Cortical Areas: Unity and Diversity (Conceptual Advances in Brain Research)

Cortical Areas: Unity and Diversity © 2002 Taylor & Francis Conceptual Advances in Brain Research A series of books f...

Author: Almut Schuez | Robert Miller

20 downloads 894 Views 32MB 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

Cortical Areas: Unity and Diversity

© 2002 Taylor & Francis

Conceptual Advances in Brain Research A series of books focusing on brain dynamics and information processing systems of the brain. Edited by Robert Miller, Otago Centre for Theoretical Studies in Psychiatry and Neuroscience, New Zealand (Editor-in-chief), Günther Palm, University of Ulm, Germany and Gordon Shaw, University of California at Irvine, USA.

Volume 1 Brain Dynamics and the Striatal Complex edited by R. Miller and J.R. Wickens Volume 2 Complex Brain Functions: Conceptual Advances in Russian Neuroscience edited by R. Miller, A.M. Ivanitsky and P.M. Balaban Volume 3 Time and the Brain edited by R. Miller Volume 4 Sex Differences in Lateralization in the Animal Brain by V.L. Bianki and E.B. Filippova Volume 5 Cortical Areas: Unity and Diversity edited by A. Schüz and R. Miller

This book is part of a series. The publisher will accept continuation orders which may be cancelled at any time and which provide for automatic billing and shipping of each title in the series upon publication. Please write for details.


Cortical Areas: Unity and Diversity

Edited by

Almut Schüz Max-Planck-Institute for Biological Cybernetics, Tübingen, Germany

and Robert Miller University of Otago, Dunedin, New Zealand

London and New York © 2002 Taylor & Francis

First published 2002 by Taylor & Francis 11 New Fetter Lane, London EC4P 4EE Simultaneously published in the USA and Canada by Taylor & Francis Inc, 29 West 35th Street, New York, NY 10001 Taylor & Francis is an imprint of the Taylor & Francis Group © 2002 Taylor & Francis Typeset in Times by Integra Software Services Pvt. Ltd, Pondicherry, India Printed and bound in Great Britain by TJ International Ltd, Padstow, Cornwall All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publishers. Every effort has been made to ensure that the advice and information in this book is true and accurate at the time of going to press. However, neither the publisher nor the authors can accept any legal responsibility or liability for any errors or omissions that may be made. In the case of drug administration, any medical procedure or the use of technical equipment mentioned within this book, you are strongly advised to consult the manufacturer’s guidelines.

British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library Library of Congress Cataloging in Publication Data A catalog record for this book has been requested ISBN 0–415–27723–X


CONTENTS

Preface

vii

List of Contributors

ix

1 Introduction: Homogeneity and Heterogeneity of Cortical Structure: A Theme and its Variations Almut Schüz

1

Part I THE EMPIRICAL STATUS OF CORTICAL MAPS 2 Cyto- and Myeloarchitectonics: Their Relationship and Possible Functional Significance Bernhard Hellwig 3 Architectonic Mapping of the Human Cerebral Cortex Katrin Amunts, Axel Schleicher and Karl Zilles 4 Topographical Variability of Cytoarchitectonic Areas Jörg Rademacher 5 Mapping of Human Brain Function by Neuroimaging Methods Rüdiger J. Seitz

15 29 53 79

Part II CORTICAL AREAS: CORRELATION WITH CONNECTIVITY 6 Regional Dendritic Variation in Primate Cortical Pyramidal Cells Bob Jacobs and Arnold B. Scheibel 7 Intrinsic Connections in Mammalian Cerebral Cortex Jonathan B. Levitt and Jennifer S. Lund 8 Thalamic Systems and the Diversity of Cortical Areas Catherine G. Cusick 9 Cortical Areas and Patterns of Cortico-Cortical Connections Jon H. Kaas v © 2002 Taylor & Francis

111 133 155 179

vi

Contents

Part III CONSTANCY AND VARIATION ACROSS SPECIES 10 The Cerebral Cortex of Mammals: Diversity within Unity Facundo Valverde, Juan A. De Carlos and Laura López-Mascaraque 11 Laminar Continuity between Neo- and Meso-Cortex: The Hypothesis of the Added Laminae in the Neocortex Robert Miller and Rupa Maitra

195

219

Part IV FUNCTIONAL EQUIVALENCE BETWEEN AREAS 12 Cross-Modal Plasticity as a Tool for Understanding the Ontogeny and Phylogeny of Cerebral Cortex Sarah L. Pallas 13 Do Primary Sensory Areas Play Analogous Roles in Different Sensory Modalities? Hubert R. Dinse and Christoph E. Schreiner 14 Plastic-Adaptive Properties of Cortical Areas Hubert R. Dinse and Gerd Boehmer

245

273 311

Part V MORPHOLOGICAL SUBSTRATES OF SEGREGATION AND INTEGRATION 15 Connectional Organisation and Function in the Macaque Cerebral Cortex Malcolm P. Young 16 The Human Cortical White Matter: Quantitative Aspects of Cortico-Cortical Long-Range Connectivity Almut Schüz and Valentino Braitenberg 17 Fundamentals of Association Cortex Stewart Shipp 18 Wheels within Wheels: Circuits for Integration of Neural Assemblies on Small and Large Scales Robert Miller

351

19 Discussion Section Robert Miller and Almut Schüz

459


377 387

423

PREFACE

Since the time when Bell and Magendie first showed the different functions of dorsal and ventral roots of the spinal cord, idea that different functions can be identified with different locations in the central nervous system has been central to attempts to understand the brain. The possibility that different psychological functions might in some way “reside” in different locations of the cerebral cortex was also an attractive idea, even when scientific study of the cerebral cortex was in its earliest infancy, as shown by the popularity of phrenology in the first half of the nineteenth century. This possibility came to have a firmer empirical basis in the latter half of that century, as a result of the studies of neurologists such as Broca, Wernicke and others. Development of ideas of cortical localization of function was given further impetus by results of cortical stimulation experiments, and, in the early twentieth century, from the study of cortical cytoarchitectonics. Nowadays, a localizationist view of the cortex is also favoured by the widespread use of functional imaging techniques. Throughout this long history, an alternative perspective has been advocated periodically, placing emphasis of the fact that many psychological functions appear not to be localized in specific cortical regions, or if they are associated with particular cortical areas, these areas are multiple, and distributed, rather than single and discrete. In the lesion studies of memory conducted by Lashley it was even concluded that functional loss depends more on the size of the lesions, rather than its exact location. A modern expression of this perspective comes from some of those using functional imaging methods, who are also concerned with widely distributed functions, and document networks of several cortical areas activated together when particular psychological functions are employed. Modern morphological work on the cerebral cortex, to which one of us has contributed also fits into this alternative tradition, cortical connectivity being described and analysed in terms of broad statistical constraints which might generalize across the whole neocortical mantle. These two perspectives might seem antithetical, but this is appearance rather than reality. It is not a contradiction to believe that some functions have a strict association with particular cortical areas, while others are based on more widely-distributed cortical regions. Which of these two perspectives emerges as prominent in an experiment depends on the way the experimenters frame their questions. In the chapters below, many aspects of this complex topic are explored. These include the actual evidence that the cortex can be subdivided into morphologically different areas, the correlation between such parcellation and patterns of connectivity of various sorts, the degree to which there is nevertheless an underlying uniformity to the cortex, generalizing across areas and between species, the functional equivalence of different areas, as well as the large-scale patterning of cortical functioning, and the overall integration of cortical vii © 2002 Taylor & Francis

viii

Preface

functions by interplay with other forebrain structures. all these issues have been discussed many times in the past. However, we believe it is timely to revisit them, and thus to put some of these long-standing debates in the context of modern evidence about the structure and function of the cerebral cortex. We would like to express thanks to a few people. Claudia Holt was very helpful in the handling of electronic material and Nicola Arndt in technical assistance with the manuscripts. In particular, we thank Valentino Braitenberg for valuable advice and discussions. The planning of this book was in part done at the Institute for Advanced Studies in Delmenhorst, Germany. R. Miller expresses his thanks to Professor Gareth Jones, of Otago University, and to the Schizophrenia Fellowship of New Zealand for continuing support, and to the Max Planck Institute for Biological Cybernetics, for support during visits to Tübingen, during the planning and development of this book. R. Miller, Dunedin A. Schüz, Tübingen April 2001


CONTRIBUTORS

Hubert R. Dinse Institute for Neuroinformatics Dept of Theoretical Biology Group Experimental Neurobiology Ruhr-University Bochum ND 04 D-44780 Bochum Germany

Katrin Amunts Institut für Medizin Forschungszentrum Jülich GmbH 52425 Jülich Germany Gerd Boehmer Institute of Physiology and Pathophysiology Gutenberg-University 55099 Mainz Germany

Jon H. Kaas Vanderbilt University Dept of Psychology 111 21st street Avenue South 301 Wilson Hall Nashville TN 37240 USA

Valentino Braitenberg Max-Planck-Institut für biologische Kybernetik Spemannstr. 38 72076 Tübingen Germany

Bernhard Hellwig Neurologische Universitätsklinik Neurozentrum Breisacher Str. 64 79106 Freiburg Germany

Catherine G. Cusick Dept of Structural and Cellular Biology and Neurosciences Programme Tulane University School of Medicine 1430 Tulane Avenue New Orleans Louisiana USA 70112

Bob Jacobs Laboratory of Quantitative Neuromorphology Dept of Psychology The Colorado College 14 East Cache La Poudre Colorado Springs CO 80903 USA

Juan A. De Carlos Instituto Cajal (CSIC) Avenida del Doctor Arce 37 28002 Madrid Spain ix © 2002 Taylor & Francis

x

Laura López-Mascaraque Instituto Cajal (CSIC) Avenida del Doctor Arce 37 28002 Madrid Spain Jennifer S. Lund Department of Ophthalmology Moran Eye Center University of Utah 50 North Medical Drive Salt Lake City UT 84132 USA Jonathan B. Levitt Dept of Biology City College of the City University of New York 138th Street & Convent Avenue New York NY10031 USA Rupa Maitra Department of Anatomic Pathology Wellington Hospital Wellington New Zealand Robert Miller Otago Centre for Theoretical Studies in Psychiatry & Neuroscience Dept of Anatomy and Structural Biology School of Medical science University of Otago PO Box 913 Dunedin New Zealand Sarah L. Pallas Dept of Biology Georgia State University PO Box 4010 Atlanta GA 30302 USA


Contributors

Jörg Rademacher Neurologische Klinik Heinrich-Heine-Universität Düsseldorf Moorenstrasse 5 40225 Düsseldorf Germany Arnold B. Scheibel Department of Neurobiology Brain Research Institute University of California Los Angeles CA 90024-1769 Axel Schleicher Institute of Neuroanatomy Heinrich Heine University 40225 Düsseldorf Germany Christoph E. Schreiner Coleman Laboratory W.M. Keck Center for Integrative Neuroscience Sloan Center for Theoretical Neurobiology University of California San Francisco San Francisco USA Almut Schüz Max-Planck-Institut für biologische Kybernetik Spemannstr. 38 72076 Tübingen Germany Rüdiger J. Seitz Dept of Neurology University Hospital Düsseldorf Moorenstrasse 5 40225 Düsseldorf Germany

Contributors

xi

Stewart Shipp Wellcome Dept of Cognitive Neurology University College Gower Street WC1E 6BT London England

Malcolm P. Young Neural Systems Group Dept of Psychology Claremont Place NE1 7RU Newcastle upon Tyne England

Facundo Valverde Instituto Cajal (CSIC) Avenida del Doctor Arce 37 28002 Madrid Spain

Karl Zilles C. and O. Vogt Institute of Brain Research Heinrich Heine University 40225 Düsseldorf Germany


Part I THE EMPIRICAL STATUS OF CORTICAL MAPS


2 Cyto- and Myeloarchitectonics: Their Relationship and Possible Functional Significance Bernhard Hellwig Neurologische Universitätsklinik, Neurozentrum, Breisacher Str. 64, 79106 Freiburg, Germany Tel: ++49 761 270 5001; FAX: ++49 761 270 5390 e-mail: [email protected]

In the human cerebral cortex, a number of cortical areas can be distinguished by anatomical methods. Two classical types of cortical parcellation have been described, based on cyto- and myeloarchitectonics. In cytoarchitectonics, the definition of areas relies on variations in the sizes and packing densities of cell bodies. Myeloarchitectonic parcellation is based on the layering, the distribution and the amount of intracortical myelinated fibres. It is shown here that cyto- and myeloarchitectonics are closely related. Two simple assumptions are sufficient to transform quantitative cytoarchitectural data into the corresponding myelin picture. The rules linking cyto- and myeloarchitectonics seem to be essentially uniform throughout the neocortex. It is also well known that characteristic functional specializations can be attributed to cortical areas. However, beyond the localization of function, the functional significance of areal variability in the cortex is largely unclear. For instance, it remains to be clarified why certain areal adaptations of the basic cortical network seem to be particularly appropriate for the execution of specific tasks. It is argued that this issue will only be understood when the wiring schemes of each area are known. Since it is difficult to infer connectivity patterns from cyto- and myeloarchitectonics, their significance for a functional interpretation of cortical anatomy seems to be limited. The paper suggests, however, possible strategies that may allow one to describe cortical architectonics in terms of connectivity. KEYWORDS: areas, connectivity, cytoarchitectonics, human cerebral cortex, myelin, myeloarchitectonics

1. INTRODUCTION It has been known for a long time, i.e. since the discovery of the stripe of Gennari in the primary visual cortex (Gennari, 1782), that the cerebral cortex is not uniform. A number of histological methods allow one to distinguish cortical areas which are defined by characteristic variations of the basic cortical architecture. Interestingly, this anatomical parcellation of the cortex is not merely descriptive, but is somehow related to cortical function. Different types of information (visual, auditory, motor, etc.) are processed in different cortical areas. As yet, this relation between structure and function has been elucidated mainly in just one respect: Certain functions can be localized in certain areas. Reaching this conclusion is an important achievement, useful, for example, for a clinical neurologist who can associate symptoms in a patient with lesions visible in a CT scan. However, 15 © 2002 Taylor & Francis

Bernhard Hellwig

16

localization of function is not the whole story. Knowing where a certain type of information is processed does not explain how this is done. The functional significance of areal variations in the cortex will not be understood until the mechanisms of information processing as well as its localization can be related to cortical anatomy. For instance, it would be interesting to know why a piece of cortex that is involved in motor control looks like the motor area, and not like the primary visual cortex. The present paper will not be able to solve this problem. However, it will consider two classical approaches in cortical parcellation, cyto- and myeloarchitectonics, and discuss whether a functional interpretation is possible beyond the mere localization of function.

1.1. Cytoarchitectonics In cytoarchitectonics, cortical areas are defined on the basis of cell body stains such as the Nissl stain. Cortical parcellation relies on variations in the sizes and packing densities of neurones leading to characteristic patterns of layering (Figure 2.1). The most prominent maps of the human cerebral cortex worked out on the basis of cytoarchitectural observa1o 1a 1b 1c

I II III

1

III

2

III

3

2 1

3

2

3

IV Va Vb

3

3

4 5a

5b 6aα

VIa 6aβ VIbα

6bα

VIbβ

6bβ

Figure 2.1. Schematic drawing of a piece of association cortex: cytoarchitectonics (left) and myeloarchitectonics (right). The stripes of Baillarger correspond to the horizontal bands of myelinated fibres in layers 4 and 5b. From Vogt and Vogt (1919).


Cyto- and Myeloarchitectonics

17

Figure 2.2. Brodmann’s map (1909) of the human cerebral cortex (lateral view).

tions are those by Brodmann (1909) (Figure 2.2) and von Economo and Koskinas (1925) (see Figure 3.1(B) in chapter by Amunts et al., this volume). It is not the aim of this paper to present all the anatomical details of each area. These can be found in the monographs by Brodmann and von Economo mentioned above as well as in a more recent treatise by Braak (1980) which combines cyto- and myeloarchitectural observations with studies on the pigmentoarchitectonics of the human cerebral cortex. However, in order to give a basic idea of the cytoarchitectural organization of the neocortex, some general principles should be mentioned. Following a suggestion by von Economo and Koskinas (1925) the different cortical areas can be collected into larger groups. Most areas, in particular the association areas, show the typical six-layered cortex schematically illustrated in Figure 2.1. They are referred to as homotypical. Areas in which six layers cannot be clearly discerned are called heterotypical. They come in two forms. First, there is the agranular cortex in which layers 2 and 4 with small, densely packed neurones are not well developed. Examples for the agranular cortex are the Brodmann areas 4 and 6, i.e. the motor and premotor areas (Figure 2.2). The second type of heterotypical cortex is the granular cortex, which is characterized by strongly developed layers 2 and 4 with many densely packed, small neurones. This type of cortex is mainly found in the primary sensory cortices, e.g. in the Brodmann areas 17, 41 and 3 (primary visual, auditory and somatosensory cortex).

1.2. Myeloarchitectonics The myeloarchitectonics of the human cerebral cortex, based on the layering, the distribution and the amount of intracortical myelinated fibres, has been described in detail by Vogt and his co-workers (e.g. Vogt, 1910, 1911; Vogt and Vogt, 1919; Strasburger, 1937;


18

Bernhard Hellwig

Hopf, 1954; Batsch, 1956). Myelin preparations show three types of intracortical fibres (Figure 2.1): (1) radial fibres (vertical to the cortical surface), (2) oblique fibres, and (3) horizontal fibres (parallel to the cortical surface). The horizontal fibres are particularly useful for cortical parcellation. In most areas they form two conspicuous horizontal bands, the so-called stripes of Baillarger (Baillarger, 1840), which are usually located in layers 4 and 5b respectively (Figure 2.1). The stripes of Baillarger vary from area to area. Again, it is not the aim of this paper to describe the areal variability of myeloarchitectonic patterns in detail. The reader is referred to the papers by Vogt and his co-workers mentioned above as well as to the treatise by Braak (1980). However, some general remarks can be made. The homotypical cortex, as it was defined for cytoarchitectonics, usually exhibits both stripes of Baillarger. In some regions, such as the frontal and temporal pole, or areas located medially in the interhemispheric cleft, only the outer stripe of Baillarger may be discernible. In the heterotypical agranular cortex the stripes of Baillarger are concealed in a dense feltwork of fibres, either completely as in the primary motor cortex or partially as in the premotor cortex where only the outer stripe of Baillarger is visible. In the heterotypical granular cortex two myelin patterns can be distinguished. On the one hand, there is the primary visual cortex with its conspicuous band of horizontal myelinated fibres in layer 4b, the so-called stripe of Gennari. On the other hand, in the primary somatosensory cortex or the primary auditory cortex both stripes of Baillarger are present, the inner one being distinctly more prominent than the outer one. It is also interesting to consider the areal variability of the total amount of myelin in the cortex. The degree of myelination diminishes with increasing distance from the primary areas. It is particularly low in the region of the frontal and temporal pole as well as in areas located medially in the interhemispheric cleft. 1.3. Cyto- and Myeloarchitectonics: Different Aspects of the Same Underlying Cortical Network? Before discussing possible functional implications of the areal variability described by cyto- and myeloarchitectonics, it seems worthwhile to consider whether there is a relation between patterns of cell bodies and patterns of myelinated fibres. Finding such a relation may elucidate aspects of the underlying cortical network. According to Brodmann (1909), maps of the human cerebral cortex based on either cyto- or myeloarchitectonics are essentially identical. This is corroborated by Sanides’ monograph (1962) on the frontal lobe of the human brain. Here, the investigation of a series of sections alternatively stained by a cell body and a myelin stain yielded only one map of areal diversity in the frontal cortex. Thus, there seems to be a close relationship between cyto- and myeloarchitectonics in the sense that they obviously reflect different aspects of the same underlying cortical network. However, what is the nature of this relation? Inspection of Figure 2.1 reveals that an answer to this question is by no means obvious. While the outer stripe of Baillarger, situated in layer 4, corresponds to densely packed, small cell bodies, the inner stripe of Baillarger, located in layer 5b, coincides with less densely packed, large cell bodies. Braitenberg (1962, 1974) put forward a hypothesis as to how cyto- and myeloarchitectonics might be related. He suggested that horizontal intracortical myelinated fibres, i.e. those fibres forming the stripes of Baillarger, correspond mainly to local axonal ramifications of pyramidal neurons, the most frequent cell type in the cerebral cortex (Braitenberg,



19

Figure 2.3. Camera lucida drawing of a Golgi-stained pyramidal cell in the cerebral cortex of the mouse. A number of horizontally directed axon collaterals originate slightly below the cell body. Bar, 50 µm.

1978; Braitenberg and Schüz, 1998). The bulk of horizontal axon collaterals of pyramidal cells leave the descending main axon 200 to 300 µm below the cell body (Figure 2.3) (cf. Cajal, 1911; Gilbert and Wiesel, 1979, 1983; Landry et al., 1980; Martin and Whitteridge, 1984; DeFelipe et al., 1986; Schwark and Jones, 1989). The pyramidal cells which in the majority of areas are most conspicuous in layers 3 and 5 would thus produce two maxima of horizontal fibres. These maxima, shifted downwards relative to layers 3 and 5 by 200 to 300 µm, could account for the two stripes of Baillarger. The assumption that horizontal myelinated fibres in the cortex consist mainly of local axonal ramifications of pyramidal cells (and not of thalamic or cortico-cortical afferent fibres) is supported by degeneration and tracer studies (Le Gros Clark and Sunderland, 1939; Fisken et al., 1975; Creutzfeldt et al., 1977; Gatter and Powell, 1978; Colonnier and Sas, 1978; Levitt et al., 1993). Starting out from Braitenberg’s hypothesis, Hellwig (1993) showed in a computational study that, provided quantitative data on the cell body picture of a certain area are given, two simple assumptions are sufficient to predict correctly the corresponding myelin picture. Part of this work is reviewed in the following section.


Bernhard Hellwig

20

2. SIMPLE RULES RELATE THE CYTO- AND MYELOARCHITECTONICS OF THE HUMAN CEREBRAL CORTEX: UNIFORMITY IN AREAL DIVERSITY 2.1. Cytoarchitectural Data It was the aim of Hellwig’s study (1993) to compute myelin pictures from quantitative data on the cytoarchitectonics of different areas. Cytoarchitectural data were taken from the treatise on the human cortex by von Economo and Koskinas (1925) which contains detailed descriptions of all areas. Three types of data were considered: (1) layer thicknesses; (2) neurone sizes in each layer (specified as the width of a cell body); (3) the volume density of neurones in each layer (specified as numbers of neurones per 0.001 mm3). 2.2. Two Basic Assumptions Two basic assumptions were used to transform von Economo’s cytoarchitectural data into myelin pictures: 2.2.1. First assumption Large neurones contribute more to the intracortical myelin content than small ones. This relation can be represented by the sigmoid curve in Figure 2.4. The assumption is hypothetical, but was inspired by observations on Nissl, myelin and Golgi stained sections through human and non-human cortices. 2.2.2. Second assumption The average distribution of horizontal axon collaterals of pyramidal neurones can be quantified by the histogram of Figure 2.5. This histogram is derived from a Golgi study on pyramidal cells in the rat visual cortex by Paldino and Harth (1977). It was used as a model of the distribution of horizontal axon collaterals with respect to the cell body. Only one modification was introduced for the computations: the histogram was scaled to the 1

myelin value

0.8 0.6 0.4 0.2 0

0

10

20

30

40

diameter of the cell body [µm] Figure 2.4.

A hypothetical curve that transforms the diameter of a neurone’s cell body into a “myelin value”.



21

number of collaterals

100 80 60 40 20 0

–200

0

200

400

600

800

distance from the cell body [µm] Figure 2.5. Modified diagram from a study by Paldino and Harth (1977) on pyramidal neurones in the rat visual cortex. Distances (vertical to the cortical surface) between the endpoints of axon collaterals and the cell body were measured (positive distances: below the cell body; negative distances: above the cell body). Note that the bulk of collaterals is located below the cell body.

thickness of each area. Note that the second assumption concerns only pyramidal neurones. For the computation this means that the few neurones in layer 1 which are all of the non-pyramidal type (e.g. Peters and Kara, 1985) were discarded. In addition, below layer 1, where the non-pyramidal neurones account for only about 15% of the whole neurone population (Peters and Kara, 1985; Braak and Braak, 1986; Braitenberg and Schüz, 1998), all neurones were considered as pyramidal cells. 2.3. Procedure and Results In all, 14 neocortical areas were chosen for this study. They comprise areas focussed on by many investigators, and include the motor cortex, the primary sensory areas or the speech centres. Moreover, they give a fair impression of the variability of myeloarchitectonic patterns across the human neocortex. Here, the results of just three areas are presented, the Brodmann areas 4, 7 and 17. The architecture of area 7, a field in the parietal association cortex, is paradigmatic for the homotypical cortex. Area 4 (primary motor cortex) and area 17 (primary visual cortex), on the other hand, represent the two extremes of the heterotypical cortex (Nissl sections shown in Figure 2.6a). Myelin pictures were computed in two steps. First, using data from von Economo and Koskinas (1925), the average size of neurones in each layer was transformed into a myelin value by means of the curve in Figure 2.4. The myelin value was then multiplied by the corresponding number of neurones per unit volume. The procedure yields, for each layer, a single value which can be considered as the layer-specific contribution to the population of horizontal myelinated fibres (Figure 2.6b). In the second step of the computation, it was taken into account that myelin is distributed along axonal arborizations. This was done by convolving the diagrams in Figure 2.6b with the histogram of Figure 2.5. This simply provides for shifting the myelin into the appropriate position (Figure 2.6c). The densities of myelin thus obtained were represented as shades of grey (Figure 2.6d) in order to facilitate comparison with real myelin preparations (Figure 2.6e). The simulated


Bernhard Hellwig

22 a

b

c

d

e

0

Area 7

1 2 3

distance from cortical surface [mm]

Area 17

4 5 0 1 2 3 4 5 0

Area 4

1 2 3 4 5 0

50

100 0

50

100

amount of myelin [%] Figure 2.6. Computation of myelin pictures for areas 4, 7 and 17 and comparison with real myelin preparations. (a) Nissl pictures. (b) First step of the computation: the layer-specific amounts of myelin are shown as a function of the cortical depth. (c) The second step of the computation: the diagrams of Figure 2.6b are convolved with the histogram of Figure 2.5. (d) Figure 2.6c, transformed into shades of grey. (e) Real myelin preparations. Bar, 1 mm.

myelin pictures are remarkably close to the real ones. In area 7, both stripes of Baillarger are visible, in agreement with Vogt’s (1911) original description. In area 17 only one band of myelinated fibres is conspicuous, the so-called stripe of Gennari (cf. Vogt and Vogt, 1919). In area 4 the comparison between simulation and reality is complicated by the fact that the myeloarchitectonic patterns differ in two subfields. The simulation is close to the anterior part of area 4 where, according to Vogt (1910), only the outer stripe of Baillarger is visible, while the inner one grades into the white matter. 2.4. Conclusion The findings presented above support the assumption that the stripes of Baillarger consist mainly of horizontal axon collaterals of pyramidal cells. The two assumptions relating cyto- to myeloarchitectonics apply also to the other areas investigated in Hellwig’s study (1993). This suggests that the distribution of horizontal axon collaterals of pyramidal neurones and the principles of their myelination are remarkably similar in different areas. Thus, there is obviously both diversity and unity in the cortex. Despite areal variability,



23

the rules linking cyto- and myeloarchitectonics seem to be essentially uniform throughout the neocortex.

3. CAN CYTO- AND MYELOARCHITECTONICS BE INTERPRETED IN FUNCTIONAL TERMS? 3.1. Connectivity and Function A functionally or computationally relevant description of cortical anatomy will focus on the connectivity between neurones. It is obvious that the wiring scheme in the cortex strongly influences how information is processed. Considering the enormous number of synapses in the cortex, connectivity certainly has to be described in a statistical way. The cortical wiring scheme can probably be adequately grasped by parameters such as connection probabilities, number of synapses involved in a connection, the amount of divergence and convergence or the relative importance of short- and long-range connections. Once these parameters are known, one should be able to specify the connections of an arbitrarily selected neurone to other neurones in the cortex in a probabilistic way. Braitenberg and Schüz (1998) have described the basic machinery of the mouse cortex on the basis of a statistical analysis of its components. To some extent, these data can be extrapolated to the human cerebral cortex. However, it is still largely unclear how the basic cortical wiring scheme varies from area to area. This leads to the question discussed in the next section: Can the variability of the connectivity scheme in different areas of the human cerebral cortex be inferred from cyto- and myeloarchitectonics and does this lead to a better understanding of the mechanisms of information processing in these areas?

3.2. Discussion In cytoarchitectonics, the local variations of size and packing density of cell bodies are used for cortical parcellation. The number of synapses on cell bodies is small, rarely exceeding 200 (Peters and Kaiserman-Abramof, 1970; White and Rock, 1980; Müller et al., 1984). This is not much compared to the overall number of synapses carried by a cortical neurone: about 8000 in the mouse and about 40 000 in the human cortex (Braitenberg and Schüz, 1998). In other words, a method that stains the cell bodies of a neurone cannot be very helpful for elucidating cortical connectivity. Connections are predominantly located in the neurophil, i.e. in those parts of the cortical tissue that remain unstained in cell body preparations. Nevertheless, a few general statements about connectivity can be made, since the size of a cell body is positively correlated to the length of its dendritic arborizations (Bok, 1959). For instance, small perikarya which are densely packed indicate that the dendritic processes are relatively short, thus occupying only a small volume. This applies to layer 4 of the primary sensory areas where the thalamic afferents arrive. The dense packing of cell bodies points to a local preprocessing of the incoming thalamic information. Some layers contain large cell bodies which are not so densely packed, e.g. layers 3 and 5 in Figure 2.1. This indicates large and richly ramified dendritic trees, i.e. information is sampled from a relatively extended piece of cortex. Pyramidal neurones in layers 3 and 5 are the origin of important long-range projections to other cortical or subcortical structures.


24

Bernhard Hellwig

Their large dendrites seem to ensure that the information which is projected contains a relatively general overview of cortical activity, and not some specialized data about the processes in small cortical patches. Beyond this, cytoarchitectonics does not tell us much about patterns of dendritic or axonal arborizatons which carry the bulk of synapses and are thus most important for cortical connectivity. In this respect, myeloarchitectonics might be more interesting because it shows patterns of intracortical fibres. The main function of myelin is probably to increase conduction velocities. However, it is doubtful that this is an important property for all intracortical axonal fibres. Many of them are so short that the actual conduction times are within the range of a few milliseconds, even if the whole variability of conduction velocities encountered in the nervous system is taken into consideration. This is unlikely to be significant (Hellwig, 1993). An important function of intracortical myelin might be its ability to insulate axonal fibres, in the sense that myelinated segments of axonal ramifications are unable to form synapses. This is an interesting property of myelin, since it means that myelination imposes a spatial structure on axonal trees: in some places they are capable of interacting with other neurones, in others they are not. The distribution of myelin over the axonal tree is probably not random. The time course of maturation in the primary visual cortex of the cat suggests that myelination is related to early learning processes. In the first postnatal weeks, there is a period of extraordinary plasticity in the visual cortex, the so-called critical period. Plasticity, for example the susceptibility to the effects of monocular deprivation, is high until some time between the sixth or eighth week after birth (Hubel and Wiesel, 1970; Olson and Freeman, 1980). On the level of pyramidal neurones this period is characterized by the emergence and refinement of axonal arborizations (Callaway and Katz, 1990). By pruning of inappropriate axon collaterals, axonal ramifications are formed in which long horizontal axonal fibres give off clusters of axon collaterals that preferably contact certain target regions, namely columns of similar orientation preference (Gilbert and Wiesel, 1989). The process of shaping axonal trees is experience-dependent (Löwel and Singer, 1992). Axonal arborizations attain an adult appearance by the end of the critical period, i.e. about 7 weeks after birth (Callaway and Katz, 1990). Interestingly, this is the time when myelination starts. The first myelinated fibres in the primary visual cortex of the cat appear by the end of the sixth postnatal week, myelination is moderate until the end of the eighth week, and then undergoes an enormous, almost explosive increase (Haug et al., 1976). In conclusion, two processes seem to coincide at the end of the critical period in the visual cortex of the cat: the termination of the experience-dependent shaping of axonal branching patterns and the onset of myelination. Observations on individual pyramidal cells suggest that the myelinated parts of axonal trees are mainly those horizontal fibre segments that interconnect clusters of collaterals (DeFelipe et al., 1986) (Figure 2.7). Thus, myelin would insulate predominantly axonal segments which failed to establish functional relations with other cortical neurones during the critical period. In other words, myelin would be a sort of memory trace, a tool to store information about early learning processes. The interpretation of myelin as a memory trace by which early experiences are fixed may explain why the overall amount of myelin is higher in the heterotypical areas, i.e. in the primary motor and sensory cortices, than in the homotypical association cortex. The primary areas are in a close relation to the outside world, and a repertoire of information



25

dendrite

myelin axon

Figure 2.7. Schematic drawing of a cortical pyramidal cell. Long myelinated horizontal axon collaterals emanate below the cell body and give off clusters of non-myelinated collaterals.

processing steps fixed by early experiences may be an efficient way to deal with ever-recurring standard tasks. In the association cortex the tasks to be expected are less predictable. Thus, a less rigid wiring scheme may be useful in which associations of all kinds can be learned. In this context, it is also interesting to note that the onset of myelination is much earlier in the primary areas than in the association cortex (Flechsig, 1920, 1927). All in all, it is, however, most difficult to interpret areal variability as revealed by myeloarchitectonics in functional terms. This is mainly due to the fact that myelin preparations, although showing axonal fibres, do not reveal the cortical wiring scheme, since those axonal fibre segments are stained that, insulated by myelin, are unable to contact other neurones. In a way, myelin preparations display the “negative” of intracortical connectivity. 3.3. Outlook It is an important task for neuroanatomists (and for neuroscientists in general) to relate structure to function. As far as the parcellation of the cortex into areas is concerned, this goal has been achieved mainly in one respect: Certain functional specializations can be attributed to certain areas. However, beyond the localization of function, the functional interpretation of areal variability in the cortex is largely unclear. In particular, it remains to be clarified why areal adaptations of the basic cortical network seem to be particularly appropriate for the execution of specific tasks. For instance, one wonders why the structure of the motor cortex is obviously useful for the control of movements, but not for other tasks such as the processing of visual information. In other words, the relation between the mechanisms of information processing and the areal variability of cortical anatomy is


26

Bernhard Hellwig

unclear. This is due to the fact that the connectivity patterns in each area are largely unknown. As pointed out above, cyto- and myeloarchitectonics are not very helpful in this respect. How can variations of wiring schemes in different areas be elucidated? Knowing the variability of neuronal arborizations in different areas would in itself be helpful. Unfortunately, this type of information is scarce. For the human cortex, one of the main sources is still Cajal’s (1911) treatise on the nervous system, which yields some qualitative, but no quantitative data on the variability of Golgi-stained neurones in different areas. More recent material is reviewed in this book in chapters by Jacobs and Scheibel (2002) and Valverde et al. (2002). In all, studying the architectonics of the cortex as revealed by Golgi or similar methods is still a worthwhile research program. An approach by which the local connectivity between pyramidal neurones in a given area can be quantitatively estimated has been suggested by Hellwig (2000). Pyramidal neurones in layers 2 and 3 of the rat visual cortex were intracellularly stained and threedimensionally reconstructed using a computer-based camera lucida system. In a computer experiment, pairs of pre- and postsynaptic neurones were formed and potential synaptic contacts, i.e. spatial contacts between axons and dendrites, were calculated. For each pair, the calculations were carried out for a whole range of distances (0 to 500 µm) between the pre- and the postsynaptic neurone, in order to describe cortical connectivity as a function of the spatial separation of neurones. It was also possible to differentiate whether neurones were situated in the same or in different cortical layers. The data thus obtained were used to compute connection probabilities, the average number of contacts between neurones or the frequency of specific numbers of contacts. It could be shown by comparison with independent data that the local cortical connectivity between pyramidal neurones estimated in this way was a good approximation to reality. In principle, this approach can be extended to other layers as well as to other areas. This makes it possible to investigate cortical architectonics in terms of connectivity. The interpretation of functional processes in cortical areas will certainly be promoted by knowledge about the underlying wiring scheme. However, data on connectivity could also be important in another context: They could actually be used to build artificial neuronal networks with a biologically realistic structure. In such networks, the areal variability of information processing could be studied. Thus, describing cortical architectonics in terms of connectivity would not just be an analytic undertaking, but could also serve as a basis for a synthetic approach.

