: A SEPARATE QUESTIONNAIRE SHOULD BE COMPLETED BY EACH PARENT

I. GENERAL DETAILS 1. Name: 2. Child’s name:

II. EDUCATIONAL DETAILS 3. 4. 5. 6.

Number of CSE, GCSE or GCE ‘O’ level passes: Number of GCE ‘A’ level passes: Qualifications gained in further or higher education: Any other qualifications:

III. OCCUPATIONAL DETAILS Please give your present occupation. (If you are unemployed or are engaged in full-time child care, please give your previous occupation.) 7. Post/Title:

443

444

Appendix H.2

IV. SELF AND FAMILY 8. Do you or any of the following members of your family have a history of reading/spelling difficulties? You Your parents Your grandparents Your brothers/sisters


9. Do you or any of the following members of your family have a history of speech difficulties? You Your parents Your grandparents Your brothers/sisters


10. Have you ever attended a speech therapy clinic? 11. Have you ever had a hearing loss?

YES/NO YES/NO

V. FINAL SECTION Please use this space if there is any more information you would like us to have.

Name:___________________________ Signed:__________________________ Date:____________________________


Appendix H.3 Questionnaire 3: Educational History

Child’s name:

Year Group:

1. Name and level of reading scheme: 2. Does the child have any difficulties with reading or spelling?

YES/NO

3. Does the child have any difficulties with any other subjects?

YES/NO

4. Is the child receiving any extra help?

YES/NO

If Yes: 4a. How much extra help a week? 4b. In what format (group, individual)? 4c. Who gives the extra help? 4d. Where does the help take place (class, out of class, home)? 5. 6. 7. 8. 9. 10.

How much phonics work does the child receive each week? Is the child statemented? YES/NO Is the child in School Action? YES/NO Does the child have an IEP? YES/NO Is the child in School Action Plus? YES/NO Has the school requested Statutory Assessment? YES/NO

445

446

Appendix H.3

11. SATS results (if appropriate) or other Reading Comprehension Writing Spelling Maths Please use this space for any comments you would like to make:

Signed:__________________________ Name:___________________________ Date:____________________________


Appendix H.4 Questionnaire 4: Speech and Language Therapy

Child’s name: Child’s date of birth: * Please report on the time period from 1st referral to Speech and Language Therapy (SLT) up until present time or provide an alternative starting date for the time period from which information is provided, e.g. when this child came onto your case load. Alternative starting date for period reported: 1. 2. 3. 4. 5. 6. 7. 8.

Date of referral to SLT service*: Date of initial appointment: Number of sessions offered during the above period*: Number of sessions attended during the above period*: Number of individual sessions*: Number of group sessions*: Approximate length of session: Type of therapy (tick as appropriate), and give % estimate of time employed on each type: ❑ ❑ ❑ ❑ ❑ ❑ ❑

phonological (e.g. minimal pairs, metaphon) oral motor skills articulatory therapy phonological awareness expressive language receptive language play skills

447

___ ___ ___ ___ ___ ___ ___

% % % % % % %

448

Appendix H.4 ❑ ❑ ❑ ❑ ❑

listening skills ___ social skills ___ parent workshop ___ parent-child interaction ___ other (please specify)________________________________

% % % % %

9. Therapy setting (tick as appropriate): ❑ ❑ ❑ ❑

clinic school language unit other (please specify)_______________________

10. How often did liaison take place between SLT and nursery/school? (tick as appropriate): ❑ ❑ ❑ ❑ ❑

Once a year Once a term Occasional None to date Other (please specify)_______________________

11. What was the management of the child during the above period*? (tick as appropriate): ❑ ❑ ❑ ❑ ❑ ❑

regular for therapy on review on waiting list planned discharge discharged other (please specify)_______________________

12. Is the child (tick as appropriate): ❑ ❑

not statemented (i.e. no plans to begin process) statemented or in receipt of School Action or other intervention

13. Any other changes/observations of note (e.g. failed hearing test, change of school, hospitalisation):

Appendix H.4

449

14. Please specify other agencies involved (tick as appropriate): ❑ ❑ ❑ ❑ ❑ ❑ ❑

ENT OT Physio Ed. Psychologist Social Worker Child Guidance Other (please specify)_______________________

15. Other comments:

Signed:__________________________ Name of therapist:_______________ Base:____________________________ Date:____________________________


