Growing Minds A Developmental Theory

GROWING MINDS
Interest in the human mind is a centuries-old fascination, dating back to Plato,

Aristotle, and Descartes. While the theories proposed about the human mind have
since advanced and evolved, the fascination remains. Growing Minds is a unique and
interdisciplinary work that guides the reader through an examination of the human
mind’s nature, performance, lifespan, and variations.
The book sets out to answer a variety of questions:
• What are the cognitive processes underlying intelligence?

• What is general and what is specific in intelligence?
• What is stable and what is changing in intelligence as children grow older?
• Why do individuals differ in intelligence, and are differences genetically determined?
• How is intelligence and intellectual development related to the genome and the
brain?
• How is intelligence related to personality?
• Can intelligence be enhanced by specific interventions?
The text is organised into three parts: the first provides a summary and evaluation
of research conducted on the human mind by experimental cognitive psychology,
differential psychology, and developmental psychology. The second presents an
overarching theory of the growing mind, showing how mind and intelligence are
at the crossroads of nature and nurture; and the third assesses the relationship
between education and intelligence.
This book is the result of decades of extensive research and culminates in the
proposal of a new overarching and integrated theory of the developing mind. For
the first time, research is gathered and combined to form a comprehensive concept
and fulfil the need for a fresh, integrative paradigm which both asks and answers
questions about the human mind from a multi-faceted perspective.
Andreas Demetriou is Professor of Psychology and President of the University of
Nicosia Research Foundation, Cyprus. He was the Minister of Education and
Culture of Cyprus, and President of the National Research Council of Cyprus.
George Spanoudis is Associate Professor in Psychology at the University

of Cyprus.
GROWING MINDS
A Developmental Theory
of Intelligence, Brain,
and Education
Andreas Demetriou and

George Spanoudis
First published 2018
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2018 Andreas Demetriou and George Spanoudis
The right of Andreas Demetriou and George Spanoudis to be identified
as authors of this work has been asserted by them in accordance with
sections 77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced
or utilised in any form or by any electronic, mechanical, or other means,
now known or hereafter invented, including photocopying and
recording, or in any information storage or retrieval system, without
permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identification and explanation
without intent to infringe.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
Names: Demetriou, Andreas, author. | Spanoudis, George, author.
Title: Growing minds : a developmental theory of intelligence, brain, and
education / Andreas Demetriou and George Spanoudis.
Description: Abingdon, Oxon ; New York, NY : Routledge, 2018. |
Includes bibliographical references.
Identifiers: LCCN 2017040878 (print) | LCCN 2017058531 (ebook) |
ISBN 9781315537375 (ebook) | ISBN 9781138689824 (hbk) |
ISBN 9781138689848 (pbk) | ISBN 9781315537375 (ebk)
Subjects: LCSH: Intellect. | Brain. | Cognition. | Learning, Psychology of.
Classification: LCC BF431 (ebook) | LCC BF431. D386 2018 (print) |
DDC 153—dc23
LC record available at https://lccn.loc.gov/2017040878
ISBN: 978-1-138-68982-4 (hbk)

ISBN: 978-1-138-68984-8 (pbk)
ISBN: 978-1-315-53737-5 (ebk)
Typeset in Bembo
by Keystroke, Neville Lodge, Tettenhall, Wolverhampton
Dedicated to our families.
Powerful contexts allowing our research to go uninterrupted

for decades.
TABLE OF CONTENTS
List of illustrations ix
Prefacexiii
Introduction1
PART I
Three traditions of research on the human mind 5
1 The experimental cognitive tradition 7
2 The differential tradition 22
3 Piaget’s theory 31
4 Neo-Piagetian theories 44
5 The development of reasoning 57
6 Awareness and knowledge about the mind 68
7 Core domains 82
viii Table of contents
PART II
An overarching theory of the growing mind 93
8 The organization of the human mind 95
9 Cycles and phases of development 121
10 Recycling and mediation of cognizance 134
11 Differentiation and binding of mental processes through

developmental cycles 147
12 Personality and emotions in the mind 162
13 Genetic, psychological, and cultural aspects of the mind 177
14 Mapping mind-brain development 191
PART III
A developmental theory of instruction 217
15 School and intellectual development 219
16 Enhancing intelligence in the laboratory 229
17 Towards a theory of instruction 247
18 Educating critical thought 273
19 Conclusions: towards an overarching theory of the

