Understanding the Prefrontal Cortex:

Selective Advantage, Connectivity, and

Neural Operations Richard Passingham
Visit to download the full and correct content document:
https://ebookmass.com/product/understanding-the-prefrontal-cortex-selective-advanta
ge-connectivity-and-neural-operations-richard-passingham/
Understanding the Prefrontal Cortex
 OX F O R D P SYC HO L O G Y SE R I E S
Editors
Mark D’Esposito Daniel Schacter
Jon Driver Anne Treisman
Trevor Robbins Lawrence Weiskrantz

1. The Neuropsychology of Anxiety J. A. Gray

2. Elements of Episodic Memory E. Tulving
3. Conditioning and Associative Learning N. J. Mackintosh
4. Visual Masking B. G. Breitmeyer
5. The Musical Mind J. A. Sloboda
6. Elements of Psychophysical Theory J.-​C. Falmagne
7. Animal Intelligence L. Weiskrantz
8. Response Times R. D. Luce
9. Mental Representations A. Paivio
10. Memory, Imprinting, and the Brain G. Horn
11. Working Memory A. Baddeley
12. Blindsight L. Weiskrantz
13. Profile Analysis D. M. Green
14. Spatial Vision R. L. DeValois and K. K. DeValois
15. The Neural and Behavioural Organization of Goal-​Directed Movements M. Jeannerod
16. Visual Pattern Analyzers N. V. S. Graham
17. Cognitive Foundations of Musical Pitch C. L. Krumhansl
18. Perceptual and Associative Learning G. Hall
19. Implicit Learning and Tacit Knowledge A. S. Reber
20. Neuromotor Mechanisms in Human Communication D. Kimura
21. The Frontal Lobes and Voluntary Action R. Passingham
22. Classification and Cognition W. K. Estes
23. Vowel Perception and Production B. S. Rosner and J. B. Pickering
24. Visual Stress A. Wilkins
25. Electrophysiology of Mind Edited by M. D. Rugg and M. G. H. Coles
26. Attention and Memory N. Cowan
27. The Visual Brain in Action A. D. Milner and M. A. Goodale
28. Perceptual Consequences of Cochlear Damage B. C. J. Moore
29. Binocular Vision and Stereopsis I. P. Howard and B. J. Rogers
30. The Measurement of Sensation D. Laming
31. Conditioned Taste Aversion J. Bures, F. Bermúdez–​Rattoni, and T. Yamamoto
32. The Developing Visual Brain J. Atkinson
33. The Neuropsychology of Anxiety, 2e J. A. Gray and N. McNaughton
34. Looking Down on Human Intelligence I. J. Deary
35. From Conditioning to Conscious Recollection H. Eichenbaum and N. J. Cohen
36. Understanding Figurative Language S. Glucksberg
37. Active Vision J. M. Findlay and I. D. Gilchrist
38. The Science of False Memory C. J. Brainerd and V. F. Reyna
39. The Case for Mental Imagery S. M. Kosslyn, W. L. Thompson, and G. Ganis
40. Seeing Black and White A. Gilchrist
41. Visual Masking, 2e B. Breitmeyer and H. Öğmen
42. Motor Cognition M. Jeannerod
43. The Visual Brain in Action A. D. Milner and M. A. Goodale
44. The Continuity of Mind M. Spivey
45. Working Memory, Thought, and Action A. Baddeley
46. What Is Special about the Human Brain? R. Passingham
47. Visual Reflections M. McCloskey
48. Principles of Visual Attention C. Bundesen and T. Habekost
49. Major Issues in Cognitive Aging T. A. Salthouse
50. Perceiving in Depth Ian P. Howard
51. The Neurobiology of the Prefrontal Cortex: Anatomy, Evolution, and the Origin of
Insight Richard E. Passingham and Steven P. Wise
52. The Evolution of Memory Systems: Ancestors, Anatomy, and Adaptations
Elisabeth A. Murray, Steven P. Wise, and Kim S. Graham
53. Understanding the Prefrontal Cortex: Selective advantage, connectivity, and neural
operations Richard E. Passingham
 Understanding
the Prefrontal Cortex
Selective Advantage, Connectivity,
and Neural Operations

R IC HA R D E . PA S SI N G HA M
Emeritus Professor of Cognitive Neuroscience
Department of Experimental Psychology
Oxford University

1
 3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
Oxford University Press is a department of the University of Oxford.
It furthers the University’s objective of excellence in research, scholarship,
and education by publishing worldwide. Oxford is a registered trade mark of
Oxford University Press in the UK and in certain other countries
© Oxford University Press 2021
The moral rights of the author have been asserted
First Edition published in 2021
Impression: 1
All rights reserved. No part of this publication may be reproduced, stored in
a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above
You must not circulate this work in any other form
and you must impose this same condition on any acquirer
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2020945296
ISBN 978–​0–​19–​884457–​0
DOI: 10.1093/​oso/​9780198844570.001.0001
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Oxford University Press makes no representation, express or implied, that the
drug dosages in this book are correct. Readers must therefore always check
the product information and clinical procedures with the most up-​to-​date
published product information and data sheets provided by the manufacturers
and the most recent codes of conduct and safety regulations. The authors and
the publishers do not accept responsibility or legal liability for any errors in the
text or for the misuse or misapplication of material in this work. Except where
otherwise stated, drug dosages and recommendations are for the non-​pregnant
adult who is not breast-​feeding
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
In memory of Alan Cowey (1935–​2012) and Lawrence Weiskrantz (1926–​2018).
‘If you want to understand function, study structure’
Francis Crick (What Mad Pursuit, 1988)
 Note on the Cover

The picture on the front cover shows that, compared to the macaque monkey brain,
it is the prefrontal cortex and other association areas (in yellow and orange) that
have expanded most in the human brain.
 Preface

Steve Wise and I published The Neurobiology of the Prefrontal Cortex in 2012. This
date might seem recent to some, but scientific monographs have a shelf life of five
years or less. This is the typical timespan over which many scientists consider such
publications to be relevant. As in other scientific disciplines, neuroscience moves
on apace, and ideas that once seemed novel and exciting soon become standard
knowledge.
I decided that it was time to write a second edition. But, as I began revising the
book, changes accumulated so rapidly that a new book took shape. That is why it
has a new title. It retains some of the organizational structure of the original book,
as well as some of the figures; and I have also made use of some of the material
in the text. So, a clear palimpsest of the original remains. But the main burden of
this book is totally new, and most of it has been written without reference to the
original.
There are several reasons why a new book is required. First, I have changed my
mind on many issues. In the original book, we attempted to frame our proposals
as clearly as possible, but the danger of clarity is that errors can become obvious.
Good scientists embrace the prospect that they might be wrong. Marvin Minsky,
for example, cultivated ‘the habit of not wanting to be right for very long.’ As he put
it: ‘If I still believe something after five years, I doubt it.’
A second reason is that the original book dealt mainly with functional neuro-
anatomy, and we described functions in general psychological terms. We argued
that anatomical inputs constrain the function that each area performs and that the
outputs convey the result to other areas. But we did not take things further by pro-
viding an explicit proposal concerning the transformation that each area performs
from its inputs to its outputs. Chapters 3–​7 discuss each area in turn, and they end
with a proposal concerning the transformation that that area performs.
A focus on input–​output transforms means that book must now consider
what can be learned from computational neuroscience. It is this field of research
that helps us to understand the transforms that are computed by cortical areas.
Beginning with Zipser and Anderson (1988), neuroscientists have produced com-
putational models that use neurophysiological data to suggest how populations of
neurons with particular properties could perform specific functions. For example,
Rolls (2016) has considered how attractor networks can account for the operations
of the cerebral cortex. It is also increasingly common in functional brain-​imaging
studies to look for activations that are related to specific parameters in a computa-
tional model (e.g. Boorman et al., 2016).
x Preface

I am not an expert in this field and am not therefore competent to present de-
tailed computational models in this book. The transforms are therefore presented
without fully worked-​out proposals as to how they are actually implemented in
the brain.
A third reason for a new book is that new methods have clarified the evolution
of prefrontal areas. For example, imaging methods have enabled anthropologists to
construct virtual endocasts from fossil crania (e.g., Long et al., 2015). These studies
have augmented the methods of comparative neuroanatomy with a direct exam-
ination of ancestral species. Studies of the pattern of anatomical connections have
provided a method for identifying which prefrontal areas are homologous in dif-
ferent species (Mars et al., 2017). And the development of phylogenetic statistics
(Smaers & Rohlf, 2016) has also been a major advance.
There is a final reason why a new book was necessary. The literature on func-
tional brain imaging has exploded in recent years, as Figure P.1 illustrates. The
purple curve shows the number of publications on the prefrontal cortex stem-
ming from research on humans, with the vast majority involving functional brain
imaging.
In the original book, we based our proposals on the animal literature. We then
included a single chapter in which we argued that the literature on the human brain
from functional imaging was consistent with these proposals. But the imaging lit-
erature has now developed to the point where it is important to consider what new
it has to tell us on its own. Only studies of the human brain itself can help us to
understand capacities such as language that are unique to humans.

2000
Publications

1000

0
1950 1970 1990 2010
Year
Humans Mice
Rats Monkeys

Figure P.1 The annual number of papers on the prefrontal cortex of humans (purple),
rats (red), mice (blue), and monkeys (orange) since the late 1940s
Reproduced from eNeuro, 5 (4), Mark Laubach, Linda M. Amarante, Kyra Swanson, and Samantha
R. White, What, if anything, is rodent prefrontal cortex? Figure 1a, Doi: https://​doi.org/​10.1523/​
ENEURO.0315-​18.2018 Copyright (c) 2018, The Authors. Licensed under CC BY 4.0.
 Preface xi

Much of the evidence comes from functional imaging. But the extent of the im-
aging literature means that I cannot hope to do more than mention a selection of
the published results. Rather than attempt a comprehensive review, I have chosen
to highlight the results that best illuminate the topic. I therefore plead guilty to
cherry-​picking throughout this book. All that I can say is that I have done my best
to avoid picking the rotten ones.
One final word of warning. This book conveys my understanding, as of now.
There is always a danger that in trying to explain things, one skates over points that
are inconvenient for the account. I hope that I have described the experimental evi-
dence in enough detail that others can come to different conclusions. After all, that
is how science progresses.

References
Boorman, E.D., Rajendran, V.G., O’Reilly, J.X., & Behrens, T.E. (2016) Two anatomically
and computationally distinct learning signals predict changes to stimulus-​outcome asso-
ciations in hippocampus. Neuron, 89, 1343–​54.
Laubach, M., Amarante, L.M., Swanson, K., & White, S.R. (2018) What, if anything, is ro-
dent prefrontal cortex? eNeuro, 5 315–​18.
Long, A., Bloch, J.I., & Silcox, M.T. (2015) Quantification of neocortical ratios in stem pri-
mates. Am J Phys Anthropol, 157, 363–​73.
Mars, R.B., Passingham, R.E., Neubert, F.X., Verhagen, L., & Sallet, J. (2017) Evolutionary
specializations of human association cortex. In Preuss, T.M., Kaas, J. (eds) Evolution of
Nervous Systems. Elsevier, New York.
Rolls, E.T. (2016) Cerebral Cortex: Principles of Operation. Oxford University Press, Oxford.
Smaers, J.B. & Rohlf, F.J. (2016) Testing species’ deviation from allometric predictions using
the phylogenetic regression. Evolution, 70, 1145–​49.
Zipser, D. & Andersen, R.A. (1988) A back-​propagation programmed network that simu-
lates response properties of a subset of posterior parietal neurons. Nature, 331, 679–​84.
 Acknowledgements