Index

ABX tasks, 47, 48 Non-word Task Full Version, 263–5 Short Form, 267–8 accents, warning about, 41 accuracy of speech, 153–72 Adams, C., 176 Advances in Speech-Language Pathology (journal), 218 aetiology, 3 aims of tasks auditory discrimination tasks ABX Non-words, 61 Complex Non-words, 57 Legal vs Illegal Words, 68 Same/Different, 50–1 auditory lexical discrimination tasks Mispronunciation Detection, 75 Mispronunciation Detection for Individual Children, 85 With and Without Pictures, 81 Words in Sentences, 88 connected speech tasks CSP Repetition Test, 139 Final Consonant Juncture Repetition Test, 145 Word-in-Sentence Repetition, 133 single-word speech Mispronunciation Self-Correction task, 122–3 Non-word Repetition Task, 116 Picture Naming Task #1, 101 Picture Naming Task #2, 106

Word Repetition Task, 110 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 157 Repetition and Spoken DDK Rates, 160 Speech Accuracy and Consistency, 168 analysis of tasks auditory discrimination tasks ABX Non-words, 62 Complex Non-words, 58 Legal vs Illegal Words, 69 Same/Different, 51–2 auditory lexical discrimination tasks Mispronunciation Detection, 76 Mispronunciation Detection for Individual Children, 86–7 With and Without Pictures, 82 Words in Sentences, 89 connected speech tasks CSP Repetition Test, 141–2 Final Consonant Juncture Repetition Test, 146–7 Word-in-Sentence Repetition, 134–8 single-word speech Mispronunciation Self-Correction task, 123 Non-word Repetition Task, 117–18 Picture Naming Task #1, 101–3

451

452

Index

analysis of tasks (Continued ) Picture Naming Task #2, 106–7 Word Repetition Task, 110–11, 112 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 157 Repetition and Spoken DDK Rates, 161 Speech Accuracy and Consistency, 169 anchor points in speech processing model, 6 apraxia of speech, 21, 57, 193 link to mutation in FOXP2 gene, 192 Aram, D., 191 Archibald, L., 155 articulation age, 49, 96, 155 assembly phase, developmental phase model, 20 assessment comparison types, 23–4 development of procedures, 210–11 interpreting data, 211–18 principles, 7–11 psycholinguistic approach, 225–7 routine procedures, gaps in, 27–8 tasks for each question, 243–5 Aston Index, 48 Auditory Discrimination and Attention Test (ADAT), 73, 80, 89 auditory discrimination tasks, 47–50 Task 1: Same/Different, 50–7, 253–8 Task 2: Complex Non-words, 57–61, 259–61 Task 3: ABX Non-words, 61–7, 263–5, 267–8 Task 4: Legal vs Illegal Words, 68–71, 269–70 auditory lexical discrimination tasks, 73–5 Task 1: Mispronunciation Detection full version, 271–8 short form, 279–82 Task 2: With and Without Pictures, 81–5 With Pictures, 287–8 Without Pictures, 283–5 Task 3: Mispronunciation Detection for Individual Children, 85–8, 291–3

Task 4: Words in Sentences, 88–93, 295–307 Baker, E., 1, 209 ‘between’ levels of processing, assessing, 11 Beving, B., 47 biased sentences, 90, 93 bilingual children, 49–50 birth order, 194, 196, 198, 199 Bishop, D., 176 bottom-up processing, 6 bottom-up tasks, 7 box-and-arrow model, 17–20, 249 Bradford, A., 98 Bridgeman, E., 32, 48, 50, 53, 55, 253, 257 Broomfield, J., 2 Bryan, A., 23, 127, 215 carers, training to identify risk signs, 225 Cassidy, B., 88 Chiat, S., 98, 98, 184 Children’s Test of Non-word Repetition, 98 chronological age, effect on task performance, 49 close juncture, 132–3 cluster sequences, 48, 50–7, 210 clustering of tasks, 7–8 Coffield, C., 156, 168, 173, 413 comparison of performance across different speech output tasks, 127–8 across different tasks, 98–100 inter-child, 24, 210–11 intra-child, 23, 210 Complex Non-words test, 57–61, 259–61 complex words, including in tasks, 210 connected speech, 131–2 Task 1: Word-in-Sentence Repetition, 133–8, 345–8 Task 2: CSP Repetition Test, 139–45, 349–53 Task 3: Final Consonant Juncture Repetition Test, 145–50, 355–60 consistency of speech repetition, 154–72