growing mind 281
References293
Index321
LIST OF ILLUSTRATIONS
Figures
1.1 Working memory (WM) mediates between ongoing perception
and past knowledge (long-term memory, LTM) on the one hand,
and behaviour on the other 9
1.2 Baddeley’s model of working memory. The shaded area represents
crystallized systems while the white area represents fluid systems 9
1.3 Wason’s selection task examining conditional reasoning 13
1.4 Baars’s theatre metaphor for conscious experience 18
2.1 The three stratum Cattell-Horn-Carroll model of intelligence 25
2.2 Examples of matrices of increasing complexity similar to those
included in Raven’s Standard Progressive Matrices 25
2.3 The van der Maas unified model of general intelligence assuming
mutual interactions between mental processes, centrality in the
power of effects of some processes compared to others, the effects of
external factors such as environmental factors that may multiply the
role of some mental process more than others, and sampling of
specific scores that are used in measurement of performance 29
3.1 The class inclusion task: what is there more of, the roses or
the flowers? 35
4.1 Development of processing speed as a function of age 45
4.2 Change in attention control from 4 to 17 years of age 46
4.3 Changes in STSS counting span as a function of counting speed 50
4.4 Reasoning attainment as a function of age and verbal working
memory level 54
4.5 The cascade model developmentally adjusted 55
6.1 Examples of theory-of-mind tasks involving appearance
transformations preserving the identity of stimuli involved 71
x List of illustrations
6.2 A model generating inferences about mental states without

assuming a ToM module 78
8.1 The four-fold model of the architecture of the mind 96
8.2 Confirmatory factor analysis model for reasoning domains,
their self-representation in cognizance, and processing efficiency
and capacity, and self-representation of the cognitive processes 98
8.3 The hierarchical model involving SCS-specific factors and a
general factor 105
8.4 Attainment of the four logical schemes as a function of age
and abstraction (concrete versus abstract) 112
8.5 Relations between reasoning and SCS reference factors and g 115
8.6 An idealized model of the structural relations between g and
each of the reference factors and between each of the reference
factors with attention control (AC), cognitive flexibility (Flex),
working memory (WM), cognizance (Cogn), and inference
(Infer) 117
8.7 The cascade model at four age phases i.e., at 4–6, 6–8, 8–11,
and 11–14 years, respectively 117
9.1 The general model of changes across cycles and phases 122
9.2 Alignment of reasoning development with a general IQ scale 131
9.3 Theoretical curves showing expected trends in mental age scores
for children with different IQs 132
10.1 Recycling in the relations between speed, working memory,
and reasoning 136
10.2 The models of the mediation of cognizance between executive
control and reasoning 141
11.1 Factorial structure across cycles 149
11.2 The general model for testing possible differentiation of mental
processes from general intelligence (g) 151
11.3 Idealized curves of the integrated integration-differentiation
logistic growth model 153
11.4 Idealized development of mental processes as a function
of developmental g (g x age) from 4–20 years of age 155
11.5 Association between SCS and general cognitive functions 159
11.6 Relation between cognitive style and ability on the three SCSs 160
12.1 The general model of the relations among general factor of
personality, fluid intelligence, and academic performance 169
13.1 (A) A model of the relations between genes, cognition, and
psychometric tests. It assumes that genes determine psychometric g,
which determines cognition and performance on tests. (B) The
watershed model showing how various parts of the brain influence
various aspects of processing speed and working memory, which
influence fluid intelligence 181
List of illustrations xi
14.1 Brain regions associated with mental processes presented in