REFERENCES Baillarger, J.G.F. (1840) Recherches sur la structure de la couche corticale des circonvolutions du cerveau. Mémoires de l’academie royale de Médecine, 8, 149–183. Batsch, G. (1956) Die myeloarchitektonische Untergliederung des Isocortex parietalis beim Menschen. Journal für Hirnforschung, 2, 225–270. Bok, S.T. (1959) Histonomy of the Cerebral Cortex. Amsterdam, London: Elsevier. Braak, H. (1980) Architectonics of the Human Telencephalic Cortex. Berlin, Heidelberg, New York: Springer Verlag. Braak, H. and Braak, E. (1986) Ratio of pyramidal cells versus non-pyramidal cells in the human frontal isocortex and changes in ratio with ageing and Alzheimer’s disease. In: D.F. Swaab, E. Fliers. M. Mirmiran, W.A. van Gool and F. van Haaren (eds), Aging of the brain and Alzheimer’s disease (Progress in Brain Research, Vol. 70), Amsterdam: Elsevier, pp. 185–212. Braitenberg, V. (1962) A note on myeloarchitectonics. Journal of Comparative Neurology, 118, 141–146.



27

Braitenberg, V. (1974) Thoughts on the cerebral cortex. Journal of Theoretical Biology, 46, 421–447. Braitenberg, V. (1978) Cortical architectonics: general and areal. In: M.A.B. Brazier and H. Petsche (eds), Architectonics of the Cerebral Cortex, New York: Raven Press, pp. 443–465. Braitenberg, V. and Schüz, A. (1998) Cortex: Statistics and Geometry of Neuronal Connectivity. Berlin, Heidelberg, New York: Springer Verlag. Brodmann, K. (1909) Vergleichende Lokalisationslehre der Großhirnrinde. Leipzig: Johann Ambrosius Barth. Cajal, S Ramón y (1911) Histologie du système nerveux de l’homme et des vertébrés. Madrid: Consejo superior de investigaciones cientificas, Instituto Ramón y Cajal. Callaway, E.M. and Katz, L.C. (1990) Emergence and refinement of clustered horizontal connections in cat striate cortex. Journal of Neuroscience, 10, 1134–1153. Colonnier, M. and Sas, E. (1978) An anterograde degeneration study of the tangential spread of axons in cortical areas 17 and 18 of the squirrel monkey (Saimiri sciureus). Journal of Comparative Neurology, 179, 245–262. Creutzfeldt, O.D., Garey, L.J., Kuroda, R. and Wolff, J.-R. (1977) The distribution of degenerating axons after small lesions in the intact and isolated visual cortex of the cat. Experimental Brain Research, 27, 419–440. DeFelipe, J., Conley, M. and Jones, E.G. (1986) Long-range focal collateralization of axons arising from corticocortical cells in monkey sensory-motor cortex. Journal of Neuroscience, 6, 3749–3766. Fisken, R.A., Garey, L.J. and Powell, T.P.S. (1975) The intrinsic, association and commissural connections of area 17 of the visual cortex. Philosophical Transactions of the Royal Society London, Series B, 272, 487–536. Flechsig, P. (1920) Anatomie des menschlichen Gehirns und Rückenmarks auf myelogenetischer Grundlage. Leipzig: Thieme. Flechsig, P. (1927) Meine myelogenetische Hirnlehre. Berlin: Springer. Gatter, K.C. and Powell, T.P.S. (1978) The intrinsic connections of the cortex of area 4 of the monkey. Brain, 101, 513–541. Gennari, F. (1782) De peculiari structura cerebri nonnullisque eius morbus. Parma. Gilbert, C.D. and Wiesel, T.N. (1979) Morphology and intracortical projections of functionally characterised neurons in the cat visual cortex. Nature, London, 280, 120–125. Gilbert, C.D. and Wiesel, T.N. (1983) Clustered intrinsic connections in cat visual cortex. Journal of Neuroscience, 3, 116–1133. Gilbert, C.D. and Wiesel, T.N. (1989) Columnar specificity of intrinsic horizontal and corticocortical connections in cat visual cortex. Journal of Neuroscience, 9, 2432–2442. Haug, H., Kölln, M. and Rast, A. (1976) The postnatal development of myelinated nerve fibres in the visual cortex of the cat. Cell and Tissue Research, 167, 265–288. Hellwig, B. (1993) How the myelin picture of the human cerebral cortex can be computed from cytoarchitectural data. A bridge between von Economo and Vogt. Journal für Hirnforschung, 34, 387–402. Hellwig, B. (2000) A quantitative analysis of the local connectivity between pyramidal neurons in layers 2/3 of the rat visual cortex. Biological Cybernetics, 82, 111–121. Hopf, A. (1954) Die Myeloarchitektonik des Isocortex temporalis beim Menschen. Journal für Hirnforschung, 1, 208–279. Hubel, D.H. and Wiesel, T.N. (1970) The period of susceptibility to the physiological effects of unilateral eye closure in kittens. Journal of Physiology (London), 206, 419–436. Jacobs, B. and Scheibel, A.B. (2002) Regional dendritic variation in primate cortical pyramidal cells. In: A. Schüz and R. Miller (eds), Cortical areas: unity and diversity (Conceptual Advances in Brain Research series), London: Taylor and Francis Publishers. Landry, P., Labelle, A. and Deschênes, M. (1980) Intracortical distribution of axonal collaterals of pyramidal tract cells in the cat motor cortex. Brain Research, 191, 327–336. Le Gros Clark, W.E. and Sunderland, S. (1939) Structural changes in the isolated visual cortex. Journal of Anatomy, 73, 563–574. Levitt, J.B., Lewis, D.A., Yoshioka, T. and Lund, J.S. (1993) Topography of pyramidal neuron intrinsic connections in macaque monkey prefrontal cortex (areas 9 and 46). Journal of Comparative Neurology, 338, 360–376. Löwel, S. and Singer, W. (1992) Selection of intrinsic horizontal connections in the visual cortex by correlated neuronal activity. Science, Washington, 255, 209–212. Martin, K.A.C. and Whitteridge, D. (1984) Form, function and intracortical projections of spiny neurones in the striate visual cortex of the cat. Journal of Physiology (London), 353, 463–504. Müller, L.J., Verwer, R.W.H., Nunes Cardoso, B. and Vrensen, G. (1984) Synaptic characteristics of identified pyramidal and multipolar non-pyramidal neurons in the visual cortex of young and adult rabbits. A quantitative Golgi-electron microscope study. Journal of Neuroscience, 12, 1071–1087. Olson, C.R. and Freeman, R.D. (1980) Profile of the sensitive period for monocular deprivation in kittens. Experimental Brain Research, 39, 17–21. Paldino, A. and Harth, E. (1977) A computerized study of Golgi-impregnated axons in rat visual cortex. In: R.D. Lindsay (ed.), Computer Analysis of Neuronal Structures, New York: Plenum Press, pp. 189–207.


28

Bernhard Hellwig

Peters, A. and Kaiserman-Abramof, I.R. (1970) The small pyramidal neuron of the rat cerebral cortex. The perikaryon, dendrites and spines. American Journal of Anatomy, 127, 321–355. Peters, A. and Kara, D.A. (1985) The neuronal composition of area 17 of rat visual cortex. II. The nonpyramidal cells. Sanides, F. (1962) Die Architektonik des menschlichen Stirnhirns. In: M. Müller, H. Spatz and P. Vogel (eds), Monographien aus dem Gesamtgebiete der Neurologie und Psychiatrie, Vol. 98, Berlin, Göttingen, Heidelberg: Springer Verlag. Schwark, H.D. and Jones, E.G. (1989) The distribution of intrinsic cortical axons in area 3b of cat primary somatosensory cortex. Experimental Brain Research, 78, 501–513. Strasburger, E.H. (1937) Die myeloarchitektonische Gliederung des Stirnhirns beim Menschen und Schimpansen. I. Teil. Myeloarchitektonische Gliederung des menschlichen Stirnhirns. Journal für Psychologie und Neurologie, 47, 461–491. Valverde, F., De Carlos, J.A. and López-Mascaraque, L. (2002) The cerebral cortex of mammals: diversity within unity. In: A. Schüz and R. Miller (eds), Cortical areas: unity and diversity (Conceptual Advances in Brain Research series), Taylor and Francis Publishers, London, New York. Vogt, C. and Vogt, O. (1919) Allgemeinere Ergebnisse unserer Hirnforschung. Journal für Psychologie und Neurologie, 25, 279–461. Vogt, O. (1910) Die myeloarchitektonische Felderung des menschlichen Stirnhirns. Journal für Psychologie und Neurologie, 15, 221–232. Vogt, O. (1911) Die Myeloarchitektonik des Isocortex parietalis. Journal für Psychologie und Neurologie, 18, 379–390. von Economo, C. and Koskinas, G.N. (1925) Die Cytoarchitektonik der Hirnrinde des erwachsenen Menschen. Wien, Berlin: Springer Verlag. White, E.L. and Rock, M.P. (1980) Three-dimensional aspects and synaptic relationships of a Golgi-impregnated spiny stellate cell reconstructed from serial thin sections. Journal of Neurocytology, 9, 615–636.


3 Architectonic Mapping of the Human Cerebral Cortex Katrin Amunts1, Axel Schleicher3 and Karl Zilles1,2,3 1

Institut für Medizin, Forschungszentrum Jülich, Germany and 2C. and O. Vogt Institute of Brain Research and 3Institute of Neuroanatomy, Heinrich Heine University, Düsseldorf, Germany Correspondance: Dr. Katrin Amunts, Institut für Medizin, Forschungszentrum Jülich, GmbH, D-52425 Jülich, Germany Tel: +49 2461 614300; FAX: +49 2461 618307 e-mail: [email protected]

The classical cyto- and myeloarchitectonic maps of the human cerebral cortex considerably influenced the concept of localization of function. Presently, these maps serve as anatomical references in functional imaging studies. However, the classical maps suffer from drawbacks such as the highly observer-dependent definition of areal borders; the fact that they present only a single aspect of architectonic organization of the cortex (e.g. only cytoarchitecture), and the lack of information on intersubject variability of location and size of a cortical area in a spatial reference system. Recent methodological progress in computerized image analysis of histological specimens, the introduction of markers which reflect various architectonic aspects of cortical organization (e.g. receptor autoradiography), and the development of warping techniques to compensate for intersubject variability of brain structure in 3D made it possible to overcome these drawbacks. We propose a new concept of architectonic mapping which is based on: (i) a definition of areal borders by using multivariate statistical analysis, and not by highly subjective judgements; (ii) a quantitative analysis of similarity and dissimilarity in architecture between cortical areas; and (iii) a multimodal characterization of cortical organization based on cyto-, myelo- and receptor-architectonic mapping. The comparison of architectonic maps with functional imaging data in a common standard reference space allows, for the first time, a direct analysis of correlations between structure and function in the living human brain, and provides new insights into the architecture of the cerebral cortex. KEYWORDS: architecture, brain mapping, human cerebral cortex, intersubject variability, transmitter receptors

1. INTRODUCTION The classical cytoarchitectonic maps of the human cerebral cortex published by Brodmann (1909), Campbell (1905), Elliot Smith (1907), von Economo and Koskinas (1925) and the Vogts (Vogt and Vogt, 1919) have recently gained considerable attention, since they present mandatory structural data for the microanatomical interpretation of functional imaging data. These maps, however, do not fulfil the requirements of an anatomical reference system for functional human brain mapping. For instance, they present only schematic, simplified drawings of a single, individual brain or hemisphere in a two-dimensional view without any descriptions of the intersubject variability of cortical architecture. The same is true for more recent architectonic maps, e.g. by the Russian school (Sarkisov et al., 1949), 29 © 2002 Taylor & Francis

30

Katrin Amunts, Axel Schleicher and Karl Zilles

Sanides (1962, 1964), Bailey and von Bonin (1951) and Braak (1979). Moreover, these maps differ between each other with respect to the number, location and extent of cortical areas (Zilles, 1990). Campbell subdivided the human cerebral cortex on the basis of cell body- and myelinstaining into 14 regions, amongst them precentral, frontal, visuo-sensory, and the auditopsychic areas (Campbell, 1905). Elliot Smith studied the regional and laminar distribution of myelinated fibers in unstained sections, and proposed a different map containing about 50 areas (Elliot Smith, 1907). Nowadays, the most widely used map is that of Brodmann, which relies on extensive studies of cell body-stained (Nissl-stain) histological sections (Brodmann, 1903, 1905, 1908, 1909). He subdivided the cortex on the basis of cytoarchitectonic criteria into approximately 40 cortical areas. Unfortunately, he never described his criteria for parcellation of most of the areas in sufficient detail. This is true in particular for so-called higher associative areas like areas of the prefrontal cortex, posterior parietal lobe, and of the inferior temporal cortex. Brodmann’s schematic surface drawing of an architectonic map was used by Talairach and Tournoux as basis of the architectonic parcellation in their stereotaxic atlas (Talairach and Tournoux, 1988). They simply transferred Brodmann’s areas to their own brain atlas by trying to identify corresponding sulcal patterns in both brains, assuming a strong association between the sulcal pattern and borders of cortical areas. Such an association, however, was already doubted by Brodmann. He mentioned that “ . . . a schematic drawing can reflect only the major spatial relationships, and therefore, precise topographical associations1 cannot be considered in general or only in a distorted manner; this is true in particular for all those cortical regions which have borders in the neighborhood of sulci and those regions which are located in the depth of such a cortical region” (Brodmann, 1908). The basis of Brodmann’s research was the working hypothesis that the cerebral cortex is composed of numerous cortical areas, each of them characterized by a distinct cytoarchitecture and function. Following this concept, the cytoarchitecture of a cortical area should be more or less constant within a cortical area, but changes considerably at its border. For example, Brodmann’s area 4 was conceptualized as the anatomical equivalent of the primary motor cortex which guides voluntary movements (Fritsch and Hitzig, 1870) and Broca’s region was regarded as the anatomical correlate of the functionally defined center of speech (Broca, 1861). Although for the vast majority of cortical areas such as microstructural-function relationship could not be rigorously tested at that time, both Brodmann and Campbell took architectonic localization of function for granted. The strict localizationist approach culminated in a map of the human cortex of Kleist (1934) in which complex functions were assigned to a distinct cytoarchitectonic area. Brodmann’s area 18 (for instance) was associated with visual attention, perception of spatial position, and eye movements toward the upper and lower visual field. Brodmann himself did not represent such an extreme localizational concept (Brodmann, 1909). In order to avoid a confusion of histological data and unproven evolutionary and functional speculations, he created his system of a “neutral” nomenclature by numbering different cytoarchitectonic areas mainly according to their dorso-ventral sequence. Older studies (Vogt and Vogt, 1919) and more recent electrophysiological studies in nonhuman primates have demonstrated that the basic idea of Brodmann was true: Neurones with similar receptive fields and

1

i.e. between sulci and areal borders [Au].


Architectonic Mapping

31

A

B

C

Figure 3.1. Cytoarchitectonic maps of the lateral surface of human brain adapted from [A] Brodmann (1909), [B] von Economo and Koskinas (1925) and [C] the Russian school (Sarkisov et al., 1949). Cytoarchitectonic areas are marked by different hatches and classified according to Brodmann’s nomenclature by Arabic numerals [A, C] or according to that of von Economo and Koskinas by letters and numerals [B]. Note differences in sulcal pattern as well as in shape and extent of the areas (e.g. in the frontal lobe with respect to areas 45, 9 and 46; compare A with C).


32


response properties lie within the same cytoarchitectonic area, as found when the same brain is sectioned and cell-stained following recording experiments, and correlations are sought between penetration sites and the cytoarchitectonic pattern. Conversely, response properties of neurones change across cytoarchitectonic borders (Luppino et al., 1991; Matelli et al., 1991; Tanji and Kurata, 1989). Although later studies extended and supplemented the maps of Brodmann and Campbell, they followed the concept of a cortical area implied by these maps. von Economo and Koskinas (1925) introduced an even more complex subdivision of the human cortex into cortical areas with regional peculiarities (= subareas). They defined cortical areas on the basis of their topography (frontal, parietal, occipital, etc.), and in terms of their cytoarchitecture and local peculiarities. As an example, area FAg is characterized by its location in the frontal lobe (“F”), an agranular cytoarchitecture (“A”) with giant pyramidal cells (“γ ”). Area FCB, e.g. has common features of area FB and FC, etc. Discussion of this concept, however, raised controversy in the scientific community. von Economo and Koskinas applied quantitative criteria (e.g. size of cells, thickness of layers) in order to provide a more precise characterization of a cortical area, to formalize the cytoarchitectonic description of cortical areas, and to make it more independent of the experience of the observer. Finally, the Russian school published another map of the human cerebral cortex which was, however, based mainly on Brodmann’s approach. Additionally, they tried to overcome one of the unsolved problems of Brodmann’s map, i.e. the neglect of intersubject variability. Their atlas considered intersubject variability in the extent and position of cortical areas by analyzing a sample of dozens of hemispheres (Filimonoff, 1932; Kononova, 1935, 1938; Sarkisov et al., 1949). The increasing number of available architectonic atlases revealed a further problem of architectonic mapping. Although all the cytoarchitectonic maps were based on the same concept of a cortical area as an architectonically distinct and homogeneous region, and all were the result of the same methodical approach, their areal patterns do not match, for example, with respect to the number of cortical areas, their relationship to sulci and gyri, as well as to the neighbouring cortical areas. Even if we compensate for interindividual differences in the macroscopical anatomy of the brains, numerous differences between the maps can hardly be explained. Thus, in the frontal lobe, area 46 has a common border with areas 44 and 45 in Brodmann’s map (Brodmann, 1909), but this border is absent in the map of the Russian school, since here area 9 separates completely area 46 from 44 and 45 (Sarkisov et al., 1949). In a more recent study, transitional areas were defined which exhibited mixed architectonic features of areas 46 and 45 (Rajkowska and Goldman-Rakic, 1995b). Considerable differences between the maps can also be found with respect to the anterior border of area 4, the extrastriate visual cortex, and the parietal cortex, where Brodmann found only a few areas, but recent observations revealed a much higher number of areas. What might be the reasons for differences between the maps? One reason concerns differences in parcellation criteria of the different observers. The most important criteria used in all studies are the density and size of nerve cells, their distribution within cortical layers, the absolute and relative thicknesses of cortical layers, the radial and horizontal arrangement of neurones, the presence of special cells (e.g. giant Betz cells of Brodmann’s area 4), and locally specific subdivisions of layers into sublayers (e.g. the subdivision of layer IV of Brodmann’s area 17 into sublayers IVA–C). For the vast majority of cortical areas, not only one, but a whole complex of criteria is used for its definition. Very often,



33

these criteria are weighted relative to each other, in a different way by each observer. In addition, the criteria are sometimes difficult to formalize objectively. This can be illustrated by the example of such a “simple” area as Brodmann’s area 4. Typical for this area are giant pyramidal cells in layer V (Betz cells), which were discovered by Betz as characteristic cells of the motor area of man, chimpanzee, other primates and dog (Betz, 1874). However, how big is a Betz cell? The height of these cells may vary between different individuals from 60–120 µm, their width from 30–60 µm. Moreover, comparably largesized cells can be found outside area 4 in the area postcentralis gigantopyramidalis (von Economo and Koskinas, 1925). Furthermore, the distance between single Betz cells increases towards subarea 4a (Geyer et al., 1996; Zilles et al., 1995). Thus, the border between area 4 and the rostrally adjoining area 6 is difficult to define on the basis of the Betz cells-criterion. If giant pyramidal cells are defined not by their absolute size, but by their relative size (i.e. comparison with cells in neighbouring areas), such cells can be also found in layers III and V of areas 44 and 45, in the extrastriate visual cortex and in the temporal cortex (Bailey and von Bonin, 1951). Consequently, a reliable definition of area 4 requires not only this, but also additional criteria, e.g. the absence of an inner granular layer. Bailey and von Bonin further followed this line of discussion and asked if there is any objective basis for a detailed cytoarchitectonic map at all. They came to the final conclusion that “ . . . vast areas are so closely similar in structure as to make any attempt at subdivisions unprofitable, if not impossible”. As a consequence, their cytoarchitectonic map is based only on a parcellation into a few main types of cortical regions: regions with numerous granular cells (koniocortex), without granular cells (agranular cortex), with large pyramids in layer III, and the allocortex, as well as 4 combinations between these main types. In contrast to the previously mentioned maps of Brodmann and others, their map does not show sharp borders but gradual transitions between areas (Bailey and von Bonin, 1951). The question arises about which cortical map is the most appropriate. Is it that of Bailey and von Bonin with 8 subdivisions, that of Campbell, Brodmann and the Russian school with about 20 to 40 subdivisions, or that of von Economo and Koskinas with about 100 areas and subareas? One way to answer this questions may be the combination of cytoarchitectonic mapping with other architectonic mapping techniques (multimodal mapping). Flechsig was the first to gave a detailed subdivision of the neocortex into 40 cortical areas by his myelogenetic method, i.e. by studying the heterochronous development of myelination in the white matter immediately below the cortex during foetal and early postnatal periods (Flechsig, 1898). The Vogts and their co-workers subdivided the human cortex on the basis of myeloarchitectonic criteria (distribution and density of myelinated axons within the cortex) into more than 150 fields (Lungwitz, 1937; Riegele, 1931; Strasburger, 1938; Vogt and Vogt, 1919; Vogt, 1919). Their map and the underlying nomenclature were quite complex and difficult to verify for other observers. This might be one reason why it did not reach general acceptance in subsequent years. More recent methods of cortical mapping, e.g. by immunohistochemistry (Bidmon et al., 1997; Campbell and Morrison, 1989; Hendry et al., 1994; Tootell and Taylor, 1995; Zilles et al., 1991c), histochemistry (Burkhalter and Bernardo, 1989; Clarke, 1994; Wong-Riley et al., 1993), pigmentoarchitecture (Braak, 1977, 1979) and regional and laminar distribution of different transmitter receptor binding sites (Dietl et al., 1987; Jansen et al., 1989; Zilles and Clarke, 1997; Zilles and Schleicher, 1995; Zilles et al., 1988, 1991d) proved to be valuable alternatives in architectonic research. Most importantly, the maps based on different histological and histochemical techniques frequently show a perfect spatial coincidence of


34


many areal borders, thus corroborating the position of an areal border by multimodal imaging. Moreover, since a single receptor may not reveal all borders demonstrated by other markers, this finding can be used to define a family of neurochemically related areas by studying the regional pattern of one transmitter receptor, and comparing its distribution with the maps revealed by other receptors or by cytoarchitecture. We think that such a multimodal concept of cortical mapping improves and supplements classical cytoarchitectonic analysis. We will show below, that the architectonic analysis of any histological or histochemical specimen can also be improved considerably by using quantitative measurements and statistically testable image analysis procedures: (i) Borders between cortical areas can be identified by observer-independent statistical analysis of local changes in cytoarchitecture (Schleicher et al., 1999). We will illustrate this approach in cytoarchitectonic specimens, although it can also be applied to receptor architectonic and myeloarchitectonic specimens (Zilles and Schleicher, 1993). (ii) We will present a method for quantifying cytoarchitectonic differences between cortical areas. This method defines the similarity or dissimilarity between cortical areas in terms of numerical distance measures. Using this approach, it can be tested statistically whether differences in cytoarchitecture (or any other architecture) are significant. Furthermore, it allows one to test the long-standing hypothesis of the gradual, rather than distinct character of the majority of cytoarchitectonic borders (Bailey and von Bonin, 1951). (iii) In contrast to previous cortical maps which were based on only one technique, multimodal architectonic analysis will be performed. We will discuss the correspondence and differences of architectonic borders which are revealed by receptor autoradiography of numerous different receptor binding sites, as well as by cytoarchitecture. Human striate and extrastriate areas, as well as Brodmann’s areas 44 and 45 (Broca’s region) will serve as examples for multimodal mapping of the cerebral cortex. (iv) We will conclude with some perspectives on the application of these maps in a threedimensional probabilistic atlas system.

2. OBSERVER-INDEPENDENT DEFINITION OF CYTOARCHITECTONIC BORDERS One of the key features of the neocortex is its organization in layers running parallel to the pial surface. Cortical layers differ by their absolute and relative widths and cell densities. The laminar pattern of a cortical area is represented by its sequence of layers, varying in cell density. Our observer-independent approach to the definition of cortical borders considers these architectonic features. It is based on the assumption that each area has a unique, homogeneous laminar pattern, which distinguishes it from those of neighbouring cortical areas. Several methods have been applied in the past for quantifying the laminar pattern. An early approach was described by Hudspeth and colleagues (1976). They analyzed optical density profiles to describe the distribution of staining intensity across cortical layers in the human primary visual cortex. Although the optical density is an easy and fast measurable parameter, it has the major disadvantage of being sensitive to differences in staining intensity of nerve cells (and of the background) in different brains and sections.



35

For technical reasons, such differences are almost inevitable in histological specimen. Variations in intensity are influenced by factors like age, clinical history, cause of death, post mortem delay, autopsy conditions, and histological techniques (Blinkov and Glezer, 1968; Haug, 1980; Skullerud, 1985; Vierordt, 1893). Based on this experience, we used the volume density of nerve cells in order to quantify the laminar pattern, a parameter with a long tradition in quantitative neurobiology (Haug, 1956; von Economo and Koskinas, 1925). It has the advantage that, within reasonable limits, it is not affected by either staining, or anisotropy (Weibel, 1979). The volume density of nerve cells was estimated as the areal fraction of all stained cellular profiles in square measuring fields of 20–30 µm and defined as gray level index (GLI) ranging from 0% to 100% (Schleicher and Zilles, 1990). Other stereological parameters (e.g. the numerical density) are also available, but we focussed on the volume density since this robust stereological parameter can be automatically estimated from existing histological series in large samples. This parameter is highly correlated with the volume density of neurones, since the density of endothelial and glial cells does not vary systematically throughout the cortical layers (Wree et al., 1982). Using a computerized image analyzer, the GLI was measured in cortical regions of interest (Figure 3.2). GLI images were achieved from which GLI profiles (= density profiles) reaching from the border between layers I and II to the border between cortex and white matter were extracted. The shape of these density profiles describes quantitatively the laminar pattern, i.e. the cytoarchitecture of a cortical area. Dissimilarities between cortical areas and their laminar patterns were reflected by differences in shape of the density profiles. The shape of a profile was numerically described by a set of ten features: the mean of the amplitude (i.e. the mean GLI; meany.o), the center of gravity in the x-direction (meanx.o), the standard deviation (sd.o), the skewness (skew.o), the kurtosis (kurt.o), and the analogous parameters for the first derivative of each profile (meany.d, meanx.d, sd.d, skew.d, kurt.d). Features are based on central moments (Dixon et al., 1988) of the original density profile, and on its first derivative, by treating the profile as a frequency distribution, whereby the cortical depth is the x-value and the GLI is the frequency value at that x-value. Features were normalized in order to weight them equally. Some features can be interpreted directly in terms of cytoarchitecture: the mean GLI increases with increasing density of cell bodies. The feature meanx.o will be smaller than 50% if the supragranular layers have a higher GLI than the infragranular layers. Vice versa, if the infragranular layers show more densely packed cell bodies than the supragranular layers, the meanx.o will be shifted to a value greater than 50%. Multivariate statistical analysis was then used in order to quantify differences in shape between profiles. The Mahalanobis distance D was used as a multivariate measure of differences in shape between neighbouring profiles for detecting cytoarchitectonic borders (Schleicher et al., 1998, 1999). The basic idea was that profiles are more or less similar in shape within a cortical area (homogeneity criterion), and the shape changes abruptly at the border of two neighbouring areas (Schleicher et al., 1995). In order to detect the position of the border, cortical regions of interest were covered by a sequence of equidistant density profiles (Figure 3.2A). The Mahalanobis distance was then calculated between two neighboring sets (= blocks) of profiles (Figure 3.2C). If these two blocks belong to one and the same area, the Mahalanobis distance was small, since differences in the laminar pattern between these two groups of profiles were small. Vice versa, if these two blocks were located exactly at opposite sides of a cortical border, the Mahalanobis distance was maximal since differences in the laminar pattern of these two groups of profiles were


36


Figure 3.2. Observer-independent definition of cytoarchitectonic borders of the visual cortex. The GLI as a measure of neuronal packing density (Wree et al., 1982) was obtained in a histological section stained for cell bodies (Zilles and Schleicher, 1980; Zilles et al., 1986; Amunts et al., 2000; Schleicher and Zilles, 1990). As a result, a GLI image [A] was produced, in which each pixel corresponds to a GLI value measured with a spatial resolution of 25 µm. Light pixels correspond to a low packing density, dark pixels to a high density. The cortical region of interest was covered by a sequence of profiles, indexed consecutively from 1 to 242 [A]. Each profile quantifies the course of the GLI from the border between layers I and II to the border between the cortex and the white matter (along a line perpendicular to the cortical surface). A multivariate distance measure, the Mahalanobis distance D, was calculated (Schleicher et al., 1998) [C]. D is a measure of difference in profile shape between neighbouring blocks of profiles; e.g. D at the position of profile 20 was calculated as the difference in shape between profiles 1–20 and profiles 21–40 [C]. Since 20 profiles of one block were compared with 20 profiles of the neighbouring block, the block size in this case was 20. D was calculated for different block sizes ranging from 8 to 24 [D]. The dots mark the positions of significant Mahalanobis distances for each block size and each position of the profile. For block size 20, significant distances were obtained from the graph of [C]. Significant values of the Mahalanobis distance are marked by red circles and lines. In this histological section, borders were quantitatively defined between areas V1 and V2d (large arrowhead at position 61), within area V2d (small arrowheads at positions 103 and 153), and between areas V2d and V3 (large arrowhead at position 173), and transferred to the original histological section [B]. The border between areas V1 and V2d corresponds to the border between Brodmann’s areas 17 and 18, that between V2d and V3 to the border between Brodmann’s areas 18 and 19 (Amunts et al., 2000; Gattass et al., 1981; Newsome and Allman, 1980; Newsome et al., 1986; Zilles and Clarke, 1997). Scal—Sulcus calcarinus. (see Color Plate 1)



37

large. After the calculation of the Mahalanobis distance for the two adjacent blocks of profiles, both blocks were shifted simultaneously by ≈128 µm (i.e. by the width of one profile) to the next position. In this manner, the Mahalanobis distance was calculated continuously for all sequential positions of all possible blocks of profiles in the region studied (Figure 3.2C). Distances were calculated for different block sizes ranging from 8 to 24 profiles per block (Figure 3.2D). They were calculated between blocks of profiles, and not between single profiles, in order to improve the signal to noise ratio. A subsequent Hotelling’s T2 test (with a Bonferroni correction of the p-values) was applied for testing the significance of each value of the Mahalanobis distance. Borders were defined at those positions of profiles where the following criteria were fulfilled. The Mahalanobis distance D is significant (Hotelling’s T2-test; α = 5%), positions with significant D were stable across different block sizes and could be followed up through neighboring histological sections. In the histological section shown in Figure 3.2, the Mahalanobis distance reaches significant values at the areal borders between areas V2d and V1 at position 61, and between areas V2d and V3 at position 173. Within area V2, the distance shows local maxima at positions 103 and 153. In our sample, internal subparcellations of V2 were associated with the presence or absence of large pyramidal cells in deep layer III. Borders within area V2 have been described in the past by several authors using cytoarchitectonic criteria (Amunts et al., 2000; von Economo and Koskinas, 1925), myeloarchitecture (Lungwitz, 1937; Sanides and Vitzthum, 1965a,b) as well as on the basis of cytochrome oxidase staining (Burkhalter and Bernardo, 1989; Clarke, 1993; Clarke and Miklossy, 1990; Gattass et al., 1997; Lewis and Olavarria, 1995; Merigan et al., 1993; Tootell and Taylor, 1995). Thus, this approach not only confirms borders between well known cytoarchitectonic areas according to Brodmann’s map, but it also detects new subdivisions. A further example is the subdivision of Brodmann’s area 4 into an anterior and a posterior part (Geyer et al., 1996). Recently, areas 3a and 3b were confirmed in cytoarchitectonic specimens (Geyer et al., 1999) using this method. These areas were first mentioned by Brodmann (1909) and later explicitly described by the Vogts in their myeloarchitectonic map (Vogt and Vogt, 1919).