Index Constable, A., 74, 81, 210 coughs and colds, higher incidence in children with speech difficulties, 196 ‘critical age hypothesis’, 176, 184 cross-sectional study aims of project, 30 assessment tasks aims of, 32–3 participants, 30–1 procedure, 33 results, 33–4 stimuli, 31, 32 CSP (Connected Speech Processes) Repetition Test, 139–45 CVC (Consonant Vowel Consonant) discrimination, 48, 210 Darili, E., 68 delayed speech, 21–2, 154–5, 192, 193 description of tasks auditory discrimination tasks ABX Non-words, 61–2 Complex Non-words, 57 Legal vs Illegal Words, 68 Same/Different, 51 auditory lexical discrimination tasks Mispronunciation Detection, 75 Mispronunciation Detection for Individual Children, 85 With and Without Pictures, 81 Words in Sentences, 88 connected speech tasks CSP Repetition Test, 139 Final Consonant Juncture Repetition Test, 145 Word-in-Sentence Repetition, 134 single-word speech Mispronunciation Self-Correction task, 123 Non-word Repetition Task, 116 Picture Naming Task #1, 101 Picture Naming Task #2, 106 Word Repetition Task, 110 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 157 Repetition and Spoken DDK Rates, 160

453

Speech Accuracy and Consistency, 169 developmental history questionnaire, 194–8, 439–42 developmental information, 197 developmental phase model, 20–3, 155 diadochokinetic (DDK) rates, 153 Diagnostic Evaluation of Articulation and Phonology (DEAP), 96–7 diagnostic labels, diversity of, 2–3 discharge criteria, specifying, 223–4 Dodd, B., 2, 98, 155, 193, 194 Down’s syndrome adolescents, accuracy and consistency tests, 156 Dry, S., 68 dysarthria, 153, 156 dyslexia, 57 dyspraxia, 98–9, 153, 156 early-years workers, training of, 224–5 Ebbels, S., 23 Eblen, R. E., 47 Edinburgh Articulation Test (EAT), 35, 49, 96, 179 educational achievement, effect of speech difficulties on, 224 educational information questionnaire, 201–2, 445–6 Ekelman, B., 191 environment-gene interaction, 192 expressive vocabulary tests, 109 family information questionnaire, 198–9, 443–4 first words, delay in speaking, 196 Fletcher, J., 187 follow-up of children after therapy, importance of, 224 Fox, A., 193 FOXP2 gene and apraxia of speech, 192 fuzziness of phonological representations, 74, 210 genetic factors, 192, 193 Goldman-Fristoe-Woodcock Auditory Skills Battery, 80 Goodman, R., 200–1 Gourlay, S., 25

454

Index

grammatical representation, 6 Gruber, F., 4, 181 Habgood, M., 153 hearing difficulties, 47, 192, 195, 206 heritability of speech difficulties, 191–2 Hodson Assessment of Phonological Patterns (HAPP-3), 97 Howard, D., 23, 127, 193, 215 hyperactivity, co-occurring with speech difficulties, 201 hypotheses about performance, 130 illegal words see legal vs illegal words inconsistent speech output, sign of arrested development, 155 input tasks, definition of, 7 intelligibility and expressive language performance, 178 main criterion for discharging clients, 224 inter-child comparison, 24, 210–11 interpretation of assessment data activity, 216–18, 216–18 auditory tasks (no pictures), 213 responses on auditory and speech tasks, 214–16 speech tasks, 212–13 interpreting children’s responses, 42–3 intervention, 29 planning, 218–21 steps in psycholinguistic approach, 227 intonation, functions of, 132 intra-child comparison, 23, 210 Joffe, V., 25 Joint Research Advisory Committee (JRAC), 30 Kamil, M., 47 Kiatkowski, J., 4, 181, 192 language and speech output problems, link to speech input problems, 179 language development, link to persisting speech difficulties, 183