Table 14.1 194
14.2 A view of the left cerebral hemisphere with the areas of cortex
numbered in accord with Brodmann’s cytoarchitectonic map 197
14.3 Representation of cognitive and brain networks according to
graph theory 205
14.4 Frontal lobe grey matter volumes according to age and SES 210
15.1 Model showing the relations between processing efficiency (speed),
cognitive constructs (intelligence and creativity), and school
performance based on several school subjects (e.g., language,
mathematics, science, etc.) 221
16.1 Stages in cognitive acceleration as implemented by Shayer 235
16.2 Distribution of cognitive developmental levels in the general
population and cognitive change as a result of intervention
(A) and transfer of training effects to school performance (B) 236
17.1 Educational priorities according to the four-fold architecture
of the mind 251
17.2 A schematic illustration of guided-alignment aiming to enable
preschool children to align maps with reality 261
Tables
4.1 Mental power demand of Piagetian stages according to
Pascual-Leone’s theory of constructive operators 48
5.1 The general characteristics of deductive reasoning at different phases 61
8.1 The three levels of organization of each specialized system
of thought 99
14.1 Locations of mental functions in the brain 195
16.1 Aims, instructions, and examples of interventions across sessions 240
16.2 Pre- and post-test percentage success and effect-sizes for
conditional reasoning study 242
PREFACE
This book is about the human mind. It presents research about the organization and
development processes involved in the human mind and about the reasons that
make some individuals better than others at using them. The book is organized into
three parts. In the first part we summarize and evaluate research conducted on the
human mind by three traditions of psychological research. The first tradition,
experimental cognitive psychology, identified mental processes such as attention,
working memory, and reasoning and specified how they work in real time. The
second tradition, differential psychology, viewed mental processes as dimensions of
individual differences and tried to identify how and why humans differ in using
these processes. The notion of general intelligence, IQ, and the tests that measure
intelligence are achievements of this tradition. The third tradition is developmental
psychology, which focused on development of the human mind and intelligence
throughout the lifespan, mapping mental possibilities in different phases of life and
specifying mechanisms of their change.
Obviously the individual mind is one and undivided. Humans use mental
processes, some better than others, and they change as they grow. Thus we will be
able to fully understand the human mind only when we have a common theory
specifying the mental processes involved, their organization and development,
and why individuals differ in how fast they develop and how far they go. This
theory is presented in the second part of the book. This part summarizes our research
integrating the cognitive, the differential, and the developmental theory of the mind
into an overarching model. This theory specifies the mental processes involved,
maps their development and interactions from birth to early adulthood, and connects
development and attainment with factors causing individual differences. This part
also integrates our theory with important lines of research focusing on the human
mind from several distinct but related perspectives: personality development, genetic
and cultural influences, and brain development. The aim is to show how mind and
xiv Preface
intelligence come at the crossroad of nature and nurture and flourish as the expression
of unique individuals.
The book’s third part is devoted to education. It summarizes research into
how education influences intelligence and intellectual development and how intel-
ligence sets the frame for education to exert its influence. It also presents experiments
designed to augment intelligence. Finally, the section uses this knowledge to advance
a complete programme for augmenting all aspects of intelligence and enhancing
critical thinking from birth to young adulthood.
This book is the magnum opus for the first author in that the research presented
spans his entire academic career. For the second author, hopefully, it is an important
step towards further development of the theory presented here. The two of us have
worked closely since the second author started his graduate studies at the Aristotle
University of Thessaloníki, Greece, more than two decades ago.
As students of the developing human mind, we are epistemologically minded.
Thus we owe an explanation as to how the historical-epistemological ideas presented
here grew. The theory presented in the book developed in three phases, the first
running from the late 1970s until the late 1980s and starting with the first author’s
PhD. In this phase, following the trends of the time concerning domain-specificity
and modularity of mental processes, the domains of thought underestimated
by Piaget were investigated and their development mapped. The second phase,
which saw the second author come on board, lasted from the early 1990s until the
beginning of the twentieth century and is delimited by two Monographs of
the Society for Research in Child Development, published in 1993 and 2002. The
aim of this phase was to bring back general mechanisms and specify how they relate
to specific domains. General mechanisms of the mind, such as attention control,
working memory, and self-awareness were central here. In a sense, this phase
predates this book: we tried to answer questions of developmental psychology using
theoretical constructs from cognitive psychology and methods from psychometric
psychology. In 2008 there was a small (fortunately) break in the first author’s
academic activity when he was appointed Minister of Education and Culture
of Cyprus. This lasted until 2011. The relations between developmental theory
and education started to form during these ministerial years, naturally so. The third
phase started in 2011 and continues. In it we redefined the relations between
domain-specific processes, general mental processes, and self-awareness. We also
started studying intelligence-personality relations, ran new experiments and used
new methods of modelling. The knowledge and ideas produced in this phase form
the backbone of the second and the third part of the book. Overall then, little is
carried from the first phase, much from the second phase, and all is embedded in
the concepts and methods of the third phase.
The study of the human mind is very rewarding. New methods and technologies
in genetics, brain science, and computer modelling allow questions to be raised
and answers generated that were unthinkable in the days of the giants of the field,
such as Charles Spearman, William James, Wilhelm Wund, or Jean Piaget. Thus it
is natural that the results presented here go beyond all of these giants. However,
Preface xv
some of their intuitions and theoretical concepts remain, underlying many of the
concepts presented here. We dedicate this book to their memory. We hope that
the book provides new answers to the questions they raised and prompts new
questions for the years to come. We would be happy if this book gave rise to
new research in all the domains covered, even if this research were to prove many
of our answers wrong or incomplete.
While the book is addressed to anyone interested in the human mind, there are
several audiences that may have a special reason to read it. First are the researchers
of human intelligence and cognitive development. As noted above, the book
presents a new theory of intelligence integrated with a theory of cognitive
development. The book is also addressed to people in education, both teachers and