I could not have written this book without the help of Steve Wise. He read all the
chapters, commented and made extensive revisions throughout the text. He con-
tributed especially to Chapter 2 because he knows much more than I do about the
evolution of the non-​human primates, and in particular about the fossil record.
I value in particular the critical comments that he made on Chapters 10 and 11: dis-
agreement can be very much more helpful than agreement. Steve also helped me by
producing many of the figures and telling me how to produce the others. I cannot
say how grateful I am for all the work he has put in.
I confess to having stolen the idea of dividing the book into parts or sections
from the book on ‘The Evolution of Memory Systems’ by Betsy Murray, Steve Wise,
and Kim Graham (2017). Their book won a prize; so, it must be a good idea.
The fundamental way of thinking about the brain on which this book is based
came out of a conversation with Klaas Stephan. It was written up a long time ago
in a paper with Klaas and Rolf Kötter on ‘The anatomical basis of functional lo-
calization in the cortex’. The idea has since been further developed by Matthew
Rushworth, Heidi Johansen-​Berg, and Rogier Mars. The motivation for this book
was to take it yet further, by changing the way in which we think about the function
of brain areas.
I have other colleagues to thank for sharing ideas and demolishing some of my
more outrageous speculations. In particular, Eleanor Maguire, Celia Heyes, and
Hakwan Lau have patiently answered my emails and questions, and put up with my
excitable nature.
Because I am long retired, it is hard to find the new literature. I am grateful in
particular to Earl Miller and Matthew Rushworth for sending me some of their
more recent papers. I have also been lucky enough to benefit from comments from
Nils Kolling, Laurence Hunt, Jill O’Reilly, and their graduate students and postdocs
who have read the whole book as a summer project during lockdown over COVID-​
19. They have had the advantage of viewing the manuscript with young eyes.
I am indebted to Larry Weiskrantz and Alan Cowey who supported me
throughout my career, and this book is dedicated to them. When I worked with
them as a postdoctoral fellow, they allowed me to work on the prefrontal cortex al-
though the grant was on vision. Those were the days.
I could not have published a book which is so lavishly illustrated had not the
authors of the relevant papers taken the trouble to send me the originals of their
figures. If the text is in danger of becoming dull, it always pays to add in a colour
figure to spice things up.
xiv Acknowledgements

Research is a collaborative project. So, I need to thank my graduate students,

postdoctoral fellows, and others who actually did the work. In various ways all have
contributed to the ideas in this book, whether they worked on prefrontal cortex
or not. I have benefited from the contributions of Katie Alcock, Sara Bengtsson,
Pierre Burbaud, Tony Canavan, Jim Colebatch, Marie-​Pierre Deiber, Julie Grezes,
Katie Hadland, Harri Jenkins, Louise Johns, Markus Jueptner, Mike Krams,
Hakwan Lau, Rogier Mars, Phil Nixon, Narender Ramnani, James Rowe, Matthew
Rushworth, Katsuyuki Sakai, Nat Schluter, Jeroen Smaers, David Thaler, and Ivan
Toni. In retirement it is their stimulating company that I miss.
I would also like to thank the editors at Oxford University Press, Martin Baum
and Charlotte Holloway, for encouraging a new book and guiding me through the
process. Apart from dealing with the manuscript, Charlotte put in an enormous
amount of work to seek the permissions for all the figures.
Finally, I would like to thank my wife Clare. I had promised her that I wouldn’t
write another book; and she has forgiven me, I think.
 Contents

List of Figures  xvii

List of Tables  xxiii
Style, Scope, and Terminology  xxv
Abbreviations  xxvii

PA RT I : F O U N DAT IO N S

1. Introduction  3

2. Evolution of the Prefrontal Cortex in Non-​human Primates  34

PA RT I I : SU BA R E A S O F T H E P R E F R O N TA L C O RT E X

3. Medial Prefrontal Cortex: Self-​Generated Actions  71

4. Orbital Prefrontal Cortex: Evaluating Resources  118

5. Caudal Prefrontal Cortex: Searching for Objects  153

6. Dorsal Prefrontal Cortex: Planning Sequences  191

7. Ventral Prefrontal Cortex: Associating Objects  236

PA RT I I I : T H E P R E F R O N TA L C O RT E X W I T H I N T H E
SYS T E M A S A W HO L E

8. Prefrontal Cortex: Abstract Rules and Attentional Performance  287

PA RT I V: T H E H UM A N P R E F R O N TA L C O RT E X

9. Evolution of the Prefrontal Cortex in the Hominins  333

10. Human Prefrontal Cortex: Reasoning, Imagination, and Planning  372
11. Human Prefrontal Cortex: Language, Culture, and Social Rules  420

Index  469
 List of Figures

P.1 The annual number of papers on the prefrontal cortex of humans

(purple), rats (red), mice (blue), and monkeys (orange) since the late 1940s x
1.1 Connectional fingerprints for the PF areas 9 and 14 6
1.2 Connectional clusters in the PF cortex 7
1.3 (A & B) Connectional fingerprints of the SMA and the rostral ventral
premotor cortex (PMvr) 9
1.4 Functional fingerprints showing selectivity for visually guided versus
memory-​guided movement sequences in the premotor cortex (lateral area 6),
the primary motor cortex (area 4), and the supplementary motor area
(medial area 6) 10
1.5 Functional fingerprints for the cortex in the anterior cingulate sulcus and
the right anterior insular cortex (Ia) 11
1.6 Comparison of anatomical and functional borders 13
1.7 (A) Area 9 (part of the granular medial PF cortex) in human subjects;
(B) Area 9 in macaques. The letters refer to the different sulci. (C) The
functional connectivity fingerprints in humans (light blue) and macaques
(purple) with the overlap shown in dark blue. (D) The summed absolute
differences between the functional coupling scores 18
1.8 Effect of a lesion on neuronal activity 21
1.9 Map of the macaque cortex 24
1.10 Maps of the human cortex (A) and macaque cortex (B) 26
1.11 The five regions of the PF cortex used in this book 28
2.1 The PF cortex and other frontal areas in macaque monkeys (A), rats (B),
and (C) bushbabies (galagos) 35
2.2 Cortical map of tree shrew brains 44
2.3 Effect of granular PF lesions in tree shrews 45
2.4 Phylogenetic reconstruction of primate social systems 47
2.5 Virtual brain endocasts of stem and crown primates 53
2.6 (A) Encephalization: Brain–​body mass relationship in Rooneyia and
other fossil primates. (B) Corticalization 54
2.7 Brain–​body mass relationship for modern strepsirrhines, modern anthropoids,
fossil anthropoids (red letters), and a fossil pongid (blue letters) 56
xviii List of Figures

2.8 Circular cladogram showing a stack of encephalization quotients (EQs)

for a series of species in each mammalian lineage, ordered by EQ value 57
2.9 Reconstructed phylogeny of encephalization quotients (EQs) for primates 58
2.10 Virtual brain endocasts of extinct catarrhines 59
3.1 Medial PF cortex in macaque monkeys (left) and humans (right), indicated
by shading 72
3.2 Homologies among agranular parts of the medial PF cortex in rodents and
anthropoids 74
3.3 Selected connections of the medial PF cortex in macaque monkeys 75
3.4 Reversal impairment for choices between two actions 83
3.5 (A) Areas that were specifically activated when monkeys watched videos of
other monkeys interacting. (B) Flat map of areas that showed a difference in
grey matter depending on the size of the group 90
3.6 Overlap between medial (default) network and social network in macaques 93
3.7 Medial network as visualized by fMRI 94
3.8 Cell activity encoding choices at feedback time for populations of cells in the
polar PF cortex (area 10) (orange) and the orbital PF cortex (area 11) (green) 103
3.9 Pictures (p) for ‘object-​in-​scenes’ task, shown for two trials or runs,
with 20 pictures per run 104
3.10 The effect of polar PF lesions in macaque monkeys 104
4.1 The orbital PF cortex in monkeys (left) and humans (right) indicated
by shading 119
4.2 Selected connections of the orbital PF cortex 120
4.3 Total selection frequency for the category of nonpreferred foods
(garlic, lemons, and monkey chow) for each experimental group across
the pre-​surgery, post-​surgery, and shuffled testing phases 123
4.4 The effects of satiation on the choice between objects that were associated
with particular foods 127
4.5 Effect of three lesions on choices in the devaluation task 129
4.6 The effect of amygdala lesions on the encoding of reward in the orbital
PF cortex 132
4.7 Action reversal and object reversal tasks 135
4.8 Impairment in object reversal set in monkeys after lesions of the central
and medial sectors of the orbital PF cortex 136
4.9 Errors made on the object reversal task, before and after the first correct choice 138
4.10 Performance on probabilistic reversals after lesions of the orbital PF cortex 140
4.11 Performance on a probabilistic reversal learning task: the 3-​arm bandit task 142
4.12 Data for object reversal learning for different groups of lesioned macaques 144
 List of Figures xix

5.1 The caudal PF cortex in the macaque monkeys (left) and humans (right) 154
5.2 Selected connections of the caudal PF cortex 156
5.3 Layout of the FEF circuit 162
5.4 Topographic maps in the frontal eye fields and posterior parietal cortex in
one subject 165
5.5 Intrinsically defined dorsal and ventral attention systems and the overlap
between them 168
5.6 A common version of the oculomotor delayed response task 172
5.7 Attention versus memory coding in the PF cortex 174
5.8 Performance on the oculomotor delayed response task for one
lesioned monkey 176
5.9 Corrective saccades after a frank error 177
5.10 Change in population vectors on an oculomotor delayed response task 179
5.11 An example of a search array on which are superimposed the scan paths
of one patient 180
6.1 The dorsal PF cortex in macaque monkeys (left) and humans (right) 192
6.2 Selected connections of the dorsal PF area 46 193
6.3 Areas 9/​46 and 46 in the human brain as identified on the basis of the
similarity of their connectional fingerprints to the same areas in a
macaque monkey 197
6.4 (A) The anterior area of the dorsal PF cortex (green) that co-​activated with
the anterior cingulate cortex. (B) The posterior area of the dorsal PF
cortex (red) that co-​activated with the parietal cortex 198
6.5 Testing procedure for the classic delayed response task in a Wisconsin
general testing apparatus (WGTA) 201
6.6 The top section shows the matching task and the bottom section the
recall task that were used in an fMRI experiment 204
6.7 The course of the BOLD signal during the tasks that were illustrated in
Figure 6.6 205
6.8 (A) The data for sites at which the spiking activity was informative.
(B) The data for sites at which the spiking activity was not informative 214
6.9 (A) The task used in an imaging experiment to test for the effect of distraction
in memory. (B) The relation between sustained activation during the delay and
the accuracy of performance on trials on which there were no distractors and
trials on which there were distractors 217
6.10 (A) The activation in the dorsal PF cortex on a task in which human subjects
decide of their own accord which finger to move. (B) The plot of the degree of
activation as a function of the equipotentiality index 220
xx List of Figures

6.11 (A) Task in which the monkeys used a handle to make sequences of four
movements. (B) Activity of cells that encoded sequences with a particular
abstract structure 222
6.12 Visual maze task 223
6.13 Population analysis of cells in the dorsal PF cortex during planning 224
7.1 The ventral PF cortex in macaque monkeys (left) and humans (right) 237
7.2 Cross section through the ventral limb of the arcuate sulcus on an MRI
scan in a macaque monkey 238
7.3 Areas in the human ventral PF cortex that correspond with areas in the
macaque monkey, as demonstrated by the functional fingerprint based on
resting state covariance 239
7.4 Selected connections of the ventral PF cortex 241
7.5 Effect of lesions of the ventral PF cortex on simultaneous matching 245
7.6 Performance across trials within problems on a series of visuo-​spatial
problems before and after ventral PF surgery 254
7.7 Stimuli used in a categorization task 257
7.8 (A) A cell in the ventral PF cortex that encoded the category ‘dog’. (B) A
cell in the ventral PF cortex that encoded the category ‘cat’ 258
7.9 A cell in the ventral PF cortex that encoded arbitrary categories, demarcated
by the lightly stippled vertical lines in Figure 7.7 260
7.10 Single cells reflecting both shape-​shape associations and motion direction
categories 263
7.11 Activations for shifting between categories, shown on inflated surface
reconstructions of the macaque monkey brain (left) and human brain (right) 270
7.12 (A) The effect of a TMS pulse over the FEF on the activation in the MT complex
when motion is relevant. (B) The effect of a TMS pulse over the FEF on the
activation in the fusiform face area when the shape of the face is relevant 273
8.1 The central neocortical hub or core as shown by graph theory 289
8.2 The organization of the feedforward (blue) and feedback (red) connections
in the neocortex 291
8.3 Organization of basal ganglia and cerebellar outputs to the cerebral cortex 293
8.4 The overlap of the projections to the striatum from different PF areas is
shown in orange. The projections from the inferior parietal area PG are
shown in green 295
8.5 PF cell encoding the abstract matching rule 300
8.6 The location of rule selective and generalist cells in the PF cortex that
coded for the general rule ‘greater than’ or ‘lesser than’ 301
8.7 Development of a learning set for visual discrimination problems, in a
selection of mammalian species 303
 List of Figures xxi