language difficulties additional to speech difficulties, 177–8 language skills vs speech as predictors of literacy development, 187 language-specific awareness, development of, 49 language therapy see speech and language therapy legal vs illegal words Auditory Discrimination Task #4, 68–71, 269–70 Leitao, S., 187 Leonard, L., 23 level of tasks auditory discrimination tasks ABX Non-words, 63 Complex Non-words, 58 Legal vs Illegal Words, 69 Same/Different, 53 auditory lexical discrimination tasks Mispronunciation Detection, 76 Mispronunciation Detection for Individual Children, 87 With and Without Pictures, 82 Words in Sentences, 89 connected speech tasks CSP Repetition Test, 142 Final Consonant Juncture Repetition Test, 147 Word-in-Sentence Repetition, 135 single-word speech Mispronunciation Self-Correction task, 124 Non-word Repetition Task, 118 Picture Naming Task #1, 103 Picture Naming Task #2, 107 Word Repetition Task, 111 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 157 Repetition and Spoken DDK Rates, 161 Speech Accuracy and Consistency, 169 Levin, A., 25 Lewis, B., 191 lexical prosodic features, 132 lexical representations, 5–6, 73 tests accessing, 109

Index lexicon (store of words), 5 literacy and monitoring after discharge from therapy, 224 measures of, 184 performance, predictors of, 187–8 Locke, A., 225 Locke, J., 47, 61, 75, 85–8 longitudinal studies, 3–4, 184 longitudinal study aims of project, 34–5 and critical age hypothesis, 175–6 assessing risk factors, 191–207 assessment tasks, 37–8, 39, 251 background information collected, 176 normative data, 38 outcome measures taken, 176 participants, 35–7 predicting outcomes, 221–2 results, 40 materials required for tasks auditory discrimination tasks ABX Non-words, 65, 263–5, 267–8 Complex Non-words, 58, 259–61 Legal vs Illegal Words, 70, 269–70 Same/Different, 54, 253–5, 257–8 auditory lexical discrimination tasks Mispronunciation Detection, 77, 271–8, 279–82 Mispronunciation Detection for Individual Children, 87, 291–3 With and Without Pictures, 83, 283–5, 287–9 Words in Sentences, 90, 295–307 connected speech tasks CSP Repetition Test, 142, 349–53 Final Consonant Juncture Repetition Test, 149, 355–60 Word-in-Sentence Repetition, 137, 345–8 single-word speech Mispronunciation Self-Correction task, 341–3 Non-word Repetition Task, 119, 331–3, 335–6, 337–9 Picture Naming Task #1, 103, 309–11, 313–14 Picture Naming Task #2, 107, 317–19

455

Word Repetition Task, 113, 321–9 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 158, 361–2 Repetition and Spoken DDK Rates, 161, 363–79, 381–95, 397–411 Speech Accuracy and Consistency, 170, 413–38 McGregor, K., 23 memory demands of tasks, 10–11, 47–8, 62, 94, 98, 134–5 Metaphon Resource Pack, 96 metaphonological phase, developmental phase model, 20 minimal pair picture tasks, 73, 88–93 mispronunciation detection tasks, 23, 31, 75–81, 85–8, 213 Mispronunciation Self-Correction task, 122–7, 341–3 models of speech processing box-and-arrow model, 17–20, 249 developmental phase model, 20–3 speech processing profile, 11–17 Morgan Barry, R., 73, 89 motor execution, box-and-arrow model, 19 motor planning, 19, 102, 141, 146 motor programs/programming, 6, 18–19, 99, 102–3, 116–17, 127, 146 naming tasks, 11, 95–109, 127–30, 155–6, 168–72, 212 Nathan, L., 110, 175, 196, 201, 222–3 National Hospital for Neurology and Neurosurgery, 30 neutral sentences, 93 Newton, C., 133, 140–1, 143, 150 non-verbal skills, link to persisting speech difficulties, 183–4 non-word repetition, 34, 98, 99, 156, 162 non-word tasks ABX task, 32, 61–7, 263–5, 267–8 complex non-words, 57–61, 259–61 legal vs illegal non-words, 68–71, 269–70 same/different, s-cluster sequences, words and non-words, 50–7, 253–8