policymakers. It may help teachers to better organize their activities in the classroom,
taking into account the understanding possibilities and difficulties of students
at different ages. The book is also addressed to scientists involved with the brain,
because it presents a theory of brain organization and growth matching mind
organization and growth. These scientists may find a frame for both interpreting
their findings and producing new ones. Finally, the book may be of interest to those
who study personality as it presents research into relations between intellectual and
personality development.
The research presented in this book was supported by many institutions and
individuals. Each of the phases was associated with a different university. The first
took place at the Aristotle University of Thessaloníki, our alma mater, the second
at the University of Cyprus, and the third at both the University of Cyprus and the
University of Nicosia. We deeply thank all three universities for generously and
hospitably providing the material support and the academic atmosphere for carrying
out the research presented here. We also gratefully thank research funding agencies
for generously financing our studies, primarily the Greek General Secretariat for
Research and Technology, the Cyprus Research Promotion Foundation, the
Johann Jacobs Foundation, and the Leventis Foundation.
We would also like to express our appreciation and thanks to colleagues who
cooperated in many of the studies presented in the book. Starting from the begin-
ning, they are: Anastasia Efklides, Maria Platsidou, Nicolaos Makris, Smaragda Kazi,
Eleftheria Gonida, Phillip Kargopoulos, and Tasos Giagkozoglou from the Aristotle
University of Thessaloníki; Antigoni Mouyi, Maria Andreou, Eleni Papageorgiou,
Anna Tourva, Rita Panaoura, and Marios Pittalis from the University of Cyprus;
Katerina Giorgalla and Valentina Zenonos from the University of Nicosia; Elena
Kazali from the Panteion University, Athens; and Demetris Tchmatzidis from
Democritus University, Thrace, Greece.
Last, but not least, we are grateful to our families who provided the supportive
framework that made our research possible. The first author needs to mention his
grandchildren—Andreas, Athina, Nicolas, and Aris, whose ages range from 2 months
to 8 years—who opened for him a new window for viewing the growing mind and
inspired some of the experiments presented in the book.
INTRODUCTION
Interest in the human mind dates back thousands of years. Philosophers such as Plato
and Aristotle, who tried to understand what knowledge and reasoning are, predate
modern cognitive science. Their answers were intended as theories of how humans
understand the world, how they build knowledge about it, and how reason fits
in with interactions with it. Later the work of cognitive philosophers—such as
Kant, Descartes, and Hume—set the frame in which modern cognitive psychology
established itself in the late nineteenth century.
Nowadays most of the theories advanced by these great thinkers are no longer
accepted. However, the questions and problems that motivated their inquiry about
the human mind are still alive, driving modern theorizing and research in cognitive
and developmental science. Plato’s theory about eternal ideas anticipated modern
theories about inherited abilities, core conceptual domains, and natural kinds. For
Aristotle, his logic was a theory of how humans understand relations between
objects; this is basically still valid as an interpretation of reasoning. Kant’s categories
of reason about quality, quantity, causality, space, and time are present in Piaget’s
theory and in every other modern cognitive scientific analysis of human thought.
We will refer to these ideas throughout this book when we try to highlight what
has stood up well to the test of time.
This book focuses on three aspects of the human mind: its nature and functioning
in the world; its changes through the lifespan; and differences in structure and function
between individuals. To understand the nature of the human mind, cognitive scientists
explore the origins of knowledge humans have about the world, such as perception
and learning, the mental processes they use to represent and store information about
the world, such as memory, and the processes they use to manipulate and extend their
knowledge about the world, such as inference. Moreover, mental processes change as
individuals grow older. In fact the human mind is a product of changes occurring
along two time-scales: the phylogenetic, which is long-lasting and operates through
2 Introduction
the evolution of our species, and the ontogenetic, which operates in the lifespan of
each human. Individuals may differ on both scales: on the phylogenetic scale, indi-
viduals may carry different genes, reflecting differences in ancestral influences; on
the ontogenetic, individuals have different developmental histories and different
environments. These differences between individuals account for their differences in
intelligence.
All humans have a mind but they do not use it in the same way. Differences
between individuals in using their mind is thought to reflect intelligence, loosely
defined as our capability to adapt in our world using our mind. The modern science
of intelligence tries to map and explain individual differences in intelligence and to
reveal why some people are more efficient than others in acquiring and using
knowledge and in facing challenges in their environment.
This book will summarize research on all three aspects of the human mind. In
the present context we use the term mind to refer to the general aspects of mental
functioning (e.g., everyone has perception, memory, and reasoning). The term
intelligence refers to the use of these processes by individuals in the real world; it
thus refers to possible differences between individuals in using mental processes for
the sake of dealing with the world. Our primary aim is to lay the ground for a
comprehensive theory that would account for all three aspects of the developing
human mind: its organization and functioning; its development through the lifespan;
and individual differences in its organization, functioning, and development.
Readers of this book will find answers to the following questions:
1. What are the cognitive processes underlying intelligence? That is, how is
information represented, connected and processed, understood, transformed
and used for the sake of learning, problem-solving, and decision making?
2. What is general and what is specific in intelligence? That is, what is common
across domains of knowledge, such as mathematics, literature, science—or skills
such as music, drawing, and dancing—and what is specific to each of them?
3. What is stable and what changes in intelligence as children grow older? Are
there general mental processes which are simply carried over the years or are
these transformed with development into different types of processes?
4. Why do individuals differ in intelligence? Are differences genetically determined
and thus inherited from parents and ancestors? What is the contribution of the
environment and culture to these differences? How do genetic and environmental
influences interact, if at all?
5. How is intelligence and intellectual development related to the genome and the
brain? Are there specific genes controlling intellectual functions and specific
patterns in brain organization and functioning which link to the different
representations and mental processes related to them? Are changes in these
patterns related to developmental changes in them?
6. How is intelligence related to personality? How do different profiles of
personality use intelligence to deal with the world and is intelligence reflected
in different aspects of personality?
Introduction 3
7. Can intelligence be enhanced by specific interventions? That is, is it possible to