3. HOW DIFFERENT ARE TWO CORTICAL AREAS IN THEIR CYTOARCHITECTURE? Whether a cortical region is homogeneous in architecture and thus constitutes a single cortical area or, alternatively, consists of two or more cortical areas, has been a matter of controversy between different observers. Consequently, the different cortical maps display a different number of cortical fields. The number reaches from 8 (Bailey and von Bonin, 1951) to more than 100 (von Economo and Koskinas, 1925; Vogt and Vogt, 1919). The analysis of interareal differences in cytoarchitectonics becomes even more complicated due to intersubject variability in architecture of the same cortical area in different brains. Cytoarchitectonic variability has been described since the early days of architectonic research (Kononova, 1938; von Economo and Koskinas, 1925). Other authors mentioned it as “considerable”, but the degree of variability was not quantified. Intersubject variability in microstructure makes it often difficult or even impossible to detect reliably subtle differences in cytoarchitecture between areas. Finally, the statement of whether several © 2002 Taylor & Francis

38


cortical areas are more similar (or different) in cytoarchitecture cannot be verified by using pure visual inspection. Thus, analysis of interareal differences has to be based on measurements. Most studies in the past relied on the measurement of single morphometric parameters within a certain sample, e.g. the dendritic length (Hayes and Lewis, 1996; Huttenlocher, 1979), cell sizes (Blinkov and Glezer, 1968; Hayes and Lewis, 1996; von Economo and Koskinas, 1925), the layer thickness (Amunts et al., 1995; Harasty et al., 1996; Zilles et al., 1986), the sizes of cortical areas, subcortical structures and fibre bundles (Andrews et al., 1997; Filimonoff, 1932; Geyer et al., 1999; Haug, 1987a; Kononova, 1935, 1938; Rajkowska and GoldmanRakic, 1995b; Stensaas et al., 1974). Stereological parameters have also been applied successfully (Brody, 1955; Gundersen et al., 1988; Haug, 1984, 1987a,b; Henderson et al., 1980; Pakkenberg and Gundersen, 1997; Schmitz et al., 1999; Terry et al., 1987; West, 1993; Zilles et al., 1986). Altogether, these parameters represent important, quantitative data of cortical microstructure. However, they often reflect only a single aspect of cortical microstructure (e.g. cell density) and do not consider that the cortex is a layered structure with local changes in cell density, size and number within cortical layers and sublayers. In a more recent study on the human frontal lobe, several cytoarchitectonic parameters of areas 9 and 46 were analyzed (Rajkowska and Goldman-Rakic, 1995a,b). Hereby, a three-dimensional counting method (Williams and Rakic, 1988) was applied to measure total cortical and relative laminar thicknesses, neuronal packing density per 0.001 mm3 in individual cortical layers, and sizes of neuronal somata in selected cortical layers. The analysis of these morphometric parameters revealed differences between both areas in the thickness of layer IV, in the packing density of neurones as well as in the size distribution of neurones. The authors concluded that objective cytometric methods can clearly distinguish two adjacent areas within the human prefrontal lobe. In this study, different morphometric parameters were treated and interpreted separately. It is also possible to combine cytoarchitectonic parameters and to analyze them by multivariate statistical analysis. Such an approach has been used by us in defining areal borders (see above). The multivariate approach offers the advantage that different morphometric parameters can be normalized by compensating for different scales and can be combined into one feature vector. The feature vectors are then used in a comprehensive statistical test. In addition, multivariate analyses (e.g. discriminant analysis) take into account the correlations – often high – between parameters and offer procedures to detect those parameters which contribute most to the dissociation between areas. We illustrate this multivariate approach and discuss its implication for a group of five cortical areas: Brodmann’s areas 6, 44, 45, V1 and V2. These areas were selected by the following considerations: (i) The terms Broca’s region and Broca’s area are based on functional concepts. They are used inconsistently with respect to cytoarchitecture. It is widely accepted that areas 44 and 45 constitute Broca’s region (Aboitiz and Garcia, 1997; Amunts et al., 1999; Kononova, 1949; Petrides and Pandya, 1994; Roland, 1993; Uylings et al., 1999), but the terms Broca’s region and Broca’s area are also applied for areas 44, 45 and 47 (Riegele, 1931; Vogt, 1910), as well as for area 44 only (Galaburda, 1980; von Economo and Koskinas, 1925). In some recent functional imaging studies, Broca’s region (or area) refers to a cortical region which includes area 44 and, sometimes, adjacent area 6 (Paulesu et al., 1993; Petrides et al., 1993). Recent fMRI studies



39

have provided evidence that the posterior part of Broca’s region, area 44, and the homologue region of the right hemisphere might be involved in imagery of movement (Binkofski et al., 1999, 2000). This would associate area 44 functionally with area 6. Originally, the definition was based on gross macroscopical markers, i.e. by gyri and sulci (Broca, 1861; Herve, 1888). In this context, we wanted to check whether areas 44 and 45 can be grouped together on the basis of cytoarchitectonic similarity. (ii) If areas 44 and 45 constitute Broca’s region, they should be more similar in cytoarchitecture to each other than to neighboring cortical areas, e.g. to the adjacent ventral part of area 6. (iii) It was expected that areas 44 and 45 would differ even more from areas V1 and V2 than from area 6, because V1 and V2 belong to the visual cortex and are characterized by a completely different organization of their input and output, reflected by different laminar patterns. (Dis-)similarities between these areas were quantified by calculating a multivariate distance measure between the density profiles in these areas. Profiles were obtained from cytoarchitectonic areas 6, 44, 45, V1 and V2 of ten human post mortem brains (Amunts et al., 1999, 2000). Fifteen to thirty profiles were obtained from three randomly selected sections of each area, hemisphere and brain. Thus, a total of about 3000 profiles were processed. Ten features were extracted from each of the profiles, as described above, for the definition of borders. In contrast to the latter approach, the Euclidean distance (and not the Mahalanobis distance) was used as multivariate distance measure. The advantage of the Euclidean distance for this type of analysis is that it is more sensitive in detecting the dissimilarity in architecture between cortical areas. i.e. the Euclidean distance measures the absolute distance between two centroids of the ten-dimensional space (= dissimilarity between areas), whereas the Mahalanobis distance depends not only on this distance, but also on the variability within an area. Thus, the Mahalanobis distance becomes smaller with increasing variance (Schleicher et al., 2000). The Euclidean distance was calculated in each individual brain for all ten possible combinations of two areas from the areas 44, 45, 6, V1 and V2 (= interareal differences). It was also calculated between profiles from corresponding areas of the left and the right hemisphere (= interhemispheric differences). Corresponding distances were averaged across the whole sample size. Multidimensional scaling (Systat® for Windows, Version 9, SPSS, USA) was applied for data reduction and visualization of distances between the cortical areas. Interhemispheric differences were tested statistically against differences between randomly selected profiles from one and the same area. The results are shown in Figure 3.3. The analysis showed a high degree of similarity in cytoarchitecture of areas 44 and 45. Both areas differed considerably from areas 6 as well as V1 and V2 (large distances between the centroids). Area 6 showed shorter distances to areas 44 and 45 than to V1 and V2, which may correspond to the close topographical and functional relationship between areas 44/45 and 6. On the basis of the cytoarchitectonic similarity of areas 44 and 45, these data provide an anatomical argument to combine areas 44 and 45, but not 44 and 6 into a region. Based on classical cytoarchitectonic descriptions, area 44 (which is dysgranular) takes a transitional position between area 45 (which is granular) and area 6 (which is agranular). This relationship is kept in the arrangement of the centroids in the graphs.


40


Figure 3.3. Dissimilarities of Brodmann’s areas 44, 45, 6, as well as areas V1 and V2 based on quantitative cytoarchitecture (K. Amunts, unpublished observations). Euclidean distances were calculated as multivariate measures of dissimilarity between profiles of different cortical areas (= interareal differences) and of the two hemispheres of one and the same area (= interhemispheric differences). Euclidean distance is based on features, which characterize the shape of the profiles of an area (see text). Multidimensional scaling was applied for visualization of interareal differences and for data reduction to a two-dimensional plane defined by dimension-1 and dimension-2 (Schleicher et al., 2000). The larger the dissimilarity in cytoarchitecture between two areas, the larger the distance between them in the graph. Error bars indicate standard errors. Results in this upper graph quantify interareal differences of the left hemisphere. Whereas areas 44 and 45 were found to be very similar in cytoarchitecture, both areas differed considerably from area 6 as well as from visual areas V1 and V2. Interhemispheric differences in cytoarchitecture (lower graph) were significant (marked by an asterisk) for areas 44 and 45, but not for areas 6, V1 and V2. The line marks the level of intersubject variability in cytoarchitecture. It was calculated as the average Euclidean distance between corresponding areas across different subjects (i.e. the distances in shape within the sample of 10 brains between all areas 44, all areas 45, etc. were calculated and then averaged across the brains and areas). The analysis supplemented previous findings on asymmetry in volume of area 44 (Amunts et al., 1999) and demonstrates that cytoarchitectonic asymmetry in areas 44 and 45 might be a microstructural correlate of brain lateralization and dominance for language, in particular.



41

Finally, interhemispheric distances between corresponding areas of both hemispheres were significant for areas 44 and 45, but not for areas 6, V1 and V2. That is, areas 44 and 45 revealed significant left/right differences in cytoarchitecture. Thus, in addition to asymmetry in volume of area 44 which was reported previously (Amunts et al., 1999; Galaburda, 1980), interhemispheric asymmetry was demonstrated at a microstructural level. This cytoarchitectonic asymmetry may contribute to the functional phenomenon of cerebral lateralization and dominance for language, in particular. Information obtained from analysis of interareal differences might be applied for creating hierarchies and families of cortical areas, as described for the cortex of nonhuman primates (Fellemann et al., 1997; Fellemann and Van Essen, 1991; Gattass et al., 1997; Hubel and Wiesel, 1972; Kaas, 1989; Kötter and Sommer, 2000; Nakamura et al., 1993; Peterhans and von der Heydt, 1993; Stephan et al., 2000; Xiao et al., 1999; Zeki, 1978; Zilles and Clarke, 1997). Multivariate distance analysis has been applied to detect interhemispheric cytoarchitectonic differences of the motor cortex (Amunts et al., 1996), its developmental changes (Amunts et al., 1997), interareal, interhemispheric and intersubject differences of Broca’s region (Amunts et al., 1999), as well interareal differences in receptor architecture of the mesial motor and premotor cortex in the macaque (Geyer et al., 1998) and of the human the somatosensory cortex (Geyer et al., 1997, 1999). A quantitative analysis of interareal differences in cytoarchitecture is also relevant for detecting architectonic differences between normal and pathologically altered cortical tissue. Finally, the criterion of similarity in architecture might be valuable with respect to a comparative analysis of homologies between humans and non-human primates (Petrides and Pandya, 1994).

4. MULTIMODAL MAPPING – CORRESPONDENCES AND DIFFERENCES BETWEEN CYTOARCHITECTONIC AND RECEPTOR ARCHITECTONIC BORDERS Architectonic analysis using multivariate statistics can be even more decisive when they incorporate other modalities of architecture, e.g. receptor architecture. The comparisons of receptor- and cytoarchitectonic maps provided evidence that in several cortical regions receptor architecture reveals similar architectonic parcellations as compared to cytoarchitecture. We will present here recent data from our analysis of the human visual cortex for discussing the correspondence between cytoarchitectonic and receptor-architectonic parcellation schemes. Numerous observations on the regional and laminar distribution of transmitter receptors in the human primary (V1) and secondary visual cortex (V2) have been published. Reports on receptors in other extrastriate areas are rare. (For an overview see Zilles and Clarke, 1997). This is due to the facts, that (i) human extrastriate areas are difficult to identify in cytoarchitectonic sections, (ii) the classical architectonic maps of the human occipital lobe show a much less detailed parcellation than corresponding maps of nonhuman primates, and (iii) receptor architectonics of the human occipital lobe require extraordinary large cryostat sections of unfixed, frozen brains, which are difficult to handle. Our own observations in nonhuman primates and human post-mortem brains, as well as recent data from the literature, provide evidence that receptor architectonic mapping is a promising approach to human brain mapping (Bonaventure et al., 2000; d’Argy et al., 1988; Geyer et al., 1997,


42


1998; King et al., 1995; Pazos et al., 1987a,b; Thoss et al., 1996; Zezula et al., 1988; Zilles et al., 1991a,b,d, 1995, 1988). As an example, results of receptor-architectonic mapping of receptor binding sites of the glutamatergic kainate receptor (ligand: [3H]kainate), the muscarinic M1 (ligand: [3H]pirenzipine) and M2 (ligand: [3H]oxotremorine-M) receptors as well as the 5-HT2 receptor (ligand: [3H]ketanserin) of human areas V1, V2v, V2d, V3d and V3a are shown (Figure 3.4, K. Zilles, unpublished observations). We used standardized protocols for the visualization of a variety of different receptors and of cytoarchitecture in a series of adjacent sections (Zilles et al., 1995), thus providing reliable data for cortical parcellation and for evaluation of areal borders. After contrast enhancement and colour coding, differences in regional and laminar receptor distributions reveal borders between visual areas which match to those in cyto- and myeloarchitecture, and provide insight into the neurochemical aspects of cortical organization in the occipital lobe. In detail, the border between areas V1 and V2 was characterized by a prominent change in density and laminar pattern in almost all ligands as well as in cyto- and myeloarchitecture (Figure 3.4A–F). In correspondence to cytoarchitectonic mapping, V1 was located on the upper and the lower bank of the calcarine sulcus and extended to the mesial surface of the cuneus (Amunts et al., 2000; Filimonoff, 1932; Stensaas et al., 1974). The cuneal extension was highly variable between different brains. For kainate receptors (Figure 3.4A), the infragranular layers displayed a higher density in V1 than in V2, whereas the supragranular layers had a lower density in V1 than in V2. The dorsal and ventral borders of V1 coincided precisely with the borders found in the autoradiographs of 5-HT2 (Figure 3.4B), M1 (Figure 3.4C) and M2 (Figure 3.4D) receptors as well as in cyto- (Figure 3.4E) and myeloarchitecture (Figure 3.4F). The border between areas V2d and V3d was clearly established by kainate, 5-HT2 and M2 receptor autoradiography, but less obviously by the M1 receptor. M2 and 5-HT2 receptor densities and laminar patterns showed a clear border between V2v and the adjacent ventral V3, whereas more subtle changes were seen in kainate and M1 receptor autoradiographs. M2 receptors revealed a distinct subdivision of area V3 into V3d and V3A, which was absent in the other receptor autoradiographs and detected only with difficulty by simple visual inspection in cytoarchitectonic sections. In the myeloarchitectonic section, this border could be identified by a higher myelin density in supragranular layers of area V3d than of V3A (Figure 3.4F). The different receptors revealed corresponding areal borders, which coincided with the cyto- or myeloarchitectonic borders. Coincidence in position of borders defined by receptor autoradiography has been proven in an observer-independent fashion using the same method as described above for cytoarchitecture (Schleicher et al., 1998). Some receptors, however, did not reveal all borders seen with other receptors. The 5-HT2 receptor and the 5-HT1A receptors (images of the regional distribution of the latter are not presented here) are examples, which display a more homogenous distribution in the occipital lobe than that of other receptors. For instance, only minor differences in receptor density were found between areas V2 and VP/V3. These receptors, however, showed a pronounced heterogeneity in the motor cortex, where they make visible the border between motor and premotor areas (Zilles et al., 1995). In contrast, the GABAa receptor is heterogeneously distributed even within area V1 (Zilles and Schleicher, 1993). It shows periodically distributed patches in layers II/IVc, which might be related to cytochrome oxidase blobs and ocular dominance columns (Hendrickson et al., 1981). © 2002 Taylor & Francis

Architectonic Mapping 3

[ H] kainate binding sites

3

M1 receptor [ H] pirenzepin binding sites

43

3

5-HT2 receptor [ H] ketanserin binding sites

3

M2 receptor [ H] oxotremorine binding sites

Figure 3.4. Regional distributions of kainate [A], 5-HT2 [B] and muscarinic M1 [C] and M2 [D] receptors (demonstrated with [3H]kainate, [3H]ketanserin, [3H]pirenzipine and [3H]oxotremorine-M binding; K. Zilles, unpublished observations) in coronal sections through the human occipital lobe. Grey value images were scaled to receptor densities, enhanced in contrast, and colour coded. The colour scales indicate the receptor densities in fmol/mg protein. The receptors are heterogeneously distributed, and therefore allow mapping of striate and extrastriate areas including V1, V2v, V2d, V3d, V3A. Borders obtained by receptor achitectonic mapping were compared with borders revealed by cytoarchitectonic [E] and myeloarchitectonic [F] mapping. Note the pronounced differences in receptor density [A–D] as well as in the laminar pattern of cell packing density [E] and myeloarchitecture [F] at the borders of area V1 to V2. The border between area V2d and V3d is most pronounced in A, B and D mapping. In E and F, it is recognizable only at higher magnification; it can be detected by the observer-independent approach for border definition. The border between V3d and V3A is clearly marked in D and F and can be verified quantitatively also in E. Thus, different architectonic techniques supplement each other and are the basis of multimodal architectonic mapping. SCal –Sulcus calcarinus, SLin—Sulcus lingualis, SCol—Sulcus collateralis, SOS—Sulcus occipitalis superior. (see Color Plate 2) © 2002 Taylor & Francis

44


Finally, the parcellations revealed by quantitative cyto-, myelo- and receptor architectonic techniques revealed a more subtle subdivision of the human occipital lobe than shown in the classical architectonic maps from the beginning of the 20th century. They seem to correspond to parcellations observed in functional imaging (DeYoe et al., 1996; Engel et al., 1997; Gulyas and Roland, 1994; Hadjikhani and Tootell, 2000; Haxby et al., 1994; Larsson et al., 1999; Sereno et al., 1995; Shipp et al., 1995; Tootell et al., 1995, 1997; Watson et al., 1993) and nonhuman primate studies (Fellemann et al., 1997; Fellemann and Van Essen, 1991; Gattass et al., 1997; Hubel and Wiesel, 1972; Kaas, 1989, 1993; Nakamura et al., 1993; Peterhans and von der Heydt, 1993; Xiao et al., 1999; Zeki, 1978; Zilles and Clarke, 1997).

5. CONCLUSION AND PERSPECTIVES Although the parcellation of the human cerebral cortex is still far from complete, receptor and quantitative cytoarchitectonic findings provide new criteria for a detailed mapping, which cannot be achieved by cytoarchitectonic analysis alone. If the borders between receptor architectonic areas are established, receptor densities for distinct areas and layers can be calculated. This information can be used to create receptor “fingerprints” of cortical areas (Geyer et al., 1998), which represent the densities of a set of several receptors in a defined cortical area as a multivariate polar plot of receptor densities. First results have shown that these fingerprints have a similar shape in functionally-related areas, but a different shape if functionally differing areas are compared. What is a cortical area? This question provides an important argument for multimodal architectonic mapping, since not all subparcellations of the cerebral cortex constitute a cortical area. For instance, the subdivision of areas V1 and V2 into blob and interblob regions (Livingstone and Hubel, 1987; Roe and Tso, 1995; Tootell et al., 1983; Wong-Riley et al., 1993) reflects differences in colour and orientation selectivity (V1) and receptive field properties (V2), but these regions do not constitute cortical areas. Additional examples are the somatotopy of the motor and somatosensory cortex, the tonotopical organisation of the auditory cortex, each of which represents a functional segregation without representing an architectonic entity. The isolated analysis of only one aspect of cortical organization without consideration of other mapping techniques, would lead to an over-parcellation of the cerebral cortex. The multimodal approach proposed here avoids this problem by providing an overview of the different hierarchical levels (e.g. cytoarchitectonic or receptor architectonic families of cortical areas) of the cortical organization. An important perspective for a functionally-relevant architectonic parcellation of the cortex arises from the integration of architectonic maps with recent PET, fMRI and MEG studies in a common spatial reference system and database, e.g., the European Computerized Human Brain Database ECHBD (Roland et al., 1999; Roland and Zilles, 1994, 1996a,b, 1998). Borders verified in an observer-independent manner for the sensorimotor (Geyer et al., 1996, 1999; Grefkes et al., 2001) and auditory cortex (Morosan et al., 2001), Broca’s region (Amunts et al., 1999), and the visual cortex (Amunts et al., 2000) have been defined in ten human post mortem brains. The areal borders were labeled in corresponding digitized histological sections, and subjected to 3-D reconstruction. These reconstructions of the histological sections and of the cytoarchitectonic areas were warped to the format of the standard reference brain of the ECHBD (Roland et al., 1999; Roland



45

and Zilles, 1996a,b; Schormann et al., 1999b). A modified “principal axes” theory and a movement model for large deformations were applied for the warping of the 3-D datasets of post mortem brains to the standard brain, in order to compensate for intersubject variability in extent, shape and sulcal pattern of the brains, and to achieve a maximal anatomical overlap between the different post mortem brains and the reference brain (Schormann et al., 1997, 1999a; Schormann and Zilles, 1997, 1998). Individual cortical maps were superimposed in the standard reference brain (Schormann et al., 1999b). The overlapping cortical areas of individual brains and their architectonic areas in the standard reference brain result in probability or population maps, which display the intersubject variability in the extent, shape and topography of cortical areas. Variability of macroscopical features (e.g. hemispheric shape or sulcal pattern) has been analyzed in numerous previous studies (Dumoulin et al., 1998; Kennedy et al., 1998; Rademacher et al., 1993; Thompson et al., 1996, 1998; Westbury et al., 1999; Zilles et al., 1997). (For a more comprehensive review concerning this aspect see chapter by Rademacher in this book.) In contrast to traditional brain atlases and to the atlas of Talairach and Tournoux (1988), the ECHBD is (i) a real 3-D representation of cortical areas, (ii) it is based on an observerindependent architectonic definition of cortical areas, and (iii) it provides quantitative information on intersubject variability of the topography of each cortical area. Using its common spatial reference system for combined analysis of PET and cytoarchitectonic data, it has been shown that the processing of both real and illusory contours activates area V2 (Larsson et al., 1999). Another example is the cytoarchitectonic subdivision of area 4 into an anterior and a posterior part, which were first observed in receptor architectonic sections (Zilles et al., 1995), then defined in cytoarchitectonic sections and superimposed with PET data, which allowed one to correlate these areas with functional differences (Geyer et al., 1996). Further combined cytoarchitectonic/functional imaging studies have been performed in the sensorimotor and visual cortices (Bodegard et al., 2000a,b; Naito et al., 1999, 2000). Since the ECHBD has an open structure, it allows the integration of additional modalities of architectonic data (e.g. receptor architectonic, myeloarchitectonic) and of further functional imaging data. In conclusion, architectonic brain mapping has become more objective and less observerdependent. Recent marker techniques, e.g. receptor-autoradiography of different receptors, immuno- and enzyme-histochemical methods have added functional meaningful information. Multimodal mapping and quantitative analysis of interareal differences promote a new and more complex concept of a cortical area. Finally, by applying recent 3-D probabilistic atlas systems, the combined analysis of architectonic maps and functional imaging studies allows the testing of the functional significance of architectonic parcellations and a systematic search for new, functionally relevant cortical areas. Thus, architectonic brain mapping has changed considerably during the last 100 years and opens exciting perspectives for the future.

ACKNOWLEDGEMENTS Published and unpublished work by the authors reviewed in this chapter was supported by the Deutsche Forschungsgemeinschaft (SFB 194/A6), the BioTech program of the EC and the Human Brain Project (P20-MHDA52176) funded by the National Institute of Mental Health, National Institute for Drug Abuse, and the National Cancer Institute. The authors


46


thank Aleksandar Malikovic, Nicola Palomero-Gallagher, Ursula Blohm, Brigitte Machus and Renate Dohm for assistance in preparing the histological sections and the autoradiographs.

REFERENCES Aboitiz, F. and Garcia, G.L. (1997) The evolutionary origin of language areas in the human brain. A neuroanatomical perspective. Brain Research Reviews, 25, 381–396. Amunts, K., Istomin, V.V., Schleicher, A. and Zilles, K. (1995) The postnatal development of the human primary motor cortex: a quantitative cytoarchitectonic analysis. Anatomy and Embryology, 192, 557–571. Amunts, K., Schlaug, G., Schleicher, A., Steinmetz, H., Dabringhaus, A., Roland, P.E. and Zilles, K. (1996) Asymmetry in the human motor cortex and handedness. Neuroimage, 4, 216–222. Amunts, K., Schmidt-Passos, F., Schleicher, A. and Zilles, K. (1997) Postnatal development of interhemispheric asymmetry in the cytoarchitecture of human area 4. Anatomy and Embryology, 196, 393–402. Amunts, K., Schleicher, A., Bürgel, U., Mohlberg, H., Uylings, H.B.M. and Zilles, K. (1999) Broca’s region revisited: Cytoarchitecture and intersubject variability. Journal of Comparative Neurology, 412, 319–341. Amunts, K., Malikovic, A., Mohlberg, H., Schormann, T. and Zilles, K. (2000) Brodmann’s areas 17 and 18 brought into stereotaxic space – where and how variable? Neuroimage, 11, 66–84. Andrews, T.J., Halpern, S.C. and Purves, D. (1997) Correlated size variations in human visual cortex, lateral geniculate nucleus, and optic tract. Journal of Neuroscience, 17, 1859–2868. Bailey, P. and von Bonin, G. (1951) The isocortex of man. Urbana: University of Illinois Press. Betz, W. (1874) Anatomischer Nachweis zweier Gehirncentra. Centralblatt für die medizinischen Wissenschaften, 12, 578–580, 594–599. Bidmon, H.J., Wu, J.Y., Godecke, A., Schleicher, A., Mayer, B. and Zilles, K. (1997) Nitric oxide synthaseexpressing neurons are area-specifically distributed within the cerebral cortex of the rat. Neuroscience, 81, 321–330. Binkofski, F., Buccino, G., Posse, S., Seitz, R., Rizzolatti, G. and Freund, H.-J. (1999) A fronto-parietal circuit for object manipulation in man: evidence from an fMRI-study. European Journal of Neuroscience, 11, 3276–3286. Binkofski, F., Amunts, K., Stephan, K.M., Posse, S., Schormann, T., Freund, H.-J., Zilles, K. and Seitz, R. (2000) Broca’s region subserves recognition of action. Human Brain Mapping, 11, 273–285. Blinkov, S.M. and Glezer, I.I. (1968) Das Zentralnervensystem in Zahlen und Tabellen. Jena: Fischer. Bodegard, A., Geyer, S., Naito, E., Zilles, K. and Roland, P.E. (2000a) Somatosensory areas in men activated by moving stimuli: cytoarchitectonic mapping and PET. NeuroReport, 11, 187–191. Bodegard, A., Ledberg, A., Geyer, S., Naito, E., Zilles, K. and Roland, P.E. (2000b) Object shape differences reflected by somatosensory cortical activation in human. Journal of Neuroscience, 20 RC 51, 1–5. Bonaventure, P., Hall, H., Gommeren, W., Cras, P., Langlois, X., Jurzak, M. and Leysen, J.E. (2000) Mapping of serotonin 5-HT(4) receptor mRNA and ligand binding sites in the post-mortem human brain. Synapse, 36, 35–46. Braak, H. (1977) The pigment architecture of the human occipital lobe. Anatomy and Embryology, 150, 229–250. Braak, H. (1979) The pigment architecture of the human frontal lobe. Anatomy and Embryology, 157, 35–68. Broca, M.P. (1861) Remarques sur le siége de la faculté du langage articulé, suivies d’une observation d’aphemie (Perte de la Parole). Bulletins et memoires de la Societe Anatomique de Paris, 36, 330–357. Brodmann, K. (1903) Beiträge zur histologischen Lokalisation der Grosshirnrinde. II: Der Calcarinustyp. Journal für Psychologie und Neurologie, II, 133–159. Brodmann, K. (1905) Beiträge zur histologischen Lokalisation der Groβhirnrinde. V: Über den allgemeinen Bauplan des Cortex pallii bei den Mammaliern und zwei homologe Rindenfelder im besonderen. Zugleich ein Beitrag zur Furchenlehre. Journal für Psychologie und Neurologie, VI, 275–303. Brodmann, K. (1908) Beiträge zur histologischen Lokalisation der Groβhirnrinde. VI: Die Cortexgliederung des Menschen. Journal für Psychologie und Neurologie, X, 231–246. Brodmann, K. (1909) Vergleichende Lokalisationslehre der Groβhirnrinde in ihren Prinzipien dargestellt auf Grund des Zellenbaues. Leipzig: Barth JA. Brody, H. (1955) Organisation of the cerebral cortex. Part 3. A study of aging in the human cerebral cortex. Journal of Comparative Neurology, 102, 551–556. Burkhalter, A. and Bernardo, K.L. (1989) Organization of cortico-cortical connections in human visual cortex. Proceedings of the National Academy of Scence, U.S.A., 13, 1916–1931. Campbell, A.W. (1905) Histological studies on the localisation of cerebral function, Cambridge: Cambridge University Press.



47

Campbell, M. and Morrison, J.H. (1989) Monoclonal antibody to neurofilament protein (SMI-32) labels a subpopulation of pyramidal neurons in the human and monkey neocortex. Journal of Comparative Neurology, 282, 191–205. Clarke, S. (1993) Callosal connections and functional subdivision of the human occipital lobe. In: B. Gulyas, D. Ottoson and P. Roland (eds), Functional organization of the human visual cortex. Oxford: Pergamon Press, pp. 137–149. Clarke, S. (1994) Modular organization of human extrastriate visual cortex: Evidence from cytochrome oxidase pattern in normal and macular degeneration cases. European Journal of Neuroscience, 6, 725–736. Clarke, S. and Miklossy, J. (1990) Occipital cortex in man: Organization of callosal connections, related cytoand myeloarchitecture, and putative boundaries of functional visual areas. Journal of Comparative Neurology, 298, 188–214. d’Argy, R., Gillberg, P.G., Stalnacke, C.G., Persson, A., Bergstrom, M., Langstrom, B., Schoeps, K.O. and Aquilonius, S.M. (1988) In vivo and in vitro receptor autoradiography of the human brain using an 11Clabelled benzodiazepine analogue. Neuroscience Letters, 85, 304–310. DeYoe, E.A., Carman, G.J., Bandettini, P., Glickman, S., Wieser, J., Cox, R., Miller, D. and Neitz, J. (1996) Mapping striate and extrastriate visual areas in human cerebral cortex. Proceedings of the National Academy of Science, U.S.A., 93, 2382–2386. Dietl, M.M., Probst, A. and Palacios, J.M. (1987) On the distribution of cholecytokinin receptor binding in the human brain: an autoradioraphic study. Synapse, 1, 169–183. Dixon, W.J., Brown, M.B., Engelman, L., Hill, M.A. and Jennrich, R.I. (1988) BMDP. Statistical software manual, Berkley: University of California Press. Dumoulin, S.O., Bittar, R.G., Kabani, N.J., Baker, C.L., LeGoualher, G., Pike, G.B. and Evans, A.C. (1998) Quantification of the variability of human area v5/MT in relation to the sulcal pattern in the parietotemporo-occipital cortex: a new anatomical landmark. Neuroimage, 7, S319-S319. Elliot Smith, G. (1907) A new topographical survey of the human cerebral cortex, being an account of the distribution of the anatomically distinct cortical areas and their relationship to the cerebral sulci. Journal of Anatomy, 41, 237–254. Engel, S.A., Glover, G.H. and Wandell, B.A. (1997) Retinotopic organization in human visual cortex and the spatial precision of functional MRI. Cerebral Cortex, 7, 181–192. Fellemann, D.J. and Van Essen, D.C. (1991) Distributed hierarchical processing in the primate cerebral cortex. Cerebral Cortex, 1, 1–47. Fellemann, D.J., Burkhalter, A. and Van Essen, D.C. (1997) Cortical connections of areas V3 and VP of Macacque monkey. Extrastriate visual cortex. Journal of Comparative Neurology, 379, 21–47. Filimonoff, I.N. (1932) Über die Variabilität der Groβhirnrindenstruktur. Mitteilung II – Regio occipitalis beim erwachsenen Menschen. Journal für Psychologie und Neurologie, 44, 2–96. Flechsig, P. (1898) Neue Untersuchungen über die Marbildung in den menschlichen Grosshirnlappen. Neurologisches Zentralblatt (Leipzig), 21, 977–996. Fritsch, G. and Hitzig, E. (1870) Über die elektrische Erregbarkeit des Grosshirns. Archiv für Anatomie, Physiologie und wissenschaftliche Medizin, Jg. 1870, 300–332. Galaburda, A.M. (1980) La region de Broca: Observations anatomiques faites un siècle après la mort de son découvreur. Revue Neurologique (Paris), 136, 609–616. Gattass, R., Gross, C.G. and Sandell, J.H. (1981) Visual topography of V2 in the macaque. Journal of Comparative Neurology, 201, 519–539. Gattass, R., Sousa, A.P., Mishkin, M. and Ungerleider, L.G. (1997) Cortical projections of area v2 in the macaque. Cerebral Cortex, 7, 110–129. Geyer, S., Ledberg, A., Schleicher, A., Kinomura, S., Schormann, T., Bürgel, U., Klingberg, T., Larsson, J., Zilles, K. and Roland, P.E. (1996) Two different areas within the primary motor cortex of man. Nature (London), 382, 805–807. Geyer, S., Schleicher, A. and Zilles, K. (1997) The somatosensory cortex of human: Cytoarchitecture and regional distributions of receptor-binding sites. Neuroimage, 6, 27–45. Geyer, S., Matelli, M., Luppino, G., Schleicher, A., Jansen, Y., Palomero-Gallagher, N. and Zilles, K. (1998) Receptor autoradiographic mapping of the mesial and premotor cortex of the macaque monkey. Journal of Comparative Neurology, 397, 231–250. Geyer, S., Schleicher, A. and Zilles, K. (1999) Areas 3a, 3b, and 1 of human primary somatosensory cortex: I. Microstructural organisation and interindividual variability. Neuroimage, 10, 63–83. Grefkes, C., Geyer, S., Schormann, T., Roland, P. and Zilles, K. (2001) Human somatosensory area 2: Observerindependent cytoarchitectonic mapping, interindividual variability, and population map. Neuroimage, 14, 617–632. Gulyas, B. and Roland, P.E. (1994) Processing and analysis of form, color and binocular disparity in the human brain: functional anatomy by positron emission tomography. European Journal of Neuroscience, 6, 725–736.


48


Gundersen, H.J.G., Bendtsen, T.F., Korbo, L., Marcussen, N., Moeller, A., Nielsen, K., Nyengaard, Pakkenberg, B., Sörensen, F.B., Vesterby, A. and West, M.J. (1988) Some new, simple and efficient stereological methods and their use in pathological research and diagnosis. Acta Pathologica Microbiologica et Immunologica Scandinavica, 96, 379–394. Hadjikhani, N.K. and Tootell, R.B.H. (2000) Projection of rods and cones within human visual cortex. Human Brain Mapping, 9, 55–63. Harasty, J., Halliday, G.M. and Kril, J.J. (1996) Reproducible sampling regimen for specific cortical regions: application to speech-related areas. Journal of Neuroscience Methods, 67, 43–51. Haug, H. (1956) Remarks on the determination and significance of the gray cell coefficient. Journal of Comparative Neurology, 104, 473–492. Haug, H. (1980) Die Abhängigkeit der Einbettungsschrumpfung des Gehirngewebes vom Lebensalter. Verhandlungen der Anatomisches Gesellschaft, 74, 699–700. Haug, H. (1984) Macroscopic and microscopic morphometry of the human brain and cortex. A survey in the light of new results. Brain Pathology, 1, 123–149. Haug, H. (1987a) Brain sizes, surfaces, and neuronal sizes of the cortex cerebri: a stereological investigation of man and his variability and a comparison with some mammals (primates, whales, marsupials, insectivores, and one elephant). American Journal of Anatomy, 180, 126–142. Haug, H. (1987b) Die menschliche Hirnrinde—Architektur und Daten. In: E.H. Graul, S. Pütter and D. Loew (eds), Natürliche und künstliche Intelligenz, Kunstfehler—Regresse—Sterbehilfe, (series: Das Gehirn und seine Erkrankungen, Vol. 17). Iserlohn: Medice, pp. 137–164. Haxby, J.V., Horrwitz, B., Ungerleider, L.G., Maisog, J.M., Pietrini, P. and Grady, C.L. (1994) The functional organization of human extrastriate cortex: a PET-rCBF study of selective attention to faces and locations. Journal of Neuroscience, 14, 6336–6353. Hayes, T.L. and Lewis, D.A. (1996) Magnopyramidal neurons in the anterior motor speech region. Archives of Neurology, 53, 1277–1283. Henderson, G., Tomlinson, B.E. and Gibson, P.H. (1980) Cell count in human cerebral cortex in normal adults throughout life using a image analysing computer. Journal of Neurological Science, 46, 113–136. Hendrickson, A.E., Hunt, S.P. and Wu, J.Y. (1981) Immunocytochemical localization of glutamic acid decarboxylase in monkey striate cortex. Nature (London), 292, 605–607. Hendry, S.H.C., Huntsman, M., Vinuela, A., Möhler, H., DeBlas, A. and Jones, E. (1994) GABA a receptor subunit immunoreactivity in primate visual cortex: Distribution in macaques and humans and regulation by visual input in adulthood. Journal of Neuroscience, 10, 2438–2450. Herve, G. (1888) La circonvolution de Broca. Etude de morphologie cerebrale. Paris: Davy, A. Hubel, D.H. and Wiesel, T. (1972) Laminar and columnar distribution of geniculo-cortical fibers in the macaque monkey. Journal of Comparative Neurology, 146, 421–450. Hudspeth, A.J., Ruark, J.E. and Kelly, J.P. (1976) Cytoarchitectonic mapping by microdensitometry. Proceedings of the National Academy of Science, U.S.A., 73, 2928–2931. Huttenlocher, P.R. (1979) Synaptic density in human frontal cortex—developmental changes and effects of aging. Brain Research, 163, 195–205. Jansen, K.L.R., Faull, R.L.M. and Dragunow, M. (1989) Excitatory amino acid receptors in the human cerebral cortex: A quantitative auroradiographic study comparing the distribution of [3H]TCP, [3H]glycine, L-[3H]glutamate, [3H]AMPA and [3H]kainic acid binding sites. Journal of Neuroscience, 32, 587–607. Kaas, J.H. (1989) Why does the brain have so many visual areas? Journal of Cognitive Neuroscience, 1, 121–135. Kaas, J.H. (1993) The organization of visual cortex in primates: problems, conclusions, and the use of comparative studies understanding the human brain. In: B. Gulyas, D. Ottoson and P.E. Roland (eds), Functional organization of the human visual cortex. Oxford: Pergamon Press, pp. 1–12. Kennedy, D.N., Lange, N., Makris, N., Bates, J., Meyer, J. and Caviness, V.J. (1998) Gyri of the human neocortex: an MRI-based analysis of volume and variance. Cerebral Cortex, 8, 372–384. King, P.R., Gundlach, A.L. and Louis, W.J. (1995) Quantitative autoradiographic localization in rat brain of alpha 2-adrenergic and non-adrenergic I-receptor binding sites labelled by [3H]rilmenidine. Brain Research, 675, 264–278. Kleist, K. (1934) Gehirnpathologie. Leipzig: Barth. Kononova, E.P. (1935) Structural variability of the cortex cerebri. Inferior frontal gyrus in adults (Russian). In: S.A. Sarkisov and I.N. Filimonoff (eds), Annals of the Brain Research Institute. Vol. I. Moscow-Leningrad: State Press for Biological and Medical Literature, pp. 49–118. Kononova, E.P. (1938) Variability in the structure of the human cerebral cortex. Frontal lobe of adult man (Russian). In: S.A. Sarkisov and I.N. Filimonoff (eds), Annuals of the Brain Research Institute. Vol. III–IV., Moscow-Leningrad: State Press for Biological and Medical Literature, pp. 213–274. Kononova, E.P. (1949) The frontal lobe (Russian). In: S.A. Sarkisov, I.N. Filimonoff and N.S. Preobrashenskaya, The Cytoarchitecture of the Human Cortex Cerebri. Moscow: Medgiz, pp. 309–343.