456

Index

nonsense sentences, 90, 92, 93 normative data auditory discrimination tasks ABX Non-words, 67 Complex Non-words, 60 Legal vs Illegal Words, 71 Same/Different, 55–6 auditory lexical discrimination tasks Mispronunciation Detection, 79 Mispronunciation Detection for Individual Children, 87–8 With and Without Pictures, 84, 85 Words in Sentences, 92–3 connected speech tasks CSP Repetition Test, 143 Final Consonant Juncture Repetition Test, 150 Word-in-Sentence Repetition, 137–8 from typically developing children, 38 single-word speech Mispronunciation Self-Correction task, 126 Non-word Repetition Task, 121–2 Picture Naming Task #1, 104 Picture Naming Task #2, 109 Word Repetition Task, 114–16 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 159 Repetition and Spoken DDK Rates, 165–7, 168 Speech Accuracy and Consistency, 171–2 using to interpret children’s responses, 42 North Thames Regional Health Authority (NTRHA)., 34 Nuffield Centre Dyspraxia Programme, 157, 205 OME (otitis media with effusion) and speech delay, 192, 193 open juncture, 132–3 oral movements, 153–4 oro-motor tasks, 157–60 orthographic representation, 6 other assessment tasks for

auditory discrimination tasks ABX Non-words, 67 Complex Non-words, 61 Same/Different, 56–7 auditory lexical discrimination tasks Mispronunciation Detection, 80–1 Mispronunciation Detection for Individual Children, 88 With and Without Pictures, 84 Words in Sentences, 93 connected speech tasks CSP Repetition Test, 145 Word-in-Sentence Repetition, 138 single-word speech Mispronunciation Self-Correction task, 126–7 Non-word Repetition Task, 122 Picture Naming Task #1, 105 Picture Naming Task #2, 109 Word Repetition Task, 116 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 160 Repetition and Spoken DDK Rates, 168 Speech Accuracy and Consistency, 172 otitis media with effusion (OME), 192, 193 output processing skills, bias of tests towards, 27–9 output tasks definition of, 7 performance comparison, 127–9 parental history of speech difficulties, 199 parental lack of concern, 223–4 parental questionnaire, 198–9, 443–4 Pascoe, M., 11, 22 PEPS-C procedure, 132 performance on tasks factors affecting, 44 pre-school children, 47–8 perinatal problems, and deviant inconsistent phonological disorder, 194 peripheral auditory processing, box-and-arrow model, 17

Index persisting speech difficulties, prediction of children who have/have not resolved their speech difficulties, 184–5 children with and without speech difficulties, 176–7 children with resolved vs persisting speech difficulties, 179–84 presence of additional language difficulty, 77–8 relationship between severity of speech difficulties, language development and speech input skills, 179 severity of speech difficulty, 177 speech input processing skills, 178–9 persisting speech difficulties and reduced therapy with age, 204–5 pervasive speech processing problems, 222 phase model see developmental phase model phenological recognition, box-andarrow model, 18 phoneme awareness, 188 phonetic discrimination, box-and-arrow model, 18 phonological and short-term working memory, and non-word repetition, 98 phonological awareness importance of, 222 measures of, 184, 185–6 phonological representations box-and-arrow model, 18 testing, 73–4 physical development, 195 picture naming speech processing route for, 102 tasks, 11, 75–93, 95–7, 101–9 planning interventions activity, 219–21 Popple, J., 27 pre and perinatal problems, risk factors for speech difficulties, 193 pre-school children accuracy and consistency in repetition, 154

457

pre-school children, performance on auditory discrimination tasks, 47–8 prediction of persisting speech difficulties, 175–88 of speech and literacy outcomes, activity, 221–2 of speech outcome, 4, 185–6 predictors of literacy performance, speech vs language skills, 187–8 prelexical phase, developmental phase model, 20 prevalence rates, 2 procedures, task administration auditory discrimination tasks ABX Non-words, 65 Complex Non-words, 58–9 Legal vs Illegal Words, 70 Same/Different, 54–5 auditory lexical discrimination tasks Mispronunciation Detection, 78–9 Mispronunciation Detection for Individual Children, 87 With and Without Pictures, 83–4 Words in Sentences, 90, 92 connected speech tasks CSP Repetition Test, 142–3 Final Consonant Juncture Repetition Test, 149 Word-in-Sentence Repetition, 137 single-word speech Mispronunciation Self-Correction task, 124 Non-word Repetition Task, 119–20 Picture Naming Task #1, 104 Picture Naming Task #2, 108 Word Repetition Task, 113–14 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 158 Repetition and Spoken DDK Rates, 162–3 Speech Accuracy and Consistency, 170 processing of tasks, analysis of, 7–8 professionals working with pre-school children, training of, 224–5