8.8 Histograms showing mean percent error in trials 2–​11 at each performance
test on discrimination learning set 305
8.9 Strategy score (performance on repeat compared with change trials) before
(black) and after (white) lesions 307
8.10 Population coding for abstract rules 308
8.11 Timing of the development of selectivity for visual objects, behavioral goals,
and actions 310
9.1 Encephalization quotients (EQs) for fossil hominins, modern humans, and
modern chimpanzees 335
9.2 CT scans of skulls of a chimpanzee, Homo heidelbergensis, Homo
neanderthalensis, and Homo sapiens 336
9.3 Granular PF cortex as a percentage of the frontal lobe, plotted versus
function of cerebral extent, in modern primates 338
9.4 Log-​log plot of the estimated volume of the granular PF cortex as a function of
non-​PF cortex in the frontal lobe using the data from Smaers et al. (2011) 339
9.5 Myeloarchitectonics in the brains of humans, chimpanzees, and macaque
monkeys 343
9.6 Regressions of the volume of the frontal pole cortex (area 10) as a function of
brain volume in selected primates 344
9.7 Expansion of the area 10 in humans 345
9.8 (A) Relative expansion of cortical regions from macaque to human brains,
shown on the human brain. (B) Relative expansion of cortical regions from
chimpanzee to human brain, shown on the chimpanzee brain 347
9.9 Log-​log plot of volume of the PF cortex as a function of an estimate of the
other association areas (see text) 348
9.10 Degree of expansion and closeness centrality 351
9.11 Areas of the brain that have changed most in their connectivity as estimated
from visualizing the major tracts using diffusion weighted imaging (DWI) 352
9.12 Arcuate fasciculus in the human and macaque monkey brain as visualized
by diffusion weighted imaging 354
9.13 Postnatal cortical surface expansion 358
10.1 The multiple-​demand system as visualized on the parcellation of the human
brain by Glasser et al. 376
10.2 A typical problem on the Raven’s Progressive Matrices 377
10.3 The relation between fluid intelligence and the volume of damage to the frontal,
parietal, or temporal lobe 379
10.4 Computerized version of the ‘Tower of London’ 384
10.5 Areas that are activated when subjects plan 387
xxii List of Figures

10.6 Activations in the left inferior caudal PF cortex (areas 44 and 45B) and
the STS during deductive reasoning 389
10.7 The accuracy of decoding using a multivoxel pattern analysis for memories
that were of events that occurred either 2 years ago or 10 years ago 392
10.8 The area in yellow shows the source of the theta oscillations as measured by
MEG while the subjects retrieved personal memories from the past 394
10.9 Indices relating the activity for hits and misses 399
10.10 ERPs recorded on the Libet task for the conditions in which the subjects
timed their intention to move (W) or the actual movement itself (M) 403
10.11 Brain regions with a significant difference between the prediction of the
subjective ratings and skin conductance reactivity 408
11.1 (A) Areas that were activated during signing and speaking. (B) Areas
that were only activated during signing 424
11.2 The arcuate fasciculus in the macaque monkeys and human brain 426
11.3 (A) The areas that were activated both when observing and when imitating.
(B) The areas that were activated when imitating was contrasted with
observing alone in green, and when observation was contrasted with
imitation in red 428
11.4 Activations for speaking, imitation, and motor inhibition in the left and
right hemisphere 429
11.5 Horizontal sections showing the areas that were under-​activated in the
brains of affected members of the KE family 430
11.6 Comparing sentences and small clauses 433
11.7 Areas of the neocortex which showed activity that was specific to naming
pictures (blue), naming by definition (red), or that showed activity for
naming in both conditions (pink) 436
11.8 Setting up the current task 442
11.9 Semantic task used by Vandenberghe et al. (1996) 443
11.10 Activations while expert stone knappers make Oldowan and Acheulean tools 446
11.11 The activations resulting from metanalyses of fMRI data for metacognition
(yellow and red) and mentalizing (green and blue) 449
11.12 The overlap for the lesions that included the medial and orbital PF cortex
in 19 patients who had had a medial meningioma removed at surgery (vmpfc) 451
 List of Tables

1.1 Prefrontal areas in human and macaque monkey brains, with area numbers
in parentheses, where applicable. 25
1.2 Groups of PF areas and a compact abbreviation for each. 29
2.1 PF areas with homologues (+) in mammals, strepsirrhine (prosimian)
primates, anthropoids, and humans. 46
9.1 Remapping factors. 340
9.2 Remapping factors. 341
9.3 Remapping factors. 349
9.4 Factors that might account for the expansion of the PF cortex in hominins. 360
10.1 Remapping factors. 393
11.1 Handedness in chimpanzees. 421
11.2 Handedness in human subjects. 421
 Style, Scope, and Terminology

I have thought it important to tell readers how the researchers obtained the results
that are summarized in this book. Accordingly, I have avoided simply stating find-
ings and conclusions as facts, followed by a long series of references in brackets.
This practice is perfectly reasonable in a paper or review article, especially in view
of the word limits imposed by many publishers. But the point of a textbook or
monograph is not to provide references, however convenient that might be. The
aim of this book is to further understanding, and this means spelling out in detail
why the results of a particular experiment lead to certain conclusions and not to
others.
Because each section describes a series of experiments in detail, to help the
reader I have added summaries at the end of each section. These provide the provi-
sional conclusions that I take to follow from these experiments. The final interpret-
ation of the results is left to the end of the chapter,
For convenience, I have adopted certain several semantic conventions:

• The word animal, when unmodified, refers to a non-​human animal.

• The word monkey, when unmodified, refers to macaque monkeys.
• I have used the phrase granular prefrontal cortex, although its architecture
is not truly granular in the same sense as the granular cortex of the primary
sensory areas.
• I use the term lesion to cover all procedures that prevent a cortical area from
functioning normally, including various forms of temporary disruption.
• The term cell is used to refer specifically to neurones throughout.
• I use the term activity to describe the rate of neuronal action potentials, com-
monly known as firing rate, discharge rate, or modulation; but I use the term
activation to describe results from functional brain-​imaging experiments be-
cause the BOLD signal is a vascular one.
• I have resisted using the term volunteers to describe people who take part in
fMRI experiments. I doubt that readers will think that these people have been
dragged kicking and screaming into the scanner. I know that the label parti-
cipants is now de rigueur, but the word subject has the advantage that it can be
used for all species, humans included.
• I will quite rightly draw the ire of anatomists by phrases that suggest that ana-
tomical connections run from or to anatomical sulci such as the intraparietal
sulcus. Of course, they run from the cortical tissue in the sulcus, not from
the sulcus itself. I also realize that fMRI activations are in the cortex of the
xxvi Style, Scope, and Terminology

sulcus, not the sulcus itself. But it becomes cumbersome to labour this point
throughout, and I have not done so.
• I use the term association cortex when it would be more accurate to use the
anatomical term homotypical cortex which is defined on the basis of cyto-
architecture. The term association cortex is an outdated functional term, but it
has the advantage that many readers will be more familiar with it; and I don’t
want to frighten the horses.
• I use the term posterior parietal cortex to distinguish it from the somatosen-
sory cortex in the parietal lobe. The association cortex of the parietal lobe lies
posterior to the primary somatosensory cortex.
• I have tried to keep the abbreviations to a minimum. It is fine to use them in
an anatomy paper but can put off readers of a book. So, I have confined the ab-
breviations to the ones that are most common. Thus, I use IPS for intraparietal
sulcus and STS for the superior temporal sulcus, but I spell out inferior frontal
sulcus in full. Similarly, though I used SMA for the supplementary motor
area, I spell out dorsal and ventral premotor cortex in full, rather than using
PFd and PMv. I toyed with using abbreviations for the different prefrontal
subareas, PFm, PFo, PFc, PFd, and PFv, but decided that I was more interested
in sales. To help the reader I define any abbreviations that I use the first time
that I mention them in a chapter. A list of all the abbreviations, together with
their definitions follows on the next page.
 Abbreviations

2-​DG 2-​deoxyglucose
ACC anterior cingulate cortex
AIP anterior intraparietal area
Area F5 rostral part of ventral premotor cortex
Area X part of a bird brain
ASL American sign language
BOLD blood oxygen-​level dependent [signal]
CMA cingulate motor area
CMAr rostral cingulate motor area
CoCoMac database of corticocortical connections in macaques
DA delayed alternation task
DR delayed response task
DREADDS designer receptors exclusively activated by designer drugs
DWI diffusion weighted imaging
ECoG electrocorticography
EEG electroencephalography
EQ encephalization quotient
ERP event-​related potential
FEF frontal eye field
FEFd dorsal frontal eye field
FEFv ventral frontal eye field
FFA fusiform face area
fMRI functional magnetic resonance imaging
FoxP2 forkhead box protein P2
GABA gamma-​aminobutryic acid, an inhibitory neurotransmitter
gACC gyral part of the anterior cingulate cortex
Hz Hertz
Ia agranular insular cortex
IFJ inferior frontal junction
IPS intraparietal sulcus
IQ intelligence quotient
Ka thousand years ago
LIP lateral intraparietal area
M1 primary motor area
Ma million years ago
MD mediodorsal nucleus of the thalamus
MEG magnetoencephalography
MEP motor evoked potential
MIP medial intraparietal area
xxviii Abbreviations

MNI Montréal Neurological Institute

MST middle superior temporal area
MT middle temporal area
MVPA multivoxel pattern analysis
N2pc ERP component linked to selective attention
NA nucleus accumbens
NMDA N-​methyl-​D-​aspartate, an excitatory neurotransmitter
ODR oculomotor delayed response task
PET positron emission tomography
PF prefrontal cortex
preSMA presupplementary motor area
ROC receiver operating characteristic
RSA representational similarity analysis
rTMS repetitive transcranial magnetic stimulation
S1 primary somatosensory cortex
S2 secondary somatosensory cortex
sACC anterior cingulate sulcus
SEF supplementary eye field
SEM standard error of the mean
SMA supplementary motor area
SOA stimulus-​onset asynchrony
STP superior temporal polysensory area
STS superior temporal sulcus
TMS transcranial magnetic stimulation
TPJ temporal-​parietal junction
Tpt temporo-​parietal area, transition between the lobes
V1 primary (striate) visual area (area 17)
V2 second visual area (area 18), an extrastriate area
V3 a low-​order area in the extrastriate visual cortex
V4 a low-​order area in the extrastriate visual cortex
WGTA Wisconsin general testing apparatus
 PART I
F OU N DAT IONS
 1
Introduction

This chapter has two purposes. The first is to explain why the book is subtitled ‘se-
lective advantage, connectivity, and neural operations’. The second is to tell the
reader why the second part of the book describes neurophysiological and neuro-
psychological experiments that were carried out on macaque monkeys. After all,
many readers would have started out with the assumption that the way to under-
stand the prefrontal (PF) cortex was to use functional brain imaging with human
subjects.

Selective Advantage

The reason why the words ‘selective advantage’ appear in the subtitle is that, as with
any other biological system, a complete understanding of the PF cortex requires
answers to four questions. Tinbergen (1951) listed them as follows:

• How did it evolve (phylogeny)?

• How does it promote fitness (selective advantage)?
• How did it develop (ontogeny)?
• How does it work (physiology)?