49

Kötter, R. and Sommer, F.T. (2000) Global relationship between anatomical connectivity and activity propagation in the cerebral cortex. Philosophical Transactions of the Royal Society, London, series B, 355, 127–134. Larsson, J., Amunts, K., Gulyas, B., Malikovic, A., Zilles, K. and Roland, P.E. (1999) Neuronal correlates of real and illusory contour perception: functional anatomy with PET. European Journal of Neuroscience, 11, 4024–4036. Lewis, J.W. and Olavarria, J.F. (1995) Two rules for callosal connectivity in striate cortex of the rat. Journal of Comparative Neurology, 361, 119–137. Livingstone, M.S. and Hubel, D.H. (1987) Connections between layer 4B of area 17 and the thick cytochrome oxidase stripes of area 18 in the squirrel monkey. Journal of Neuroscience, 7, 3371–3377. Lungwitz, W. (1937) Zur myeloarchitektonischen Untergliederung der menschlichen Area praeoccipitalis (Area 19 Brodmann). Journal für Psychologie und Neurologie, 47, 607–638. Luppino, G., Matelli, M., Camarda, R.M., Gallese, V. and Rizzolatti, G. (1991) Multiple representations of body movements in mesial area 6 and the adjacent cingulate cortex: An intracortical microstimulation study in the macaque monkey. Journal of Comparative Neurology, 311, 463–482. Matelli, M., Luppino, G. and Rizzolatti, G. (1991) Architecture of superior and mesial area 6 and the adjacent cingulate cortex in the macaque monkey. Journal of Comparative Neurology, 311, 445–462. Merigan, W.H., Nealey, T.A. and Maunsell, J.R. (1993) Visual effects of lesions of cortical area v2 in macacques. Journal of Neuroscience, 13, 3180–3191. Morosan, P., Rademacher, J., Schleicher, A., Amunts, K., Schormann, T. and Zilles, K. (2001) Human primary auditory cortex: Cytoarchitectonic subdivisions and mapping into a special reference system. Neuroimage, 13, 684–701. Naito, E., Ehrsson, H.H., Geyer, S., Zilles, K. and Roland, P.E. (1999) Illusory arm movements activate cortical motor areas: a positron emission tomography study. Journal of Neuroscience, 19, 6134–6144. Naito, E., Kinomura, S., Geyer, S., Kawashima, R., Roland, P.E. and Zilles, K. (2000) Fast reaction to different sensory modalities activate common fields in the motor areas, but the anterior cingulate cortex is involved in the speed of reaction. Journal of Neurophysiology, 83, 1701–1709. Nakamura, H., Gattass, R., Desimone, R. and Ungerleider, L.G. (1993) The modular organization of projections from areas V1 and V2 to areas V4 and TEO in macaques. Journal of Neuroscience, 13, 3681–3691. Newsome, W.T. and Allman, J.M. (1980) Interhemispheric connections of the visual cortex in the owl monkey, Aotus trivirgatus, and the bushbaby. Journal of Comparative Neurology, 194, 209–233. Newsome, W.T., Maunsell, J.H.R. and Van Essen, D.C. (1986) Ventral posterior visual area of the macaque: visual topography and areal boundaries. Journal of Comparative Neurology, 252, 139–153. Pakkenberg, B. and Gundersen, H.J.G. (1997) Neocortical neuron number in humans: Effect of sex and age. Journal of Comparative Neurology, 384, 312–320. Paulesu, E., Frith, C.D. and Frackowiak, R.S.J. (1993) The neural correlates of the verbal component of working memory. Nature (London), 362, 342–345. Pazos, A., Probst, A. and Palacios, J.M. (1987a) Serotonin receptors in the human brain—III. Autoradiographic mapping of serotonin-1 receptors. Neuroscience, 21, 97–122. Pazos, A., Probst, A. and Palacios, J.M. (1987b) Serotonin receptors in the human brain—IV. Autoradiographic mapping of serotonin-2 receptors. Neuroscience 21, 123–139. Peterhans, E. and von der Heydt, R. (1993) Functional organization of area V2 in the alert monkey. European Journal of Neuroscience, 5, 509–524. Petrides, M., Alivisatos, B., Meyer, E. and Evans, A.C. (1993) Functional activation of the human frontal cortex during the performance of verbal working memory tasks. Proceedings of the National Academy of Science, U.S.A., 90, 878–882. Petrides, M. and Pandya, D.N. (1994) Comparative architectonic analysis of the human and the macaque frontal cortex. In: F. Boller and J. Grafman (eds), Handbook of Neuropsychology. New York: Elsevier, pp. 17–58. Rademacher, J., Caviness, J., Steinmetz, H. and Galaburda, A.M. (1993) Topographical variation of the human primary cortices: implications for neuroimaging, brain mapping, and neurobiology. Cerebral Cortex, 3, 313–329. Rajkowska, G. and Goldman-Rakic, P.S. (1995a) Cytoarchitectonic definition of prefrontal areas in the normal human cortex: I. Remapping of areas 9 and 46 using quantitative criteria. Cerebral Cortex, 5, 307–322. Rajkowska, G. and Goldman-Rakic, P.S. (1995b) Cytoarchitectonic definition of prefrontal areas in the normal human cortex: II. Variability in locations of areas 9 and 46 and relationship to the Talairach coordinate system. Cerebral Cortex, 5, 323–337. Riegele, L. (1931) Die Cytoarchitektonik der Felder der Broca’schen Region. Journal für Psychologie und Neurologie, 42, 496–514. Roe, A.W. and Tso, D.Y. (1995) Visual topography in primate v2: multiple representation across functional stripes. Journal of Neuroscience, 15, 3689–3715. Roland, P.E. (1993) Brain activation. New York: Wiley-Liss. Roland, P.E. and Zilles, K. (1994) Brain atlases – a new research tool. Trends in Neuroscience, 17, 458–467.


50


Roland, P.E. and Zilles, K. (1996a) Functions and structures of the motor cortices in human. Current Opinion in Neurobiology, 6, 773–781. Roland, P.E. and Zilles, K. (1996b) The developing European computerized human brain database for all imaging modalities. Neuroimage, 4, S39–S47. Roland, P.E. and Zilles, K. (1998) Structural divisions and functional fields in the human cerebral cortex. Brain Research Reviews, 26, 87–105. Roland, P.E., Fredriksson, J., Svensson, P., Amunts, K., Cavada, C., Hari, R., Cowey, A., Crivello, F., Geyer, S., Kostopoulos, J., Mazoyer, B., Poppelwell, D., Schleicher, A., Schormann, T., Seppa, M., Uylings, H., deVos, K. and Zilles, K. (1999) ECHBD, A Database for functional-structural and functional-functional relations. Neuroimage, 8, 128. Sanides, F. (1962) The architecture of the human frontal lobe and the relation to its functional differentiation. International Journal of Neurology, 5, 247–261. Sanides, F. (1964) The cyto-myeloarchitecture of the human frontal lobe, and its relation to phylogenetic differentiation of the cerebral cortex. Journal für Hirnforschung, 6, 269–282. Sanides, F. and Vitzthum, H.G. (1965a) Die Grenzerscheinungen am Rande der menschlichen Sehrinde. Deutsche Zeitschrift für Nervenheilkunde, 187, 708–719. Sanides, F. and Vitzthum, H.G. (1965b) Zur Architektonik der menschlichen Sehrinde und den Prinzipien ihrer Entwicklung. Deutsche Zeitschrift für Nervenheilkunde, 187, 680–707. Sarkisov, S.A., Filimonoff, I.N. and Preobrashenskaya, N.S. (1949) Cytoarchitecture of the human cortex cerebri (Russ.). Moscow: Medgiz. Schleicher, A. and Zilles, K. (1990) A quantitative approach to cytoarchitectonics: analysis of structural inhomogeneities in nervous tissue using an image analyser. Journal of Microscopy, 157, 367–381. Schleicher, A., Amunts, K., Geyer, S., Simon, U., Zilles, K. and Roland, P.E. (1995) A method of observerindependent cytoarchitectonic mapping of the human cortex. Human Brain Mapping, Suppl. 1, 77. Schleicher, A., Amunts, K., Geyer, S., Kowalski, T. and Zilles, K. (1998) An observer-independent cytoarchitectonic mapping of the human cortex using a stereological approach. Acta Stereologia, 17, 75–82. Schleicher, A., Amunts, K., Geyer, S., Morosan, P. and Zilles, K. (1999) Observer-independent method for microstructural parcellation of cerebral cortex: a quantitative approach to cytoarchitectonics. Neuroimage, 9, 165–177. Schleicher, A., Amunts, K., Geyer, S., Kowalski, T., Schormann, T., Palomero-Gallagher, N. and Zilles, K. (2000) A stereological approach to human cortical architecture: Identification and delineation of cortical areas. Journal of Chemical Neuroanatomy, 20, 31–47. Schmitz, C., Schuster, D., Niessen, P. and Korr, H. (1999) No difference between estimated mean nuclear volumes of various types of neurons in the mouse brain obtained on either isotrophic uniform random sections or conventional frontal or sagittal sections. Journal of Neuroscience Methods, 88, 71–82. Schormann, T. and Zilles, K. (1997) Limitations of the principle axes theory. IEEE Transactions in Medical Imaging, 16, 942–947. Schormann, T. and Zilles, K. (1998) Three-dimensional linear and nonlinear transformations: an integration of light microscopical and MRI data. Human Brain Mapping, 6, 339–347. Schormann, T., Dabringhaus, A. and Zilles, K. (1997) Extension of the Principal axis theory for the determination of affine transformations. In Informatik aktuell, edited by Anonymous, pp. 384–391. Springer. Schormann, T., Henn, S. and Zilles, K. (1999a) Ein sequentielles System zur Anpassung medizinischer Bilddaten. In Proceedings of the DAGM, Berlin: Springer. Schormann, T., Posse, S. and Zilles, K. (1999b) The new reference brain of the ECHBD database. Neuroimage, 9, 40. Sereno, M.I., Dale, A.M., Reppas, J.B., Kwong, K.K., Belliveau, J.W., Brady, T.I., Rosen, B.R. and Tootell, R.B.H. (1995) Borders of multiple visual areas in humans revealed by functional magnetic resonance imaging. Science (Washington), 268, 889–893. Shipp, S., Watson, J.D.G., Frackowiak, R.S.J. and Zeki, S. (1995) Retinotopic maps in human prestriate visual cortex: the demarcation of areas V2 and V3. Neuroimage, 2, 125–132. Skullerud, K. (1985) Variations in the size of the human brain. Acta Neurologica Scandinavica, 71, 1–94. Stensaas, S.S., Eddington, D.K. and Dobelle, W.H. (1974) The topography and variability of the primary visual cortex in man. Journal of Neurosurgery, 40, 755. Stephan, K.E., Hilgetag, C.C., Burns, G.A., O’Neill, M.A., Young, M.P. and Kotter, R. (2000) Computational analysis of functional connectivity between areas of primate cerebral cortex. Philosophical Transactions of the Royal Society, London series B, 355, 111–126. Strasburger, E.H. (1938) Vergleichende myeloarchitektonische Studien an der erweiterten Brocaschen Region des Menschen. Journal für Psychologie und Neurologie, 48, 477–511. Talairach, J. and Tournoux, P. (1988) Coplanar stereotaxic atlas of the human brain, Stuttgart: Thieme. Tanji, J. and Kurata, K. (1989) Changing concepts of motor areas of the cerebral cortex. Brain Development, 11, 374–377.



51

Terry, R.D., De Teresa, R. and Hansen, L.A. (1987) Neocortical cell counts in normal human adult aging. Annals of Neurology, 21, 530–539. Thompson, P., Schwartz, C. and Toga, A.W. (1996) High-resolution random mesh algorithms for creating a probabilistic 3D surface atlas of the human brain. Neuroimage, 3, 19–34. Thompson, P.M., Moussai, J., Zohoori, S., Goldkorn, A., Khan, A.A., Mega, M.S., Small, G.W., Cummings, J.L. and Toga, A.W. (1998) Cortical variability and asymmetry in normal aging and Alzheimer’s disease. Cerebral Cortex, 8, 492–509. Thoss, V.S., Piwko, O. and Hoyer, D. (1996) Somatostatin receptors in the rhesus monkey brain: localization and pharmacological characterization. Naunyn Schmiedebergs Archives of Pharmacology, 353, 648–440. Tootell, R.B.H., Silverman, M.S., De Valois, R.J. and Jacobs, G.H. (1983) Functional organization of the second cortical visual area im primates. Science, Washington, 220, 737–739. Tootell, R.B.H., Reppas, J.B., Kwong, K.K., Malach, R., Born, R.T., Brady, T.J., Rosen, B.R. and Belliveau, J.W. (1995) Functional analysis of human MT and related visual cortical areas using magnetic resonance imaging. Journal of Neuroscience, 15, 3215–3230. Tootell, R.B.H. and Taylor, J.B. (1995) Anatomical evidence for MT and additional cortical visual areas in humans. Cerebral Cortex, 5, 39–55. Tootell, R.B.H., Mendola, J.D., Hadjikhani, N.K., Ledden, P.J., Liu, A.K., Reppas, J.B., Sereno, M.I. and Dale, A.M. (1997) Functional analysis of V3A and related areas in human visual cortex. Journal of Neuroscience, 17, 7060–7078. Uylings, H.B.M., Malofeeva, L.I., Bogolepova, I.N., Amunts, K. and Zilles, K. (1999) Broca’s language area from a neuroanatomical and developmental perspective. In: P. Hagoort and C. Brown (eds), Neurocognition of language processing, Oxford: Oxford University Press, pp. 319–336. Vierordt, H. (1893) Anatomische, physiologische und physikalische Daten und Tabellen zum Gebrauch für Mediziner, Jena: Fischer Verlag. Vogt, C. (1910) Die myeloarchitektonische Felderung des menschlichen Stirnhirns. Journal für Psychologie und Neurologie, 15, 221–238. Vogt, C. and Vogt, O. (1919) Allgemeinere Ergebnisse unserer Hirnforschung. Journal für Psychologie und Neurologie, 25, 292–398. Vogt, O. (1919) Die myeloarchitektonische Felderung des menschlichen Stirnhirns. Journal für Psychologie und Neurologie, XV, 221–232. von Economo, C. and Koskinas, G.N. (1925) Die Cytoarchitektonik der Hirnrinde des erwachsenen Menschen, Berlin: Springer. Watson, J.D.G., Frackowiak, R.S.J. and Zeki, S. (1993) Functional separation of colour and motion centres in human visual cortex. In: B. Gulyas, D. Ottoson and P.E. Roland (eds), Functional organization of the human visual cortex, Oxford: Pergamon Press, pp. 317–328. Weibel, E. (1979) Stereological methods. Vol 1: Practical methods for biological morphometry, London: Academic Press. West, M.J. (1993) New stereological methods for counting neurons. Neurobiology of Aging, 14, 275–285. Westbury, C.F., Zatorre, R.J. and Evans, A.C. (1999) Quantifying variability in the planum temporale: A probability map. Cerebral Cortex, 9, 392–405. Williams, R.W. and Rakic, P. (1988) Three-dimensional counting: an accurate and direct method to estimate numbers of cells in sectioned material. Journal of Comparative Neurology, 278, 344–352. Wong-Riley, M.T., Hevner, R.F., Cutlan, R., Earnest, M., Egan, R., Frost, J. and Nguyen, T. (1993) Cytochrome oxidase in the human visual cortex: distribution in the developing and the adults brain. Visual Neuroscience, 10, 41–58. Wree, A., Schleicher, A. and Zilles, K. (1982) Estimation of volume fractions in nervous tissue with an image analyzer. Journal of Neuroscience Methods, 6, 29–43. Xiao, Y., Zych, A. and Fellemann, D.J. (1999) Segregation and convergence of functionally defined V2 thin stripe and interstripe compartment projections to area V4 of Macaques. Cerebral Cortex, 9, 792–804. Zeki, S. (1978) Functional specialization of the visual cortex of the rhesus monkey. Nature (London), 274, 423–428. Zezula, J., Cortés, R., Probst, A. and Palacios, J.M. (1988) Benzodiazepine receptor sites in the human brain: autoradiographic mapping. Neuroscience, 25, 771–795. Zilles, K. (1990) Cortex. In: G. Paxinos (ed.), The human nervous system. San Diego: Academic Press, pp. 757–802. Zilles, K. and Schleicher, A. (1980) Quantitative Analyse der laminären Struktur menschlicher Cortexareale. Verhandlungen der Anatomisches Gesellschaft, 74, 725–726. Zilles, K. and Schleicher, A. (1993) Cyto- and myeloarchitecture of human visual cortex and the periodical GABA-A receptor distribution. In: B. Gulyas, D. Ottoson and P. Roland (eds), Functional organization of the human visual cortex. Oxford: Pergamon Press, pp. 111–120. Zilles, K. and Schleicher, A. (1995) Correlative imaging of transmitter receptor distributions in human cortex. In: W. Stumpf and H. Solomon (eds), Autoradiography and correlative imaging. San Diego: Academic Press, pp. 277–307.


52


Zilles, K. and Clarke, S. (1997) Architecture, connectivity and transmitter receptors of human extrastriate visual cortex. Comparison with non-human primates. In: K.S. Rockland, J.H. Kaas and A. Peters (eds), Cerebral Cortex. Vol. 12. New York: Plenum Press, pp. 673–742. Zilles, K., Schleicher, A., Rath, M. and Bauer, A. (1988) Quantitative receptor autoradiography in the human brain. Histochemistry, 90, 129–137. Zilles, K., Werner, L., Qü, M., Schleicher, A. and Gross, G. (1991a) Quantitative autoradiography of 11 different transmitter binding sites in the basal forebrain region of the rat—evidence of heterogenity in distribution patterns. Neuroscience, 42, 473–481. Zilles, K., Gross, G., Schleicher, A., Schildgen, S., Bauer, A., Bahro, M.S., Zech, K. and Kolassa, N. (1991b) Regional and laminar distribution of alpha-adrenoreceptors and their subtypes in human and rat hippocampus. Neuroscience, 40, 307–320. Zilles, K., Hajos, F., Kalman, M. and Schleicher, A. (1991c) Mapping of glial fibrillary acidic protein-immunoreactivity in the rat forebrain and mesencephalon by computerized image analysis. Journal of Comparative Neurology, 308, 340–355. Zilles, K., Qü, M., Schröder, H. and Schleicher, A. (1991d) Neurotransmitter receptors and cortical architecture. Journal für Hirnforschung, 32, 343–356. Zilles, K., Schlaug, G., Matelli, M., Luppino, G., Schleicher, A., Qü, M., Dabringhaus, A., Seitz, R. and Roland, P.E. (1995) Mapping of human and macaque sensorimotor areas by integrating architectonic, transmitter receptor, MRI and PET data. Journal of Anatomy, 187, 515–537. Zilles, K., Schleicher, A., Langemann, C., Amunts, K., Morosan, P., Palomero-Gallagher, N., Schormann, T., Mohlberg, H., Bürgel, U., Steinmetz, H., Schlaug, G. and Roland, P.E. (1997) A quantitative analysis of sulci in the human cerebral cortex: development, regional heterogeneity, gender difference, asymmetry, intersubject variability and cortical architecture. Human Brain Mapping, 5, 218–221. Zilles, K., Werners, R., Büsching, U. and Schleicher, A. (1986) Ontogenesis of the laminar structures in area 17 and 18 of the human visual cortex. A quantitative study. Anatomy and Embryology, 174, 129–144.


4 Topographical Variability of Cytoarchitectonic Areas Jörg Rademacher Neurologische Klinik, Heinrich-Heine-Universität, Düsseldorf, Moorenstrasse 5, 40225, Düsseldorf, Germany Tel: (0049)2461-612107; FAX: (0049)211-81-18485; e-mail: [email protected]

There is growing interest in the pattern of convolutions in the human brain, in the context of modern neuroimaging studies of relationships between structure and function, because the sulcal landmarks can be reliably visualized in vivo by magnetic resonance imaging techniques. It is generally assumed that the pattern of gyri and sulci is a morphological feature of the brain which is strictly related to the structural constraints of functional organization. However, the assumption that macroanatomic topographic landmarks may serve as a guide to functional imaging is problematic, because such a system depends upon constant relationships of cytoarchitectonic field boundaries to the gyral pattern and sulci of the brain. Although consistent correlations between the positions of certain architectonic field boundaries and the primary brain sulci have been reported occasionally, the classical templates, including the Brodmann map, do not provide such information. At the very best, these templates include general guidelines for using macroanatomic landmarks that provide the framework for specific cytoarchitectonic areas. They lack information about the course and size of gyri and sulci, and do not permit prediction of how these landmarks relate to the architectonic areas. Thus, the early students of cortical cytoarchitecture do not explain the extent to which cortical areas correlate with the individual gyral and sulcal pattern. Individual anatomic variations have not been systematically charted, but there is evidence suggesting that, at least for some cytoarchitectonic areas, there is considerable variability. For practical purpose, one can distinguish “class 1 variability”, which is closely predictable from the gyral and sulcal landmarks, and “class 2 variability” which is not predictable from the visible landmarks. The latter introduces a significant error to the localization of function, if the mapping technique is based on gross anatomical landmarks. KEYWORDS: brain development, brain mapping, cerebral cortex, cytoarchitecture, gyrification, morphometry

1. INTRODUCTION The gyrencephalic human cerebral cortex is distributed over a folded cerebral surface, thereby allowing for a larger area of cortical surface in the same volume than in case of a lissencephalic brain with a smooth cerebral surface (Zilles, 1990). Early neuroanatomical and electrophysiological studies supported the hypothesis that cytoarchitectonically-defined cortical areas may represent distinct functional units, and that the cerebral sulci may coincide with the areal borders (Brodmann, 1909; Vogt and Vogt, 1919). Consequently, the Brodmann map has become the most influential anatomical reference system for the analysis of structure-function correlations in the human brain. It is used as a twodimensional template for a cytoarchitectonically-based parcellation of the cerebral cortex, which is defined by regional heterogeneities of the cellular and laminar organization. 53 © 2002 Taylor & Francis

54

Jörg Rademacher

Modern neuroimaging techniques, including positron emission tomography and functional magnetic resonance imaging, subdivide the human cerebral cortex with increasing spatial resolution. For analysis, the resulting foci of activation are related to visible macroanatomic landmarks, because even the most advanced imaging protocols do not permit direct visualization of the laminar heterogeneities which define the cytoarchitectonic pattern. It is the general assumption of many brain mapping studies that these gyral and sulcal landmarks coincide with the borders of cytoarchitectonic areas, as shown in the Brodmann map. It has also been hypothesized that the intrasulcal cortices may play a distinctive role in higher cognitive processing, because the most rapid changes in neuronal activity are frequently observed in the sulcal fundi (Markowitsch and Tulving, 1995). However, the available knowledge about the precise relationship between the topography of specific cytoarchitectonic fields and the sulcal and gyral patterns in the general population is not adequate compared to the demands of structural and functional brain mapping techniques (Rademacher et al., 1992; Zilles et al., 1995; Geyer, 1996; van Essen et al., 1998; Roland and Zilles, 1998; Schormann and Zilles, 1998; Amunts et al., 1999). Anatomic variability may obscure and distort structure-function relationships in several ways, if the range of individual macroanatomic variations (Eberstaller, 1890; Cunningham, 1892; Galaburda et al., 1987; Steinmetz et al., 1989; Ono et al., 1990; Paus et al., 1996; Penhune et al., 1996; Thompson et al., 1996; Westbury et al., 1999) and microanatomic variations (Tables 4.1 and 4.2) is underestimated. First, interindividual gross anatomic variation regarding the course and the extent of the major (primary) gyri and sulci may lead to mismatching of anatomy with function (Steinmetz et al., 1990). Caution has to be urged where map locations from a coordinate system such as the Talairach atlas (Talairach and Tournoux, 1988) or the Damasio atlas (Damasio and Damasio, 1989) are referenced to a “standard” brain. This caution is needed especially because the transfer of Brodmann areas to a three-dimensional reference system represents a “best guess” topography, derived from Brodmann’s two-dimensional template of one hemisphere and is not based on systematic anatomical data. Second, while there is considerable overlap in the cytoarchitectonic parcellation of the human brain between the published brain maps (Brodmann, 1909; von Economo and Koskinas, 1925; Sarkisov et al., 1949; Braak, 1980), there is also considerable variation (Zilles, 1990). These discrepancies between schemes of parcellation in the identification of areal borders may result from the use of descriptive non-quantitative criteria by different authors, but they can also be expected to reflect microanatomic variability. Such variation will probably increase the range of mismatching between anatomy and function. Third, interhemispheric asymmetries have been shown to exist for many macroanatomically or cytoarchitectonically defined cortical regions, both in the general population and in individuals (Geschwind and Levitsky, 1968; Galaburda et al., 1978; Falzi et al., 1982; Eidelberg and Galaburda, 1984; Steinmetz et al., 1990; Witelson and Kigar, 1992; Ide et al., 1996; Hutsler et al., 1998; Amunts et al., 1999; Rademacher et al., 1999). In contrast, the classical cytoarchitectonic maps and the Talairach atlas are based on the assumption that one hemisphere always represents the topographical mirror image of its contralateral “homologue”. As a consequence, asymmetries of functional activation are difficult to interpret. They may reflect different cognitive strategies between the hemispheres, if bilateral but topographically discrepant foci map onto different cytoarchitectonic areas. However, they may also represent identical functional units, if they simply follow asymmetries in cytoarchitectonic topography and map onto the same areas.


Topographical Variability of Cytoarchitectonic Areas

55

In the following sections evidence will be presented on the topographical relationship between the visible gyri and sulci and the borders of cytoarchitectonic areas, including developmental aspects and experimental data on gyrus formation. Essentials of gyrus formation will be described first, because they relate to the more general principles that underlie this aspect of cortical organization. Results from the making of maps are discussed thereafter. This dual strategy is suggested to represent a useful concept for imaging studies, because one may want not only to construct detailed anatomic maps, but also to characterize the principles that govern them (Friston, 1998).

2. GYRUS FORMATION Knowledge of the phylogenetic and ontogenetic mechanisms of gyrus formation may help to improve understanding of the relationship between sulcal landmarks and architectonic areal borders, as well as the biological range of variations. Several questions arise: Do these anatomic parameters vary independently or do they show a strict covariation? Are there systematic differences in the degree of correlation between distinct cortical regions? If so, does this difference reflect organizational principles, for example, a difference between the primary somatosensory areas and the association areas? Is there a fundamental difference between the primary (high incidence rate) and tertiary (low incidence rate) brain sulci with respect to cytoarchitectonic areal borders? Can cytoarchitectonic (and functional) homologies or asymmetries be deduced from the sulcal topography? To what extent are these patterns genetically determined? The ultimate answers to these questions cannot be given yet, but the increasing research interest in these issue, which is important to neuroimaging, have already provided interesting new insights. Cortical folding has been proposed to represent the brain’s solution to the problem of packing a phylogenetically increasing cortical surface into the restricted volume of the cerebral vault (Zilles et al., 1988; Welker, 1990). Similar to the general phylogenetic pattern from lissencephalic to gyrencephalic brains (Zilles et al., 1989), the human cortex changes from a smooth surface to a highly convoluted structure during ontogenesis (Chi et al., 1977; Armstrong et al., 1995). The continuous increase in gyrification begins in ontogenetic week 22 after the majority of neurones have finished their migration into the cortex. In general, the degree of cortical folding appears to be closely associated with the size of the brain, if different species are compared (Armstrong et al., 1995). Nevertheless, anatomical data from primates have shown that the size of individual cytoarchitectonic areas cannot be predicted simply by regression analysis on the basis of brain weight (Holloway, 1979) suggesting that additional intrinsic mechanisms may be active. In the human brain, interindividual differences in the size of a given cytoarchitectonic area have a wide range, up to a factor ten (Filimonoff, 1932; Stensaas et al., 1974; Galaburda et al., 1978; Rademacher et al., 1993; Rajkowska and Goldman-Rakic, 1995; Amunts et al., 1999) and obviously cannot be explained by the size of the individual brain. It has also been speculated that gyrus formation correlates with the topography of cytoarchitectonic areas and their respective borders, which tend to coincide with the course of the cerebral sulci (Welker, 1990). In this model, gyri and sulci (i.e. macroanatomic parameters) are interpreted as the expression of specific structural principles and constraints reflecting intracortical organization (Sisodiya et al., 1996). In fact, the degree of cortical folding found in adult human brains appears to be a rather constant phenomenon,


56

Jörg Rademacher

with distinct and stable local changes in the anterior-to-posterior direction, showing the highest values over the prefrontal and parieto-temporo-occipital association cortices and the lowest values over the primary sensory cortices (Zilles et al., 1988). This finding could imply that the cytoarchitectonic patterns show a similar constancy in their distribution and that their relationship to the cerebral gyri is fixed genetically. However, the formation of cytoarchitectonically defined cortical areas in the brains of mammals does not necessarily lead to sulcus formation. Lissencephalic monkeys show the same main architectonic subdivisions as do other primate brains, including the human cortex (Brodmann, 1909), and the highly convoluted brains of dolphins or whales do not reflect an increase in cytoarchitectonic areas. Phylogenetically, cytoarchitectonic areal formation and the development of a convoluted brain are not interdependent. They probably represent distinct processes with partially overlapping genetic and epigenetic mechanisms, thereby allowing for variations not only of their respective patterns, but also of their relationship to each other. Is this phylogenetic evidence paralleled by similar insights from studies of ontogenesis? Shortly after birth, the convolutedness of the human brain reaches its adult configuration while the brain itself continues to grow (Chi et al., 1977). Consequently, the postnatal changes in the degree of cortical folding match those of brain growth thereby maintaining a constant degree of convolutedness. On the basis of this observation, Armstrong et al. (1995) have made a qualitative distinction between the primary and secondary brain sulci, which together establish the degree of convolutedness that characterizes the human brain, and the tertiary sulci which appear postnatally and only maintain the degree of convolutedness that was previously established. Interestingly, only the topography of the deep and ontogenetically early primary sulci can be reliably observed in all brains (Ono et al., 1990) and this appears to be strongly determined genetically (Lohmann et al., 1999). However, the heritability of the precise overall gyral pattern may be less than 20% (Bartley et al., 1997), and surprising discordances for sulcal shape have been found among monozygotic twins (Steinmetz et al., 1995). Thus, one may postulate that an ontogenetic process exists that can produce profound morphological shifts as determined by random environmental factors without much genetic change. If this principle is also valid for the relationship between sulci and cytoarchitectonic borders, one would expect a topographical coincidence in only a few examples, i.e. by chance. In contrast, the available evidence shows that the rate of a topographical overlap between these two anatomical parameters is much higher than would be expected, if such relationship was governed by chance (Sanides, 1962; Rademacher et al., 1993; White et al., 1997a,b; Amunts et al., 1999; Geyer et al., 1999). However, dissociations between the sulcal topography and the borders of the cytoarchitectonic areas would not be surprising. In this context, it is a major challenge to analyze how genetically determined intrinsic control and non-genetically determined extrinsic control have a complementary or concurrent influence on the final amount of overlap between the cytoarchitectonic topography and the individual gyral pattern. Three standard theories have been proposed. First, the “mechanical model” of brain development of brain convolutions (Richman et al., 1975) which can explain the general degree of cortical folding, but not the distinct placement and orientation of gyri and sulci (Armstrong et al., 1995). In brief, the mechanical model assumes that the relative amount of cortical folding is the result of the mechanical forces internal to the cortex and generated by having two differently sized cortical strata (i.e. supragranular layers I–III vs granular and infragranular layers IV–VI). Second, a “tensionbased” theory of mammalian brain morphogenesis has been proposed that stresses the



57

importance of mechanical tension along axons and dendrites in the white matter (van Essen, 1997). According to this model, tension along axons can explain how and why the cortex folds in a distinct pattern. The underlying phylogenetic (and ontogenetic) principle would be to keep total axonal length low, thus contributing to the compactness of neural connectivity in the adult brain. Third, an active process termed “gyrogenesis” is postulated, whereby internal growth processes including cytoarchitectonic differentiation, ingrowth of thalamic and cortical afferents, selective neuronal death and progressive myelination move the gyral crowns outward (Rakic, 1988; Kostovic and Rakic, 1990; Welker, 1990). The concept of gyrogenesis suggests that the formation of gyri and sulci must bear a close relationship to the cytoarchitectonic parcellation of the cerebral cortex. It has also been hypothesized to explain better the individual placement, orientation, and depth of the cerebral convolutions (Armstrong et al., 1995) and this has been supported by experimental data on gyrus formation (Smart and McSherry, 1986; Ferrer et al., 1988). The comparative experimental evidence for genetic control of these mechanisms responsible for mammalian cortical development has been presented recently by Rubenstein and Rakic (1999). In agreement with the concept of gyrogenesis, it takes the complex and ordered array in functionally distinct cytoarchitectonic areas as depending on species-specific interactions between intrinsic properties of cortical cells and connectivity between cortical or subcortical structures (Rubenstein et al., 1999). Defective genes are recognized as a cause for cortical abnormalities at the microanatomic (cytoarchitecture) and macroanatomic (gyri and sulci) level (Raymond et al., 1995). For example, bilateral enucleation in the foetal monkey leads to a decrease of Brodmann area 17 (primary visual cortex) and to an increase of Brodmann area 18 (visual association cortex), paralleled by the induction of supplementary sulci and gyri of the occipital lobe (Dehay et al., 1996). The modified borders of the striate cortex often coincided with a new and deep sulcus. While these and the aforementioned results support the concept that the architectonic subdivisions and the sulcal pattern may show a high degree of anatomical (and functional) plasticity, it also supports the hypothesis that there is some degree of covariation between the cytoarchitectonic borders and the framework of sulcal landmarks. To date, it is impossible to clarify the precise impact of the genotype on these biological parameters. Quantitative analyses of the anatomical phenotype are mandatory to improve our understanding of the relationship between cytoarchitecture and macroanatomy.