improve the ability of developing individuals to represent and processes relations,
make better choices and decisions, and handle more complex problems? How
can specific interventions be implemented in the context of education, where
individual needs must be reconciled with the needs of many other individuals?
What is involved in intelligence?

The human mind emerges from the functioning of the human brain. The human
brain is an extremely complex system that has evolved to enable humans to know
the world and cope with its demands for the sake of survival and continuation. We
are connected with the world through our senses. Vision provides access to the
visible aspects of the world. The space we move in, the objects we see, their colour,
shape, size, and number, their positions relative to each other, and their interactions,
are all known through our eyes and the parts of the brain that are responsible for
representing and processing the information collected by the eyes. Hearing provides
access to sound. All sorts of sounds produced by the natural environment and other
living beings, including humans, can become known through the ears and the parts
of the brain that are responsible for recording and processing sound. The other
senses, such as smell, taste, and touch, provide other kinds of knowledge about the
world. Information coming from each channel has properties that are specific to it.
Through time the human brain has evolved structures and processes specializing
both in the processing of the various types of information provided by the senses
and in the integration of different kinds of information into holistic and systematic
representations enabling accurate understanding, appropriate decision making, and
efficient problem-solving (Jung & Haier, 2007).
The description above applies to all other organisms with a brain, to degrees
varying with the complexity of their brain. Humans, and probably some other
animals, may draw knowledge and solutions to problems from stored pools of
knowledge and ideas. These are cultural products that were made possible initially
because humans evolved to have a communication system that enables them to
share knowledge and ideas. This is language. Humans acquired language at least
100,000 years ago (Sampson et al., 2009); much later, about 3,200 years ago (Daniels
& Bright, 1996), they rendered knowledge and ideas virtually eternal by inventing
arbitrary symbol systems that stand for language, primarily writing. As a result,
knowledge and problem-solving become collective and, in principle, always
available and always revisable. The stored knowledge of different cultures reflects the
history of each of them. Education is the institution that was developed through
the millennia as a means for the induction of each individual into his or her own
culture (Mithen, 1996).
The human mind has been the focus of three research traditions in psychology
since the late nineteenth century: the cognitive tradition focused on cognitive
functions and processes (Sternberg, 2011); the differential tradition focused on
individual differences in intellectual ability (Hunt, 2011; Mackintosh, 2011); and
4 Introduction
the developmental tradition focused on cognitive development (Case, 1985; Piaget,

1970). Because of their varying priorities, the three traditions’ research questions,
methods, and theoretical constructs differed, especially in the early years. With time,
however, the three traditions interacted with and informed each other in theory and
methods. This interaction led to considerable convergences in how the human
mind is analysed and studied. According to all traditions, the human mind (i) learns
actively from interactions with the world, (ii) represents its interactions in mental
models which are used to (iii) guide further encounters, especially when they are
new and unexpected. The success in dealing with unexpected encounters is a
measure, in all traditions, of the overall functional efficiency of the mind, gauged
vis-à-vis the current situation (in the cognitive tradition), other individuals (the
differential tradition), or the age of the individual (the developmental tradition). All
traditions are also reductionist: they always try to reduce more complex processes
(such as intelligence or thinking) to simpler processes (such as speed or accuracy in
encoding information) to explain our capacity to learn or think (Demetriou et al.,
2010, 2011, 2012). The first part of this book summarizes the fundamental
assumptions of each tradition. The aim is to lay the ground for our theory, which
builds on the strengths of each tradition in order to provide an integrated description
of the organization and development of the human mind.
PART I
Three traditions of research
on the human mind
1
THE EXPERIMENTAL COGNITIVE
TRADITION
The experimental tradition focused primarily on the more dynamic on-line aspects
of mental functioning. The aim was to explain how humans: (i) perceive the world
and choose information that is relevant at a given moment; (ii) make sense of or
understand the information perceived; (iii) solve the problems encountered; and
(iv) store and organize their knowledge and experience about the world.
Information processing models have dominated cognitive psychology since the
early 1950s. According to these models, humans usually operate under conditions
of uncertainty caused by redundant, conflicting, or incongruent information relative
to a goal. It is a truism that always we see much more than what is needed at a given
moment. For instance, when looking for someone in an open space we often see
many other persons or objects around. To find the person we are looking for
requires searching for a particular face and bodily outline, quickly rejecting others
that may look alike. Also, to make a final choice, we often need to fill in gaps
of information, as when the person we are looking for is partially occluded
by somebody else. In this case we match what we see with what we remember by
drawing on memory or inference. Likewise, when in a conversation people often
need to make decisions about the meaning of what was heard or the intentions of
their partners. Often there may be conflicting information that must be interpreted,
such as when one has the feeling that the partner means something other than what
she or he says. To meet their goals, humans must be able to focus attention and
process goal-relevant information efficiently, filtering out goal-irrelevant information
within the constraints of the fast flow of a conversation and filling in gaps by
inference. In effect, controlled attention, speed of processing, working memory,
and inference are considered important in registering information, understanding,
learning, and problem-solving.
8 Three traditions of research on the mind
Mechanisms of attention and inhibition