The literature on the PF cortex is mainly devoted to the fourth question, though
it tackles the third one as well. But it rarely addresses the first two. Yet, they hold an
important key to understanding the human PF cortex. The reason is that charac-
teristics that are unique to humans evolved by co-​opting and elaborating mechan-
isms that existed in ancestral primates. We cannot properly understand the human
PF cortex if we do not appreciate that the fundamental organization of the human
PF cortex is the same as in other anthropoid primates, the monkeys and apes.
Chapter 2 explains that some parts of the PF cortex first appeared in early pri-
mates and others came along during the evolution of the anthropoids, the monkeys
and apes. If we take into account the life and times of the ancestral species in which
particular PF areas first appeared, we can obtain clues about what these areas did
then and the selective advantage that they conferred.
It truly matters that humans are primates. Unlike other mammals, primates
forage by reaching and grasping food with their hands. This involves the coord-
ination of hand and eye. Specialized areas evolved to support this, including the

Understanding the Prefrontal Cortex. Richard E. Passingham, Oxford University Press. © Oxford University Press 2021.
DOI: 10.1093/​oso/​9780198844570.003.0001
4 Introduction

anterior parietal area (AIP) and the ventral premotor cortex. If either the inferior
parietal cortex (Stepniewska et al., 2009) or the ventral premotor cortex (Graziano
et al., 2002) is stimulated electrically in non-​human primates, the hand reaches
towards the mouth irrespective of its initial starting position. Shadmehr and Wise
(2005) provide a full description of the complex mechanisms that are involved in
reaching for a target.
However, as Chapters 6 and 7 demonstrate, it is the PF cortex that generates the
goal or target for searching with the eye and reaching with the hand. The PF cortex
sends connections not only to the frontal eye field (FEF) but also to the premotor
areas that are specialized for the movements of the hand and arm, though not the
leg. It is these mechanisms that have been co-​opted and elaborated to generate new
goals in humans. These goals are as different as choosing the item that completes a
series on tests of non-​verbal intelligence (IQ) or generating the verbs that are ap-
propriate given a particular noun.
Unfortunately, the tasks that comparative psychology and psychometrics have
devised to study intelligence were devised without consideration of the ecological
conditions in which human and other primates evolved. Yet Chapter 2 argues that
it is critical to understand that primates originally evolved to forage in the fine-​
branch niche, and that later many old-​world monkeys and apes came to the ground
to forage. However, the fruit and leaves were patchy; heat stress meant that the
periods for foraging were short; and because there were periods in which there was
a shortage of particular foods, there was a premium on behavioural flexibility and
the ability to solve new problems rapidly.
This was particularly true during the evolution of the hominins (Chapter 9). This
means that to understand how humans came to have unique capacities, it is im-
portant to appreciate the selection pressures that encouraged them. For example,
we cannot fully understand how it is that humans can reflect on the intentions of
others without realizing that the survival of our ancestors depended critically on
cooperation.
However, we cannot study the brains of ancestral hominins or the primates from
which they evolved. And unfortunately, as a later section explains, we have very
little evidence on the brains of chimpanzees, our closest ancestor. Therefore, much
of this book is devoted to trying to understand the PF cortex of macaque monkeys.
Of course, humans are not descended from either chimpanzees or macaque
monkeys. The last common ancestor of macaques and chimpanzees lived be-
tween 25 and 30 million years ago (Stevens et al., 2013). Since that time, the lines
leading to modern macaques and chimpanzees have themselves undergone fur-
ther change. Nonetheless, studies of macaque monkeys can provide a clue, how-
ever indirect, that allows us to infer the PF mechanisms that the common ancestors
of monkeys and humans were likely to have possessed.
 Connectivity 5

Summary

The human PF cortex cannot be understood without recognizing that it evolved by

co-​opting and elaborating mechanisms that existed in the brains of our primate an-
cestors. Its fundamental organization is the same as in monkeys and apes.

Connectivity

The word ‘connectivity’ in the subtitle means anatomical connectivity. As Devlin

and Poldrak (2007) have pointed out, many regard anatomy as tedious. The word
anatomy can conjure up the memory of classes in which brains were cut up and a
host of labels had to be learned so as to distinguish between the different areas.
This sort of anatomy is indeed tedious. But understanding how the brain is wired
up is not, because it provides clues as to mechanisms. Working out the structure of
deoxyribonucleic acid (DNA) proved crucial to understanding how genetic ma-
terial is copied from parents to offspring. No wonder Francis Crick (1988) wrote ‘if
you want to understand function, study structure’.
When in later life Crick turned his attention to the brain, he looked first for evi-
dence on the anatomical connections of the human brain. He was disturbed to dis-
cover how ‘backward’ human neuroanatomy was at the time (Crick & Jones, 1993).
Fortunately, the situation has changed with the development of diffusion-​weighted
tractography (DWI) and the study of the covariance between the activations in dif-
ferent brain areas while the subjects are at rest.
Modern work on the connections between the areas of the neocortex began with
the studies by Pandya and Kuypers (1969) and Jones and Powell (1970) on ma-
caque monkeys. However, the methods have become more sensitive, and the pic-
ture has changed. For example, Markov et al. (2014) injected tracers into 29 of the
91 areas of the macaque neocortex. They found 1,615 paths between areas, and
roughly a third of them had not been described before.
One way of visualizing the anatomical inputs and outputs of an area is to
chart them on polar plots. These can be regarded as ‘connectional fingerprints’
(Passingham et al., 2002). The analogy is with the polar plots produced by Zilles
et al. (2001) for the receptor densities in each cortical area. The circumference
shows the areas that are connected with the featured region, and the radius plots
the strength of each connection, ranging from one to three.
Passingham et al. (2002) plotted these using the data CoCoMac (cocomac.g-​
node.org/​). This is a database for the cortical (Co) connections (Co) of macaque
(Mac) monkeys. In the version available at the time, it contained data from 413
studies with 39,748 connectional entries. A second edition called CoCoMac
2 has been developed from the Donders Institute, and it includes many more
connections.
6 Introduction

By using the data from CoCoMac 1, Passingham et al. concluded that each cor-
tical area has a unique set of inputs and outputs. They plotted the strength of the
connections in polar coordinates for various PF and premotor areas. Figure 1.1
shows two examples for the PF areas 9 and 14.
After constructing connectional fingerprints, Passingham et al. (2002) then
used multidimensional scaling to prove that each area had a unique pattern of in-
puts and outputs. As Figure 1.2A (next page) illustrates, the analysis showed that
all the PF areas lie at different points in a ‘connectional space’. This is evidence that
no two areas have exactly the same overall pattern of connections. The distance
between two areas in this space (Figure 1.2A) is a measure of the difference in the
connectivity.

Area 14 Area 9
10 10
9 3 11 9 11

8B 2 12 8B 12
1

Afferents 8A 13 8A 13

46 14 46 14

45 24 45 24
25 25

10 10
9 11 9 11

8B 12 8B 12

Efferents 8A 13 8A 13

46 14 46 14

45 24 45 24
25 25

Figure 1.1 Connectional fingerprints for the PF areas 9 and 14

Each plot shows the PF areas on the circumference that are connected with the
featured area, with the intensity of the projection along the radius: light (1), moderate
(2), or heavy (3). Projections within an area are not included.
Reproduced from Passingham R.E., Stephan K.E., & Kotter R. The anatomical basis of functional
localization in the cortex. Nat Rev Neurosci, 3 (8), 606-​16, Figure 1, Doi: 10.1038/​nrn893 Copyright ©
2002, Springer Nature.
 Connectivity 7

(A) (B) (C)

Figure 1.2 Connectional clusters in the PF cortex

(A) Plot of two connectional dimensions with the area denoted next to each
point. (B) Hierarchical clustering along a connectional dimension, denoted as d1.
(C) Clusters colour coded to match (A) and (B).
Reproduced from Passingham R.E., Stephan K.E., & Kotter R. The anatomical basis of functional
localization in the cortex. Nat Rev Neurosci, 3 (8), 606–​16, Figure 2, Doi: 10.1038/​nrn893 Copyright ©
2002, Springer Nature.

Systems

Passingham et al. (2002) then used hierarchical cluster analysis to identify clus-
ters of regions with similar intrinsic PF connections. Figure 1.2B shows five clus-
ters within the PF cortex. Averbeck and Seo (2008) identified similar clusters,
basing their analysis on the afferent connections to the PF cortex. In both ana-
lyses, a cluster of dorsal areas contrasts with a cluster of ventral areas (Figure 1.2C).
The same conclusion was reached by Blumenfeld et al. (2014) who have produced
open-​source software so that users can plot the connections of any cortical area
based on the CoCoMac database.
Considering the neocortex as a whole, Passingham et al. (2002) suggested that
there are ‘families’ of areas or ‘systems’ that shared a similar, though not identical,
pattern of connections. More recently, Markov et al. (2013) came to a similar con-
clusion based on an analysis of the whole neocortical connectome. For example,
they identified a large-​scale cortical network that includes the posterior parietal
cortex and the PF cortex. Markov et al. regarded these two regions as part of a ‘core’
network, with other groups of areas such as the occipital and temporal components
of the ventral visual system being situated on the periphery.
The terms ‘systems’ or ‘networks’ are sometimes used loosely in the func-
tional magnetic resonance imaging (fMRI) literature to refer to the constellation
of activations found in a particular experiment. However, these terms only have a
meaning if they are restricted to those areas that can be shown to be closely con-
nected anatomically.
8 Introduction

Summary

Each cortical area has a unique pattern of anatomical inputs and outputs. These can
be plotted as connectional fingerprints.

Neural Operations

The final words in the subtitle are ‘neural operations’. Having established that each
area has a unique pattern of inputs and outputs, the next step is to understand how
an area processes the incoming information so as to transform it into the influence
that it exerts on other areas via its outputs.
Much of the fMRI literature is devoted to identifying the ‘function’ of an area.
Yet, the functional labels that are attached to areas come from psychology. Thus,
an area might be said to be involved in ‘attention’, ‘decision making’, or ‘conflict
monitoring’. But these labels say nothing about the neural operation that an area
performs.

Function

There are other problems with using functional labels. Many have been inherited
from common sense psychology (James, 1890). Some are vague such as ‘execu-
tive’, one of the labels in BrainMap (Lancaster et al., 2012). Others such as ‘working
memory’ are appropriate for systems, not single areas. And there is a danger that
the functional labels are allocated on the basis of studying just one or two tasks.
To avoid this danger, Passingham et al. (2002) introduced the concept of ‘func-
tional fingerprints’. The idea was that the function of an area could be characterized
by producing polar plots in which a series of tasks were shown round the circum-
ference, and the degree of activation on each task plotted on the radius. However, at
the time the fMRI literature was not rich enough to provide an illustration.
So instead, Passingham et al. plotted data from neurophysiology. They took five
properties of cell activity: auditory or visual responses; proprioceptive or cuta-
neous responses; a muscle-​like pattern of activity; a temporal correlation of activity
with movements; and persistent delay-​period activity.
Figure 1.3C (next page) shows functional fingerprints for two premotor areas,
the supplementary motor area (SMA) and the ventral premotor cortex. It shows
that they differ in the relative frequency of the various cell classes. For example, a
higher proportion of the cells in the SMA have somatosensory responses. For com-
parison, the figure also shows that the two areas differ in their connectional finger-
prints (Figure 1.3A and B).
 Neural Operations 9

Supplementary motor area Ventral premotor area

(SMA) (PMvr)

(A) M1 M1

3
2
prePMd PMd prePMd PMd
Afferents

preSMA PMvc preSMA PMvc

PMvr = 2 SMA = 2

(B)
M1 M1

prePMd PMd prePMd PMd

Efferents

preSMA PMvc preSMA PMvc

PMvr = 2 SMA = 2

(C) Delay-period activity Delay-period activity

80%
Auditory Auditory
Movement 60% Movement
& visual & visual
time time
inputs inputs
Physiology

Movement Proprioceptive & Movement Proprioceptive &

type cutaneous inputs type cutaneous inputs

Figure 1.3 (A & B) Connectional fingerprints of the SMA and the rostral ventral
premotor cortex (PMvr)
Abbreviations: M1, primary motor cortex (area 4); PMd, dorsal premotor cortex
(caudal dorsal area 6); PMvc (caudal ventral area 6); pre-​SMA, presupplementary
motor area (rostral medial area 6); pre-​PMd, rostral part of the dorsal premotor cortex
(rostral dorsal area 6). The strength of connections between the areas depicted in the
two columns is presented beneath each polar plot. C = Functional fingerprints for the
same two areas.
A and B: Reproduced from Passingham, R.E. & Wise, S.P. The Neurobiology of Prefrontal Cortex, p. 16,
Figure 1.9a and b Copyright © 2012, Oxford University Press.
C: Adapted from Passingham R.E., Stephan K.E., & Kotter R. The anatomical basis of functional
localization in the cortex’, Nat Rev Neurosci, 3 (8), 606–​16, Doi: 10.1038/​nrn893 Copyright © 2002,
Springer Nature.
10 Introduction