458

Index

profiles of speech processing skills, 8–9 profiling sheets, 12–16 questions to ask, 9–10 prosodic features, 132 psycholinguistic analysis see analysis of tasks psycholinguistic approach aim of, 7 assessment principles, 7–11 basic structure of input/output, 5–6 intervention steps, 225–7 speech and literacy difficulties, 2–5 speech processing profiles, 11–17 theoretical models box-and-arrow model, 17–20, 249 developmental phase model, 20–3 using in practice and research, 23–5 psycholinguistic assessment see assessment psycholinguistic procedures, administering children who can be tested, 41 children who should not be tested, 42 see also procedures, task administration regional accents, 41 user training, 40–1 psycholinguistic tasks, using in practice, 209 psychosocial risk factors, 193 Psychosocial Strengths and Difficulties Questionnaire, 200–1 questionnaires assessing risk factors Developmental History, 194–8 Educational Information, 201–2, 445–6 Family Information, 198–9, 443–4 Psychosocial Strengths and Difficulties Questionnaire, 200–1 Speech and Language Therapy, 202–5, 447–9 questions for the speech processing profile, 9–10 tasks for each question, 243–5 questions to consider auditory discrimination tasks ABX Non-words, 67 Complex Non-words, 60–1 Legal vs Illegal Words, 71

Same/Different, 56 auditory lexical discrimination tasks Mispronunciation Detection, 79–80 Mispronunciation Detection for Individual Children, 88 With and Without Pictures, 84 Words in Sentences, 93 connected speech tasks CSP Repetition Test, 143–5 Final Consonant Juncture Repetition Test, 150 Word-in-Sentence Repetition, 138 single-word speech Mispronunciation Self-Correction task, 126 Non-word Repetition Task, 122 Picture Naming Task #1, 104–5 Picture Naming Task #2, 109 Word Repetition Task, 116 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 160 Repetition and Spoken DDK Rates, 167 Speech Accuracy and Consistency, 172 rate of speech repetition, 155–72 reading difficulties, family, 199 real word repetition, 34, 99, 130, 156, 162 Rees, R., 211 regional accents, warning about, 41 Renfrew Action Picture Test (RAPT), 178 repetition tasks, 97–8, 110–22, 153–6 research, training in, 225 ‘resolved speech difficulties’, defining, 180–1, 222–3 resolved vs persisting speech difficulties defining ‘resolved’, 180–1 language skills, 183 non-verbal skills, 183–4 phonological awareness and literacy measures, 184 speech input and output, 181–3 tasks differentiating the subgroups, 181, 182

Index respiratory health problems, 196 response rates, 155–6 risk factors assessing, 191–4 using questionnaires, 194–205 identifying signs, 225 Roy, P., 97 Rudegeair, R., 47 same/different tasks Auditory Discrimination Task #1, 50–7 reduced version, 253–5 shortest form, 257–8 cross-sectional study, 32 Schissel, R., 47 school learning, and hidden speech difficulties, 224 scoring guidelines, 315 scoring of tasks auditory discrimination tasks ABX Non-words, 66–7 Complex Non-words, 60 Legal vs Illegal Words, 70–1 Same/Different, 55 auditory lexical discrimination tasks Mispronunciation Detection, 79 Mispronunciation Detection for Individual Children, 87 With and Without Pictures, 84 Words in Sentences, 92 connected speech tasks CSP Repetition Test, 143 Final Consonant Juncture Repetition Test, 149 Word-in-Sentence Repetition, 137 single-word speech Mispronunciation Self-Correction task, 125–6 Non-word Repetition Task, 120–1 Picture Naming Task #1, 104 Picture Naming Task #2, 108 Word Repetition Task, 114 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 158–9 Repetition and Spoken DDK Rates, 163–5 Speech Accuracy and Consistency,