Controlled attention includes processes enabling individuals to stay focused on
information of interest while filtering out irrelevant information (MacLeod, 1991;
Neill, Valdes, & Terry, 1995). For instance, attention drives the search for particular
characteristics when we try to find a person in the crowd. In laboratory situations
the Stroop phenomenon is the paradigmatic example of the conditions requiring
efficient handling of conflicting information. In the classic version of the task,
participants are presented with cards showing colour words printed either in black
ink or in an ink colour which is different from the colour denoted by the word itself
(for example, the word “red” printed in blue ink). The participants may be examined
under several conditions. Of primary interest is the condition of having to read the
colour words which are printed in black ink and the condition of having to name
the ink colour of words where meaning and ink colour differ (for instance, say blue
when the word is “red”). Under these conditions, word reading is much faster
(43.30 seconds for a card with 100 words) than naming the ink colour (110.3
seconds). The time difference between these two conditions has been ascribed to
the interference of the dominant aspect of the stimuli (i.e., the tendency to read a
word) with the processing of the weaker but goal-relevant aspect (i.e., the naming
of the ink colour). This difference is taken as a measure of inhibition, which is the
basic component of controlled attention (MacLeod, 1991; Stroop, 1935).
Speed of information processing was always important in theories of information
processing. This is because we usually operate under conditions whereby the
information available changes quickly. Thus it is important to be able to perceive
and recognize a particular stimulus before it disappears or is overtaken by other
stimuli. Speed of information processing basically refers to the time needed to
recognize a stimulus or execute a relevant mental operation. Usually, in tests of
speed of information processing, the individual is asked to recognize a simple
stimulus as quickly as possible, such as a letter, a geometrical figure, or a word in
one’s native language. Under these conditions, speed of processing indicates the
time needed to record and give meaning to information. Traditionally the faster an
individual can recognize a stimulus the more efficient this individual is considered
(Jensen, 1998, 2006; Posner & Raichle, 1997; MacLeod, 1991). A fast speed is
considered good for problem-solving because decisions are often made under
situations changing rapidly; an understanding of current information must be
completed before it is flooded by incoming information.
Mechanisms of representation and processing

It was noted above that understanding requires connecting current information with
information already possessed. This requires holding current information active for
as long as required to process relations with relevant knowledge from the past.
Working memory is the bridge between perception and our knowledge store, or
between concepts in the knowledge store (see Figure 1.1). Working memory
The experimental cognitive tradition 9
FIGURE 1.1 Working memory (WM) mediates between ongoing perception and past
knowledge (long-term memory, LTM) on the one hand, and behaviour on the other
enables a person to hold information in an active state while integrating it with other
information until the current problem is solved. A common measure of working
memory is the maximum amount of information and mental operations that the
mind can efficiently activate simultaneously. For instance, remember the second
last word of each of several sentences, remember the sum of several arithmetic
operations, remember where an object was located in a succession of scenes, etc.
There is extensive evidence that understanding, learning, and problem-solving
are positively related to the capacity of working memory. The assumption is that
enhanced working memory increases the connections that can be built between bits
of the newly encountered information or between this information and information
already stored in long-term memory. Thus enhanced working memory capacity
enables us to consider more options in understanding a concept, construct a new
concept, or invent solutions to problems.
Baddeley’s (1990, 2000, 2012) model, which received extensive empirical and
theoretical scrutiny, is widely regarded as a good approximation to the architecture
of working memory (see Figure 1.2). It posits that working memory consists of
FIGURE 1.2 Baddeley’s model of working memory. The shaded area represents
crystallized systems while the white area represents fluid systems
a central executive, two specialized storage systems, and an integrative episodic buffer.
The central executive is an attentional control system monitoring and coordinating
the operation of the two slave systems and coordinating information in working
memory with information in long-term memory.
The phonological loop involves a short-term phonological buffer and a subvocal
rehearsal loop. The first stores verbal information as encountered; information in
this buffer decays rapidly. The second counteracts this decay by refreshing memory
traces through rehearsal. The faster rehearsal is, the more the information that can
be held in the phonological loop. The visuo-spatial sketchpad is responsible for the
retention and manipulation of visual or spatial information. The two slave systems
draw on partially different resources. As a result, each is amenable to interference
from system-specific information that does not affect the other system. That is, the
phonological loop is affected by interference from verbal but not visuo-spatial
information; the visuo-spatial sketchpad is affected by visuo-spatial but not verbal
information (Shah & Miyake, 1996). However, these systems are interrelated and
information from one can be translated into the code of the other through rehearsal
guided by the central executive.
The episodic buffer is “a limited-capacity temporary storage system that is capable
of intergrading information from a variety of sources” (Baddeley, 2000, p. 421) into
unitary multi-dimensional representations using a multi-modal code. It integrates
information from the other working memory components and the long-term
memory into more complex structures, such as scenes or episodes. It serves as a
mediator between subsystems with different codes, such as words, visual images, or
number digits. The limited capacity of the central executive affects the integration and
maintenance of information within the episodic buffer. The process of retrieving
and binding information from multiple sources and modalities is primarily based on
conscious awareness.
Baddeley’s model allows for both specificity and generality in cognitive
functioning. Specificity is defined in terms of the modality in which information
is received (that is, acoustic or visual) and the ensuing symbol systems, which
handle information presented in these modalities (that is, language versus mental
imagery). Generality is ensured by the episodic buffer and the central executive.
The episodic buffer ensures the communication and production of integrated
mental products; the capacity of the central executive sets the general constraints
under which the two slave systems can function. Miller’s (1956) famous paper
suggested that the capacity of working memory of the normal human adult is
7 units of information plus or minus 2. Later the capacity of working memory was
considered to be lower, between 3–5 units, with the difference between the two
figures related to capacity needed for the operation of executive processes (Cowan,
2010). We will see below that these aspects of working memory influence
intelligence: differences between individuals in working memory are related to
differences in their ability to integrate information and reason with it. They also
relate to cognitive development, as the capacity of working memory increases
with age.
Mechanisms of integration: association, inference,