3. CORTICAL MAPS A comprehensive analysis of the relationship of cytoarchitectonic areal borders to the individual gyral pattern and surrounding sulci needs to be based on larger series of brain specimens in order to consider both interhemispheric and interindividual variations. The most relevant anatomic studies which fullfill this criterion are summarized in Tables 4.1 and 4.2. Based on this criterion the present selection of cytoarchitectonic areas is focused on the primary areas, their direct neighbours, and the language regions. There is a lack of comparable data relating to other brain regions such as the superior parietal lobe, the inferomedial temporal lobe, and others. To evaluate the findings, two classes of variability were distinguished. Class 1 variability is closely predictable from visible landmarks. It reflects topographical variations which can be predicted by the framing gyri and sulci even if these landmarks are variable in a stereotactic system. It does not degrade the


58

Jörg Rademacher

confidence of mapping of a system keyed to landmarks visible in MR images. Class 2 variability, is not predictable from visible landmarks, thereby degrading the confidence of mapping. The incidence rates in % (IR) of the relevant sulci (for their topography see Figures 4.1 and 4.2) were taken from the atlas of Ono et al. (1990) which is based on 25 autopsy specimen brains. Given the individual variability described above, attention has also been payed to differences in the location of specific cytoarchitectonic areas between the classical maps. The systematic presentation follows well-established criteria of subdividing the cerebral cortex. On the basis of neuroanatomical studies in primates including the human brain, the cortex is subdivided into the frontal, parietal, temporal and occipital lobes which contain the primary sensorimotor areas and the association areas (Zilles, 1990). Cytoarchitectonic areas are labeled according to the Brodmann (1909) nomenclature (i.e. numbers), unless otherwise stated. In addition, the nomenclature of von Economo and Koskinas (1925) is given in brackets (i.e. letters). Anatomical descriptions are kept short, and extensive definitions can be found in the original literature (see Tables 4.1 and 4.2; Ono et al., 1990; Rademacher et al., 1992). 3.1. Frontal Lobe: Primary Motor Cortex 3.1.1. Cytoarchitectonic criteria and macroanatomic landmarks Area 4 (area FAg) is distinctive for the presence of giant pyramidal cells in layer V, relatively large cells in sublayer IIIc, and a low cell packing density through all layers without sharp laminar definition. Related macroanatomic landmarks (Figures 4.1 and 4.2) include the precentral gyrus which lies between the precentral (IR 100%) and the central sulci (IR 100%). On the medial hemispheric surface, the paracentral lobule extends from the ascending and descending paracentral sulci (IR 92%) anteriorly to the terminal upswing of the cingulate sulcus (IR 100%) posteriorly. 3.1.2. Class 1 variability The identification of area 4 with an observer-independent method (Schleicher et al., 1999) has shown that it always occupies the posterior bank of the precentral gyrus (Geyer et al., 1996) thereby supporting earlier observations (Rademacher et al., 1993). Its presence on the exposed surface of the lateral convexity is restricted to the dorsal-most portion of the precentral gyrus. The superior frontal sulcus represents the inferior level at which area 4 may be found on the convexity surface. There is no extension of area 4 onto the postcentral gyrus and little or no extension onto the superior frontal gyrus, so that the central and precentral sulci represent the posterior and anterior borders of dorsolateral area 4, respectively. More ventrally, area 4 is exclusively localized in the depth of the central sulcus where it shows a constant extent on its anterior bank (White et al., 1997a). Medially, area 4 is located in the central portion of the paracentral lobule, anterior to the termination of the central sulcus. The bihemispheric surface topography of area 4 shows roughly symmetric patterns (shapes) in individual brains (Rademacher et al., 1993; White et al., 1997a,b). It has also been shown that certain features visible in MR images provide reliable evidence for structural-functional organization. The functionally-defined motor hand area can be localized to a knob on the middle third of the precentral gyrus (Yousry et al., 1997) and the asymmetry in the depth of the central sulcus may represent a morphological



59

Figure 4.1. Sulci of the lateral brain surface. aar, anterior ascending ramus of the Sylvian fissure; ag, angular sulcus; ahr, anterior horizontal ramus of the Sylvian fissure; ao, anterior occipital sulcus; ce, central sulcus; if, inferior frontal sulcus; im, intermediate sulcus; ip, intraparietal sulcus; it, inferior temporal sulcus; lo, lateral occipital sulcus; par, posterior ascending ramus of the Sylvian fissure; phr, posterior horizontal ramus of the Sylvian fissure; poc, postcentral sulcus; prc, precentral sulcus; sf, superior frontal sulcus; st, superior temporal sulcus.

Figure 4.2. Sulci of the medial brain surface. ca, callosal sulcus; calc, calcarine sulcus; cc, corpus callosum; ce, central sulcus; ci, cingulate sulcus; ma, marginal sulcus; pa, paracingulate sulcus; po, parietooccipital sulcus; sp, subparietal sulcus; sr, superior rostral sulcus.


60

Jörg Rademacher

correlate of asymmetry in the size of the motor cortex and hand preference (Amunts et al., 1996). 3.1.3. Class 2 variability While the general topography of area 4 appears to be predictable from the gyral landmarks, the precise borders for example between primary motor area 4 and premotor area 6 differ between hemispheres and do not always match macrostructural landmarks (Rademacher et al., 1993; White et al., 1997b). Medially, on the paracentral lobule, the ventral border of area 4 varies considerably with respect to the cingulate sulcus and there is no constant limiting sulcal landmark for the anterior border of area 4. The ascending and descending paracentral sulci are mostly located within area 6. Laterally, the intrasulcal size of area 4 on the anterior lip of the central sulcus shows individual variations by a factor of up to two. Consequently, the total cortical volume of area 4 cannot be inferred reliably from the gyral pattern. Similarily, the subdivision of area 4 into areas 4a and 4p, which has been recognized by Geyer et al. (1996) is not indicated by sulcal borders. 3.1.4. Differences between classical maps On the dorsal portion of the precentral gyrus area 4 correlates well with Sanides’ motor area 42 (Sanides, 1962). In contrast, the ventral portions of their homologues in the maps of Brodmann and von Economo and Koskinas have been depicted with a much larger surface extent towards the Sylvian fissure. This discrepancy exceeds the range of the observed anatomic variations as described above and may result from the authors’ intention to visualize the intrasulcal extent of area 4 on the two-dimensional brain templates. Medially, area 4 may be relatively large and extends down to the cingulate sulcus (Brodmann, 1909) or it may be relatively small (i.e. 90%) of the planum temporale (von Economo and Horn, 1930). Heschl’s sulcus represents in most cases its anterior border and posteriorly, the end of the horizontal ramus of the Sylvian fissure may serve as the border for its maximal variation zone (Rademacher et al., 1999). Area 42 does not extend onto the insular cortex medially and it does not occupy the convexity of the superior temporal gyrus laterally (Figure 4.3). Area 22 covers a cortical stripe of varying extent along the lateral border of the planum temporale. Its major portion is always localized on the convexity of the superior temporal gyrus where the superior temporal sulcus represents its ventral border. The posterior end of the horizontal portion of the Sylvian fissure marks the posterior border of area 22. Anteriorly, area 22 always reaches the coronal level which passes through the anterolateral end of Heschl’s sulcus. 3.7.3. Class 2 variability The precise cytoarchitectonic borders of areas 42 and 22 do not coincide with gyral or sulcal landmarks. Bordering on the medial boundary of area 42 parainsular prokoniocortex occupies the planum temporale without a limiting sulcus (Galaburda and Sanides, 1980). Similarily, lateral and immediately adjacent to area 42, area 22 is found without a limiting sulcus. The anterior and posterior borders of area 22 are also not directly delineated by brain sulci. In individual brains, area 22 may cross the coronal levels which indicate the anterolateral end of Heschl’s sulcus or the posterior end of the horizontal segment of the Sylvian fissure. In addition, the shape of areas 42 and 22 varies considerably and does not always follow the expected rostrocaudal orientation as described by Galaburda and Sanides (1980). The cortical volumes of area 42 vary by up to a factor of five and these intersubject differences cannot generally be inferred from the size of the planum temporale (Rademacher et al., 1999). In the modern literature, gross left-right asymmetry of the planum temporale has first been described by Geschwind and Levitsky (1968). In a meta-analysis of 520 adult brains from various studies, a left area larger than the right was found in 74% of all brains, with left/right ratios ranging from 1.20 to 2.0 (values from Steinmetz et al., 1990). From their cytoarchitectonic study in four brains Galaburda et al. (1978) concluded that there is a strong correlation between asymmetry of the planum temporale and asymmetry of area 22 which explains the full structural asymmetry. However, a leftward asymmetry has also been observed for area 42 in more than two thirds of cases in another brain series while area 22 showed leftward asymmetry in only half of the brains (Rademacher et al., 1999). On the planum temporale, area 42 was considerably larger than area 22 thus indicating that macroanatomic asymmetries may not simply be reflections of the underlying asymmetry of one architectonic area. 3.7.4. Differences between classical maps In general, there is good overlap between the classical maps in the topography of area 42 on the planum temporale and area 22 on the middle and posterior third of the superior


70

Jörg Rademacher

temporal gyrus (Brodmann, 1909; von Economo and Koskinas, 1925). Anteriorly, at the level of Heschl’s gyrus and posteriorly, at the caudal end of the planum temporale, area 22 extends into the depth of the Sylvian fissure. In contrast, Galaburda and Sanides (1980) also depicted a portion of area 22 (area Tpt) which covers the parietal operculum and postulated a parietal extension of the auditory region. Topographically, this region is identical with the posterior portion of area 40 (area PFcm). Witelson et al. (1995) reported that area 22 may in fact occupy the terminal ascending segment of the Sylvian fissure. These discrepancies are further obscured by the finding of macroanatomic Sylvian fissure asymmetries and planum temporale variability which have usually not been referred to in the cytoarchitectonic maps (Witelson and Kigar, 1992; Ide et al., 1996; Westbury et al., 1999). In addition the parcellation of cytoarchitectonic areas on the planum temporale varies between different brain maps. For example, area 42 has been subdivided into three parakoniocortical areas (i.e. caudo-dorsal, internal and external) by Galaburda and Sanides (1980). The external parakoniocortex (so-defined) is supposed to share a major portion of the superior temporal gyrus with area 22. Therefore, the new parcellation scheme includes changes in the number of areas per macroanatomic unit and significant differences in the local topography which cannot be explained by the range of biological variability. Rivier and Clarke (1997) have described six auditory areas which are located in the posterior temporal region. Significant controversies over the structural organization of the auditory cortex can be expected to emerge if results from myeloarchitectonics (Hopf, 1954), pigmentoarchitectonics (Braak, 1978), and cytochrome oxidase staining (Rivier and Clarke, 1997) are included in a comparison between different areal maps of the superior temporal lobe. 3.8. Occipital Lobe: Primary Visual Cortex 3.8.1. Cytoarchitectonic criteria and macroanatomic landmarks Area 17 (area OC) is defined by its characteristic laminar composition with the granular layer IV having three distinct sublaminae IVa, IVb, and IVc (stripe of Gennari). This pattern ceases abruptly at its border with area 18. Macroanatomically, the calcarine sulcus (IR 100%) is the ventral border of the cuneus’ and the parieto-occipital sulcus (IR 100%) is its anterior border (Figure 4.2). The cuneal point marks the intersection of both sulci. The sagittal sulci of the cuneus (IR 46%) and the intralingual sulci (IR 34%) course horizontally across the cuneus and the lingual gyrus, respectively. The collateral sulcus (IR 100%) represents the boundary between lingual gyrus and fusiform gyrus. On the cerebral convexity, the lateral occipital sulcus (IR 96%) takes a sagittal course (Figure 4.1). 3.8.2. Class 1 variability Typically, area 17 is almost exclusively (>90%) localized posteriorly to the cuneal point on the medial occipital brain surface, where it lies within the region outlined by the parietooccipital sulcus anteriorly and the occipital pole posteriorly (Rademacher et al., 1993) (Figure 4.4). Bilaterally, approximately two thirds of striate cortex are buried in the calcarine sulcus, confirming earlier results (Stensaas et al., 1974). With the relative size of the intracalcarine portion of area 17 compared to the total amount of striate cortex being relatively constant, the length and depth of the calcarine sulcus can serve as an indicator for the overall size of area 17 (Filimonoff, 1932). Direct visualization and identification of



71

Figure 4.4. Surface rendering of the medial occipital lobe. Left and right (L and R) cytoarchitectonic areas 17 (red) and 18 (green); dorsal is at the top; arrowheads, parietooccipital sulcus; (modified from Amunts et al., 2000). (see Color Plate 3)

the calcarine sulcus are also mandatory for studies of structural-functional mapping, because its variability zone in a stereotactic reference system measures 2 cm in the vertical axis (Steinmetz et al., 1990). 3.8.3. Class 2 variability Given the diversity of patterning of the cuneal, lingual, collateral and lateral occipital sulci, it requires some surmise to distinguish the homologies between different hemispheres. The exact topography of area 17 with regard to the occipital gross anatomical landmarks and the absolute amount of area 17 show a considerable amount of individual variation (Filimonoff, 1932; Stensaas et al., 1974; Rademacher et al., 1993; Amunts et al., 2000). For example, functional activation which maps onto the cuneus may relate either to area 17 or to area 18, depending on the individual anatomy (Figure 4.4). Relative to the size of the cuneus, the dorsal portion of area 17 may constitute between 12% and 74% of its area (Rademacher et al., 1993). Also, the varying relationship between the cuneal and lingual portions of area 17, showing up to fivefold individual differences, cannot be inferred from the surface relief (Amunts et al., 2000). The total size of area 17 appears to be symmetrical between the hemispheres for individual brains (Rademacher et al., 1993; Amunts et al., 2000), but in a minority of cases there may be a relevant asymmetry, with differences between sides of up to 400% (Stensaas et al., 1974). Interhemispheric differences have been described for the spatial position of area 17, which is located more medially and caudally on the left than on the right. Whether area 17 extends onto the lateral convexity or not cannot be deduced from the sulcal pattern. When present, this portion usually comprises less than 10% of the total striate cortex (Rademacher et al., 1993). 3.8.4. Differences between classical maps The Brodmann map shows an area 17 which takes a sagittal course parallel to the calcarine sulcus from the cuneal point anteriorly to the occipital pole posteriorly. The von


72

Jörg Rademacher

Economo and Koskinas map shows the same anterior-to-posterior extent of area 17, but its shape is triangular with the base at the occipital pole. In contrast with the Brodmann map, which describes a close relationship between the cuneal and lingual extent of area 17 and the course of the cuneal and lingual sulci respectively, the von Economo and Koskinas map excludes such an overlap between limiting sulci and cytoarchitectonic topography. In general, the modern maps show more irregular patterns in the shape of area 17 than the classical maps (Rademacher et al., 1993; Amunts et al., 2000) (Figure 4.4). 3.9. Occipital Lobe: Visual Association Cortex 3.9.1. Cytoarchitectonic criteria and macroanatomic landmarks Areas 18 and 19 (areas OB and OA) represent larger parts of the extrastriate visual cortex. Area 18 has a very prominent layer III and a lower cell packing density than area 17. There is no stripe of Gennari. Delineation of the border between areas 18 and 19 by visual inspection alone is problematic in Nissl-stained sections (Zilles, 1990). Filimonoff (1932) reported that the most typical difference is a more compact layer III with less radial stripes in area 18. At the parietal border, the remaining radial stripes of area 19 disappear and there are larger cells in layer III. At the border with area 37 there is a characteristic increase in the cell packing density of layer V. Recently, area 18 has been analyzed on the basis of quantitative cytoarchitecture and multivariate statistics (Amunts et al., 2000). Macroanatomically, the occipitotemporal sulcus (IR 100%) marks the border between the fusiform gyrus and the inferior temporal gyrus on the ventral brain surface. On the dorsolateral brain convexity the external parieto-occipital sulcus (IR 98%) separates the superior parietal lobe and the occipital lobe. The terminal downward projection of the intraparietal sulcus (IR 100%) and the anterior occipital sulcus (IR 98%) mark the approximate border between the inferior parietal lobe and the occipital lobe (Figure 4.1). The lateral occipital (IR 96%) sulci take a horizontal or oblique course across the lateral occipital lobe and, when present, the transverse occipital sulcus (IR 62%) is characterized by its vertical extent near the occipital pole. 3.9.2. Class 1 variability Typically, areas 18 and 19 are localized dorsally and ventrally to area 17 on the medial hemispheric surface. Anteriorly, the parieto-occipital sulcus is a reliable border of area 18 (Amunts et al., 2000) (Figure 4.4) and it may also be so for area 19 (Filimonoff, 1932; Braak, 1980). Ventrally, the collateral sulcus frequently represents the border between area 19 and area 37. Considerable amounts of areas 18 and 19 are localized on the lateral brain surface near the occipital pole with the external parieto-occipital sulcus and the terminal descending segment of the intraparietal sulcus as the anterior borders of area 19 (Filimonoff, 1932). 3.9.3. Class 2 variability There are no sulcal landmarks on the ventral and inferolateral brain surface which represent the precise macroanatomic or cytoarchitectonic border between the occipital and the temporal cortices. On the lateral brain surface, the lateral occipital sulci do not represent



73

useful landmarks for the borders of areas 18 and 19 (Filimonoff, 1932; Amunts et al., 2000). Similarily, the anterior extent of area 19 onto the parieto-occipital sulcus and the intraparietal sulcus varies considerably (Filimonoff, 1932). For area 18, topographical variations appear to be most prominent on the medial brain surfaces (Amunts et al., 2000). Compared to area 17, the topographical variability of area 18 is considerably larger (Amunts et al., 2000) and the extent of area 18 is underestimated in the stereotaxic atlas of Talairach and Tournoux (1988). On the ventral brain surface, the occipitotemporal sulcus does not appear to coincide systematically with cytoarchitectonic areal borders (Braak, 1980). The mean volumes of area 18 do not differ significantly between the hemispheres (Amunts et al., 2000), but individual size differences of up to 100% have been reported (Heinze, 1954). It is not known whether the consistent differences in gyrification between the occipital association cortex (higher degree of cortical folding) and the occipital pole (lower degree of cortical folding) may serve as a marker for the indirect localization of visual association areas 18 and 19 (Gebhard et al., 1993). 3.9.4. Differences between classical maps Typically, the parieto-occipital sulcus is depicted as the anterior border of the visual association cortices on the medial brain surface and the external parieto-occipital sulcus and the posterior intraparietal sulcus are presented as the anterior border of the visual association cortices on the lateral brain surface. However, there are also differences between the classical studies (Brodmann, 1909; von Economo and Koskinas, 1925; Sarkisov et al., 1949). While the Brodmann map shows that the superior half of the parieto-occipital sulcus is bordered by area 19 and its inferior half is bordered by area 18, the other maps indicate that more than 80% of the length of the parieto-occipital sulcus borders area 19. The relative size of area 18 with respect to area 19 on the cuneus shows all possible variations including symmetry (Sarkisov et al., 1949), asymmetry in favour of area 18 (Brodmann, 1909) and asymmetry in favour of area 19 (von Economo and Koskinas, 1925). These differences and asymmetries can be expected to result from the normal range of variations (Filimonoff, 1932; Amunts et al., 2000). While the Brodmann map shows the sagittal lingual sulcus as the dorsal border of area 18 and the collateral sulcus as the ventral border of area 18, such a clear relationship between the sulcal landmarks and the cytoarchitectonic borders cannot be deduced from the other classical or modern maps (Amunts et al., 2000). Even more important, the concept of a single area 19 is not supported by the modern literature on human brain mapping (Clarke and Miklossy, 1990; Zilles, 1990; Zeki, 1993). Braak (1980) subdivided the same region into 10 pigmentoarchitectonically defined areas. With few exceptions, the borders of the occipital association cortices do not show a systematic relationship to the local sulcal pattern (Amunts et al., 2000).

4. ANATOMIC LANDMARKS AND FUNCTIONAL ZONES Since cortical areas reflect the organization of the cerebral cortex, the issue of parcellating the cortex is fundamental for human brain mapping (Roland and Zilles, 1994). As described above, the gyral and sulcal landmarks frequently indicate the approximate location of a cortical area in the individual brain (class 1 variability), but the precise cytoarchitectonic borders do not match reliably with gyral or sulcal landmarks (class 2 variability).


74

Jörg Rademacher

The aforementioned data from various brain maps indicate that neither do the macro- and microanatomic parameters vary independently nor do they show a strict covariation. In this context, there is a fundamental difference between the primary (high incidence rate) and tertiary (low incidence rate) brain sulci with respect to cytoarchitectonic areal borders. With varying frequencies, only the former showed relevant relationships with areal borders. The data also appeared to reflect other systematic differences, in that there was a higher degree of class 1 variability for the primary somatosensory areas, and a higher degree of class 2 variability for the association areas. A higher biological variability of the association cortices compared to relatively strict patterns for the primary cortices may reflect a useful ontogenetic principle which allows for competition in the final development of individual cortical organization. Interindividual discrepancies have been observed for various anatomic parameters such as areal topography, size, laterality and parcellation. Between individual brains these differences could vary up to a factor 10, the functional implications of such enormous variations being unknown. In this context, it seems to be at least problematic, if cytoarchitectonic and functional homologies or asymmetries are to be deduced from the sulcal topography and the size of regions of interest as defined macroanatomically. Functional imaging studies have traditionally relied on a “standard” topography derived from a single brain (Damasio and Damasio, 1989) or on stereotactic reference systems (Talairach and Tournoux, 1988) for estimating the location of regions of interest. Substantial anatomical variation must be ignored by these approaches, with inevitable inconsistencies and mismatching between anatomy and function (Steinmetz et al., 1990). The significance of intersubject variability in the topography of functional signal changes remains unclear, because it may reflect either individual differences in cognitive strategy, or in anatomic topography. Obviously, a probabilistic atlas of the human brain including macroanatomic and microanatomic population maps could account for the variance in position of structures between individuals and hemispheres (Roland and Zilles, 1994; Maziotta et al., 1995; Zilles et al., 1995). Population maps are defined as three-dimensional representations of cytoarchitectonic areas in standard anatomical format in a population of subjects. Probabilistic cytoarchitectonic maps provide overlay maps for each area showing the degree of interindividual microstructural variability in any point of the brain’s threedimensional space (Morosan et al., 1996; Amunts et al., 1999, 2000; Geyer et al., 1999; Rademacher et al., 1999). The hypothesis that the cerebral sulcal patterns provide practically no information about the underlying anatomical organization of the neocortex (Killackey, 1995) is not supported by these studies. However, also cytoarchitectonic and macroanatomic parcellations are only interesting if they can be attributed to cerebral function. For example, many brains show asymmetries in the size of a multitude of gyri and of cytoarchitectonic areas. Despite these findings, the mean asymmetry scores are not so frequently asymmetrical, and the large variability of individual side differences indicates that cortical morphological asymmetry may be present even in the absence of clear functional asymmetry (Hutsler et al., 1998). Other problems relate to the homologies between different parcellation systems. The cytoarchitectonically based Brodmann map (1909) depicts about 50 cortical areas while the myeloarchitectonically based Vogt map shows up to 200 cortical areas (Vogt and Vogt, 1919). Similarily, it is not known how these subdivisions relate to those of Braak’s pigmentoarchitectonic method (Braak, 1980). In the case of the human visual association



75

areas, the classical cortical maps do not describe as many subdivisions as in the wellestablished map of the macaque (Zeki, 1993). In contrast, functional and anatomical (Clarke and Miklossy, 1990) evidence suggests that the human primary visual cortex (V1) is also surrounded by multiple functionally distinct association areas which can be mapped onto the gyral pattern and the primary sulci. The visual colour area (V4) maps onto the posterior portion of the fusiform gyrus (Allison et al., 1993) and the visual motion area (V5) bears a constant relationship to the posterior segment of the inferior temporal sulcus and the lateral occipital sulcus (Watson et al., 1993). This shows that discrepancies exist between the results of any pair of mapping techniques. Therefore, it seems reasonable to combine cytoarchitectonic and macroanatomic parcellation systems in order to attribute functional properties to distinct cortical areas.

REFERENCES Allison, T., Begleiter, A., McCarthy, G., Roessler, E., Nobre, A.C. and Spencer, D.D. (1993) Electrophysiological studies of color processing in human visual cortex. Electroencephalography and Clinical Neurophysiology, 88, 343–355. Amunts, K., Schlaug, G., Schleicher, A., Steinmetz, H., Dabringhaus, A., Roland, P.E. and Zilles, K. (1996) Asymmetry in the human motor cortex and handedness. NeuroImage, 4, 216–222. Amunts, K., Schleicher, A., Bürgel, U., Mohlberg, H., Uylings, H.B.M. and Zilles, K. (1999) Broca’s region revisited: cytoarchitecture and intersubject variability. Journal of Comparative Neurology, 412, 319–341. Amunts, K., Malikovic, A., Mohlberg, H., Schormann, T. and Zilles, K. (2000) Brodmann’s areas 17 and 18 brought into stereotaxic space—Where and how variable? NeuroImage, 11, 66–84. Armstrong, E., Schleicher, A., Omran, H., Curtis, M. and Zilles, K. (1995) The ontogeny of human gyrification. Cerebral Cortex, 5, 56–63. Bartley, A.J., Jones, D.W. and Weinberger, D.R. (1997) Genetic variability of human brain size and cortical gyral patterns. Brain, 120, 257–269. Braak, H. (1978) On magnopyramidal temporal fields in the human brain—probable morphological counterparts of Wernicke’s sensory speech region. Anatomy and Embryology, 152, 141–169. Braak, H. (1979) The pigment architecture of the human frontal lobe. I. Precentral, subcentral and frontal region. Anatomy and Embryology, 157, 35–68. Braak, H. (1980) Architectonics of the Human Telencephalic Cortex. Berlin: Springer-Verlag. Brodmann, K. (1909) Vergleichende Lokalisationslehre der Grosshirnrinde. Leipzig: Barth. Chi, J.G., Dooling, E.C. and Gilles, F.H. (1977) Gyral development of the human brain. Annals of Neurology, 1, 86–93. Clarke, S. and Miklossy, J. (1990) Occipital cortex in man: organization of callosal connections, related myeloand cytoarchitecture, and putative boundaries of functional visual areas. Journal of Comparative Neurology, 298, 188–214. Clarke, S. and Rivier, F. (1998) Compartments within human primary auditory cortex: evidence from cytochrome oxidase and acetylcholinesterase staining. European Journal of Neuroscience, 10, 741–745. Cunningham, D.J. (1892) Contribution to the Surface Anatomy of the Cerebral Hemispheres. Dublin: Royal Irish Academy. Damasio, H. and Damasio, A.R. (1989) Lesion Analysis in Neuropsychology. New York: Oxford University Press. Dehay, C., Giroud, P., Berland, M., Killackey, H. and Kennedy, H. (1996) Contribution of thalamic input to the specification of cytoarchitectonic cortical fields in the primate: effects of bilateral enucleation in the fetal monkey on the boundaries, dimensions, and gyrification of striate and extrastriate cortex. Journal of Comparative Neurology, 367, 70–89. Eberstaller, O. (1890) Das Stirnhirn. Ein Beitrag zur Anatomie der Oberfläche des Gehirns. Wien: Urban & Schwarzenberg. Eidelberg, D. and Galaburda, A.M. (1984) Inferior parietal lobule. Divergent architectonic asymmetries in the human brain. Archives of Neurology, 41, 843–852. Falzi, G., Perrone, P. and Vignolo, L.A. (1982) Right-left asymmetry in anterior speech region. Archives of Neurology, 39, 239–240. Ferrer, I., Hernandez-Marti, M., Bernet, E. and Galofre, E. (1988) Formation and growth of the cerebral convolutions. I. Postnatal development of the median-suprasylvian gyrus and adjoining sulci in the cat. Journal of Anatomy, 160, 89–100.


76

Jörg Rademacher

Filimonoff, I.N. (1932) Über die Variabilität der Grosshirnrindenstruktur. Mitteilung II: Regio occipitalis beim erwachsenen Menschen. Journal of Psychology and Neurology, 44, 1–96. Friston, K.J. (1998) Imaging neuroscience: principles or maps? Proceedings of the National Academy of Sciences U.S.A., 95, 796–802. Galaburda, A.M. (1980) La région de Broca: observations anatomiques faites un siècle après la mort de son découvreur. Revue Neurologique, 136, 609–616. Galaburda, A. and Sanides, F. (1980) Cytoarchitectonic organization of the human auditory cortex. Journal of Comparative Neurology, 190, 597–610. Galaburda, A.M., Sanides, F. and Geschwind, N. (1978) Human brain: cytoarchitectonic left-right asymmetries in the temporal speech region. Archives of Neurology, 35, 812–817. Galaburda, A.M., Corsiglia, J., Rosen, G.D. and Sherman, G.F. (1987) Planum temporale asymmetry, reappraisal since Geschwind and Levitsky. Neuropsychologia, 25, 853–868. Gebhard, R., Zilles, K. and Armstrong, E. (1993) Gross anatomy and gyrification of the occipital cortex in human and non-human primate. In: B. Gulyas, D. Ottoson and P.E. Roland (eds), Functional Organization of the Human Visual Cortex. Oxford: Pergamon Press, pp. 101–109. Geschwind, N. and Levitsky, W. (1968) Human brain: Left-right asymmetries in temporal speech region. Science, Washington, 161, 186–187. Geyer, S., Ledberg, A., Schleicher, A., Kinomura, S., Schormann, T., Bürgel, U., Klingberg, T., Larsson, J., Zilles, K. and Roland, P.E. (1996) Two different areas within the primary motor cortex of man. Nature, London, 382, 805–807. Geyer, S., Schleicher, A. and Zilles, K. (1999) Areas 3a, 3b, and 1 of human primary somatosensory cortex. NeuroImage, 10, 63–83. Heinze, G. (1954) Zytoarchitektonische Untergliederung der Area occipitalis. Journal für Hirnforschung, 1, 173–189. Holloway, R.L. (1979) Brain size, allometry, and reorganization: toward a synthesis. In: M.E. Hahn, C. Jensen and B.C. Dudek (eds), Development and Evolution of Brain Size. New York: Academic Press, pp. 59–88. Hopf, A. (1954) Die Myeloarchitektonik des Isocortex temporalis beim Menschen. Journal für Hirnforschung, 1, 208–279. Hutsler, J.J., Loftus, W.C. and Gazzaniga, M.S. (1998) Individual variation of cortical surface area asymmetries. Cerebral Cortex, 8, 11–17. Ide, A., Rodríguez, E., Zaidel, E. and Aboitiz, F. (1996) Bifurcation patterns in the human Sylvian fissure: hemispheric and sex differences. Cerebral Cortex, 6, 717–725. Killackey, H.P. (1995) Evolution of the human brain: a neuroanatomical perspective. In: M.S. Gazzaniga (ed.), The Cognitive Neurosciences. Cambridge: MIT Press, pp. 1243–1253. Kononova, E.P. (1949) The frontal lobe [in Russian]. In: S.A. Sarkisov, I.N. Filimonoff and N.S. Preobrashenskaya (eds), The Cytoarchitecture of the Human Cortex Cerebri, Moscow: Medgiz, pp. 309–343. Kostovic, I. and Rakic, P. (1990) Developmental history of the transient subplate zone in the visual and somatosensory cortex of the macaque monkey and human brain. Journal of Comparative Neurology, 297, 441–470. Lohmann, G., von Cramon, Y. and Steinmetz, H. (1999) Sulcal variability of twins. Cerebral Cortex, 9, 754–763. Markowitsch, H.J. and Tulving, E. (1995) Cognitive processes and cerebral cortical fundi. NeuroReport, 15, 413–418. Mazziotta, J.C., Toga, A.W., Evans, A., Fox, P. and Lancaster, J. (1995) A probabilistic atlas of the human brain: theory and rationale for its development. NeuroImage, 2, 89–101. Morosan, P., Schleicher, A., Schormann, T. and Zilles, K. (1996) Cytoarchitectonic mapping of cortical areas on the first transverse temporal gyrus and intersubject variability. NeuroImage, 3, S141. Ono, M., Kubik, S. and Abernathey, C.D. (1990) Atlas of the Cerebral Sulci. Stuttgart-New York: Georg Thieme Verlag. Paus, T., Tomaiuolo, F., Otaky, N., MacDonald, D., Petrides, M., Atlas, J., Morris, R. and Evans, A.C. (1996) Human cingulate and paracingulate sulci: pattern, variability, asymmetry, and probabilistic map. Cerebral Cortex, 6, 207–214. Penhune, V.B., Zatorre, R.J., MacDonald, J.D. and Evans, A.C. (1996) Interhemispheric anatomical differences in human primary auditory cortex: probabilistic mapping and volume measurement from magnetic resonance scans. Cerebral Cortex, 6, 661–672. Pfeifer, R.A. (1920) Myelogenetisch-anatomische Untersuchungen über das kortikale Ende der Hörleitung. Abhandlungen der mathematisch physikalischen Klasse der sächsischen Akademie der Wissenschaften, 37, 1–54. Rademacher, J., Galaburda, A.M., Kennedy, D.N. and Caviness, Jr., V.S. (1992) Human cerebral cortex: localization, parcellation, and morphometry with magnetic resonance imaging. Journal of Cognitive Neuroscience, 4, 352–374. Rademacher, J., Caviness, Jr., V.S., Steinmetz, H. and Galaburda, A.M. (1993) Topographical variation of the human primary cortices and its relevance to brain mapping and neuroimaging studies. Cerebral Cortex, 3, 313–329.



77

Rademacher, J., Werner, C., Schleicher, A. and Zilles, K. (1999) Planum temporale and language: cytoarchitectonic variability. NeuroImage, 9, S1045. Rajkowska, G. and Goldman-Rakic, P.S. (1995) Cytoarchitectonic definition of prefrontal areas in the normal human cortex. II. variability in locations of areas 9 and 46 and relationship to the Talairach coordinate system. Cerebral Cortex, 5, 323–337. Rakic, P. (1988) Specification of cerebral cortical areas. Science, Washington, 241, 170–176. Raymond, A.A., Fish, D.R., Sisodiya, S.M., Alsanjayri, N., Stevens, J.M. and Shorvon, S.D. (1995) Abnormalities of gyration, heterotopias, tuberous sclerosis, focal cortical dysplasia, microdysgenesis, dysembryoplastic neuroepithelial tumour and dysgenesis of the archicortex in epilepsy. Clinical, EEG and neuroimaging features in 100 adult patients. Brain, 118, 629–660. Richman, D.P., Stewart, R.M., Hutchison, J.W. and Caviness, V.S. (1975) Mechanical model of brain convolutional development. Science, Washington, 189, 18–21. Riegele, L. (1931) Die Cytoarchitektonik der Felder der Brocaschen Region. Journal of Psychology and Neurology, 42, 496–514. Rivier, F. and Clarke, S. (1997) Cytochrome oxidase, acetylcholinesterase, and NADPH-diaphorase staining in human supratemporal and insular cortex: evidence for multiple auditory areas. NeuroImage, 6, 288–304. Roland, P.E., and Zilles, K. (1994) Brain atlases—A new research tool. Trends in Neurosciences, 17, 458–467. Roland, P.E. and Zilles, K. (1998) Structural divisions and functional fields in the human cerebral cortex. Brain Research Brain Research Reviews, 26, 87–105. Rubenstein, J.L.R., Anderson, S., Shi, L., Miyashita-Lin, E., Bulfone, A. and Hevner, R. (1999) Genetic control of cortical regionalization and connectivity. Cerebral Cortex, 9, 524–532. Rubenstein, J.L.R. and Rakic, P. (1999) Genetic control of cortical development. Cerebral Cortex, 9, 521–523. Sanides, F. (1962) Die Architektonik des menschlichen Stirnhirns. Berlin: Springer-Verlag. Sarkisov, S.A., Filimonoff, I.N. and Preobrashenskaya, N.S. (1949) Cytoarchitecture of the Human Cortex Cerebri [in Russian]. Moscow: Medgiz. Schleicher, A., Amunts, K., Geyer, S., Simon, U., Zilles, K. and Roland, P.E. (1999) Observer-independent method for microstructural parcellation of cerebral cortex: a quantitative approach to cytoarchitectonics. NeuroImage, 9, 165–177. Schormann, T. and Zilles, K. (1998) Three-dimensional linear and nonlinear transformations: an integration of light microscopical and MRI data. Human Brain Mapping, 6, 339–347. Schulze, H.A.F. (1962) Quantitative Untersuchungen zur Frage der individuellen Variation und der Hemisphärendifferenzen der corticalen Areale des unteren Parietalläppchens. Journal für Hirnforschung, 5, 345–376. Sisodiya, S., Free, S., Fish, D. and Shorvon, S. (1996) MRI-based surface area estimates in the normal adult human brain: evidence for structural organisation. Journal of Anatomy, 188, 425–438. Smart, I.H. and McSherry, G.M. (1986) Gyrus formation in the cerebral cortex of the ferret. II. Description of the internal histological changes. Journal of Anatomy, 147, 27–43. Steinmetz, H., Rademacher, J., Huang, Y., Hefter, H., Zilles, K., Thron, A. and Freund, H.-J. (1989) Cerebral asymmetry: MR planimetry of the human planum temporale. Journal of Computer Assisted Tomography, 13, 996–1005. Steinmetz, H., Fürst, G. and Freund, H.-J. (1990) Variation of perisylvian and calcarine anatomic landmarks within stereotaxic proportional coordinates. American Journal of Neuroradiology, 11, 1123–1130. Steinmetz, H., Rademacher, J., Jäncke, L., Huang, Y., Thron, A. and Zilles, K. (1990) Total surface of temporoparietal intrasylvian cortex: diverging left-right asymmetries. Brain and Language, 39, 357–372. Steinmetz, H., Herzog, A., Schlaug, G., Huang, Y. and Jäncke, L. (1995) Brain (a)symmetry in monozygotic twins. Cerebral Cortex, 5, 296–300. Stengel, E. (1930) Morphologische und cytoarchitektonische Studien über den Bau der unteren Frontalwindung bei Normalen und Taubstummen. Ihre individuellen und Seitenunterschiede. Zeitschrift für Experimentelle und Angewandte Psychologie, 130, 630–677. Stensaas, S.S., Eddington, D.K. and Dobelle, W.H. (1974) The topography and variability of the primary visual cortex in man. Journal of Neurosurgery, 40, 747–755. Talairach, J. and Tournoux, P. (1988) Coplanar Stereotaxic Atlas of the Human Brain. Stuttgart: Thieme. Thompson, P.M., Schwartz, C., Lin, R.T., Khan, A.A. and Toga, A.W. (1996) Three-dimensional statistical analysis of sulcal variability in the human brain. Journal of Neuroscience, 16, 4261–4274. van Essen, D.C. (1997) A tension-based theory of morphogenesis and compact wiring in the central nervous system. Nature, London, 385, 313–318. van Essen, D.C., Drury, H.A., Joshi, S. and Miller, M.I. (1998) Functional and structural mapping of human cerebral cortex: solutions are in the surfaces. Proceedings of the National Academy of Sciences, U.S.A., 95, 788–795. Vogt, O. and Vogt, C. (1919) Ergebnisse unserer Hirnforschung. Journal of Psychology and Neurology, 25, 277–462. von Economo, C. and Koskinas, G.N. (1925) Die Cytoarchitektonik der Hirnrinde des erwachsenen Menschen. Berlin: Springer Verlag.