By using multidimensional scaling, Passingham et al. were also able to show

that the functional fingerprints of the SMA and the ventral premotor cortex were
unique, as were the fingerprints for the other motor and premotor areas. This was
shown by the fact that they lay at different locations in two-​dimensional space. The
conclusion is that functional fingerprints are unique to an area, just like connec-
tional fingerprints.
To show that the differences can be one of degree, Passingham et al. (2002) also
plotted the preference of cells for movement sequences that were either performed
from memory or based on visual cues (Mushiake et al., 1991). The histograms in
Figure 1.4 show the results for the SMA and the premotor cortex, with activity clas-
sified from 1 to 7. Cells in class 1 had complete specificity for the visual task; those
in class 7 had complete specificity for the memory-​guided task. A classification of
4 indicates activity that did not differ statistically between the two tasks. The SMA

Dorsal premotor cortex Primary motor cortex Supplementary motor area

1 1
30%
7 2 7 2
20%

6 3 6 3
50 50 50
5 4 5 4
40 40 40
Percent of cells

30 30 30

20 20 20

10 10 10

0 0 0
1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 5 6 7
Visually Memory
guided guided

Figure 1.4 Functional fingerprints showing selectivity for visually guided versus
memory-​guided movement sequences in the premotor cortex (lateral area 6), the
primary motor cortex (area 4), and the supplementary motor area (medial area 6)
Cells in class 1 show complete specificity for visually guided sequences; cells in class 7
show complete specificity for memory-​guided sequences. Cells in the other five classes
show intermediate properties. The polar plots above the left and right bar graphs show
the same data as the bar graphs.
Reproduced from Mushiake H., Inase M., & Tanji J. Neuronal activity in the primate premotor,
supplementary, and precentral motor cortex during visually guided and internally determined
sequential movements. J Neurophysiol, 66 (3), 705–​18, Doi: 10.1152/​jn.1991.66.3.705 Copyright ©
1991, The American Physiological Society, with permission.
 Neural Operations 11

showed a preponderance of activity for the memory-​guided task, with cells in the
lateral premotor cortex showing the opposite bias.
Figure 1.4 shows these data in the form of functional fingerprints. The polar
plots at the top of figure show the same data as the bar charts below them, with the
cell class (1–​7) plotted around the circumference and the proportion in each class
along the radius.
Given the rapid rise in the number of studies using fMRI, functional finger-
prints can now be plotted on the basis of activations. Figure 1.5 presents an ex-
ample. Based on the BrainMap database, the plots compare the range of tasks over
which fMRI experiments have revealed activations in the anterior insular cortex
(Ia) and the cortex in the anterior cingulate sulcus (Sporns, 2014).
Unfortunately, the functional labels used in the BrainMap database are crude
(Lancaster et al., 2012). Nonetheless, the plots in Figure 1.5 are adequate to show
that these two areas have different functional fingerprints, even though the areas
are interconnected (Hutchison et al., 2012). For example, the anterior insula cortex

Dorsal anterior cingulate cortex Anterior insular cortex

Exe Ima Exe Ima

Vis Vis
Som Inh Som Inh
Aud Motl Aud Motl
Sad Obs Sad Obs

Hap ACC Pre Hap Ia Pre

Fear Att Fear Att

Dis LanS Dis LanS

Ang LanO Ang LanO
Rea MemW Rea MemW
Mem Mem

Figure 1.5 Functional fingerprints for the cortex in the anterior cingulate sulcus and
the right anterior insular cortex (Ia)
The green shaded area gives the estimated involvement of the cortical area in each
kind of task or emotion plotted on the circumference of each circle, based on
fMRI activation studies. The magenta and dark green lines reflect the upper and
lower bounds of these estimates. Abbreviations, in clockwise order beginning at
12 o’clock: Exe, executing movements; Ima, imagining movements; Inh, inhibiting
movements; Motl, motor learning; Obs, observing movements; Pre, planning
movements; Att, attention; LanS, semantic aspects of language; LanO, other
(nonsemantic) aspects of language; MemW, working memory; Mem, memory; Rea,
reasoning; Ang, anger; Dis, disgust; Hap, happiness; Sad, sadness; Aud, auditory
perception; Som, somatosensory perception; Vis, visual perception.
Adapted from Sporns O. Contributions and challenges for network models in cognitive neuroscience.
Nat Neurosci, 17, 652–​60, https://​doi.org/​10.1038/​nn.3690 Copyright © 2014, Springer Nature.
12 Introduction

(Ia) has stronger somatosensory activations compared to most other features

plotted for that area. And the anterior cingulate cortex (ACC) has stronger activa-
tions that relate to motor execution. These differences reflect the inputs and out-
puts of both areas. The Ia has strong somatosensory inputs whereas the ACC does
not; and the ACC has direct outputs to the motor cortex whereas the Ia has none
(Saleem et al., 2008).
Multivoxel pattern analysis (MVPA) and representational similarity analysis
(RSA) can also be used to show that different areas encode different information. In
an fMRI study, Woolgar et al. (2011) taught human subjects two incompatible sets
of stimulus–​response mappings. The subjects saw one of four stimuli and pressed
one of four response keys. The colour of the background specified the appropriate
set of mappings between each stimulus and a response key.
Woolgar et al. then used MVPA to study four aspects of cortical encoding: the
mapping rule, the position of the stimulus, the response, and the colour of the
background. Their analysis showed that, for example, an activation in the ventral
PF cortex coded for the rule, and an activation in the medial parietal cortex (area
7m) coded for the position of the stimulus.
Similar methods can be used with neurophysiological data. Hunt et al. (2018)
recorded from cells in the orbital and dorsal PF cortex and the dorsal bank of the
anterior cingulate sulcus. The task involved attention-​guided search and choice.
By using RSA, Hunt et al. were able to demonstrate a triple dissociation among
the contributions of these three areas. Specifically, they found that: (1) activity in
the orbital PF reflected attention-​guided comparisons of value; (2) activity in the
dorsal PF also reflected these comparisons, but it did not store them, unlike the
orbital PF; and (3) that activity in the sulcus of the ACC reflected both choice com-
mitment and action selection.

The Relation between Function and Connectivity

The fundamental claim made by Passingham et al. (2002) was that localization of
function depends on the unique pattern of connections of each area. However,
they did not present any data to prove this.
Mars et al. (2018) have now done so. They analysed data from the Human
Connectome Project. This enabled them to detect where there were changes in the
pattern of connections as visualized by resting-​state covariance, and to see if these
changes matched the borders between areas as visualized by task-​related fMRI
activations.
Figure 1.6 (next page) shows the result. There was a remarkable degree of overlap
between the borders as established by these two methods.
This demonstration would be even more convincing if the anatomical borders
were determined by using DWI to visualize the connections. It was Johansen-​Berg
 Neural Operations 13

Figure 1.6 Comparison of anatomical and functional borders

Using data from the Human Connectome Project, Mars et al. based cortical borders on
changes in connectivity profiles as determined using resting-​state activation (black) or
by similarities in fMRI activations during tasks (red).
Reproduced from Mars R.B., Passingham R.E., & Saad J. Connectivity and function in cognitive
neuroscience: From areal fingerprints to abstract spaces. T Cog Neurosci, 22, 1026–​37, Figure 2, https://​
doi.org/​10.1016/​j.tics.2018.08.009 Copyright © 2018, The Authors. Licensed under CC BY 4.0.

et al. (2004) who were the first to demonstrate that the borders between areas can
be visualized by using DWI. They were able to detect the border between the pre-​
SMA and the SMA by finding a change in the overall pattern of connections.
To validate the method, it is critical that the borders as revealed by DWI should
correspond to the borders as identified by a study of the cytoarchitecture of the
same areas. Klein et al. (2007) used Broca’s area as a test. It turned out that there was
good agreement on the boundary between Broca’s area 44 and 45 as determined by
the two methods.
Beckmann et al. (2009) used DWI to detect the borders between nine subareas
in the cingulate cortex. This allowed them to test whether the anatomical borders
corresponded to functional border. To do this, they carried out a meta-​analysis
of 171 studies in which fMRI had been used to identify the functions of the dif-
ferent subareas. It proved possible to relate differences between areas as established
by DWI to differences in functional specialization. Though this study did not plot
functional fingerprints, it nonetheless confirmed that functional localization de-
pends on connectional fingerprints.

Transformation

Though functional fingerprints can be shown to be unique to an area, they are

simply a way of using polar plots to summarize the responsiveness of an area.
They do not tell us about the flow of information from inputs to outputs. In other
words, they do not tell us about the neural operations that occur in the area.
This is constrained by the pattern of inputs. But it also depends on the intrinsic
14 Introduction

micro-​circuitry of the area, that is the percentage of cells of particular types and the
way in which they are interconnected (Douglas & Martin, 2004).
It is this circuitry that performs the transformation from inputs to outputs.
Chapters 3–​7 therefore make proposals concerning the way in which the inputs to
different PF subareas are combined and transformed so as to generate the outputs.
These proposals are given at the end of each of these chapters, and they are printed
in italics. They are mainly framed at the computational and algorithmic level as
defined by Marr (1982). In other words, they attempt to describe what operation is
performed rather than how it is implemented.
To give an example, Chapter 5 deals with the frontal eye-​field, an area that is in-
volved in identifying the targets for saccadic eye movements. Neurophysiological
studies have shown that some of cells respond to visual stimuli, others are
visuo-​motor, and yet others are motor (Schall, 1995). This suggests that these
cell types are involved in the transformation from visual inputs to motor out-
puts. However, it is quite another enterprise to present a precise computational
account of how the intrinsic micro-​circuitry implements that transformation
(Heinzle et al., 2007).
The proposals at the end of Chapters 3–​7 might be thought to imply that the
transformation that each area performs is static. We know, of course, that this is
not the case. For example, in an electroencephalogram (EEG) study Stephani et al.
(2020) have demonstrated that the responses of the primary somatosensory cortex
(SI) are variable, depending on the excitability of the system, and in particular os-
cillatory activity in the alpha band.
Clearly, therefore, the proposals are over-​simplified. Research is needed on the
extent to which the transformation that an area performs is dynamic, depending
on the internal state of the area, the specific inputs at the time, and the current task
or rule. Chapter 11 describes an experiment which shows that the PF cortex sends
outputs to different areas depending on the task that the subject is going to perform
(Sakai & Passingham, 2006).

Summary

Localization of function depends on the fact that each area has a unique pattern of
inputs and outputs. The challenge is to understand the transformation that each
area performs between these inputs and outputs.