and reasoning
There are several mechanisms of information integration. Association is a more or
less automatic mechanism that associates stimuli or responses on the basis of their
physical proximity in time or space or their sharing of common characteristics.
Various types of learning, such as Pavlovian classical conditioning and Skinnerian
operant conditioning, are based on association. In Pavlovian classical conditioning
a particular physical stimulus, a bell ringing, takes the properties of another stimulus,
food, and then causes the response naturally evoked by this other stimulus, such as
salivation at the sight of food, because it occurred just before food appeared. In
Skinnerian operant conditioning, a stimulus is associated with a response if the latter
appears soon after the former, as when a particular response is rewarded and learned
or punished and avoided.
Integration by association may be the basis for inference, because it provides
the raw materials for it. Additionally, inference is a more self-directed form of
association where links between stimuli are established on the basis of rules
encoding what is wrong and right based on past experience. Generally speaking,
inference comprises processes that enable thinkers to transfer meaning from one
representation to another. This transfer normally occurs on the basis of properties
which are present in both the initial (or base) representation and the target
representation.
Reasoning is thinking that involves inference. In reasoning, common properties
are used as an intermediary between two representations; such that properties character-
izing the base representation (apart from common properties) are also ascribed to the
target representation. For instance, a four-year-old child thoughtfully concludes: “If
it doesn’t break when I drop it, it’s a rock. . . . It didn’t break. It must be a rock”
(DeLoache, Miller, & Pierroutsakos, 1998). The two things to be connected here are
the notion of the rock and the present object, their common property being that they
did not break when dropped. By virtue of this common property, the property of
“rockyness” and other derivative properties (such as rocks are heavy, hard, etc.) may
also be transferred to the present object.
There are several types of reasoning, the two most inclusive types being inductive
and deductive. Inductive reasoning is a freer kind of reasoning which can involve any
kind of representation, such as perceptions, mental images, and propositions of one’s
language. Moreover, in inductive reasoning inference goes from the particular to
the general or from the particular to the particular and the conclusion is not
necessary but only probable. An example is: “The dogs I met bark; therefore all
dogs bark.” Obviously this conclusion is only probable because in the future we
may meet dogs who do not bark. Analogical reasoning is inductive reasoning
applied on relationships rather than similarity between the objects themselves. The
classic structure of problems which require analogical reasoning is represented by
the following formula: a : b :: c : d. For instance, Athens is for Greece what London
is for the UK (Holland, Holyoak, Nisbett, & Thagard, 1989).
Deductive reasoning is a more constrained kind of reasoning. First, it only