459

170–1 segmental juncture features, 132–3 self-correction of speech errors, 122–7, 210, 341–3 semantic representation, 6 sentence repetition tasks CSP Repetition Test, 139–45 Final Consonant Juncture Repetition Test, 145–50 Word-in-Sentence Repetition, 133–8 severe speech problems, therapy resources, 205 severity of speech difficulty, 177 short-term memory, 48, 62, 62, 63, 66, 98, 134–5 short-term normalisation, 4, 181 Shriberg, L., 4, 181, 192, 193 sibling studies, 191–2, 198, 199 Simpson, S., 222–3 single-word speech Mispronunciation Self-correction, 122–7, 341–3 Non-word Repetition task, 116–22, 337–9 picture naming, 95–7 Picture Naming Tasks, 101–9, 309–11, 313–14 repetition tasks, 97–8 task comparisons, 98–100 Word Repetition Tasks, 110–16, 321–9 Snowling, M., 32, 48, 50, 53, 55, 253, 257 Speake, J., 225 speech accuracy, 153–6 speech accuracy, rate and consistency Task 1: Oral Movements and Silent DDK Rates, 157–60, 361–2 Task 2: Repetition and Spoken DDK Rates non-words, 381–95 syllable sequences, 397–411 Task 3: Speech Accuracy and Consistency, 168–72, 413–38 speech and language therapy discharge from, 223–4 duration of, 204–5 questionnaire, 202–5, 447–9 Speech Assessment and Interactive Learning System (SAILS), 80 speech delay, subtypes of, 193

460

Index

speech input and output problems, link to persisting speech difficulties, 181–3 speech input processing skills, 178–9 speech output difficulties, link to speech input problems, 179 speech output processing, DDK (diadochokinetic) tasks, 153–4 speech output tasks, comparison of performance, 127–9 speech processing profiles aim of, 11–12 construction of, 12–17 see also level of tasks pre-psycholinguistic assessment, 28 questions for, 9–10 template, 247 speech production, connected speech, 131–2 speech production tasks, guidelines for scoring, 315 speech vs language skills as predictors of literacy outcome, 187–8 speech/non-speech discrimination, box-and-arrow model, 17 spelling problems, 86, 177, 180–1, 186, 187–88, 195, 198, 201 spoken language problems, defining, 2–3 Stackhouse, J., 7, 11, 22, 49, 57, 74, 81, 153, 155–6, 210 standardised assessments used in longitudinal study, 251 Statutory Attainment Tests (SATs), 177, 187, 201 stimuli appropriateness for younger children, 48 cross-sectional study, 31, 32 design of, 23 introducing more challenging, 38, 49, 74, 100, 223 stimuli for tasks auditory discrimination tasks ABX Non-words, 63–5, 66 Complex Non-words, 58 Legal vs Illegal Words, 69–70 Same/Different, 53–4 auditory lexical discrimination tasks Mispronunciation Detection, 76–7, 78

Mispronunciation Detection for Individual Children, 87 With and Without Pictures, 82–3 Words in Sentences, 89–90, 91 connected speech tasks CSP Repetition Test, 142 Final Consonant Juncture Repetition Test, 147–8 Word-in-Sentence Repetition, 135–7 single-word speech Mispronunciation Self-Correction task, 124 Non-word Repetition Task, 118–19 Picture Naming Task #1, 103 Picture Naming Task #2, 107 Word Repetition Task, 112–13 speech accuracy, rate and consistency tasks Oral Movements and Silent DDK Rates, 157–8 Repetition and Spoken DDK Rates, 161, 162 Speech Accuracy and Consistency, 170 strengths and weaknesses box-and-arrow model, 19, 20 identifying, 3, 10–11 inter-child perspective, 24 supra-lexical prosodic features, 132 surface speech difficulties, 211 syllable sequence discrimination/repetition, 49, 57, 58, 153, 154, 155 task measuring, 160–8, 397–411 systematic simplification phase, developmental phase model, 20 task comparisons, 98–101 tasks for each question in psycholinguistic assessment, 243–5 teachers identifying subtle spoken language problems, 224 support for, 225 terminology for speech difficulties, 2–3 Test of Word Finding, 96 theoretical speech processing models, 17–23 therapist lack of concern about child’s speech development, 223–4

Index therapy see speech and language therapy top-down processing, 6, 7 top-down tasks, 7 training in use of psycholinguistic procedures, 40–1 of parents and teachers, 224–5 twin studies, 191 typically developing children control group in longitudinal study, 35–6 for inter-child comparisons, 210–11 performance on DDK tasks, 153–4 use of tasks auditory discrimination tasks ABX Non-words, 62 Complex Non-words, 57–8 Legal vs Illegal Words, 68 Same/Different, 51 auditory lexical discrimination tasks Mispronunciation Detection, 75–6 Mispronunciation Detection for Individual Children, 85–6 With and Without Pictures, 81–2 Words in Sentences, 88–9 connected speech tasks CSP Repetition Test, 139–41 Final Consonant Juncture Repetition Test, 145–6 Word-in-Sentence Repetition, 134 single-word speech Mispronunciation Self-Correction task, 123 Non-word Repetition Task, 117 Picture Naming Task #1, 101 Picture Naming Task #2, 106 Word Repetition Task, 110 speech accuracy, rate and consistency tasks