Macaque Monkeys

Understanding the transformation that each PF area performs requires a speci-

fication of the connections and an account of the neurophysiological activity in
that area.
 Macaque Monkeys 15

Connectivity

Some readers might think that DWI and resting-​state covariance are wholly ad-
equate for specifying the connections and that fMRI can replace the methods that
have been developed for studying function in animals. So, it is necessary to explain
why Chapters 3–​8 depend mainly, though not entirely, on experiments that have
been carried out with macaque monkeys.
It is true that DWI and resting-​state covariance provide information about con-
nections, but the problem is that these methods have limitations. DWI depends
on the fact that the MRI signal can be sensitized to the diffusion of water along
axon bundles (Basser & Ozarlsan, 2009). Resting-​state covariance measures the
degree to which the activation of different areas follows the same time course.
Together, these methods have been used to produce a ‘connectome’ of the neo-
cortex (Bullmore & Sporns, 2009; Behrens & Sporns, 2012).
However, DWI has the limitation that it can produce false-​positive results
(Maier-​Hein et al., 2017) and though methods have been developed to prune these
(Roberts et al., 2017), it is not clear to what extent they are successful. Also, as cur-
rently used, DWI cannot trace fibres into the cortical grey matter (Reveley et al.,
2015). To overcome this problem, researchers seed the white matter under a cor-
tical area. However, there is a danger of seeing fibres of passage by doing this, and
thereby producing misleading findings.
Resting-​state covariance or ‘functional connectivity’ is assumed to be related to
anatomical connectivity (Bullmore & Sporns, 2009; O’Reilly et al., 2013). However,
it has a major limitation, in that there is no way of knowing whether the connec-
tions are mono-​or poly-​synaptic.
The advantage of using tracers such as fluorescent diamidino yellow is that we
can be sure that the connections that are identified are mono-​synaptic. Diamidino
yellow is an example of a retrograde tracer, meaning that if it is injected into area B
it travels from the terminals of the cells in area B to the cell bodies in area A.
Of course, if the aim is to characterize a system, tracers can be used to visu-
alize poly-​synaptic connections. For example, viruses have been used to study the
projections of the basal ganglia and cerebellum to the neocortex via the thalamus
(Bostan & Strick, 2018). There are also magnesium-​enhanced tracers that are para-
magnetic and therefore detectable by MRI, and these have been used to visualize
circuits (Murayama et al., 2006). However, they cannot be used for the human brain
because it would not be ethical to inject tracers of any sort into human subjects. For
this reason, the experiments have been conducted with macaque monkeys.

Activity versus Activation

Understanding how the system processes or transforms information requires

knowledge of the cellular mechanisms. But the BOLD signal recorded during
16 Introduction

fMRI is a vascular one. It depends indirectly on the fact that when cells in an area
become active there is an increase in the flow of the arterial blood bringing oxygen
and glucose.
The typical spatial resolution of the signal is in the order of millimetres, and
this is adequate for visualizing activations that are located within neocortical areas.
With higher magnetic strength and improved coils, it is now possible to achieve
spatial resolutions that are close to the size of the functional patches that have been
demonstrated in macaque monkeys. For example, there are patches in the inferior
temporal cortex that respond to object-​like stimuli, and these are roughly 0.4mm
in size (Tanaka, 1997; Wang et al., 1998). Similarly, in the PF cortex there are func-
tional patches for spatial and motion direction that are roughly 0.7mm in size
(Masse et al., 2017).
However, to understand how areas perform their functions, it is necessary to
record from the cells or neurones themselves, and neuronal cells measure in the
order of microns.
It is true that recordings can be taken in patients, either with arrays of electrodes
on the cortex (Forseth et al., 2018) or by implanting microelectrodes during sur-
gery (Mukamel & Fried, 2012). But this can only be done when there is clinical jus-
tification. Furthermore, if recordings are taken in patients during surgery, it is only
possible to record over relatively short periods, whereas recordings can be taken
over months when animals such as macaque monkeys are used.

Macaques as a Model

There are several advantages of using macaque monkeys. They are highly adapt-
able animals. They live in a very wide range of habitats from Afghanistan to India,
Indonesia, Japan, and China. This means that they adapt well to laboratory condi-
tions. They are also able to learn complex cognitive tasks, even if it can take months
to train them.
However, there are two important reservations. First, macaques cannot be
treated as if they are ‘representative’ of monkeys. There are seventy-​nine genera
of primates, and they are remarkably diverse (Preuss, 2000). Just consider, for ex-
ample, the golden lion tamarin (Leontopithecus rosalia), the hamadryas baboon
(Papio hamadryas), and the red-​faced spider monkey (Ateles paniscus).
Next, humans are, of course, much more closely related to common chimpan-
zees (Pan troglodytes) and bonobos (Pan paniscus) (Staes et al., 2019). But it has not
been felt to be ethically acceptable to carry out anatomical tracing or cell recording
experiments in chimpanzees.
However, fortunately, there is some brain imaging data on chimpanzees using
positron emission tomography (PET), and this is mentioned in Chapters 3 and 11.
Another random document with
no related content on Scribd:
time you enter the peerage, my dear. Which will be home first, you or
Henry?”
Letitia tore a leaf out of her sketch-book, which she still carried,
and wrote a note for her husband in case of his immediate return.
The earl charged himself with it, as she had no time to go back and
seal it; and putting her arm within his own, led her to the gate where
the carriage was to meet her. He thought, as she did, that it was best
to avoid the risk of encountering anybody who might look for an
explanation.
“Farewell, my dear,” said he, as the carriage stopped. “We shall be
glad to see you back again; meanwhile, all success to your
measures!”
“How good you are to trust me for meaning something better than
folly, as I see you do!” said Letitia, with tearful eyes. “This looks so
like a madcap expedition!”
“When I have seen you do a foolish thing, my dear, I will believe
that you may do another. Till then, my faith is strong. Nay, give me a
happier smile before you go. Has your power ever failed you at
need? I do not know what you expect from it, but I will venture to
predict that it will not now fail you for the first time.”
Before the carriage had well cleared the gate, it stopped again at
the earl’s command. He appeared at the window to say,
“It never occurred to me to ask whether I can be of use by going
with you. Say that you wish it, and I am ready, this moment.”
“You are kind; but I do not wish it:” and again the carriage rolled
on.
With a beating heart, Letitia made her inquiry at the door of her
town-house. Lord F—— was not there. He had gone down into the
country,—(not to Weston),—that afternoon, leaving a letter to be
forwarded to her, which had been put into the post-office some hours
before. Letitia’s best hope was over. It was midnight;—too late to go
to her lawyer. She gave orders to be driven to her sister’s, thinking it
better to alarm her a few hours sooner than to risk any loss of time or
of counsel.
She carried little new alarm into Maria’s abode. There were lights
seen in the windows, and Maria herself was up and dressed. This
was the second night that she had not gone to rest, for it was the
second that Waldie was absent without notice, or any intimation
where he might be found. The unhappy wife flew to the door on
hearing the carriage-wheels. When she saw her sister and Thérèse
alone alight, she assumed a forced calmness of manner, as if
bracing herself up to bear the worst. Letitia judged it best to use no
disguise, from which Maria had suffered all too much already.
Inwardly moved by the downcast look of perplexity with which the
tidings were received, she told of Waldie’s appearance at Weston, of
his errand,—if errand it might be called,—and intended return. It was
some relief to Maria to suppose him engaged in town, providing for
the approaching crisis, instead of being kept away by any of the
horrible causes which she could not prevent from filling her
imagination by turns.
The lawyer, Mr. Bland, was—not much to his content—called away
from his breakfast and newspaper, the next morning, by the ladies,
whom, being ladies, he could not think of keeping in waiting till he
had made himself master of all the news. Coldly and solemnly he sat
himself down to listen to their affair, and prepared himself with his
snuff-box to get over as well as he could the tedium of hearing a
business statement from women. He would have cut the matter short
near the beginning, with the assurance of the impossibility of raising
securities for so large an amount before two o’clock; but Letitia
would not be silenced. She showed that she understood the case,
pointed out the advantage that might accrue to all parties from the
transaction, and indicated such satisfactory means of ascertaining
whether the speculation could in reality fail, if the proper funds were
provided, that the surly Mr. Bland was won over to promise that he
would see what could be done; whereupon the ladies immediately
left him, promising to return in four hours, to convey him and his
securities to the place where the business was to be transacted.
“Where shall we go?” asked Maria. “What can we do with
ourselves for these long four hours?”
“If you have courage to go with me,” replied her sister, “you will
find ample employment for the time. If not, we part here, and I advise
you to take a country drive to refresh yourself. I am going into the
depths of the city to find up a money-lender, who has proved a very
convenient help to certain young gentlemen of lord F——’s
acquaintance. One may as well try to have two strings to one’s bow,
since the worst that can happen is to be laughed at, as women are
every day when they propose to meddle with business.”
“Is this the worst that can happen?” asked the timid Maria. “Do you
understand the law in such matters? I would not have you involved,
Letitia, even to save us.”
“Trust me for doing nothing that my husband would not have me
do,” replied Letitia. “Will you come? Our dress tells nothing, does it?
It might belong to anybody, from a milliner to a maid of honour. Will
you trust yourself with me?”
Maria gave herself up to her sister’s guidance. They quitted the
carriage about half a mile from the house they were in search of.
“I know the lane,” observed Letitia, “but not the number. We must
venture a guess upon the house. I will make no inquiries.”
They walked two or three times along the narrow and dark lane, all
the dwellings of which appeared to Maria equally desolate and
unpromising; but her sister, who had fixed on one from the
beginning, was confirmed in her opinion by seeing half a pint of blue
milk taken in at the front door, while a fruiterer’s boy, carrying a
covered basket, through whose sides might be discerned the richest
of grapes, turned into a court which led to the back of the premises.
“Blue milk in public for the serving man’s breakfast,” said Letitia,
“and purple grapes in private for the master’s luncheon. This suits
the man exactly. This must be the place.”
So saying, she walked up as the milkman made way, and asked
for Mr. Simeon. The wizened, sly-looking old serving-man replied
that Mr. Simeon was engaged on business. Perhaps the ladies had
mistaken this place for the shop in —— street. Only the wholesale
jewellery business was carried on here. No; they wanted Mr.
Simeon, and would wait till he was at liberty. After several messages
backwards and forwards, the ladies were beckoned in, with
apologies for the parlour not being at liberty. A dingy wareroom
having been passed, it was next required of them to mount a sort of
ladder into what they supposed would prove a loft, but was in reality
a counting-house, so dark that it appeared questionable whether any
business could be carried on at any hour of any season without
lamps. Maria would have sunk down on the first chair, if chair there
had been: and in the absence of any, was fain to perch herself on
the high stool, which afforded little rest for want of a footstool. Letitia,
who was always conscious of inward enjoyment when in strange
scenes and circumstances, peered round in the gloom to make her
observations. It was well that she kept to herself her remarks on
chests and padlocks, on the flask which stood on a corner shelf, and
on the bareness of the whole place, which left nothing but the said
flask which could be carried away: it was well that she made no
audible remarks on these things, as some one was present before
either she or her sister was aware. Mr. Simeon had entered by an
unseen door, and his compliments to the ladies were the first
intimation of his presence. She observed a manœuvre to get them