78

Jörg Rademacher

von Economo, C. and Horn, L. (1930) Über Windungsrelief, Maße und Rindenarchitektonik der Supratemporalfläche, ihre individuellen und ihre Seitenunterschiede. Zeitschrift für Neurologie und Psychiatrie, 130, 678–757. Watson, J.D.G., Myers, R., Frackowiack, R.S.J., Hajnal, J.V., Woods, R.P., Mazziotta, J.C., Shipp, S. and Zeki, S. (1993) Area V5 of the human brain: evidence from a combined study using positron emission tomography and magnetic resonance imaging. Cerebral Cortex, 3, 79–94. Welker, W. (1990) Why does cerebral cortex fissure and fold? A review of determinants of gyri and sulci. In: E.G. Jones and A. Peters (eds), Cerebral Cortex, Vol 8B, Comparative Structure and Evolution of Cerebral Cortex, Part II. New York: Plenum Press, pp. 3–136. Westbury, C.F., Zatorre, R.J. and Evans, A.C. (1999) Quantifying variability in the planum temporale: a probability map. Cerebral Cortex, 9, 392–405. White, L.E., Andrews, T.J., Hulette, C., Richards, A., Groelle, M., Paydarfar, J. and Purves, D. (1997a) Structure of the human sensorimotor system. I: morphology and cytoarchitecture of the central sulcus. Cerebral Cortex, 7, 18–30. White, L.E., Andrews, T.J., Hulette, C., Richards, A., Groelle, M., Paydarfar, J. and Purves, D. (1997b) Structure of the human sensorimotor system. II: Lateral symmetry. Cerebral Cortex, 7, 31–47. Witelson, S.F. and Kigar, D.L. (1992) Sylvian fissure morphology and asymmetry in men and women: bilateral differences in relation to handedness in men. Journal of Comparative Neurology, 323, 326–340. Witelson, S.F., Glezer, I.I. and Kigar, D.L. (1995) Women have greater density of neurons in posterior temporal cortex. Journal of Neuroscience, 15, 3418–3428. Yousry, T.A., Schmid, U.D., Alkadhi, H., Schmidt, D., Peraud, A., Buettner, A. and Winkler, P. (1997) Localization of the motor hand area to a knob on the precentral gyrus. A new landmark. Brain, 120, 141–157. Zeki, S. (1993) The multiple visual areas of the cerebral cortex. In: S. Zeki (ed.), A Vision of the Brain. Oxford: Blackwell Scientific Publications, pp. 87–93. Zilles, K. (1990) Cortex. In: G. Paxinos (ed.), The Human Nervous System. San Diego: Academic Press, pp. 757–802. Zilles, K., Armstrong, E., Schleicher, A. and Kretschmann, H.-J. (1988) The human pattern of gyrification in the cerebral cortex. Anatomy and Embryology, 179, 173–179. Zilles, K., Armstrong, E., Moser, K.H., Schleicher, A. and Stephan, H. (1989) Gyrification in the cerebral cortex of primates. Brain, Behavior and Evolution, 34, 143–150. Zilles, K., Schlaug, G., Matelli, M., Luppino, G., Schleicher, A., Qü., Seitz, R. and Roland, P.E. (1995) Mapping of human and macaque sensorimotor areas by integrating architectonic, transmitter receptor, MRI and PET data. Journal of Anatomy, 187, 515–537.


5 Mapping of Human Brain Function by Neuroimaging Methods Rüdiger J. Seitz Department of Neurology, Heinrich-Heine-University Düsseldorf Correspondence: Department of Neurology, University Hospital Düsseldorf Moorenstraße 5, D-40225 Düsseldorf Tel: 0049-211 81-18974; FAX: 0049-211 81-18485 e-mail: [email protected]

This chapter gives an account of functional neuroanatomy of the human brain as revealed by neuroimaging methods. Metabolic autoradiography and optical imaging are the foundations of functional brain mapping, providing robust insights into the topography and temporal dynamics of cerebral metabolism in laboratory animals. The tomographic imaging methods provide excellent localising information about activation-related haemodynamic changes, but cannot resolve the temporal evolution of the underlying electromagnetic brain activity. Further, imaging of human brain function is hampered by inter-subject variability of brain structure and function as well as by the “partial volume effect” inherent in tomographic imaging. Nevertheless, the neuroimaging techniques allow one to study brain function non-invasively in healthy people, making feasible the construction of a physiological atlas of the different aspects of brain function. New technical developments have opened avenues for exploiting the temporal aspect of brain activity and analysing functional– structural relationships with reference to microanatomy. Examples will be described showing a good correspondence of specific activation studies in healthy subjects to lesion studies in patients with circumscribed neurological deficits. In contrast, activations which are abnormal in terms of quantitation and localisation have been observed in pathological conditions, suggesting that cerebral reorganisation can affect functional brain maps. Finally, evidence will be presented showing that the cerebral activations are influenced by learning and follow remarkable temporal modulations upon task performance. It is concluded that, the cortical representations of function appear as distributed computation nodes with widespread cortical and subcortical connections, processing information in task-related networks. KEYWORDS: autoradiography, brain activation, brain lesions, cerebral blood flow, cerebral metabolizm, cerebral reorganisation, deoxyhaemoglobin, electromagnetic activity, optical imaging, partial volume effect, structural variability, tomographic imaging

1. INTRODUCTION Functional neuroimaging is a powerful technology for mapping human brain function, and has undergone enormous development during the past decade (Fox, 1997; Seitz et al., 2000). Most widely used are measurements of stimulation-related haemodynamic changes. These changes can be assessed with measurements of the regional cerebral blood flow (rCBF) using positron emission tomography (PET), and of the blood oxygenation level-dependent changes (BOLD) using functional magnetic resonance imaging (fMRI). These tomographic 79 © 2002 Taylor & Francis

80

Rüdiger J. Seitz

imaging tools can localize brain activity changes with relatively good spatial resolution of approximately 5 to 9 mm (Frackowiak et al., 1994; Calamante et al., 1999). The temporal resolution of PET and fMRI, however, is relatively poor, being in the range of approximately 6 s to 1 min, due, respectively, to the tracer kinetics and the haemodynamic characteristics of the measurements. Nevertheless, the reconstructed tomographic imaging data allow one to detect activity changes occurring simultaneously in different parts of the brain, including the different parts of the cerebral cortex, subcortical structures as the basal ganglia and thalamus, and the cerebellum. It should be born in mind, however, that the observed haemodynamic changes represent only indirect measures of brain activity. Although, under physiological conditions, there is a tight coupling of activation-related metabolic and haemodynamic changes to increases in neural activity (Fox et al., 1984; Blomqvist et al., 1994; Bandettini et al., 1997; Hoge et al., 1999), bioelectric neural activity has a time-course in the range of several milliseconds, faster than the haemodynamic measures by three orders of magnitude. Therefore, one of the assumptions underlying functional imaging with PET and fMRI is that a state of activation has to be kept constant over a sufficiently long period of time in order to capture functional changes in the different parts of the brain, during a condition approaching a steady state. Bioelectric activity of the human brain can be recorded directly from the surface of the head using electroencephalography (EEG) and magnetoencephalography (MEG), the latter measuring the magnetic fields induced by electrical current flow. Temporal resolution of these techniques lies in the range of milliseconds, optimally reflecting the dynamics of brain activity (Hari and Lounasmaa, 1989; Näätänen et al., 1994). In comparison, spatial resolution is relatively poor, being determined by the number and distribution of the recording electrodes or sensors covering the head. The electrical potentials or magnetic fields recorded on the surface of the head, respectively with EEG and MEG, do not however reflect the localization of the real electrical activity in the brain, because the regionally-varying degrees of volume conductance in the cerebrospinal fluid compartment, the meninges, and in particular the skull severely distort the recorded data. Therefore, spatial analysis of the recorded data is based on biomathematical models that explain the data, recorded from the surface of the brain statistically, in terms of intracerebral sources (Wood et al., 1985; Romani and Rossini, 1988; Scherg, 1990; Kristeva et al., 1991; Snyder, 1991; de Peralta et al., 1997). In other words, the uncertainty of localization of the calculated source, the so-called inverse problem, critically depends on the model assumptions of head shape and volume conduction. Recently, information obtained from structural magnetic resonance imaging (MRI) has been used to create realistic head models for the analysis of the cortical generators of bioelectric activity as recorded with EEG (Dale and Sereno, 1993; Gevins et al., 1994; Marin et al., 1998). Magnetic fields are virtually unaffected by volume conduction. Therefore, even sources as deep in the brain as the thalamus or the hippocampal formation can be picked up by MEG (Ribary et al., 1991; Ebersole, 1997). However, dependent on the type of measuring devices, radial or tangential magnetic fields remain undetected in the MEG measurements, and this can profoundly effect the interpretation of the recorded data (Hämäläinen et al., 1993). Furthermore, due to the limited statistical power inherent in the biomathematical models, only a limited number of equivalent dipoles can be identified to explain the recorded data. In consequence, either well-determined experimental stimuli, such as electrical stimuli which evoke potentials, or simple, phase-locked sensory stimuli or movements have been studied extensively in brain research. More recently, localization of functional activity changes, as demonstrated by


Functional Brain Mapping

81

rCBF or fMRI measurements, as well as anatomical constraints, have been used for interpreting as well as restricting the “search space” in MEG recordings (Heinze et al., 1994; George et al., 1995; Gerloff et al., 1996; Kinsces et al., 1999; Korvenoja et al., 1999). Nevertheless, the basic assumption underlying these measures is that the task-specific, regional neuronal activity has the same temporal and spatial pattern across successive trials during the event-related recordings of EEG and MEG. Thereby, these methods will allow one to capture the temporal sequence of events in the brain. In spite of their methodological limitations, the neuroimaging tools can provide new insights into the topographic and temporal organization of human brain functions. Changes of brain activity were originally conceptualized as increased activity in a taskspecific stimulation condition, as compared to a specific control condition (Raichle, 1987). Later, task-specific decreases of brain activity also came into focus (Seitz and Roland, 1992a; Drevets et al., 1995; Shulman et al., 1997). Such categorical comparisons were based on the simplified model of a hierarchical implementation of activity in the human brain, that could be disentangled by psychological subtraction (Petersen et al., 1988; Kosslyn et al., 1995). Recently, more complex experimental designs have been addressed analytically to assess the effect of a number of different effective factors. In these factorial designs, main effects of each variable, as well as the interaction between these variables in the different psychophysical tasks, can be measured explicitly (Price et al., 1997). Also, biomathematical approaches have been developed to account for coherent brain activity in different brain regions, in relation to defined task conditions (Alexander and Moeller, 1994; Friston, 1994; McIntosh and Gonzalez-Lima, 1994; Büchel et al., 1997). These statistical approaches can be subdivided into those that are hypothesis-driven and region-based, thus restricting the search space by imposing an external model (McIntosh et al., 1994; Azari et al., 1999) and those using “omnibus” statistics, evaluating the functional connectivity in the entire set of image data (Friston et al., 1993; Seitz et al., 2001). Most recently, fMRI has been developed further to accommodate also activity recordings in an event-related fashion (D’Esposito et al., 1997; Buckner et al., 1998; Friston et al., 1998; Beauchamp et al., 1999). These measurements combine high spatial resolution with high temporal resolution (Menon and Kim, 1999). Common to these different types of image data analysis, however, is the fundamental idea that human brain function can be localized in topographically defined maps. The physiological foundations, technical limitations of functional neuroimaging, and the perspectives of the functional parcellation of the human brain will be discussed in this chapter.

2. PHYSIOLOGICAL FOUNDATIONS Microelectrode recordings in non-human primates have provided evidence for the modular organization of the cerebral cortex. Hubel and Wiesel (1963) as well as Mountcastle and collaborators (1969) were the first to report that neuronal populations assemble in vertical units extending from the upper to the lower laminae of the cerebral cortex. These so-called cortical columns are spatially distinct, representing receptive fields of the different sensory modalities or motor output modules (Hubel et al., 1978; Phillips et al., 1988; Fetz, 1993). A cortical area represents a local network of neuronal populations which are anatomically organized in cortical units, as evident physiologically by their correlated neuronal activity (Peters and Kara, 1987; Gray et al., 1990). In primary cortical areas these units


82

Rüdiger J. Seitz

follow a topographical map of their receptive fields (van Essen, 1985; Kaas, 1993). In motor cortex, however, representations of neighbouring units are intermingled such that sub-regions represent parts of local networks with highly organized patterns of corticocortical and cortico-subcortical afferents, and efferent connections allowing for specificity of movement rather than of muscles (Schieber and Hibbard, 1993; Lemon, 1988). More recently, it has been shown that correlated firing in neighbouring neurones, as recorded with multielectrodes, provides higher-order features of neural activity and thus far more information than the mere neuronal firing rate (Maynard et al., 1999). Furthermore, evidence was obtained showing that adjacent receptive fields are not invariant, but can be reshaped in relation to experimental manipulations. For example, repetitive stimulation of a receptive unit induces a local expansion of the corresponding cortical area, while conversely denervation results in shrinkage and even loss of the corresponding cortical representation (Jenkins and Merzenich, 1987; Eysel, 1992; Nudo et al., 1996; Kaas and Florence, 1997). In the motor cortex, the locally interconnected sub-fields provide avenues for reorganization subsequent to local cortical damage (Stepniewska et al., 1993; Weiss and Keller, 1994). Autoradiography provides quantitative means of studying the regional glucose consumption (rCMRGLc), the rCBF, and neurotransmitter and neuroreceptor distributions in laboratory animals. Such studies, which were pioneered by Sokoloff (Sokoloff et al., 1977) revealed that the cerebral cortex metabolizes the glucose analogue deoxyglucose in a highly organized, stimulus-related pattern (Kennedy et al., 1976; Juliano et al., 1981). Figure 5.1 shows a column-like cortical labelling in the striate cortex, with the highest intensity in cortical layer IV, during visual stimulation of one eye. In comparison to binocular whole-field visual stimulation, which induced a homogenous labelling of the occipital cortex, the regular metabolic pattern appeared to correspond to the ocular dominance columns. Exceptions were the areas with exclusively monocular input probably representing the blind spot (arrows). Likewise, cortical columns in somatosensory cortex were mapped autoradiographically in relation to simple and complex somatosensory stimuli (Juliano et al., 1983). Thus, it became obvious that metabolic mapping provided a means of studying the functional organization of the brain in vivo, with high spatial resolution. Also, autoradiography was used to demonstrate cortical plasticity in response to experimental brain lesions and peripheral deafferentation (Dietrich et al., 1985; Gilman et al., 1987; Kossut et al., 1988; Welker et al., 1992). It should be emphasized that these autoradiographic recordings demonstrate the total amount of glucose consumption of brain tissue, which is brought about by the neurones, the surrounding glial cells and to a small part by the endothelium of the intracerebral vessels. In addition to the topographic information, autoradiography also provided information about the driving forces of enhanced metabolic activity. Evidence was obtained showing that labelling intensity of glucose metabolism correlated with stimulus intensity (Kadekaro et al., 1985; Yarowsky et al., 1983), as did the intensity of glucose metabolism with the rCBF (Kuschinski et al., 1981; Cremer et al., 1983; Ueki et al., 1988). More specifically, regional metabolic activity was shown to reflect potassium ion fluxes at synaptic clefts (Mata et al., 1980; Kadekaro et al., 1985). Apart from neurones, astrocytes are predominantly responsible for metabolizing glucose (Magistretti and Pellerin, 1999). Recently, it was shown that excitation and inhibition are spatially closely related in the cerebral cortex, both resulting in a graded increase of glucose metabolism (Brühl and Witte, 1995). By applying optical imaging techniques, it was found that electrically active cortical areas induce a locally enhanced blood flow and enhanced oxygen consumption (Frostig et al., 1990; Malonek



83

A

B

C

5.0 mm Figure 5.1. Autoradiographic mapping of the ocular dominance columns using 2D-deoxy-glucose in the rhesus monkey. The coronal sections at the level of the striate cortex show intensive metabolic activity during binocular vision (A), low metabolic activity during bilateral visual deprivation (B), and a patterned metabolic activity after right eye occlusion (C). Note the maximal metabolic activity in cortical Layer IV and the alternate dark and light striations during one-eye visual stimulation. The arrows indicate the location of the blind spot with only monocular input. Taken from Kennedy et al. (1976) with permission.

and Grinvald, 1996). Thus, glucose, the only energy substrate of the primate brain, is metabolized in the presence of oxygen, which is supplied rapidly to the area of enhanced brain activity by a local rise in rCBF, that exceeds quantitatively and spatially the area of enhanced metabolism (Grinvald et al., 1991). In addition, due to the high temporal resolution of the


84

Rüdiger J. Seitz

optical imaging techniques it became possible to uncover the sequential events of oxygen supply, oxygen extraction, and haemodynamic response after stimulation onset (Frostig et al., 1990). Accordingly, oxygen transport from the cerebral blood vessels into brain tissue is the initial event followed by a graded haemodynamic response. In accord with theoretical predictions these results support the view that a local rise in blood flow is such that it allows sufficient regional oxygen supply (Kislyakov and Ivanov, 1986; Buxton and Frank, 1997). These experimental recordings with nearly microscopical resolution opened detailed insight into the spatially and temporally organized metabolic-hemodynamic changes related to brain activity. The mediators for these events are manifold, including potassium ions, nitrous oxide, and possibly lactate but their individual contributions are still unclear (Erecinska and Silver, 1989; Iadecola, 1993). The functional imaging techniques, such as positron emission tomography (PET) and functional magnetic resonance imaging (fMRI), provide means to study the active human brain. However, they present only indirect indicators of neural activity, as they measure the stimulation-related haemodynamic response. PET activation studies rely essentially on the tracer technology measuring the local task-specific cerebral accumulation of the rCBF tracer throughout the entire brain (Raichle, 1987). Depending on the biomathematical quantification model chosen, sampling intervals of 40 to 100 seconds are used in rCBF studies. Tracers labelled with positron-emitting isotopes such as [15O]-water or [15O]butanol allow one to estimate the rCBF in quantitative units by computation with the time-activity curve of the arterial tracer concentration (Herzog et al., 1996) or semiquantitatively after normalizing the cerebral tracer uptake to the global normal reference value of 50 ml/100g/min (Friston et al., 1994; Votaw et al., 1999). With respect to the neuronal processes that evolve in the time frame of milliseconds, an inherent limitation of the rCBF technique is therefore the long duration of the measuring time needed. In fMRI, the endogenous blood oxygenation level-dependent (BOLD) contrast is used as an indirect marker of cerebral blood flow changes (Calamante et al., 1999). Due to the enhanced blood flow in activated brain areas the amount of diamagnetic oxygenated blood becomes locally enhanced (see above). Thus, the paramagnetic deoxy-haemoglobin decreases which results in a signal increase in the fMRI images. In proportion to the fast evolving haemodynamics (Sitzer et al., 1994; Deppe et al., 2000), this signal builds up in some 8 seconds which can be followed with fast MR sequences including echo-planar imaging (Stehling et al., 1991; Frahm et al., 1992; Bandettini et al., 1997). While image evaluation of rCBFPET capitalizes on identifying significant changes that persist in the steady-state during the scanning interval compared to the control steady-state, fMRI exploits the consistency of activation-related changes over a couple of subsequent activation-control cycles. The areas that survive the different steps of image analysis (see also above) are converted to pseudocoloured hot spots, which thereafter can be superimposed on co-registered structural MR images or anatomical templates (Steinmetz et al., 1992a; Frackowiak, 1994). These hot spots indicate those areas in the brain that are specifically activated by a given task. However, such an area is not uniquely specialized for this task but rather may subserve also related operations. Nevertheless, the more routinely a task is performed, the more focal are the areas of cerebral activation (Roland, 1993). Interestingly, there seems to be a good correlation between brain electrical activity and the haemodynamic response as measured with rCBF-PET and fMRI, both during physiological stimulation (Heinze et al., 1994; Gerloff et al., 1996; Korvenoja et al., 1999) and under pathological conditions (Volkmann et al., 1998; Krakow et al., 1999). Nevertheless,



85

the mode of stimulation has been shown to affect the hemodynamic response profoundly. In the visual cortex, a relation between the haemodynamic response and the stimulation frequency was observed (Fox et al., 1984; Hoge et al., 1999). In contrast, evoked electrical brain activity, as measured with EEG and MEG, becomes more clearly discernible in relation to an increasing interstimulus interval (Ibánez et al., 1995; Schnitzler et al., 1999). Recently, it was shown by simultaneous electrophysiological recordings and fMRI scanning during visual stimulation in the primate that the BOLD signal closely correlates with the local field potentials but not with the activity of single neurones (Logothetis et al., 2001). These data support the idea that neuroimaging can record information processing in the human brain due to concerted activity of neuronal populations. While these findings hold not only for the different lobes of the cerebral cortex and most likely also for the cerebellum, task-related activations may be elusive in the basal ganglia, thalamus, and brainstem structures. This failure may be due to the diversity of neurotransmitter systems, the higher anatomical packing density of diversified functional units in these structures (Chesselet et al., 2000; Schmahmann, 2000), as well as to the need to apply specialized approaches to image analysis (Bohm et al., 1991; Kleinschmidt et al., 1994). In addition, the reader should be alerted to the observation that there are gender differences of the rCBF and oxygenation response to physiological stimuli (Kastrup et al., 1999). Finally, it should be pointed out that the regional activation-induced changes of metabolism and blood flow, though strongly linked to each other (Roland et al., 1987; Fox et al., 1988; Seitz and Roland, 1992a; Blomqvist et al., 1994; Hoge et al., 1999), have different time constants of return to baseline, that can be picked-up by neuroimaging techniques (Madsen et al., 1999).

3. INTER-INDIVIDUAL VARIABILITY OF THE HUMAN BRAIN The most critical issue concerning the transposition of the results of animal experiments to tomographic imaging of the human brain is the inter-individual variability of human brain anatomy. This factor influences group image data which have been conceptualized as probabilistic representations of brain functions as well as the anatomic correspondence of activation foci related to identical tasks among different subjects. Inter-individual variability of the human brain has recently come into focus again in relation to in vivo morphometry from MRI studies of healthy subjects. Qualitative analysis of MRI images had already revealed profound inter-individual differences in gyrus formation, even in the brains of monozygotic twins (Steinmetz et al., 1995). By in vivo MRI, it was shown by quantitative means that gyral configuration of the human brain is highly variable among different individuals, and largely determined by epigenetic factors (Steinmetz et al., 1992b; Amunts et al., 1997; Kennedy et al., 1998). Also, there are interhemispheric differences within and between the sexes, probably related to differences in handedness, skills, and capacities (Schlaug et al., 1995; Gur et al., 1999). Still, after proportional scaling of the brains of different individuals into stereotactic reference space, inter-individual variability of the major cerebral sulci at the cerebral surface has a range of 1–2 cm (Steinmetz et al., 1989). These in vivo studies on high-resolution MR images validated previous findings by Talairach and colleagues obtained from post-mortem investigations (Talairach et al., 1967). Likewise, other approaches of spatial standardization using non-linear transformation procedures in addition to linear (affine) ones have also revealed residual inter-subject variability of a similar magnitude, both in terms of structural


86

Rüdiger J. Seitz

and functional inter-subject variability (Evans et al.., 1988; Seitz et al., 1990; Friston et al., 1994; Thurfjell et al., 1995; Roland et al., 1997). This is illustrated in Figure 5.2. After spatial standardization using affine and non-linear transformations, residual variability of the central sulcus was approximately 4 mm. The structural variability was similar to the variability of the rCBF increases related to sequential finger movements after applying the same spatial standardization parameters to the rCBF images as to the MR images. This residual functional variability was shown to be of such a magnitude that the areas of activation related to finger movement in eight of nine different subjects overlapped in only one pixel (Schlaug et al., 1994). In addition, there is considerable inter-subject variability of the cerebral activation pattern, that appeared to be related to the individual task performance (Schlaug et al., 1994). Similar data were also reported by other groups using other methods for normalization of spatial image data (Hunton et al., 1996; Hasnain et al., 1998). Thus, large rCBF changes in tomographic images appeared to be consistent across subjects, this being different from small rCBF changes that had a relatively large inter-subject variability (Figure 5.2). This finding is a reflection of the observation that the variance of rCBF changes is not stationary within the pixel matrix (Grabowski et al., 1996). That is, in the case of high mean rCBF due to consistent performance across the subjects, the interindividual location of the activation focus is more similar than in case of low mean rCBF related to low and less consistent performance. The problem of inter-subject variability cannot be discussed without mentioning the limited spatial resolution, the so-called partial volume effect in tomographic images. The partial volume effect obscures anatomic resolution and attenuates quantitation of measured activity (Mazziotta et al., 1981; Bohm et al., 1991). Figure 5.3 shows that the partial volume effect severely attenuates the signals recorded with functional imaging: The smaller the object the smaller the signal. The activity of objects with a dimension greater than three-times the optimal image resolution is recovered at 95% of the true activity only. The problem of the partial volume effect is of an even greater implication for imaging the human brain, because the cerebral cortex, and subcortical structures, in particular, are far smaller than the image resolution of current rCBF and fMRI image data. In addition, the cerebral gyri are lined by the cerebral sulci that contain inactive cerebrospinal fluid which further accentuates the partial-volume effect in functional imaging data. The same holds because of the close neighbourhood of the basal ganglia, thalamus and brain stem nuclei to the ventricular system. Figure 5.3 also shows that the smaller the signal compared to the image noise, the greater is the underestimation of the signal in magnitude and spatial dimension due to the partial volume effect (Knorr et al., 1993). Both, image noise and the partial volume effect may contribute to false negatives in functional imaging. Thus, the true activity changes in cerebral cortex and subcortical nuclei are underestimated by functional imaging, even if the partial volume effect is in the range of a few pixels, as is the case in fMRI (Calamante et al., 1999). Consequently, the partial volume effect adds to inter-subject variability of brain configuration, by diminishing stimulation-induced signal changes in functional imaging data. Thus, group image data underestimate the real changes, calculated by taking the peak changes in the individual subjects, by about of 30% (Seitz et al., 1990). Further, Figure 5.2 shows that inter-subject averaging may result in a slightly different position of the mean activation centre compared to the arithmetic mean of the peak changes in the individual subjects. Thus, inter-subject averaging can affect the resulting data not only quantitatively but also qualitatively with respect to anatomic validity. The danger of mislocalization



87

Figure 5.2. Influence of the inter-individual variability on the delineation of an activation area in spatially standardized images of a group of subjects. (A) A great mean rCBF change has a small inter-individual variance (expressed as standard error of mean rCBF change) in identical pixel locations suggesting consistency of the activation-induced mean rCBF changes across subjects. Note that pixels with low activation-induced rCBF changes are quite variable (including also the largest SEM), reflecting a high degree of functional variability across subjects. (B) Comparison of structural and functional variability expressed as standard deviations. The tip of the central sulcus (left) had a range of variability of 4 mm compared to the variability of the mean rCBF increase related to finger movements of 5 to 10 mm (right). The arithmetic mean (♦) of the peak activations in the spatially standardized images of the individual subjects (●) deviated from the location of the peak of the mean rCBF increase in the statistical group image (䊊) by approximately 3 mm. For further details see Seitz et al. (1990).


88

Rüdiger J. Seitz

Figure 5.3. Influence of the functional image data on the delineation of an activation area. (A) Profile of the mean rCBF increase along a continuous row of pixels in the left motor cortex in a group of healthy subjects during right-hand finger movements. The mean rCBF increase at a threshold of p < 0.01 (28.7%) is lower than the peak change (41.2%) and smaller in extent than the entire area of mean rCBF increase. The entire area of the mean rCBF change had an increase of 23.4% compared to the resting control state. (B) Attenuated recovery of real activity in individual data in relation to object size and image noise. Real activity is recovered at approximately 95% for large objects with a mean background activity of 10% (open columns). Increasing image noise progressively impairs object detection and object recovery progressively: 60% background activity (hatched columns), 78% background activity (black column). Object size is given in relation to image resolution (FWHM, full width at half maximum). A complete description of the simulation study is given by Knorr et al. (1993).


Functional Brain Mapping A

89

B

Figure 5.4. Superimposition of the mean rCBF increases related to finger movement imagery as compared to preparation in 10 subjects (A) as yellow area onto the cytoarchitectonic subarea 4a of left motor cortex (B). Common areas of overlap of the anatomical site of cytoarchitectonic areas 4a (red) and 4p (violet) of two brains (n = 2) after standardization into reference space of the Human Brain Atlas (Roland et al., 1997). The centres of gravity (COG) are indicated. Note, the additional activation in the supplementary motor area and the lack of activation in subarea 4p of motor cortex (A). The location of the central sulcus is indicated as undulated yellow (A) and black (B) line, respectively. Preliminary data have been communicated by Stephan et al. (1995b) and were kindly provided by the authors. Left in the image is the lateral surface of the brain, right in the image the interhemispheric cleft. (see Color Plate 4)

increases with the partial volume effect in the image data. It is well recognized that heavy image smoothing may enhance the detectability of functional changes, but at the expense of creating virtual activation centres by merging of small closely adjacent real activity changes (Worsley et al., 1992). Increasing the partial volume effect artificially by image smoothing may therefore result in activation areas in possibly unexpected locations. To get a more realistic view of human brain function, in particular in cognitive tasks, it is mandatory to use improved methods of image standardization and optimal image resolution, as is feasible in the latest generation of PET scanners, and in high-Tesla fMRI scanners. These technical refinements will enhance the tomographic reflection of true activity changes, reducing the liability to false negatives and false positives. What will remain is the inter-individual variability due to anatomic differences in the microscopical dimension. As is highlighted in the chapter by Rademacher in this volume, there is not even a definite correspondence of cytoarchitectonic and chemoarchitectonic areas with macroanatomic landmarks in the human brain. This has been shown for the central sulcus, the lateral and the calcarine fissure (Rademacher et al., 1993). Accordingly, inter-subject variability also has a microscopical dimension which can be revealed only when macroscopical variability can be fully accounted for by spatial standardization procedures. To this end, new methods have been developed which allow one to transform different brains into the same reference with a point-to-point correspondence (Christensen et al., 1994; Schormann and Zilles, 1998; Fischl et al., 1999). These processes allow one to create maps of true microanatomic variability between different individuals, thus increasing the areas of rCBF overlap in different subjects in group image data, and improving the signal-to-background relation in activated areas dramatically (Schormann, personal communication).


90

Rüdiger J. Seitz

By these means it will also become possible to study functional subdivisions of cortical areas. For example, electrophysiological evidence in the somatosensory domain suggests that the cortical areas 3a and 2 are concerned with proprioceptive information, while areas 3b and 1 are processors of cutaneous input (Kaas, 1993). Likewise, area 4 of the motor cortex was shown to be made up by two cytoarchitectonically different sub-fields (Geyer et al., 1996). Co-registration of these cytoarchitectonic data with functional imaging maps suggest that area 4p in the depth of the central sulcus is concerned with motor execution while area 4a at the lateral outer aspect of the precentral gyrus is more concerned with preparatory aspects of movement (Figure 5.4). A question closely related to the issue of subdivisions within areas is whether identical neuronal populations have the capacity to assemble in a task-related manner to different subsets affording different functions. In conclusion, at the microscopical and microelectrode level, respectively, the issue of structural-functional correspondence is unsettled. Further discussion of this topic is given in the chapter by Amunts et al., in this volume.