461 Oral Movements and Silent DDK Rates, 157 Repetition and Spoken DDK Rates, 160–1 Speech Accuracy and Consistency, 169

Vance, M., 61, 75 Vargha-Khadem, F., 192 ‘verbal dyspraxia’, 48, 156 visual problems, 195, 196 Walker, J., 155 Waters, D., 12, 19 Weiner’s Phonological Process Analysis, 96 Wellington, W., 27 Wells, B., 7, 11, 22, 74, 81, 101, 133, 150, 210 Wepman’s Auditory Discrimination Test, 48 What do tests really test? questions, 7, 40 whole word phase, developmental phase model, 20, 21, 22–3, 155 Williams, N., 98 Williams, P., 153, 155–6 ‘within’ levels of processing, assessing, 10–11 word-finding difficulties, 74, 95–6 Word Finding Vocabulary Test, 96, 109 word-in-sentence tasks, 88–93, 92–3, 133–8, 345–8 word stress, lexical prosodic feature in English language, 132 words, lexical representation of, 6 working memory, 98 z scores, calculating, 42–3

Other Titles in this Series

Children’s Speech and Literacy Difficulties: A Psycholinguistic Framework, Book 1 Joy Stackhouse and Bill Wells Paperback ISBN: 1-86156-030-3 Children’s Speech and Literacy Difficulties: Identification and Intervention, Book 2 Edited by Joy Stackhouse and Bill Wells Paperback ISBN: 1-86156-131-8 Persisting Speech Difficulties in Children: Children’s Speech and Literacy Difficulties, Book 3 Michelle Pascoe, Joy Stackhouse and Bill Wells Paperback ISBN: 978-0-470-02744-8

Compendium of Auditory and Speech Tasks: Children's Speech and Literacy Difficulties, Book 4

Persisting speech difficulties in children

Speech Physiology, Speech Perception, and Acoustic Phonetics

Speech acoustics and phonetics

Speech and Language Processing

Meaning and Speech Acts

Speech and language processing

Speech Production and Perception

Speech and Writing

Speech synthesis and recognition

Language and Speech Processing

Speech and Theology

Alcohol and Speech

Alcohol and Speech

Speech And Reading

Speech Physiology, Speech Perception, and Acoustic Phonetics (Cambridge Studies in Speech Science and Communication)

Speech Recognition

Super Star Speech, Speech Therapy Made Simple

Gesture and Speech (October Books)

Speech Processing and Soft Computing

Construal (Language, Speech, and Communication)

Robust speech recognition and understanding

Clear Speech: Practical Speech Correction and Voice Improvement

Freedom of Speech

The Ark of Speech

Figures of Speech

Perturbed Speech

Speech Acts and Conversational Interaction

Robust speech recognition and understanding

Applied speech and audio processing

Compendium of Auditory and Speech Tasks: Children's Speech and Literacy Difficulties, Book 4

Persisting speech difficulties in children

Speech Physiology, Speech Perception, and Acoustic Phonetics

Speech acoustics and phonetics

Speech and Language Processing

Meaning and Speech Acts

Speech and language processing

Speech Production and Perception

Speech and Writing

Speech synthesis and recognition

Language and Speech Processing

Speech and Theology

Alcohol and Speech

Alcohol and Speech

Speech And Reading

Speech Physiology, Speech Perception, and Acoustic Phonetics (Cambridge Studies in Speech Science and Communication)

Speech Recognition

Super Star Speech, Speech Therapy Made Simple

Gesture and Speech (October Books)

Speech Processing and Soft Computing

Construal (Language, Speech, and Communication)

Robust speech recognition and understanding

Clear Speech: Practical Speech Correction and Voice Improvement

Freedom of Speech

The Ark of Speech

Figures of Speech

Perturbed Speech

Speech Acts and Conversational Interaction

Robust speech recognition and understanding

Applied speech and audio processing

Recommend Documents