placed opposite the little light there was, and disappointed it; being
disposed to reconnoitre the person with whom she was about to
deal. She was surprised to find him a well-made, middle-aged man,
whose countenance, as far as she could see, corresponded with his
address, which was mild and courteous. She explained, without
delay, that her business was to ascertain on what terms so many
thousand pounds could be borrowed for a month.
On no terms which were not sanctioned by the law of the realm.
Perhaps the ladies were aware of the law?
Letitia replied that the same terms might suit the present case as
had been agreed upon by Mr. Simeon for loans of five, ten, and forty
thousand pounds, at such, and such and such dates.
This proof of some knowledge of his transactions caused the
money-lender to pause and attentively consider his guests; after
which he observed, as if half to himself, that debts of honour were
troublesome things, and especially to ladies, to whom ways and
means were less open than to gentlemen. Letitia supposed that Mr.
Simeon knew best, from the nature of his business; but she had
believed that gaming was obsolete among ladies. She knew no
ladies who were addicted to play. Simeon’s further remarks glanced
upon unpaid jewellery, the flight of Chancery-wards to the continent,
and divers other suppositions, all of which were baffled by one or
other sister, who did not choose to allow occasion for any scandal
against themselves, in case of the present transaction becoming
known. Letitia cut the investigation short by requesting to look at the
statute which regulates the rate of interest on monies lent, and which
she concluded to be in the possession of a money-lender. It was
brought, and with it a taper, by whose light Mr. Simeon was enabled
to perplex himself still further about the quality of his fair visitors.
“It is an unjust law, madam, a cruel law, worthy only of the
Mahomedans, who call it a sin to lend monies on interest; but it is the
law....”
“And must therefore be obeyed, Mr. Simeon. The forfeit—‘the
treble value of the monies, or other things, so lent, bargained, &c.’—I
wonder they do not ordain the treble value of silks and sugars to be
forfeited when the price rises. As well one commodity as another.”
“Ah, madam, that would raise the prices unconscionably. People
must have commodities; and if they cannot get them by a straight-
forward course, they must have their little plans and managements.
There is risk and trouble in such plans; and for this the planners
must be paid. So much being added, the prices would rise
unconscionably.”
“That is to say, sir, that we are to pay you unconscionably, if you
can make a little plan to furnish us with this money. Let us hear your
terms, supposing we can furnish you with unquestionable security.”
Mr. Simeon seemed disposed, however, to descant a little longer
on the hardship of the law, which not only, he observed, obliged him
to be wary and even apparently rigid in his proceedings,—not only
was a perpetual and most injurious hinderance in the way of
commerce,—not only showed that the makers of the statute did not
understand the office of a circulating medium,—not only brought the
holy law of Moses (by which the taking of interest was falsely
supposed to be forbidden) into contempt,—but had actually brought
two charming ladies from their native regions of refinement into a
dark hole quite unworthy of their presence! He was recalled to
business, and obliged to state the rate of interest he would receive
through one of the circuitous and safe methods which necessity has
invented. He was not sufficiently aware with whom he had to deal.
“Your terms, Mr. Simeon, would suit a time when money is scarce;
whereas you know as well as I that it is plentiful, and that the rate of
profit has not for many years been so low as at present.”
 Mr. Simeon endeavoured to mystify her by pointing out that the
kind of profit in question had nothing to do with other profits, the
lending of money being an unique case. It would not do.
“Consider interest in what light you will, sir, it comes to this.
Interest is the nett profit on capital, and that nett profit cannot but be
low in the present state of the market. There is a money-lending
market, as you well know, though your department of it is
discountenanced; and we are not in such a hurry but we can walk
through it and learn what terms some of your neighbours have to
offer. Our object is gained in finding that you can advance what is
wanted.”
Mr. Simeon shook his head, and observed that the securities were
not yet before him, that he had entered into large engagements
already this morning, and that there were sundry other difficulties in
the way of a conclusion of the bargain. To which the ladies replied
that both parties had better take time to consider; and that a
messenger should wait on Mr. Simeon at three o’clock to put an end
to the treaty, or conduct him to the place where the securities would
be waiting for him. To this the man of money agreed, only requesting
to appoint a later hour, on account of prior engagements.
The ladies were urged to refresh themselves with some rare
foreign wine, to accept an escort home, and to do or permit many
other things which might afford a chance of their revealing
themselves: but in vain.
On leaving the place, Maria proposed making a circuit to join the
carriage.
“Why?” asked her sister. “We have done nothing to be ashamed
of.”
“Why then conceal your name?”
“Simply because it had nothing to do with the business, our errand
being merely exploratory; and it might have altered the terms in a
way injurious to your husband. Now that our errand is done, let them
follow us and see who we are, if they like.”
“But the errand itself!”
“Is anything but a pleasant one, certainly; but my conscience is at
ease as to my share of it. We keep the letter of the statute, you
know, and that is enough. No one is bound to keep the spirit of a bad
law, since evasion is the only means of bringing on its repeal. As for
the usury laws,—they have been repeatedly condemned by
committees of the legislature; and the more they are evaded, the
better is the chance of getting rid of them. Do not you see this? Do
not you see that perpetual evasion of any law is a sufficient proof of
its badness?”
“You have such courage!” exclaimed Maria. “All I wish for is to get
through life as quietly as I can, and bring up my children to do the
same.”
“Beware of teaching them blind obedience, Maria,” said Letitia,
when once more seated in the carriage; “your girls equally with your
son. Obedience, by all means; but a rational, discriminating, and
therefore loving and hearty obedience to the public laws as well as to
those of your own house. Your little ones will learn hereafter that
your object in forbidding them to set foot on the hearth-rug in your
absence, is to guard them from being burned. Let them learn at the
same future time the purposes of the laws under which they live, that
they may be ready to do their part in that renovation of the system
which is required as years roll on. If you would not have your
children retain a superstitious dread of a hearth-rug through life,
neither would you have them cling to laws enacted in the infancy of
the state, and inappropriate to its present condition.”
“Implicit obedience is at least safe,” observed Maria.
“Safe to a certain point, but no farther. If you continue the law of
the hearth-rug for twenty years to come, your obedient children will
never be burned by crossing it; but do you suppose they will not by
that time have discovered other means of getting the warmth they
wish for? They will creep under it; they will creep round it; they will
jump over it. So is it, and so should it be with absurd, antiquated
laws.”
“Who is to judge which are absurd and which sound?”
“The bulk of the subjects of them. A sound law can never be
evaded by more than a solitary simpleton here and there, against
whom society will rise up; since it is the paramount interest of society
to keep good laws in effectual operation. When the time comes for
the bulk of society to approve and enforce the usury laws, you and I
will pay no more visits to Mr. Simeon. Till then, or till their repeal, let
there be opposition to the spirit and grudging obedience to the letter,
unless we are prepared for the consequences of a breach of both.”
“Not I, nor, I hope, anybody belonging to me,” replied Maria. “O,
Letitia, what o’clock is it? I cannot trust my watch.”
“Far enough from two o’clock, my dear. So you will not be amused,
even with talk about the usury laws. Well? I will keep all drowsy
subjects to lull you to sleep with to-night, when all will be settled;—all
redeemed, I trust; and when you will own at last that watching has
nearly worn you out.”
Mr. Bland looked as immitigably solemn as ever, when he
appeared at his own door on the carriage stopping. He would have
had the ladies wait the result at his house; but Letitia’s business was
not finished till she had ascertained whether Simeon’s help would be
wanted or not. Mr. Bland was obliged to let his law papers be tossed
into her lap, and to edge in his stick and portly person as well as he
could. He had been busy since the morning interview, and had fully
satisfied himself in the matter of the spices; but he said to himself,
while being whirled along, that the affair could hardly be brought to a
satisfactory conclusion, since a woman had so much to do in it. If it
had not been for the Earl’s recommendation of the case, he would
have eschewed the whole matter; and the oddest thing was that his
lordship did not say whether he was himself informed of the
particulars.
“Is Mr. Waldie here?” inquired the trembling wife, in a choking
voice, of one of the clerks, who appeared when the carriage
stopped.
“He is, madam; but particularly engaged at present, except——”
“Except to this gentleman,” said Letitia, handing Mr. Bland’s card
with her own, which brought an immediate request that the party
would alight.
Mr. Waldie was in the act of shutting somebody into an inner room
when his wife appeared at the door. He looked pale and worn, but
composed and active. He received his wife and her sister as if
nothing extraordinary had happened, stated that the money would be
forthcoming if the securities were so; and went straight to business
with Mr. Bland.
 As soon as satisfied that all was likely to be well, the ladies
proposed to withdraw into the inner room, and await the issue.
“That room? No; not there, my dear,” said he. “Yet you will not
mind my other man of business being there. He will not be in your
way long.”
So they were ushered into the apartment where stood—Mr.
Simeon.
“You will be saved the trouble of another excursion at four o’clock,
Mr. Simeon,” observed Letitia. “We have only to regret having
consumed some of your time already this day. You will hardly see us
again till we have debts of honour to pay, or a Chancery elopement
to provide for.”
Mr. Simeon considered himself a gainer by the transaction in
proportion to the honour his poor counting-house had enjoyed; an
honour the more precious for its being confined within his own
breast. He knew his duty too well to reveal what had passed.
“Do as you please about that,” replied Letitia. “You and Mr. Waldie
must agree about your keeping Mr. Waldie’s secrets; but, for my part,
I have none. You owe neither honour nor duty to me, aware, as you
no doubt are, that I did not come to borrow money on my own
account.”
Mr. Simeon merely mentioned the temptation of talking about the
affair, because it was really an extraordinary case. Not that it was a
rare thing for ladies to want money; but that they usually employed
agents to procure it. How indeed should it be otherwise? since not
one woman in five thousand understood even the forms of business;
and these solitary exceptions were in a class which had no dealings
with moneylenders. On this, followed a series of narratives of fair
ones’ difficulties for want of cash, which amused Letitia exceedingly,
from the romance of adventure which was mixed up with the most
sordid borrowing transactions. The heroines were only A. B. and C.;
but they became real personages in Letitia’s imagination on the
instant; and she was almost sorry when Mr. Simeon was called into
the next room to review his securities and perform his promises.
It seemed an age before Mr. Waldie threw open the door,
announcing that all was well. He briefly thanked Letitia for having
saved him; urged them to return home and rest from their anxieties;
and was only sorry that he could not accompany them, or even
promise to follow them for some days, as he should be incessantly
occupied till the expected cargo was secured. He perceived that his
wife’s countenance fell on hearing this, and rallied her; asking what
there was now left to be afraid of?—She did not know, but——
“She is worn out,” said Letitia. “I will take care that she shall recruit
herself, and wait patiently, unless you try her too long.—You may be
quite easy,” she continued to her sister, when Waldie’s last grave
smile dismissed them. “All is safe, with him as well as with his affairs.
How calm he is! How entirely himself! He will speculate no more,
believe me.”
Maria shook her head, as her tears fell fast. It was not only that
her nerves and spirits were shaken by what she had gone through.
Her confidence was utterly overthrown, and she felt the present relief
to be no more than a respite.
 Chapter VIII.