4. LOCALIZATION OF HUMAN BRAIN FUNCTION Study of localization of human brain function has a long tradition. Until the advent of computed tomography (CT) this could only be done retrospectively by correlating neurological deficits with brain lesions at autopsy. CT provided for the first time a means of making in vivo correlations of brain lesions with clinical deficits, allowing for prospective analyses. However, since only a few Hounsfield units differentiate tissue compartments within the brain, and because of the partial volume effect produced by the skull and the geometrically complicated base of the skull, anatomical resolution is compromised in CT. Thus, small brain lesions, particularly in white matter structures and the brain stem, are usually elusive in CT scans. The lesion-based approach for studying the brain was dramatically improved by MRI which is far more sensitive and less prone to the confounds of the partial-volume effect than CT (Young et al., 1982; Prichard and Brass, 1992). Accordingly, MRI has become the method of choice for study of correlations between lesions and deficits. Brain lesions, however, do not follow functional divisions, but develop within the pathogenetic framework of the underlying disease. For example, brain infarctions have an individual configuration following the territories of the cerebral arteries or their branches (Bogousslavski et al., 1986; Ringelstein et al., 1992). Thus, superimposition of brain lesions of different patients introduces noise into the group data. Nevertheless, a common area of lesion overlap is expected to demonstrate a brain area critical for a certain function. It was therefore surprising that lesion mapping of syndromes like hemineglect was shown not to be functionally revealing but rather resulted in a reflection of the territory of the middle cerebral artery (Kertezs and Ferro, 1984). Similarly, motor aphasia was not a sufficient description of deficit to pinpoint the representative speech area in the brain (Poeck et al., 1984; Alexander et al., 1990). One explanation for this failure could be the distributed localization of brain function involving networks of different brain structures—a concept pioneered by Mesulam (1981, 1990). Accordingly, only damage to a critical node within such a network interferes with the function of the network. However, there is a large number of different nodes within such a network, subserving different sub-functions and allowing for partial rewiring after damage to one or a few of these nodes. Combined



91

behavioural, electrophysiological and anatomical studies have shown the multiplicity and diversity of such networks, for example for voluntary control of action (Rizzolatti et al., 1998). This concept would explain variability of lesions across different subjects presenting with a closely similar deficit. However, as was evidenced recently, these functional deficits have to be fairly well defined in neurophysiological or neuropsychological terms to reflect a cortical module or a critical node. Examples are the clinically similar but clearly differentiable syndromes of mirror agnosia, mirror ataxia, and visuomotor ataxia (Rondot et al., 1977; Binkofski et al., 1999a). Thus, slightly different lesion locations are characterized by slightly different patterns of neurophysiological or neuropsychological deficits, which can be mapped to non-overlapping brain lesions even within one lobe. A different approach to localizing human brain function is to stimulate the brain or to record brain activity directly during open brain surgery. Careful documentation of observed brain lesions and the results of intra-operative cortical stimulation studies have been the basis for the elaboration of the topographical organization of motor and sensory representations by Foerster (1936) and subsequently by Penfield and Jasper (1954). More recently such a systematic approach was used to map brain structures relevant for production of speech and music (Ojemann et al., 1989; Creutzfeld and Ojemann, 1989; Haglund et al., 1992). This type of investigation has, however, a severe inherent confounding factor, which is the fact that studies are performed on abnormal brains. As lesions interfere with function, compensatory mechanisms are evoked, resulting in functional reorganization which is greater the earlier the lesion was acquired and the more slowly it expands (Seitz and Azari, 1999; Gadian et al., 1999). Recently, hypothesis-driven stimulation of the normal cortex using transcranial magnetic stimulation (TMS) has been shown to be appropriate to map the cortical representations of individual finger muscles in human motor cortex (Wassermann et al., 1996; Classen et al., 1998a). Quite differently, TMS has also been used as a probe to interfere with brain function. Thereby, the role of posterior brain areas for mediating visuospatial functions and visual imagery can be evaluated (Beckers and Hömberg, 1991; Paus et al., 1996; Kosslyn et al., 1999). The localizing capacity of TMS is limited however, due to the uncertainty of the exact location and spatial extent of the stimulated part of the brain. The most direct approach to the study of human brain function is to measure brain activity during brain work. This has become possible since the advent of the functional imaging techniques such as PET, fMRI, and MEG. In most experimental designs and data analyses of functional neuroimaging, a hierarchical concept of functional representation has been assumed (Petersen et al., 1988; Kosslyn et al., 1995). This approach follows the idea that neuropsychological processes can be identified by subtracting from a more complex task, a similar task in which the cerebral computations differ by one aspect. Thereby, functional units subserving this additional demand should become identifiable. In an ideal situation, task A would require areas 1 and 2 and task B areas 1 and 3, while the control task used in both comparisons would engage area 1 to the same degree. It is clear, however, that area 1 represents a specific activation in task A as in task B thus being elusive by direct comparisons of tasks A and B. Conversely, since area 1 is of critical importance for both tasks, it could be demonstrated as an area of a main effect by a conjunction analysis (Friston, 1997; Price et al., 1997). When area 1 is activated in a different manner in task A and B, there is a region-specific interaction. When delineating activation areas in functional imaging data, it is usually tacitly assumed that the activation area reflects a functional brain unit activated by a given task. However, stimulation-induced activity changes


92

Rüdiger J. Seitz

obtained by matrix statistics of groups of subjects are critically influenced by the image noise (Figure 5.3). The peak mean activity change exhibits the point with the greatest significance level, while the area of change at a predefined significance threshold is smaller than the entire area of change (Lueck et al., 1989). As evident from Figure 5.3, definition of what is considered to represent the stimulation-related change is a critical determinant of the spatial dimension and magnitude of this change. The currently available image processing methods yield astonishingly similar results with an advantage for larger compared to small samples (Grabowski et al., 1996; Missimer et al., 1999). In addition to the categorical type of data analysis, multivariate statistics have been applied for accounting for the concept of distributed cerebral functions, in the sense of task-related networks involving sub-units in different cortical and subcortical areas. Approaches of this sort without a priori assumptions are correlation and principal components analyses, requiring additional statistical testing to make inferences about the identified patterns (Friston et al., 1993; Alexander and Moeller, 1994; Friston, 1994). These results can be displayed as image data, but do not provide any clue about the direction of information flow among the constituting areas, or about the strength of connectivity among them. For this goal, the approaches of structural equation modelling can be applied. These types of analysis incorporate external models to describe inter-regional interactions (McIntosh and Gonzalez-Lima, 1994; Büchel and Friston, 1997). These methods are, however, biased by the model-related selection of areas included in the calculation. Also, in these approaches the number of regions should be less than the number of study subjects, since otherwise the calculated solutions are not stable.

5. CORRESPONDENCE OF LESION AND ACTIVATION STUDIES It is widely accepted that brain lesions, by definition, interfere with brain function but do not necessarily show the site where a certain brain function is represented. On the contrary, brain imaging usually shows a number of brain areas activated in relation to a certain task. This becomes evident from neuroimaging studies on visual information processing (see Gulyas, 1997, for review). Accordingly, network approaches to analysis of image data have provided evidence that a specific function involves an entire network of brain structures. For example, object vision involves a number of striate and extrastriate cortical areas (Horwitz, 1994; Büchel and Friston, 1997). The question, though, is whether a certain brain structure which is essential for a given function can be identified by the correspondence of lesion and activation studies. A lesion to such a structure would then lead to a permanent deficit and coincide with the activation site related to the corresponding activation paradigm. In the motor system, it is well established that the mid-dorsal portion of the precentral gyrus contains the cortical representation of finger and hand movements of the opposite side of the body. The first evidence for this was obtained from intra-operative stimulation studies showing that low threshold electrical stimulation of this cortical area induces muscle twitches of the contralateral forearm (Foerster, 1936; Penfield and Boldrey, 1953). Likewise, in neuroimaging studies, circumscribed rCBF increases were observed in this location during hand and finger movements of the contralateral arm (Seitz and Roland, 1992b; Rao et al., 1993; Stephan et al., 1995; Seitz et al., 1996, 1997; Rijntjes et al., 1999). More importantly, even muscle relaxation activates the same cortical areas (Toma



93

et al., 1999). There are anatomical landmarks such as the omega-shaped configuration of the central sulcus on axial MR images, as well as the endpoint of the superior frontal fissure touching the precentral sulcus at this point, to indicate the location of the motor hand representation (Ebeling et al., 1992; Yousry et al., 1995). Conversely, combined morphological and electrophysiological evidence was obtained recently showing that the degree of destruction of part of the precentral gyrus or its corticospinal projection is a quantitative indicator of both the loss of individual finger movements of the contralateral hand, and the capacity for recovery (Binkofski et al., 1996; Pendlebury et al., 1999). It should also be emphasized that lesion effects are topographically specific. For instance, leg movements were affected to a lesser degree, or not at all, when precentral lesions affected arm function, and vice versa (Schneider and Gautier, 1994; Azari et al., 1996). Furthermore, motor impairment does not correlate with somatosensory impairment of the same body part (Azari et al., 1996). These results have their counterpart in the rCBF increases in the frontomesial cortex related to leg movements and in the postcentral gyrus related to somatosensory stimulation in healthy volunteers (Fox et al., 1987a; Fukuyama et al., 1997). Similarly, bimanual co-ordination was shown to be significantly impaired in patients with circumscribed lesions of the mid-part of the cingulate gyrus, while activation studies in healthy volunteers demonstrated large areas of activation both in the supplementary motor area and in the mid-cingulate cortex, probably involving the cingulate motor area as well as the lateral premotor cortex (Stephan et al., 1999). Quantitative investigations revealed persistent motor deficits in patients or primates with lesions within the precentral gyrus in spite of apparently complete clinical recovery (Friel and Nudo, 1998; Seitz et al., 1999). Apparently, clinical restoration goes along with measurable minute deficits of the arm contralateral and ipsilateral to the lesion, and is often accompanied by subtle, but clear-cut changes of task performance related to alternative strategies of task performance (Winstein and Pohl, 1995). It should be emphasized that cortical lesions subsequent to focal ischemia affect not only the local cortical machinery but also their cortical and cortico-subcortical afferents and efferents. This widespread effect of cortical lesions can be visualized by metabolic mapping (Seitz et al., 1994, 1999). Consequently, it can be argued that persistent neurological impairments result not only from the lesion itself but also from suppression or deafferentation of the affected cortical node from other parts of the network (von Giesen et al., 1994; Classen et al., 1995). By contrast, circumscribed experimental lesions of cortex sparing subcortical fibre tracts remain free from associated remote effects (Buchkremer-Ratzmann et al., 1997). Similar observations have also been made in the visual system, showing a mirror-like correspondence of the retinotopic activation of the visual cortex upon visual stimulation and the central or peripheral homonymous visual field defects after circumscribed lesions of the calcarine cortex or the optic tract (Fox et al., 1987b; Zihl and von Cramon, 1985). The correspondence of lesion and activation studies also appears to hold for higher order visual areas. Specifically, there appears to be a close correspondence between clinical deficits in motion and colour perception related to circumscribed occipito-temporal brain lesions and the cerebral activations upon motion perception and colour vision in healthy volunteers (Zihl et al., 1991; Watson et al., 1993; Ungerleider and Haxby, 1994; Tootell et al., 1995). Recently, it was demonstrated that such a close correspondence of lesions and sites of activation is also pertinent for the parietal lobe. Evidence was obtained that unilateral disturbances of forming a precision grip formation for grasping are associated with lesions


94

Rüdiger J. Seitz

Figure 5.5. Topographic correspondence of an area of lesion overlap among different patients and the site of activation in healthy volunteers. The patients had a severe impairment of object prehension, while the healthy subjects specifically activated the same region during prehension movements as contrasted with reaching (Binkofski et al., 1998). The lesion and the activation foci are localized in the lateral bank of the cortex lining the anterior part of the intraparietal sulcus (IPS) thus probably corresponding to the anterior intraparietal area. Note that the lesions of the patients differed in extent (shaded area). CS = central sulcus.

of the parietal cortex lining the anterior portion of the intraparietal sulcus (Binkofski et al., 1998). Conversely, healthy volunteers required to perform grasping movements showed a specific activation of the cortex lining the interior part of the intraparietal sulcus in surprisingly close correspondence to the cortical lesions inducing the grasping deficit (Figure 5.5). Similarly, lesions of this superior parietal lobule induce severe deficits of object manipulation, resulting in tactile agnosia (Binkofski et al., 2001). Conversely, neuroimaging studies in healthy volunteers showed circumscribed activations of the superior parietal cortex in a closely corresponding location, related to tactile information processing during exploration of large geometric objects, as well as of textures (Seitz et al., 1991; O’Sullivan et al., 1994; Binkofski et al., 1999b). Lesions of the inferior parietal cortex including the parietal operculum have been shown to abolish the ability to identify objects upon tactile exploration, a syndrome which was termed object agnosia (Caselli et al., 1993). Recently, fMRI data in healthy volunteers provided evidence that the inferior parietal cortex, including the secondary somatosensory area, becomes activated during tactile exploration and identification of large complex geometric objects (Binkofski et al., 1999b). These data support the notion that the secondary somatosensory area is involved in processing intrinsic object information necessary for object recognition. A specific location for speech production was demonstrated recently in a study on language recovery, after post-stroke aphasia (Heiss et al., 1999). Patients with frontal and subcortical infarctions recovered well within 8 weeks after stroke, while patients with an infarction involving the superior temporal cortex did not. The cortical activation patterns related to a verb generation task showed activation of the superior temporal cortex bilaterally, and of the right inferior frontal cortex in the patients who recovered, whereas the patients who did not recover showed activation of the inferior frontal cortex in both hemispheres but only of the right superior temporal cortex. These patterns demonstrated the impact of the left superior temporal cortex for language recovery in post-stroke aphasia. They correspond to the activation patterns related to verb generation in healthy volunteers, as sampled across different imaging centres (Poline et al., 1996). Of particular interest in this connection are also the observations by Friston et al. (1991) who showed that the superior temporal activations were negatively correlated to the dorsolateral prefrontal



95

activations, suggesting a modulatory action of the dorsolateral prefrontal cortex on the temporal language areas. Conversely, a remote functional disturbance in the superior temporal cortex impaired language abilities in patients with left medial temporal lobe epilepsy (Arnold et al., 1996). These examples indicate that circumscribed structural lesions (as well as functional irritations remote from the brain lesions) can interfere with functions that are represented in exactly these locations as evident from functional imaging. In other words, lesion studies can provide specific information about critical cortical nodes, whereas corresponding activation studies demonstrate a number of activated cortical areas inclusive of the critical area as implicated from lesion studies. Accordingly, the notion that cerebral lesions interfere with the network required to produce a neuronal function, while the network becomes visible as a whole in neuropsychological activation studies in healthy volunteers, is supported by the evidence available from combined lesion and activation studies.

6. DISCREPANCY OF LESION AND ACTIVATION STUDIES In contrast to complete destruction of a cortical module, a partial lesion of this module can be compensated by reorganization of the remaining parts (Nudo et al, 1996; Xerri et al., 1998). Similarly, a lesion to one node within a cortical network can be compensated by reshaping of the remaining parts of the network. Such mechanisms of plastic reorganisation are supposed to underlie the clinical recovery of function. If so, cerebral lesions and activation studies are expected to be discrepant. Examples include slowly progressive brain lesions. Indeed, evidence was obtained showing that low grade gliomas distort the representation of a given brain function to such an extent that it is mapped to an atypical location (Wunderlich et al., 1998). This kind of cortical reorganization took place predominantly within a functional unit, such as the motor cortex. Morphometric measures performed in parallel to the activation studies revealed that the functional displacement exceeded the mass effect of the growing tumor (Figure 5.6). In cortical lesions that are acquired in utero, neonatally or during early infancy, reorganization of function has been shown even to exceed the limits of functional borders, allowing for abnormal functional representation in grossly abnormal locations including even the contralateral cerebral hemisphere (Müller et al., 1998; Seitz and Azari, 1999). In temporal lobe epilepsy in which a structural abnormality evolved during adolescence, there may be also a discrepancy between a circumscribed cerebral lesion and the clinical deficit. It was shown that patients with left temporal lobe epilepsy exhibiting atrophy and sclerosis of the hippocampal formation are impaired in verbal recall, verbal fluency and verbal intelligence (Frisk and Milner, 1990; Tranel, 1991). The metabolic changes related to this disease condition occurred not only in the mesiotemporal region but also in remote locations in the superior temporal and inferior frontal language areas and in prefrontal cortex (Arnold et al., 1996; Jokeit et al., 1997). Network approaches to analysis of the data revealed large-scale abnormalities in these patients involving a weakened relationship of the bilateral prefrontal cortex but strengthened connectivity of the language areas and the right thalamus (Figure 5.6). These observations are in accord with the finding that patients with left temporal lobe epilepsy enjoy recovery of their language functions following selective anterior temporal lobectomy and thus excision of the irritative lesion (Regard et al., 1994).


96

Rüdiger J. Seitz

Figure 5.6. Local and large-scale reorganization of functional representations in chronic brain lesions. (A) Three-dimensional vector displacements of the rCBF increases in the affected cerebral hemisphere of five patients with precentral gliomas. Origin corresponds to the location of motor hand area as determined for the normal hand in the contralesional hemisphere. The displacement of the vectors from origin indicates the structural displacement of the central sulcus. d = dorsal, f = frontal, l = lateral, o = occipital, m = medial, v = ventral. (B) Abnormal metabolic interactions in patients suffering from left mesiotemporal epilepsy and impairment of verbal abilities. Compared to healthy controls, weak interhemispheric coupling of prefrontal cortex and enhanced coupling between the metabolically depressed inferior frontal and superior temporal cortex and the unaffected right thalamus. The individual image data have been spatially standardized to the stereotactice space (Talairach and Tournoux, 1988) to allow for inter-individual comparisons.

Another condition in which brain lesions compromise brain function in a remote fashion is neglect. In motor neglect the somatosensory afferent and motor efferent projections as well as the corresponding primary cortical processing areas were shown to be functional. However, circumscribed lesions in quite variable locations interfered with the network mediating conscious behaviour, resulting in a widespread pattern of metabolic deficits including premotor, parietal, cingulate and thalamic locations (von Giesen et al., 1994). Interestingly, this behavioural deficit regressed as the exaggerated cortical inhibition of the motor cortical apparatus after stroke normalized, as evidenced by transcranial magnetic stimulation (Classen et al., 1997). These findings have been supplemented by network analysis approaches showing that post-stroke motor recovery involves large-scale networks inclusive of brain structures remote from the infarct lesion. Within the motor system there was a resetting of the pathway from cerebellum to thalamus and supplementary motor area in the sub-acute post-infarct period (Azari et al., 1996). Furthermore, a network including thalamus and visual cortex, active with a more prolonged time course after stroke, suggested supramodal compensation of the deficit (Seitz et al., 1999). In acute stroke magnetic resonance imaging is the method of choice for demonstrating perfusion abnormalities (Prichard, 1992; Calamante et al., 1999). It should be pointed out that in such situations conventional and even modern neuroimaging techniques (as for instance diffusion-weighted imaging) may reveal only small lesions or even be normal (Tong et al., 1998; Neumann-Haefelin et al., 1999). In these situations, the affected cerebral cortex is electrically inexcitable as demonstrated in humans and in animal experiments, remaining profoundly abnormal for a long period of time (Dominkus et al., 1990; Heald et al., 1993; Bolay and Dalkara, 1998). Specifically, after incomplete



97

ischemia, the motor cortex may be excitable with transcranial magnetic stimulation in patients who recovered from stroke, while the haemodynamic response related to functional activation remains negative (Seitz et al., 1998). Initial data suggest that this electricalhaemodynamic decoupling develops in the subacute stage after infarction (Binkofski et al., 1999c). At present it is unclear whether, and if so, when the electrical-haemodynamic coupling will become restored again. As outlined above, the assumption underlying rCBF and fMRI measurements is that brain function (as characterized by the changing demands of electric activity upon regionally enhanced metabolism) can be picked up indirectly by the haemodynamic response. Changes of the threshold of neuronal excitability have however been shown to remain elusive in such measurements, while they can be recognized by transcranial magnetic stimulation and MEG recordings (Schnitzler et al., 1995; Stephan and Frackowiak, 1996). Specifically, it was shown that imagery of motor activity fails to induce rCBF changes in the motor cortex of a magnitude, spatial extent or significance comparable to that in premotor, cingulate, and parietal cortical areas (Stephan et al., 1995a; Porro et al., 1996; Roth et al., 1996; Seitz et al., 1997). Rather, only a portion in the lateral subarea of motor cortex appears to be activated (Figure 5.4). Nevertheless, excitability of the motor cortex after motor imagery was enhanced, threshold for excitation being reduced at the same time (Schnitzler et al., 1995; Stephan and Frackowiak, 1996). Likewise, visual imagery induces only minute haemodynamic changes in primary visual cortex (Roland and Gulyas, 1994). Furthermore, it was shown that transcranial magnetic stimulation over the occipital cortex eliminated visual imagery while it was ineffective in posterior temporal application (Kosslyn et al., 1999). These data show that haemodynamic measurements may only partially reflect activity changes in the brain, adding to possible dissociations of lesion and activation studies.

7. CORTICAL REPRESENTATIONS OF FUNCTION The cortical representations of function have been shown to be affected remarkably by physiological learning. Using fMRI it was shown that learning of sequential finger movements induces a spatial increase of the cortical finger representation as learning progressed (Karni et al., 1995). Group data of motor learning provided evidence of a common area of representation of hand-finger movement, with a clear rate effect on the rCBF increase, and irrespective of the degree of learning of finger movement (Seitz and Roland, 1992; Grafton et al., 1992; Jenkins et al., 1994). Using transcranial magnetic stimulation, it was however demonstrated that the threshold for cortical excitability varies in relation to task acquisition. The excitable area of motor cortex was most extensive when the task was explicitly and routinely performed (Pascual-Leone et al., 1994, 1995). Moreover, it was shown that these changes coded movement direction (Classen et al., 1998b). Also, during skill learning the motor cortical activation becomes progressively lateralized (Seitz and Roland, 1992b). This change was accompanied by a temporal reorganization of EMG activity from the typical three-phasic burst discharges related to individual finger movements to a two-burst discharge pattern after learning, probably reflecting the establishment of a tremor-like activity in the cortico-subcortical loop. Event-related fMRI has provided evidence that the cortical representations of function may also be modulated in a highly organized fashion in the temporal dimension. For


98

Rüdiger J. Seitz

example, it was shown that in working memory conditions the haemodynamic response, in those cortical areas concerned with stimulus perception, preceded those in areas related to response generation, while prefrontal areas showed enhanced haemodynamic responses during the entire delay period between stimulus presentation and the response execution (D’Esposito et al., 1997; Menon and Kim, 1999). Likewise, the perception of ambiguous stimuli resulted in antagonistic response curves in different visual sub-fields, which were related to the actual perception of the stimuli (Kleinschmidt et al., 1998). As outlined in this chapter, activation studies in healthy subjects and lesion studies in patients with well-defined clinical deficits provide means for elucidating the modular organization of the human brain. Thus, construction of a physiological atlas of the different brain functions appears feasible. One should bear in mind that anatomical afferents and efferents converge at the observed, activated nodes allowing the processing of information gathered from different sources. This view of a task-related assembly of cerebral networks comprising different critical nodes would apply to the so-called primary areas and also to the higher order associative cortices. The view is also capable of accounting for plastic topographical and temporal reorganization related to physiological learning and to adaptation to pathological conditions. Finally, this view would explain the relative paucity of corticofugal efferent and corticopetal afferent projections in the presence of a plethora of cortical-cortical connections, which can be estimated from the large dimension of cerebral white matter compared to the small diameter of the pyramidal tract, the posterior spinal columns, and the optic nerves.

REFERENCES Alexander, M.P., Naeser, M.A. and Palumbo, C. (1990) Broca’s area aphasia: aphasia after lesions including the frontal operculum. Neurology, 40, 353–362. Alexander, G.E. and Moeller, J.R. (1994) Application of the subprofile model to functional imaging in neuropsychiatric disorders: A principal component approach to modeling brain function in disease. Human Brain Mapping, 2, 79–94. Amunts, K., Schmidt-Passos, F., Schleicher, A. and Zilles, K. (1997) Postnatal development of interhemispheric asymmetry in the cytoarchitecture of human area 4. Anatomy and Embryology (Berlin), 196, 393–402. Arnold, S., Schlaug, G., Niemann, H., Ebner, A., Lüders, H., Witte, O.W. and Seitz, R.J. (1996) Topography of interictal metabolic depressions in mesiotemporal lobe epilepsy. Neurology, 46, 1422–1430. Azari, N.P., Binkofski, F., Pettigrew, K.D., Freund, H.-J. and Seitz, R.J. (1996) Enhanced regional cerebral metabolic interactions in thalamic circuitry predicts motor recovery in hemiparetic stroke. Human Brain Mapping, 4, 240–253. Azari, N.P., Arnold, S., Schlaug, G., Niemann, H., Ebner, A., Lüders, H., Witte, O.W. and Seitz, R.J. (1999) Reorganized patterns of metabolic interactions in temporal lobe epilepsy. Neuropsychologia, 37, 625–636. Bandettini, P.A., Kwong, K.K., Davis, T.L., Tootell, R.B.H., Wong, E.C., Fox, P.T., Belliveau, J.W., Weisskoff, R.M. and Rosen, B.R. (1997) Characterization of cerebral blood oxygenation and flow changes during prolonged brain activation. Human Brain Mapping, 5, 93–109. Beauchamp, M.S., Haxby, J.V., Jennings, J.E. and DeYeo, E.A. (1999) An fMRI version of the FarnsworthMunsell 100Hue test reveals multiple color-sensitive areas in human ventral occipitotemporal cortex. Cerebral Cortex, 9, 257–263. Beckers, G. and Hömberg, V. (1991) Impairment of visual perception and visual short term memory scanning by transcranial magnetic stimulation of occipital cortex. Experimental Brain Research, 87, 421–432. Binkofski, F., Seitz, R.J., Arnold, S., Claßen, J., Benecke, R. and Freund, H.-J. (1996) Thalamic metabolism and integrity of the pyramidal tract determine motor recovery in stroke. Annals of Neurology, 39, 460–470. Binkofski, F., Dohle, C., Posse, S., Stephan, K.M., Hefter, H., Seitz, R.J. and Freund, H.-J. (1998) Human anterior intraparietal area subserves prehension. A combined lesion and functional MRI activation study. Neurology, 50, 1253–1259.



99

Binkofski, F., Buccino, G., Dohle, C., Seitz, R.J. and Freund, H.-J. (1999a) Mirror agnosia and mirror ataxia constitute different parietal lobe disorders. Annals of Neurology, 46, 51–61. Binkofski, F., Buccino, G., Posse, S., Seitz, R.J., Rizzolatti, G. and Freund, H.-J. (1999b) A fronto-parietal circuit for object manipulation in man: evidence from an fMRI study. European Journal of Neuroscience, 11, 3276–3286. Binkofski, F., Fink, G.R., Wittsack, H.-J., Buccino, G., Seitz, R.J. and Freund, H.-J. (1999c) Lesion location determines cortical activation patterns in stroke recovery. Neuroimage, 9, S726. Binkofski, F., Kunesch, E., Dohle, C., Seitz, R.J., Freund, H.-J. (2001) Tactile Apraxia: unimodal apractic disorder of tactile object exploration associated with parietal lobe lesions. Brain, 124, 132–144. Blomqvist, G., Seitz, R.J., Sjögren, I., Halldin, C., Stone-Elander, S., Widen, L., Solin, O. and Haaparanta, M. (1994) Regional cerebral oxidative and total glucose consumption during rest and activation studied with positron emission tomography. Acta Physiologica Scandinavica, 151, 29–43. Bohm, C., Greitz, T., Seitz, R.J. and Eriksson, L. (1991) Specification and selection of regions of interest (ROIs) in a computerized brain atlas. Journal of Cerebral Blood Flow and Metabolism, 11 (Suppl 1), A64–A68. Bogousslavsky, J., Regli, F. and Assal, G. (1986) The syndrome of unilateral tuberothalamic artery territory infarction. Stroke, 17, 434–441. Bolay, H. and Dalkara, T. (1998) Mechanisms of motor dysfunction after transient MCA occlusion: persistent transmission failure in cortical synapses is a major determinant. Stroke, 29, 1988–1994. Brühl, C. and Witte, O.W. (1995) Cellular activity underlying altered brain metabolism during focal epileptic activity. Annals of Neurology, 38, 414–420. Büchel, C. and Friston, K.J. (1997) Modulation of connectivity in visual pathways by attention: cortical interactions evaluated with structural equation modelling and fMRI. Cerebral Cortex, 7, 768–778. Buchkremer-Ratzmann, I. and Witte, O.W. (1997) Extended brain disinhibition following small photothrombotic lesions in rat frontal cortex. NeuroReport, 8, 519–522. Buckner, R.L., Goodman, J., Burock, M., Rotte, M., Koutstaal, W., Schacter, D., Rosen, B. and Dale, A.M. (1998) Functional-anatomic correlates of object priming in humans revealed by rapid presentation of event-related fMRI. Neuron, 20, 285–296. Buxton, R.B. and Frank, L.R. (1997) A model for the coupling between cerebral blood flow and oxygen metabolism during neural stimulation. Journal of Cerebral Blood Flow and Metabolism, 17, 64–72. Calamante, F., Thomas, D.L., Pell, G.S., Wiersma, J. and Turner, R. (1999) Measuring cerebral blood flow using magnetic resonance imaging techniques. Journal of Cerebral Blood Flow and Metabolism, 9, 701–735. Caselli, R.J. (1993) Ventrolateral and dorsomedial somatosensory association cortex damage produces distinct somesthetic syndromes in humans. Neurology, 43, 762–771. Chesselet, M.F. (2000) Mapping the basal ganglia. In: A.W. Toga and J.C. Mazziotta (eds), Brain Mapping: the systems. Academic Press: San Diego, pp. 177–206. Christensen, G.E., Rabbitt, R.D. and Miller, M.I. (1994) 3-D brain mapping using deformable neuroanatomy. Physics in Medicine and Biology, 39, 609–618. Classen, J., Kunesch, E., Binkofski, F., Hilperath, F., Schlaug, G., Seitz, R., Glickstein, M. and Freund, H.-J. (1995) Subcortical origin of visuomotor apraxia. Brain, 118, 1365–1374. Classen, J., Schnitzler, A., Binkofski, F., Werhahn, K.J., Kim, Y.-S., Kessler, K.R. and Benecke, R. (1997) The motor syndrome associated with exaggerated inhibition within the primary motor cortex. Brain, 120, 605–619. Classen, J., Knorr, U., Werhahn, K.J., Schlaug, G., Kunesch, E., Cohen, L.G., Seitz, R.J. and Benecke, R. (1998a) Output mapping of human central motor representation: multiple modalities and multiple spatial scales. Journal of Physiology (London), 512, 163–179. Classen, J., Liepert, J., Wise, S.P., Hallett, M. and Cohen, L.G. (1998b) Rapid plasticity of human cortical movement representation induced by practice. Journal of Neurophysiology, 79, 1117–1123. Cremer, J.E., Cunninham, V.J. and Seville, M.P. (1983) Relationsphips between extraction and metabolism of glucose, blood flow, and tissue blood volume in regions of rat brain. Journal of Cerebral Blood Flow and Metabolism, 3, 291–302. Creutzfeld, O. and Ojemann, G. (1989) Neuronal activity in the human lateral temporal lobe. I. Responses to speech. Experimental Brain Research, 77, 451–475. Dale, A.M. and Sereno, M.I. (1993) Improved localization of cortical activity by combining EEG and MEG with MRI cortical surface reconstruction: a linear approach. Journal of Cognitive Neuroscience, 5, 162–176. Deppe, M., Knecht, S., Papke, K., Lohmann, H., Fleischer, H., Heindel, W., Ringelstein, E.B. and Henningsen, H. (2000) Assessment of hemispheric language lateralization: a comparison between fMRI and fTCD. Journal of Cerebral Blood Flow and Metabolism, 20, 263–268. de Peralta Menendez, R.G., Hauk, O., Andino, S.G., Vogt, H. and Michel, C. (1997) Linear inverse solutions with optimal resolution kernels applied to electromagnetic tomography. Human Brain Mapping, 5, 454–467. D’Esposito, M., Aguirre, G.K., Zarahn, E. and Ballard, D. (1997) Functional MRI studies of spatial and nonspatial working memory. Cognitive Brain Research, 7, 1–13. Dietrich, W.D., Ginsberg, M.D., Busto, R. and Smith, D.W. (1985) Metabolic alterations in rat somatosensory cortex following unilateral vibrissal removal. Journal of Neuroscience, 5, 874–880.


100

Rüdiger J. Seitz

Dominkus, M., Grisold, W. and Jelinek, V. (1990) Transcranial electrical motor evoked potentials as a prognostic indicator for motor recovery in stroke patients. Journal of Neurology Neurosurgery and Psychiatry, 53, 745–748. Drevets, W.C., Burton, H., Videen, T.O., Snyder, A.Z., Simpson, J.R.Jr. and Raichle, M.E. (1995) Blood flow changes in human somatosensory cortex during anticipated stimulation. Nature (London), 373, 249–252. Ebeling, U., Schmid, U.D., Ying, H. and Reulen, H.J. (1992) Safe surgery of lesions near the motor cortex using intra-operative mapping techniques: a report on 50 patients. Acta Neurochirurgia, 119, 23–28. Ebersole, J.S. (1997) Magnetoencephalography/magnetic source imaging in the assessment of patients with epilepsy. Epilepsia, 38 (Suppl 4), S1–S5. Erecinska, M. and Silver, I.A. (1989) ATP and Brain Function. Journal of Cerebral Blood Flow and Metabolism, 9, 2–19. Evans, A.C., Beil, C., Marrett, S., Thompson, C.J. and Hakim, A. (1988) Anatomical-functional correlation using an adjustable MRI-based region of interest atlas with positron emission tomography. Journal of Cerebral Blood Flow and Metabolism, 8, 513–530. Eysel, U.T. (1992) Cortical plasticity: remodelling receptive fields in sensory cortices. Current Biology, 2, 389–391. Fetz, E.E. (1993) Cortical mechanisms controlling limb movement. Current Opinion in Neurobiology, 3, 932–939. Fischl, B., Sereno, M.I., Tootell, R.B.H. and Dale, A.M. (1999) High-resolution intersubject averaging and a coordinate system for the cortical surface. Human Brain Mapping, 8, 272–284. Foerster, O. (1936) Motorische Felder und Bahnen. In: O. Bumke and O. Foerster (eds), Handbuch der Neurologie, Vol. 6, Allgemeine Neurologie. Berlin: Julius Springer Verlag, pp. 1–357. Fox, P.T. and Raichle, M.E. (1984) Stimulus rate dependence of regional cerebral blood flow in human striate cortex, demonstrated by positron emission tomography. Journal of Neurophysiology, 51, 1109–1120. Fox, P.T., Burton, H. and Raichle, M.E. (1987a) Mapping human somatosensory cortex with positron emission tomography. Journal of Neurosurgery, 67, 34–43. Fox, P.T., Miezin, F.M., Allman, J.M., van Essen, D.C. and Raichle, M.E. (1987b) Retinotopic organization of human visual cortex mapped with positron emission tomography. Journal of Neuroscience, 7, 913–922. Fox, P.T., Raichle, M.E., Mintun, M.A. and Dence, C. (1988) Nonoxidative glucose consumption during focal physiologic neural activity. Science, (Washington), 241, 462–464. Fox, P.T. (1997) The growth of human brain mapping. Human Brain Mapping, 5, 1–2. Frackowiak, R.S.J. (1994) Functional mapping of verbal memory and language. Trends in Neuroscience, 17, 109–115. Frahm, J., Bruhn, H., Merboldt, K.-D. and Hänicke, W. (1992) Dynamic MR imaging of human brain oxygenation during rest and photic stimulation. Journal of Magnetic Resonance Imaging, 2, 501–505. Friel, K.M. and Nudo, R.J. (1998) Recovery of motor function after focal cortical injury in primates: compensatory movement patterns used during rehabilitative training. Somatosensory and Motor Research, 15, 173–189. Frisk, V. and Milner, B. (1990) The role of the left hippocampal region in the acquisition and retention of story content. Neuropsychologia, 28, 349–359. Friston, K.J., Frith, C.D., Liddle, P.F. and Frackowiak, R.S.J. (1991) Investigating a network model of word generation with positron emission tomography. Proceedings of the Royal Society of London, B, 244, 101–106. Friston, K.J., Frith, C.D., Liddle, P.F. and Frackowiak, R.S.J. (1993) Functional connectivity: the principalcomponent analysis of large (PET) data sets. Journal of Cerebral Blood Flow and Metabolism, 13, 5–14. Friston, K.J. (1994) Functional and effective connectivity in neuroimaging: a synthesis. Human Brain Mapping, 2, 56–78. Friston, K.J., Holmes, A.P., Worsley, K.J., Poline, J.-P., Frith, C.D. and Frackowiak, R.S.J. (1994) Statistical parametric maps in functional imaging: a generalized linear approach. Human Brain Mapping, 2, 189–210. Friston, K.J. (1997) Imaging cognitive anatomy. Trends in Cognitive Science, 1, 21–27. Friston, K.J., Fletcher, P., Josephs, O., Holmes, A., Rugg, M.D. and Turner, R. (1998) Event-related fMRI: Characterizing differential responses. Neuroimage, 7, 30–40. Frostig, R.D., Lieke, E.E., Ts’o, D. and Grinvald, A. (1990) Cortical functional architecture and local coupling between neuronal activity and the microcirculation revealed by in vivo high-resolution optical imaging of intrinsic signals. Proceedings of the National Academy of Science, U.S.A., 87, 6082–6086. Fukuyama, H., Ouchi, Y., Matsuzaki, S., Nagahama, Y., Yamauchi, H., Ogawa, M., Kimura, J. and Shibasaki, H. (1997) Brain functional activity during gait in normal subjects: a SPECT study. Neuroscience Letters, 228, 183–186. Gadian, D.G., Mishkin, M. and Vargha-Khadem, F. (1999) Early brain pathology and its relation to cognitive impairment: the role of quantitative magnetic resonance techniques. In: P. Stefan, F. Andermann, S.A. Shorvon and P. Chauvel (eds), Plasticity and epilepsy. (Advances in Neurology, Volume 81). New York: Raven Press, pp. 307–315. Gevins, A., Cutillo, B., DuRousseau, D., Le, J., Leong, H., Martin, N., Smith, M.E., Bressler, S., Bricket, P., McLaughlin, J., Barbero, N. and Laxer, K. (1994) Imaging the spatiotemporal dynamics of cognition with high-resolution evoked potential methods. Human Brain Mapping, 1, 101–116. George, J.S., Aine, C.J., Mosher, J.C., Schmidt, D.M., Ranken, D.M., Schlitt, H.A., Wood, C.C., Lewine, J.D., Sanders, J.A. and Belliveau, J.W. (1995) Mapping function in the human brain with magnetoencephalo-