CONSEQUENCES.

Those who had educated Waldie were partly answerable for his
propensity to speculation, which arose more from a restless ambition
than from a desire of overgrown wealth. The foundation of the
fortunes of his family was laid by an ancestor who, a few hundred
years ago, introduced a new manufacture, which he had learned
abroad, into this country. Though, from having to take workmen
away from other manufactures, and to engage them to learn his own,
he gave higher wages than any of his neighbours, his profits were
also very great, as his article bore a high price in the market, in the
way that new articles of convenient manufacture generally do. If
profits may be said (as they are by some said) to be as much the
reward of labour as wages,—that is, the reward of present
superintendence, and of the labour out of which arose the capital
employed, it is certain that this first rich member of the Waldie family
reaped as large a proportionate reward as his workmen; for long
after their wages and his profits were lowered by his silk stuffs
becoming more common, and the difficulty of getting workmen being
less, he continued to grow rich from his having more capital to
employ in bringing him profits. If every hundred pounds did not
produce seventy-five, he had, in course of time, for every such
hundred, five that brought 50 per cent., and afterwards fifteen that
brought 25 per cent.; so that he continued to grow rich, just as
individuals and countries may in these days, if accumulation
proceeds faster than profits fall. His descendants for some
generations carried on this manufacture, for which there was a
permanent demand, and so steady a one that the variations in its
price arose only from the variations in the prices of other things, and
not from changes of fashion. Now and then the price of provisions
fell, which enabled these manufacturers to lower their men’s wages,
and enjoy larger profits, till the time came for profits to fall also; and
sometimes the reverse happened, when the price of provisions rose.
Sometimes complaints were made against the largeness of their
profits, when the fact was that they gave precisely the same
proportion of their produce to their workmen as before, but there
were more workmen to divide this proportion; which was no fault of
their masters. With these few variations, the family continued to
prosper, being for the most part content with the ordinary rate of
profits, and making up by continual accumulation for their gradual
fall;—that fall which must take place wherever the supply of food is
restricted. They were all proud of the ancestor who had founded the
wealth of their family, and sent his praises down from generation to
generation. The present Mr. Waldie had been early accustomed to
listen to them, and impressed with the idea that it was time for
somebody else to be adding glory to the family, as it had become
much less distinguished in these times of improvements than in the
first days of its wealth. He quitted the manufacture, and became a
merchant, thinking that this occupation would afford better
opportunities for the gratification of his ambition than the
straightforward old manufacture. No wonder he was tempted by
schemes which promised a higher than extraordinary rate of profits!
No wonder he dealt in articles whose extremely varying prices were
determined by other than the usual circumstances,—which prices he
hoped to catch at the highest, and then to have done with the article!
No wonder that he guessed respecting such uncertain
circumstances as the changes of fashion, wet and dry seasons, the
extent of particular crops on the other side the globe; and then
proceeded to act upon these guesses! Sometimes he was right,
sometimes wrong. Sometimes he made five thousand pounds at a
stroke, sometimes lost ten in a season. Much as his capital was
lessened on the whole by his speculations, this was by no means the
worst result of his proceedings. Like all other gamesters, he became
so fond of the excitement, that, much as he often suffered from it, he
could no longer live without it; and the domestic influence which is
the most powerful means of winning a man from bad habits of any
kind, was not so powerful in Waldie’s case as if his first affections
had not been disappointed. Attentive as he always was to his wife
when with her, kind as he had till lately been to his children,
vehement as were the fancies he took, now to Portugal laurels, now
to tall trees, now to bay-windows or new drawing-room furniture, his
happiness was not in his home, but in the heats and chills of his
hopes, in city news of disasters at sea, of changeable weather, of
new inventions or improvements of manufactured articles, and of
political changes,—of anything that might affect his speculations. His
poor wife knew nothing for a long while of his unfortunate ventures,
though she heard enough of exultation over his good ones. She
believed that he must be growing enormously rich, and sighed over
the idea, since it seemed that the richer he became, the less
pleasure he took in his home. When he first began to alter his tone,
(which he did very suddenly,) when he talked one day of bringing up
their children to provide for themselves, and moving into a small
house in the city, and the next of purchasing some splendid estate,
and again of giving up his carriages and sending away half his
servants, she was confounded; not knowing how much to believe, or
to wish to believe; whether to suppose poverty to be in prospect, or
her husband to have lost the soundness of his judgment. It was now
some comfort to know how their affairs stood, though she would
rather have heard it from her husband than from Letitia. She had
long seen that Waldie’s family ambition could not be gratified.
Whichever way the scale might turn at last, whether he left his family
in poverty or magnificence, it was impossible that his memory should
be honoured like that of the ancestor who had prospered by uniting
prudence and industry with his zeal of enterprise. Whether all was to
be swallowed up in Kentish hops and Russian tallow, or all
redeemed and even doubled by India spices, those of Waldie’s
descendants who knew his history, would, in either case, pity or
despise him as a gambler.—How much of pity his fate would call for,
not even the most alarmed imaginations of his timid wife had fully
conceived. She had fancied him, over and over again, in gaol, in
poverty; the idea of suicide even had flashed across her mind; but
that which actually happened took her more by surprise than arrest,
or ruin, or death by his own hand.
Two anxious days were passed by the sisters in expectation of the
decision of Waldie’s affairs, and still he did not appear. Notes came
two or three times a day from himself or from his clerk, who wrote at
his desire, requesting Mrs. Waldie not to leave home, as her
husband did not know how soon he might be with her, to take a few
days’ repose on the conclusion of his business. Letitia heard from
her husband also on his arrival at home, and from the earl, both
desiring her not to leave her sister till she could do so with comfort;
which, in Letitia’s mind, meant till Waldie should have come home.
On the third morning arrived this extraordinary note.
“My own dearest Maria” (substituted for “Letitia,” scratched out)
Coming, coming, coming, as rich as Crœsus! Light the bonfire.
Ring the bells. Hurrah! Spices for ever! Coming, coming, coming!
F. W.”
“I do wish Waldie would control his spirits a little,” said Maria,
showing the note to her sister, and then looking as if she would fain
have withdrawn it. “How can he bring himself to write in such a
way?”
Letitia had nothing to say at the moment; not even congratulations
on the wealth of Crœsus having crowned all these vicissitudes. She
asked for the children. They were gone out with Thérèse and their
own nurse-maid. She offered a turn in the shrubbery; but Maria was
not, she presently saw, strong enough to walk. She threw open the
bay-windows, and beckoned her sister to come and be refreshed by
the feel of the mild autumn air, the bloom of the autumn roses, and
the tranquil beauty of the green prospect. There they sat, watching
and working, letting drop a few words now and then, but keeping up
nothing like conversation, and looking out as often as a horseman
might be seen through the trees, or a carriage heard in the road. At
last, the sound of a horse’s hoofs reached them, far too rapid for
safety, they were sure; and immediately Waldie was seen on
horseback, approaching at tremendous speed, with something white
before him, which proved to be his two elder children.
“Mercy! Mercy!” cried his wife, putting her hands before her eyes.
“Thank God! the gate is open. They are in! Safe!” exclaimed
Letitia, as the horseman wheeled round the corner and up to the
window, checking his steed so suddenly as to throw it on its
haunches, setting the pebbles flying in all directions, and mingling
his loud hurrah with the laughter of the younger child, who saw
nothing but fun in all this. The elder one was convulsed with terror.
 It was well some one was on the spot; for Waldie threw down both
the children as if they had been mere bundles of clothes. They were
caught,—not, however, without so much slight bruising as called
forth their cries to add to the confusion.
“O Waldie,” shrieked his wife, “what are you about?”
“Look, look, look!” he cried, flourishing his whip over his head,
clapping spurs to his horse, and trampling the beds, walks, and lawn
alike, and finishing by making his horse leap high and still higher
shrubs. He finished by fixing his eye upon the greenhouse, as if he
contemplated a leap there too.
“Mr. Waldie,” said Letitia, in her steadiest tone, “what are you
doing?”
In a moment he was off, had flung the bridle on the neck of the
sweating and trembling horse, and was by her side, swearing deep
oaths that she had ever governed his life and ever should govern it.
With her in wealth, as with her in poverty, he would....
Maria had rushed into the house upon this, but not the less did
Letitia by eye, and gesture, and word, command him from her, and
prevail for the moment. He obeyed when she pointed his way into
the house, and she was still standing, faint in body and spirit, with
the poor children clinging to her, when Thérèse came in from the
road, breathless, and sinking with terror. When she saw the children
safe, she burst into tears: she had feared that she might have to
answer for their lives, from not having had presence of mind to
evade Waldie’s vehement desire that the children should have a ride.
—Her mistress gave her a few directions, which she hastened into
the house to execute, and Letitia, after giving the servants a charge
to take the children into the nursery and keep them there, repaired to
her sister. She found Maria lying across the bed, groaning in heart-
breaking grief.
“Sister!” said she, gently, after watching silently beside her for a
few moments—“Sister, your husband wants you. He is ill, fearfully ill;
and who should tend him but you?—Nay; why this despair? A brain
fever ... all may be well....”
“Letitia; do not deceive me. It is mockery to attempt it.”
“Maria, if I wished to deceive you, I dare not. What I have done will
prove this. Thérèse is packing up, and I am going in half an hour. It
grieves me to leave you; but I must go.”
“O yes, yes; you must go.”
At this moment, there was a tremendous knocking at the room
door, which was luckily fastened. It was Waldie, still calling upon
Letitia, who would not answer. Maria dared not. The knocking went
on till there seemed some probability of the door giving way, when,
perhaps from having his attention diverted by the servants, the
madman quitted his object, and ran down stairs.
“Yes, yes: you must go,” repeated Maria, bitterly.
Letitia could forgive the tone in which this was spoken.
“Listen, Maria, what you must do. Command yourself, and go and
tell the butler that his master has a brain-fever, and desire him not to
quit your husband’s side but at your bidding. Have the children kept
away, and, if possible, stay with your husband, looking and speaking
like yourself, till some one comes to relieve you of your charge. I will
immediately send proper advice and help from town.—Farewell,
sister. I shall not come again till you send for me. As soon as I can
be of any comfort, send. My husband will wish it.”
“But Waldie will insist on going with you. He will never let you drive
off. He will....”
“All this is provided against. I can plead an errand near the
turnpike, and shall go out with Thérèse by the little shrubbery gate.
The carriage will overtake us. Do not detain me. Farewell.”
Letitia said nothing about removing the children. She thought that
if, as was probable, Waldie’s state should prove such as to render
Maria’s presence improper, her children would be her best comfort.
—In a few minutes, Maria saw, but diverted her husband from
seeing, Letitia and Thérèse hastening from the back of the house
through the shrubbery, and disappearing down the road. It was with
a strange mixture of bitter and yearning feelings that the unhappy
wife witnessed such a conclusion as this of a visit which had been
planned and endured for her sake.
There was ample time in after years for the sisters to explain, and
forgive, and renew the confidence which had been unshaken till this
day. Waldie was never more an impediment to their intercourse. He
was kept under close restraint from the hour after Letitia’s departure,
when he insisted on searching every corner of the house for her, and
was frantic at having sought in vain, up to the moment when, after
years, first of madness and then of imbecility, death released himself
and his friends from the burden of his existence.
More than once Maria tremblingly asked the confidential physician
whether her sister’s presence was likely to be of any service; and
almost rejoiced to be answered with a decided negative.
It was perfectly true that Waldie had become, as is commonly said,
as rich as Crœsus. But what to Maria was all the splendour in which
she might have henceforth lived, if she had so chosen? What to her
was the trebling of the fortunes of her children? As a compensation
for the love which had been disappointed, the domestic hopes which
had been rudely overthrown, these things were nothing, though there
were some in the world who called them prosperity.
 Chapter IX.

EACH FOR ALL.

Lady F—— remained a few hours in London to learn the physician’s

opinion of Waldie’s state, and to give notice at home of her
approach. She had no rest, in town or on the road, from the visions
which haunted her of what she had lately seen. Waldie’s
countenance of fierce glee was for ever before her; his raised voice
startled her imagination perpetually. She had no repose till her
husband met her some miles from Weston, suffered her to alight at
the park-gates, and invited her to wander with him to the ruin, and
through the autumnal woods, to her beloved seat beside the stream
that fed the lake. Refreshed and composed, she joined her guests at
the dinner table, and was warmly welcomed back again: not the less
so for no one but the earl and lady Frances having an idea what had
caused her absence. All were ready with that delicate homage which
may be supposed to have been as gratifying in its way to Letitia as it
is to many who relish a grosser flattery than she would ever endure.
All were ready with tidings of her protegés, from pheasants to men
and women. One could assure her that a very favourite plant had not
suffered from the frosts of the night after she left Weston. Another
had tasted the cream of her dairy; a third admired her bantams; a
fourth amused himself with Nanny White; a fifth conversed with the
old sexton; and lady Frances herself condescended to hope that that
good girl, Thérèse, had not been left behind in London. She was
such a treasure! Thereby hung a confession, afterwards given in
private, that Philips was really very much spoiled, and becoming a
great trouble. Her manners were anything but improved, to say
nothing of her temper. Miss Falconbridge, whom she knew to be as
intimate as a sister with lady Frances, had taken a fancy to lady
Frances’s style of hair; and as the easiest way of gratifying her, lady
Frances had ordered Philips to dress Miss Falconbridge’s hair the
day before; whereupon Philips sent word through Miss
Falconbridge’s maid that she must beg to decline the honour! Lady
Frances had insisted, and her maid in some sort obeyed: but never
was anything seen so absurd as the young lady’s head. What was
lady Frances to do? To part with Philips was altogether impossible;
and to bear with her now was scarcely less so. Letitia could not
answer for what she should do if compelled to retain such a person
as Philips: she could only appeal to her own management of
Thérèse as a proof of how easy a matter it is to make a valuable
friend out of a hired attendant.
“O yes! by taking the trouble of educating her, no doubt. But that is
a task I could not submit to. That reminds me—how does Thérèse
get on with politics? I remember her one day, so eloquent about the
revolution her father remembers, and the prospect of another
revolution, and the glory of having seen Lafayette.”
“She knows more than she would probably have learned in the
very heart of Paris. She has left off assuring me that all the kings of
France have been royalists.”
“I suppose it is for the sake of keeping her innocent of some things
which lady’s maids learn soon enough that you let her read and talk
politics as she does?”
“Partly; and partly with a more direct view to my own interest. It will
be of very great consequence to me that she should be, not only
pure in her conduct, but well educated up to as high a point as I can
carry her.”
“Ah! you mean for the sake of your little heir. I see Thérèse is as
busy about the preparations as if she had taken her office upon her
already. But you began your care of Thérèse from the day you knew
her, she tells me.”
“I did; and so I should do still, if there were no heir in prospect.
Should I be justified, think you, in placing any one where I myself
order the circumstances which are to form her character, and at the
same time neglecting to order those circumstances well?—It is
perfectly true that, in engaging servants, we undertake a great task.
In the case of Thérèse, however, the task has been all pleasure.”
“Well, for your reward, I suppose you will keep her always. You will
not let her marry, I conclude; or, if she marries, will insist on her
remaining with you. It would be too hard to lose all your pains.”