101

graphy, anatomical magnetic resonance imaging, and functional magnetic resonance imaging. Journal of Clinical Neurophysiology, 12, 406–431. Gerloff, C., Grodd, W., Altenmüller, E., Kolb, R., Naegele, T., Klose, U., Voigt, K. and Dichgans, J. (1996) Coregistration of EEG and fMRI in a simple motor task. Human Brain Mapping, 4, 199–209. Geyer, S., Ledberg, A., Schleicher, A., Kinomura, S., Schormann, T., Bürgel, U., Klingberg, T., Larsson, J., Zilles, K. and Roland, P.E. (1996) Two different areas within the primary motor cortex of man. Nature (London), 382, 805–807. Gilman, S., Dauth, G.W., Frey, K.A. and Penney, J.B. (1987) Experimental hemiplegia in the monkey: basal ganglia glucose activity during recovery. Annals of Neurology, 22, 370–376. Grabowski, T.J., Frank, R.J., Brown, C.K., Damasio, H., Boles Ponto, L.L., Watkins, G.L. and Hichwa, R.D. (1996) Reliability of PET activation across statistical methods, subject groups, and samples sizes. Human Brain Mapping, 4, 23–46. Grafton, S.T., Mazziotta, J.C., Presty, S., Friston, K.J., Frackowiak, R.S.J. and Phelps, M.E. (1992) Functional anatomy of human procedural learning determined with regional cerebral blood flow and PET. Journal of Neuroscience, 12, 2542–2548. Gray, C.M., Engel, A.K., König, P. and Singer, W. (1990) Stimulus-dependent neuronal oscillations in cat visual cortex: receptive field properties and feature dependence. European Journal of Neuroscience, 2, 607–619. Grinvald, A., Frostig, R.D., Siegel, R.M. and Bartfeld, E (1991) High-resolution optical imaging of functional brain architecture in the awake monkey. Proceedings of the National Academy of Science, U.S.A., 88, 11559–11563. Gulyas, B. (1997) Functional organization of human visual cortical areas. In: K.S. Rockland, J.H. Kaas and A. Peters (eds), Cerebral Cortex, Vol 12, New York: Plenum Press, pp. 743–775. Gur, R.C., Turetsky, B., Matsui, M., Yan, M, Bilker, W., Hughett, P. and Gur, R.E. (1999) Sex differences in brain gray and white matter in healthy young adults: correlations with cognitive performance. Journal of Neuroscience, 15, 4065–4072. Haglund, M.M., Ojemann, G.A. and Hochman, D.W. (1992) Optical imaging of epileptiform and functional activity in human cerebral cortex. Nature (London), 358, 668–671. Hämäläinen, M., Hari, R., Ilmoniemi, R.J., Knuutila, J. and Lounasmaa, O.V. (1993) Magnetoencephalography—theory, instrumentation, and applications to noninvasive studies of the working human brain. Reviews of Modern Physics, 65, 413–497. Hari, R. and Lounasmaa, O.V. (1989) Recording and interpretation of cerebral magnetic fields. Science, (Washington), 244, 432–436. Hasnain, M.K., Fox, P.T. and Woldorff, M.G. (1998) Intersubject variability of functional areas in the human visual cortex. Human Brain Mapping, 6, 301–315. Haxby, J.V., Horwitz, B., Ungerleider, L.G., Maisog, J.M., Pietrini, P. and Grady, C.L. (1994) The functional organization of human extrastriate cortex: a PET-rCBF study of selective attention to faces and locations. Journal of Neuroscience, 14, 6336–6353. Heald, A., Bates, D., Cartlidge, N.E.F., French, J.M. and Miller, S. (1993) Longitudinal study of motor conduction time following stroke. 2. Central motor conduction measured within 72 h after stroke as a predictor of functional outcome at 12 months. Brain, 116, 1371–1385. Heinze, H.J., Mangun, G.R., Burchert, W., Hinrichs, H., Scholz, M., Münte, T.F., Gös, A., Scherg, M., Johannes, S., Hundeshagen, H., Gazzaniga, M.S. and Hillyard, S.A. (1994) Combined spatial and temporal imaging of brain activity during visual selective attention in humans. Nature (London), 372, 543–546. Heiss, W.D., Kessler, J., Thiel, A., Ghaemi, M. and Karbe, H. (1999) Differential capacity of left and right hemispheric areas for compensation of poststroke aphasia. Annals of Neurology, 45, 430–438. Herzog, H., Seitz, R.J., Tellmann, L. and Müller-Gärtner, H.-W. (1996) Quantification of regional cerebral blood flow with 15O-butanol and positron emission tomography in humans. Journal of Cerebral Blood Flow and Metabolism, 16, 645–649. Hoge, R.D., Atkinson, J., Gill, B., Crelier, G.R., Marrett, S. and Pike, G.B. (1999) Linear coupling between cerebral blood flow and oxygen consumption in activated human cortex. Proceedings of the National Academy of Science, U.S.A., 96, 9403–9408. Horwitz, B. (1994) Data analysis paradigms for metabolic-flow data: combining neural modeling and functional neuroimaging. Human Brain Mapping, 2, 112–122. Hubel, D.H. and Wiesel, T.N. (1963) Shape and arrangement of columns in cat’s striate cortex. Journal of Physiology (London), 165, 559–568. Hubel, D.H., Wiesel, T.N. and Stryker, M.P. (1978) Anatomical demonstration of orientation columns in macaque monkey. Journal of Comparative Neurology, 177, 361–380. Hunton, D.L., Miezin, F.M., Buckner, R.L., van Mier, H.I., Raichle, M.E. and Petersen, S.E. (1996) An assessment of functional-anatomical variability in neuroimaging studies. Human Brain Mapping, 4, 122–139. Ibánez, V., Deiber, M.P., Sadato, N., Toro, C., Grissom, J., Woods, R.P., Mazziotta, J.C., and Hallett, M. (1995) Effects of stimulus rate of regional cerebral blood flow after median nerve stimulation. Brain, 118, 1339–1351.


102

Rüdiger J. Seitz

Iadecola, C. (1993) Regulation of the cerebral microcirculation during neural activity: is nitric oxide the missing link? Trends in Neuroscience, 16, 206–214. Jenkins, W.M. and Merzenich, M.M. (1987) Reorganization of neocortical representations after brain injury: a neurophysiological model of the bases of recovery from stroke. In: F.J. Seil, E. Herbert and B.M. Carlson (eds), Neural regeneration. (Progress in Brain Research, Vol. 71). Amsterdam: Elsevier, pp. 249–266. Jenkins, I.H., Brooks, D.J., Nixon, P.D., Frackowiak, R.S.J. and Passingham, R.E. (1994) Motor sequence learning: a study with positron emission tomography. Journal of Neuroscience, 14, 3775–3790. Jokeit, H., Seitz, R.J., Markowitsch, H.J., Neumann, N., Witte, O.W. and Ebner, A. (1997) Prefrontal asymmetric interictal glucose hypometabolism and cognitive impairment in patients with temporal lobe epilepsy. Brain, 120, 2283–2294. Juliano, S.L., Hand, P.J. and Whitsel, B.L. (1981) Patterns of increased metabolic activity in somatosensory cortex of monkeys macaca fascicularis, subjected to controlled cutaneous stimulation: a 2-deoxyglucose study. Journal of Neurophysiology, 46, 1260–1284. Juliano, S.L., Hand, P.J. and Whitsel, B.L. (1983) Patterns of metabolic activity in cytoarchitectural area SII and surrounding cortical fields of the monkey. Journal of Neurophysiology, 50, 961–980. Kaas, J.H. (1993) The functional organization of somatosensory cortex in primates. Anatomische Anziege, 175, 509–518. Kadekaro, M., Crane, A.M. and Sokoloff, L. (1985) Differential effects of electrical stimulation of sciatic nerve on metabolic activity in spinal cord and dorsal root ganglion in the rat. Proceedings of the National Academy of Science, U.S.A., 82, 6010–6013. Karni, A., Meyer, G., Jezzard, P., Adams, M.M., Turner, R. and Ungerleider, L.G. (1995) Functional MRI evidence for adult motor cortex plasticity during motor skill learning. Nature (London), 377, 155–158. Kastrup, A., Li, T.-Q., Glover, G.H., Krüger, G. and Moseley, M.E. (1999) Gender differences in cerebral blood flow and oxygenation response during focal physiological neural activity. Journal of Cerebral Blood Flow and Metabolism, 19, 1066–1071. Kennedy, C., DesRosiers, M.H., Skurada, O., Shinohara, M., Reivich, M., Jehle, W. and Sokoloff, L. (1976) Metabolic mapping of the primary visual system of the monkey by means of the autoradiographic [14C]deoxyglucose technique. Proceedings of the National Academy of Science, U.S.A., 73, 4230–4234. Kennedy, D.N., Lange, N., Makris, N., Bates, J., Meyer, J. and Caviness, Jr., V.S. (1998) Gyri of the human neocortex: an MRI-based analysis of volume and variance. Cerebral Cortex, 8, 372–384. Kertezs, A. and Ferro, J.M. (1984) Lesion size and location in ideomotor apraxia. Brain, 107, 921–933. Kincses, W.E., Brain, C., Kaiser, S. and Elbert, T. (1999) Modeling extended sources of event-related potentials using anatomical and physiological constraints. Human Brain Mapping, 8, 182–193. Kislyakov, Y.Y. and Ivanov, K.P. (1986) O2 transport in cerebral microregions (mathematical simulation). Journal of Biomechanical Enginineering, 108, 28–32. Kleinschmidt, A., Merboldt, K.D., Hänicke, W., Steinmetz, H. and Frahm, J. (1994) Correlational imaging of thalamocortical coupling in the primary visual pathway of the human brain. Journal of Cerebral Blood Flow and Metabolism, 14, 952–957. Kleinschmidt, A., Büchel, C., Zeki, S. and Frackowiak, R.S.J. (1998) Human brain activity during spontaneously reversing perception of ambiguous figures. Proceedings of the Royal Society London, B, 265, 2427–2433. Knorr, U., Weder, B., Kleinschmidt, A., Wirrwar, A., Huang, Y., Herzog, H. and Seitz, R.J. (1993) Identification of task-specific rCBF changes in individual subjects: validation and application for positron emission tomography. Journal of Computer Assisted Tomography, 17, 517–528. Korvenoja, A., Huttunen, J., Salli, E., Pohjonen, H., Martinkauppi, S., Palva, J.M., Lauronen, L., Virtanen, J., Ilmoniemi, R.J. and Aronen, H.J. (1999) Activation of multiple cortical areas in response to somatosensory stimulation: combined magnetoencephalographic and functional magnetic resonance imaging. Human Brain Mapping, 8, 12–27. Kosslyn, S.M., Alpert, N.M. and Thompson, W.L. (1995) Identifying objects at different levels of hierarchy: a positron emission tomography study. Human Brain Mapping, 3, 107–132. Kosslyn, S.M., Pascual-Leone, A., Felician, O., Camposano, S., Keenan, J.P., Thompson, W.L., Ganis, G., Sukel, K.E. and Alpert, N.M. (1999) The role of area 17 in visual imagery: convergent evidence from PET and rTMS. Science (Washington), 284, 167–170. Kossut, M., Hand, P.J., Greenberg, J. and Hand, C.L. (1988) Single vibrissal cortical column in SI cortex of rat and its alterations in neonatal and adult vibrissa-deafferent animals: a quantitative 2DG study. Journal of Neurophysiology, 60, 829–852. Krakow, K., Woermann, F.G., Symms, M.R., Allen, P.J., Lemieux, L., Barker, G.J., Duncan, J.S. and Fish, D.R. (1999) EEG-triggered functional MRI of interictal epileptiform activity in patients with partial seizures. Brain, 122, 1679–1688. Kristeva, R., Cheyne, D. and Deecke, L. (1991) Neuromagnetic fields accompanying unilateral and bilateral voluntary movements: topography and analysis of cortical sources. Electroencephalography and Clinical Neurophysiology, 81, 284–298.



103

Kuschinsky, W., Suda, S. and Sokoloff, L. (1981) Local cerebral glucose utilization and blood flow during metabolic acidosis. American Journal of Physiology, 241, H772–H777. Lemon, R. (1988) The output map of the primate motor cortex. Trends in Neuroscience, 11, 501–506. Logothetis, N.K., Pauls, J., Augath, M., Trinath, T. and Oeltermann, A. (2001) Neurophysiological investigation of the basis of the fMRI signal. Nature, 412, 150–157. Lueck, C.J., Zeki, S., Friston, K.J., Deiber, M.-P., Cope, P., Cunningham, V.J., Lammertsma, A.A., Kennard, C. and Frackowiak, R.S.J. (1989) The colour centre in the cerebral cortex of man. Nature (London), 340, 386–389. Madsen, P.L., Cruz, N.F., Sokoloff, L. and Dienel, G.A. (1999) Cerebral oxygen/glucose ratio is low during sensory stimulation and rises above normal during recovery: excess glucose consumption during stimulation is not accounted for by lactate efflux from or accumulation in brain tissue. Journal of Cerebral Blood Flow and Metabolism, 19, 393–400. Magistretti, P.J. and Pellerin, L. (1999) Cellular mechanisms of brain energy metabolism and their relevance to functional brain imaging. Philosophical Transactions of the Royal Society, London, series B, 354, 1155–1163. Malonek, D. and Grinvald, A. (1996) Interactions between electrical activity and cortical microcirculation revealed by imaging spectroscopy: implications for functional brain mapping. Science (Washington), 272, 551–554. Marin, G., Guerin, C., Baillet, S., Garnero, L. and Meunier, G. (1998) Influence of skull anisotropy for the forward and inverse problem in EEG: simulation studies using FEM on realistic head models. Human Brain Mapping, 6, 250–269. Mata, M., Fink, D.J., Gainer, H., Smith, C.B., Davidsen, L., Saviki, H., Schwartz, W.J. and Sokoloff, L. (1980) Activity-dependent energy metabolism in rat posterior pituitary primarily reflects sodium pump activity. Journal of Neurochemistry, 34, 213–215. Maynard, E.M., Hatsopoulos, N.G., Ojakangas, C.L., Acuna, B.D., Sanes, J.N., Normann, R.A. and Donoghue, J.P. (1999) Neuronal interactions improve cortical population coding of movement direction. Journal of Neuroscience, 19, 8083–8093. Mazziotta, J.C., Phelps, M.E., Plummer, D. and Kuhl, D.E. (1981) Quantitation in positron emission computed tomography: 5. physical-anatomical effects. Journal of Computer Assisted Tomography, 5, 734–743. McIntosh, A.R. and Gonzalez-Lima, F. (1994) Structural equation modeling and its application to network analysis in functional brain imaging. Human Brain Mapping, 2, 23–44. McIntosh, A.R., Grady, C.L., Ungerleider, L.G., Haxby, J.V., Rapoport, S.I. and Horwitz, B. (1994) Network analysis of cortical visual pathways mapped with PET. Journal of Neuroscience, 14, 655–666. Menon, R.S. and Kim, S.-G. (1999) Spatial and temporal limits in cognitive neuroimaging with fMRI. Trends in Cognitive Science, 3, 207–216. Mesulam, M.M. (1981) A cortical network for directed attention and unilateral neglect. Annals of Neurology, 10, 309–325. Mesulam, M.M. (1990) Large-scale neurocognitive networks and distributed processing for attention, language, and memory. Annals of Neurology, 28, 597–613. Missimer, J., Knorr, U., Maguire, R.P., Herzog, H., Seitz, R.J., Tellman, L. and Leenders, K.L. (1999) On two methods of statistical image analysis. Human Brain Mapping, 8, 245–258. Mountcastle, V.B., Talbot, W.H., Sakata, H. and Hyvärinen, J. (1969) Cortical neuronal mechanisms in fluttervibration studied in unanaesthetized monkeys: Neural periodicity and frequency discrimination. Journal of Neurophysiology, 32, 452–484. Müller, R.-A., Rothermal, R.D., Behen, M.E., Muzik, O., Mangner, Rj. and Chugani, H.T. (1998) Differential patterns of language and motor reorganization following early left hemispheric lesion. Archives of Neurology, 55, 1113–1119. Näätänen, R., Ilmoniemi, R.J. and Alho, K. (1994) Magnetoencephalography in studies of human cognitive brain function. Trends in Neuroscience, 17, 389–395. Neumann-Haefelin, T., Wittsack, H.-J., Wenserski, F., Siebler, M., Seitz, R.J., Mödder, U. and Freund, H.-J. (1999) Diffusion- and perfusion-weighted MRI. The DWI/PWI mismatch region in acute stroke. Stroke, 30, 1591–1597. Nudo, R.J., Wise, B.M., SiFurentes, F. and Milliken, G.W. (1996) Neural substrates for the effects of rehabilitative training on motor recovery after ischemic infarct. Science (Washington), 272, 1791–1794. Ojemann, G., Ojemann, J., Lettich, E. and Berger, M. (1989) Cortical language localization in left, dominant hemisphere. An electrical stimulation mapping investigatin in 117 patients. Journal of Neurosurgery, 71, 316–326. O’Sullivan, B.T., Roland, P.E. and Kawashima, R. (1994). A PET study of somatosensory discrimination in man: Microgeometry versus macrogeometry. European Journal of Neuroscience, 6, 137–148. Pascual-Leone, A., Grafman, J. and Hallett, M. (1994) Modulation of cortical motor output maps during development of implicit and explicit knowledge. Science (Washington), 263, 1287–1289. Pascual-Leone, A., Wassermann, E.M., Sadato, N. and Hallett, M. (1995) The role of reading activity on the modulation of motor cortical outputs to the reading hand in Braille readers. Annals of Neurology, 38, 910–915. Paus, T., Marrett, S., Worsley, K. and Evans, A. (1996) Imaging motor-to-sensory discharges in the human brain: an experimental tool for the assessment of functional connectivity. Neuroimage, 4, 78–86.


104

Rüdiger J. Seitz

Pendlebury, S.T., Blamire, A.M., Lee, M.A., Style, P. and Matthews, P.M. (1999) Axonal injury in the internal capsule correlates with motor impairment after stroke. Stroke, 30, 956–962. Penfield, W. and Boldrey, E. (1953) Somatic motor and sensory representation in the cerebral cortex of man as studied by electrical stimulation. Brain, 60, 389–443. Penfield, W. and Jasper, H. (1954) Epilepsy and the functional anatomy of the human brain. Little, Brown, Boston. Peters, A. and Kara, D.A. (1987) The neuronal composition of area 17 of rat visual cortex. IV. The organization of pyramidal cells. Journal of Comparative Neurology, 260, 573–590. Petersen, S.E., Fox, P.T., Posner, M.I., Mintun, M. and Raichle, M.E. (1988) Positron emission tomographic studies of the cortical anatomy of single-word processing. Nature (London), 331, 585–589. Phillips, J.R., Johnson, K.O. and Hsiao, S.S. (1988) Spatial pattern representation and transformation in monkey somatosensory cortex. Proceedings of the National Academy of Science, U.S.A., 85, 1317–1321. Poeck, K., deBleser, R. and Graf v. Keyserlingk, D. (1984) Neurolinguistic status and localization of lesion in aphasic patients with exclusively consonant-vowel recurring utterances. Brain, 107, 199–217. Poline, J.B., Vandenberghe, R., Holmes, A.P., Friston, K.J. and Frackowiak, R.S.J. (1996) Reproducibility of PET activation studies: lessons from a multi-center European experiment. EU concerted action on functional imaging. Neuroimage, 4, 34–54. Porro, C.A., Francescato, M.P., Cettolo, V., Diamond, M.E., Baraldi, P., Zuiani, C., Bazzocchi, M. and di Prampero, P.E. (1996) Primary motor and sensory cortex activation during motor performance and motor imagery: a functional magnetic resonance imaging study. Journal of Neuroscience, 16, 7688–7698. Prichard, J.W. and Brass, L.M. (1992) New anatomical and functional imaging methods. Annals of Neurology, 32, 395–400. Price, C.J., Moore, C.J. and Friston, K.J. (1997) Subtractions, conjunctions, and interactions in experimental design of activation studies. Human Brain Mapping, 5, 264–272. Rademacher, J., Caviness, V.S., Steinmetz, H. and Galaburda, A.M. (1993) Topographical variation of the human primary cortices: implications for neuroimaging and brain mapping studies. Cerebral Cortex, 3, 313–329. Raichle, M.E. (1987) Circulatory and metabolic correlates of brain function in normal humans. In: F. Plum (ed.), Handbook of Physiology—Nervous System Volume V, Higher Functions of the brain. Bethesda, M.D.: American Physiological Society, pp. 643–674. Rao, S.M., Binder, J.R., Bandettini, P.A., Hammeke, T.A., Yetkin, F.Z., Jesmanowicz, A., Lisk, L.M., Morris, G.L., Mueller, W.M., Estkowski, L.D., Wong, E.C., Haughton, V.M. and Hyde, J.S. (1993) Functional magnetic resonance imaging of complex human movements. Neurology, 43, 2311–2318. Regard, M., Cook, N.D., Wieser, H.G. and Landis, T. (1994) The dynamics of cerebral dominance during unilateral limbic seizures. Brain, 117, 91–104. Ribary, U., Ioannides, A.A., Singh, K.D., Hasson, R., Bolton, J.P., Lado, F., Mogilner, A. and Llinas, R.R. (1991) Magnetic field tomography of coherent thalamocortical 40-Hz oscillations in humans. Proceedings of the National Academy of Science, U.S.A., 88, 11037–11041. Rijntjes, M., Dettmers, C., Buchel, C., Kiebel, S., Frackowiak, R.S. and Weiller, C. (1999) A blueprint for movement: functional and anatomical representations in the human motor system. Journal of Neuroscience, 19, 8043–8048. Ringelstein, E.B., Binieck, R., Weiller, C., Ammeling, B., Nolte, P.N. and Thron, A. (1992) Type and extent of hemispheric brain infarctions and clinical outcome in early and delayed middle cerebral artery recanalization. Neurology, 42, 289–298. Rizzolatti, G., Luppino, G. and Matelli, M. (1998) The organization of the cortical motor system: new concepts. Electroencephalography and Clinical Neurophysiology, 106, 283–296. Roland, P.E., Eriksson, L., Stone-Elander, S. and Widén, L. (1987) Does mental activity change the oxidative metabolism of the brain? Journal of Neuroscience, 7, 2373–2389. Roland, P.E. (1993) Brain activation. John Wiley, New York. Roland, P.E. and Gulyas, B. (1994) Visual imagery and visual representation. Trends in Neuroscience, 17, 281–287. Roland, P.E., Geyer, S., Amunts, K., Schormann, T., Schleicher, A., Malikovic, A. and Zilles, K. (1997) Cytoarchitectural maps of the human brain in standard anatomical space. Human Brain Mapping, 5, 222–227. Romani, G.L. and Rossini, P. (1988) Neuromagnetic functional localization: principles, state of the art, and perspectives. Brain Topography, 1, 5–21. Rondot, P., DeRecondo, J. and Ribadeu Duma, J.L. (1977) Visuomotor ataxia. Brain, 100, 355–376. Roth, M., Decety, J., Raybaudi, M., Massarelli, R., Delon-Martin, C., Segebarth, C., Morand, S., Gemignani, A., Décorps, M. and Jeannerod, M. (1996) Possible involvement of primary motor cortex in mentally simulated movement: a functional magnetic resonance imaging study. NeuroReport, 7, 1280–1284. Scherg, M. (1990) Fundamentals of dipole source potential analysis. In: F. Grandori, M. Hoke and G.L. Romani (eds), Auditory evoked magnetic fields and electrical potentials. (Advances in Audiology, Vol. 6), Basel: Karger, pp. 40–69. Schieber, M.H. and Hibbard, L.S. (1993) How somatotopic is the motor cortex hand area? Science (Washington), 261, 489–492.



105

Schlaug, G., Knorr, U. and Seitz, R.J. (1994) Inter-subject variability of cerebral activations in acquiring a motor skill. A study with positron emission tomography. Experimental Brain Research, 98, 523–534. Schlaug, G., Jäncke, L., Huang, Y. and Steinmetz, H. (1995) In vivo evidence of structural brain asymmetry in musicians. Science, Washington, 267, 699–701. Schmahmann, J.D. (2000) Cerebellum and Brainstem. In: A.W. Toga and J.C. Mazziotta (eds), Brain Mapping: the systems. San Diego: Academic Press, pp. 207–259. Schneider, R. and Gautier, J.-C. (1994) Leg weakness due to stroke. Site of lesions, weakness patterns and causes. Brain, 117, 347–354. Schnitzler, A., Witte, O.W., Cheyne, D., Haid, G., Vrba, J. and Freund, H.J. (1995). Modulation of somatosensory evoked magnetic fields by sensory and motor interferences. NeuroReport, 6, 1653–1658. Schnitzler, A., Volkmann, J., Enck, J., Frieling, T., Witte, O.W. and Freund, H.-J. (1999) Different cortical organization of visceral and somatic sensation in humans. European Journal of Neuroscience, 11, 305–315. Schormann, T. and Zilles, K. (1998) Three-dimensional linear and nonlinear transformations: an integration of light microscopical and MRI data. Human Brain Mapping, 6, 339–347. Seitz, R.J., Bohm, C., Greitz, T., Roland, P.E., Eriksson, L., Blomkvist, G., Rosenkvist, G. and Nordell, B. (1990) Accuracy and precision of the computerized brain atlas programme (CBA) for localization and quantification in positron emission tomography. Journal of Cerebral Blood Flow and Metabolism, 10, 443–457. Seitz, R.J., Roland, P.E., Bohm, C., Greitz, T. and Stone-Elander, S. (1991) Somatosensory discrimination of shape: tactile exploration and cerebral localization. European Journal of Neuroscience, 3, 481–492. Seitz, R.J. and Roland, P.E. (1992a) Vibratory stimulation increases and decreases the regional cerebral blood flow and oxidative metabolism: a positron emission tomography (PET) study. Acta Neurologica Scandinavica, 86, 60–67. Seitz, R.J. and Roland, P.E. (1992b) Learning of finger movement sequences: a combined kinematic and positron emission tomography study. European Journal of Neuroscience, 4, 154–165. Seitz, R.J., Schlaug, G., Kleinschmidt, A., Knorr, U., Nebeling, B., Wirrwar, A., Steinmetz, H., Benecke, R. and Freund, H.-J. (1994) Remote depressions of cerebral metabolism in hemiparetic stroke: topography and relation to motor and somatosensory functions. Human Brain Mapping, 1, 81–100. Seitz, R.J., Schlaug, G., Knorr, U., Steinmetz, H., Tellmann, L. and Herzog, H. (1996) Neurophysiology of the human supplementary motor area: positron emission tomography. In: H.O. Lueders (ed.), Supplementary sensorimotor area. (Advances in Neurology, Vol. 70). New York: Raven Press, pp. 167–175. Seitz, R.J., Canavan, A.G.M., Yagüez, L., Herzog, H., Tellmann, L., Knorr, U., Huang, Y. and Hömberg, V. (1997) Representations of graphomotor trajectories in the human parietal cortex: evidence for controlled processing and automatic performance. European Journal of Neuroscience, 9, 378–389. Seitz, R.J., Höflich, P., Binkofski, F., Tellmann, L., Herzog, H. and Freund, H.-J. (1998) Role of the premotor cortex for recovery from middle cerebral artery infarction. Archives of Neurology, 55, 1081–1088. Seitz, R.J. and Azari, N.P. (1999) Postlesional neuroplasticity in man. In: P. Stefan, F. Andermann, S.A. Shorvon and P. Chauvel (eds), Plasticity and epilepsy. (Advances in Neurology, Volume 81). New York: Raven Press, pp. 37–47. Seitz, R.J., Azari, N.P., Knorr, U., Binkofski, F., Herzog, H, and Freund, H.-J. (1999) The role of diaschisis in stroke recovery. Stroke, 30, 1844–1850. Seitz, R.J., Volkmann, J. and Witte, O.W. (2000) Functional Mapping of the Human Brain. The Neuroscientist, 6, 75–76. Seitz, R.J., Knorr, U., Azari, N.P. and Weder, B. (2001) Cerebral networks in sensorimotor disturbances. Brain Research Bulletin, 54, 299–305. Shulman, G.L., Fiez, J.A., Corbetta, M., Buckner, R.L., Miezin, F.M., Raichle, M.E. and Petersen, S.E. (1997) Common blood flow changes across visual tasks. II. Decreases in cerebral cortex. Journal of Cognitive Neuroscience, 9, 648–663. Sitzer, M., Knorr, U. and Seitz, R.J. (1994) Cerebral hemodynamics during sensorimotor activation in humans. Journal of Applied Physiology, 77, 2804–2811. Sokoloff, L., Reivich, M., Kennedy, C., Des Rosiers, M.H., Patlak, C.S., Pettigrew, K.D., Sakurada, O. and Shinohara, M. (1977) The [14C]deoxyglucose method for the measurement of local cerebral glucose utilization: theory, procedure, and normal values in the conscious and anesthetized albino rat. Journal of Neurochemistry, 28, 897–916. Snyder, A.Z. (1991) Dipole source localization in the study of EP generators: a critique. Electroencephalography and Clinical Neurophysiology, 80, 321–325. Stehling, M.K., Turner, R. and Mansfield, P. (1991) Echo-planar imaging: magnetic resonance imaging in a fraction of a second. Science, Washington, 254, 43–50. Steinmetz, H., Fürst, G. and Freund, H.-J. (1989) Cerebral cortical localization: application and validation of the proportional grid system in MR imaging. Journal of Computer Assisted Tomography, 13, 10–19. Steinmetz, H., Huang, Y., Seitz, R.J., Knorr, U., Herzog, H., Hackländer, T., Kahn, T. and Freund, H.-J. (1992a) Individual integration of positron emission tomography and high-resolution magnetic resonance imaging. Journal of Cerebral Blood Flow and Metabolism, 12, 919–926.


106

Rüdiger J. Seitz

Steinmetz, H., Jäncke, L., Kleinschmidt, A., Schlaug, G., Volkmann, J. and Huang, Y. (1992b) Sex but no hand difference in the isthmus of the corpus callosum. Neurology, 42, 749–752. Steinmetz, H., Herzog, A., Schlaug, G., Huang, Y. and Jäncke, L. (1995) Brain (a)symmetry in monogzygotic twins. Cerebral Cortex, 5, 296–300. Stephan, K.M., Fink, G.R., Passingham, R.E., Silberzweig, D., Ceballos-Baumann, A.O., Frith, C.D. and Frackowiak, R.S.J. (1995a) Functional anatomy of the mental representation of upper extremity movements in healthy subjects. Journal of Neurophysiology, 73, 373–386. Stephan, K.M., Dabringhaus, A., Fink, G.R., Geyer, S., Passingham, R.E., Schormann, T., Knorr, U., Seitz, R.J., Roland, P.E., Frith, C.D., Frackowiak, R.S.J. and Zilles, K. (1995b) Motor imagery and motor performance: PET activations compared to cytoarchitectonic data. Human Brain Mapping, Suppl 1, 301. Stephan, K.M. and Frackowiak, R.S.J. (1996) Motor imagery—anatomical representation and electrophysiological characteristics. Neurochemical Research, 21, 1105–1116 Stephan, K.M., Binkofski, F., Halsband, U., Dohle, C., Wunderlich, G., Schnitzler, A., Tass, P., Posse, S., Herzog, H., Sturm, V., Zilles, K., Seitz, R.J. and Zilles, K. (1999) The role of ventral medial wall motor areas for bimanual coordination: a combined lesion and activation study. Brain, 122, 351–368. Stepniewska, I., Preuss, T.M. and Kaas, J. (1993) Architectonics, somatotopic organization, and ipsilateral cortical connections of the primary motor area (M1) of Owl monkey. Journal of Comparative Neurology, 330, 238–271. Talairach, J., Szikla, G., Tournoux, P., Prossalentis, A., Bordas-Ferrer, M., Covello, L., Joacob, M., Mempel, A., Buser, P. and Bancaud, J. (1967) Atlas d’anatomie stéréotaxique du télencéphale. Paris: Masson. Talairach, J. and Tournoux, P. (1988) Co-planar stereotaxic atlas of the human brain. Thieme, Stuttgart, New York. Thurfjell, L., Bohm, C., Bengtsson, E. (1995) CBA—an atlas based software tool used to facilitate the interpretation of neuroimaging data. Computer Methods and Programs in Biomedicine, 47, 51–71. Toma, K., Honda, M., Hanakawa, T., Okada, T., Fukuyama, H., Ikeda, A., Nishizawa, S., Konishi, J. and Shibasaki, H. (1999) Activities of the primary and supplementary motor areas increase in preparation and execution of voluntary muscle relaxation: an event-related fMRI study. Journal of Neuroscience, 19, 3527–3534. Tong, D.C., Yenari, M.A., Albers, G.W., O’Brien, M., Marks, M.P. and Moseley, M.E. (1998) Correlation of perfusion- and diffusion-weighted MRI with NIHSS score in acute (

Cortical Areas: Unity and Diversity (Conceptual Advances in Brain Research)

The Female Brain, Second Edition (Conceptual Advances in Brain Research)

Time and the Brain (Conceptual Advances in Brain Research, Vol 3)

Unity and Diversity of Languages

Unity and Diversity in Language Use

Retrotransposition, Diversity and the Brain (Research and Perspectives in Neurosciences)

Progress in Brain Research Volume 119 Advances in Brain Vasopressin

Global Change and Protected Areas (Advances in Global Change Research)

Conserving Plant Genetic Diversity in Protected Areas

Complex Brain Functions: Conceptual Advances in Russian Neuroscience

Complex Brain Functions: Conceptual Advances in Russian Neuroscience

A Theory of the Basal Ganglia and Their Disorders (Conceptual Advances in Brain Research)

Retrotransposition, Diversity and the Brain

The flagellates: unity, diversity, and evolution

Advances in Brain Imaging

Biology: The Unity and Diversity of Life

Brain, Mind and Consciousness: Advances in Neuroscience Research

Unity and Diversity: Local Cultures and Identities in China

Brain, Mind and Consciousness: Advances in Neuroscience Research

Conceptual Similarity across Sensory and Neural Diversity

Diversity and Unity in Early North America (Rewriting Histories)

Advances in Research and Theory

Cortical Circuits

Emotional Cognition: From Brain to Behavior (Advances in Consciousness Research)

Advances in Food Research

Advances in Nanofibre Research

Progress in Brain Research Volume 47 Hypertension and Brain Mechanisms

Advances in Cancer Research

Advances in Giardia Research

Advances in Mathematics Research

Advances in Conceptual Modeling: ER '99

Cortical Areas: Unity and Diversity (Conceptual Advances in Brain Research)

The Female Brain, Second Edition (Conceptual Advances in Brain Research)

Time and the Brain (Conceptual Advances in Brain Research, Vol 3)

Unity and Diversity of Languages

Unity and Diversity in Language Use

Retrotransposition, Diversity and the Brain (Research and Perspectives in Neurosciences)

Progress in Brain Research Volume 119 Advances in Brain Vasopressin

Global Change and Protected Areas (Advances in Global Change Research)

Conserving Plant Genetic Diversity in Protected Areas

Complex Brain Functions: Conceptual Advances in Russian Neuroscience

Complex Brain Functions: Conceptual Advances in Russian Neuroscience

A Theory of the Basal Ganglia and Their Disorders (Conceptual Advances in Brain Research)

Retrotransposition, Diversity and the Brain

The flagellates: unity, diversity, and evolution

Advances in Brain Imaging

Biology: The Unity and Diversity of Life

Brain, Mind and Consciousness: Advances in Neuroscience Research

Unity and Diversity: Local Cultures and Identities in China

Brain, Mind and Consciousness: Advances in Neuroscience Research

Conceptual Similarity across Sensory and Neural Diversity

Diversity and Unity in Early North America (Rewriting Histories)

Advances in Research and Theory

Cortical Circuits

Emotional Cognition: From Brain to Behavior (Advances in Consciousness Research)

Advances in Food Research

Advances in Nanofibre Research

Progress in Brain Research Volume 47 Hypertension and Brain Mechanisms

Advances in Cancer Research

Advances in Giardia Research

Advances in Mathematics Research

Advances in Conceptual Modeling: ER '99

Recommend Documents