involves verbal statements or propositions, or other symbols which stand for
propositions. Thus inference in deductive reasoning is the process which transfers
meaning from one set of propositions (the premises) to other propositions (the
conclusion). Second, in deductive reasoning the inferential process always proceeds
from the general (the premises) to the specific (the conclusion), and the conclusion
follows necessarily from the premises. This occurs because in deductive reasoning
the premises must be accepted as given, thus the conclusion is mandatory. An
example is given below:
Dogs bark
Max is a dog
________________________
Therefore, Max barks.
Obviously, once both premises of this argument are accepted as true the conclusion
is necessarily true, no matter what kind of animal Max is or what we know about
him. This is more clearly apparent in an argument where the premises are not
consistent with reality:
Dogs fly
Max is a dog
________________________
Max flies
Obviously, in this example we have to accept that Max flies, given the premises,
even if we know that dogs do not fly.
Both inductive and deductive reasoning implicate a number of different
varieties, each of which comprises several inferential processes. For example,
inductive reasoning involves statistical reasoning, which focuses on probabilities
and analogical reasoning, which focuses on relational similarity. Deductive
reasoning involves categorical reasoning and conditional reasoning. Categorical
reasoning is based on class relations. For instance, if a specific property characterizes
a class (e.g., dogs bark) then this property necessarily goes to the members of the
class (all sorts of dogs). Conditional reasoning examines the relationships between
“if . . . then” types of propositions. Conditional reasoning is important for intelli-
gent functioning because it allows integration and evaluation of information
(Johnson-Laird & Khemlani, 2014).
Conditional reasoning is grounded on four logical schemes, slowly mastered
throughout childhood and adolescence (Markovits & Vachon, 1990; Moshman,
2011; Müller, Overton, & Reene, 2001): modus ponens (MP), modus tollens
(MT), affirming the consequent (AC) and denying the antecedent (DA). Two of
the schemes, MP and MT, are decidable and rather easy to grasp because all
information needed for a conclusion is present in the premises. In MP, if one
accepts that “if A then B” and “A occurs”, one must also accept that “B necessarily
occurs”. In MT, if B did not occur it necessarily follows that A did not occur. The
remaining two, AC and DA, are not decidable because the conclusion depends on
information not given in the premises. Specifically, in AC, if B occurs it does not
follow that A would also occur, because a third, non-specified factor may be
involved. In DA, it does not follow that B would not occur if A does not occur,
because a third factor may cause B. Thus these two schemes are called “logical
fallacies” because they may deceive the thinker into drawing a conclusion that is
not tenable. We will see in the chapters following that MP and MT are attained
early in development, at 7–9 years, by practically everyone. The two fallacies are
not mastered before the age of 11–12 years and then no more than about one third
of adults can handle them systematically (Gauffroy & Barrouillet, 2009; Johnson-
Laird & Wason, 1970; Markovits, 2014; Moshman, 2011; Overton, 1990; Ricco,
2010; Wason & Evans, 1975).
Wason’s selection task is a famous problem demonstrating these processes. This
task involves four cards marked with a letter on one side and a number on the other.
For instance, the four cards have an A, D, 3, and 7, respectively, as shown in Figure
1.3. The participants were told that the “following rule applies to the four cards and
may be true or false: If there is an A on the one side of the card, then there is a 3 on the
other side of the card.” They were then asked to indicate which cards must be turned
over to decide if the rule is true or false. The correct answer is A and 7. A is relevant
because it is stated in the rule. If there is anything but three on the other side the rule
would be proved false. Obviously, this is a test of the MP argument. D is irrelevant
because the rule does not refer to cards with letters other than A. The card with
3 appears relevant but it is not because the rule specifies what must follow if A occurs
and does not state what must follow if 3 occurs. Thus even if an A does not appear
on the other side of the “3” card the rule is not contradicted, because the rule does
not state what should be marked on the back side of 3. This is the affirmation of the
consequent (i.e., the AC fallacy), as explained above. Finally, the card with 7 is
FIGURE 1.3 Wason’s selection task examining conditional reasoning

relevant because it might have an A on the other side. This is not permitted by the
rule, because if there is an A there should be a 3. Thus turning the 7 card is relevant
because it can falsify the rule (this captures MT in its relation with MP). Most adult
subjects involved in the original experiment selected the A and the “3” cards,
indicating that they operated as though the “if A then 3” rule were equivalent to the
“if 3 then A” rule, which of course is not the case for the reasons already explained.
There have been several theories about the nature and operation of reasoning. At
the one extreme, several theories argue that reasoning operates on the basis of rules
which somehow mimic the rules of logic (Rips, 1994, 2001). For instance, in the
examples above, MP obeys a rule stating that “‘if A à B’ is true, any A is necessarily
B”. Thus, in this case, once it is accepted as true that “dogs fly” it is necessarily true
that any dog flies. According to the rules theory of reasoning, there is a rule
underlying each of the four conditional reasoning arguments outlined above.
At the other extreme, the mental model theory is dominant. Mental models are
representations of situations with a strong iconic component somehow depicting
the represented situation to the thinker. For instance, in the premise “All dogs
bark”, the mental model would depict a dog barking. This theory claims that people
build mental models of the elements and relations involved in arguments or episodes
and proceed to manipulate and combine these models to draw conclusions or make
decisions. For instance, in the argument above about barking dogs, the model of
the “dog Max” who also “barks” is associated with the general model of “barking
dogs” and a conclusion is derived that is consistent. According to Johnson-Laird,
who advanced the mental model theory of reasoning, reasoning is based on the
construction of mental models based on the premises and general knowledge: in
MP, a model of Max barking. Each model represents what is true in a possibility
that may be derived from a premise. A conclusion is taken as valid if it holds for all
models of the premises. There is research showing that reasoners also construct
mental models of counter-examples that would falsify the premises. Not finding any
counter-examples strengthens the belief in the validity of the conclusion. There is
evidence that, the fewer the models needed by an argument to make the conclu-
sion, the easier the argument. For instance, the MP argument would need fewer
models than the AC argument; the AC argument would also need models about
possibilities other than those stated in the basic premises. The conclusion selected is
the one where no counter-example is found. For instance, no model of a non-
barking dog may be conceived (Johnson-Laird, 1983; Johnson-Laird & Khemlani,
2014). Errors occur when reasoners do not consider all possible models. For this
reason, working memory is related to performance on these tasks, the Wason’s task
included: more complex arguments require considering more models, which thus
require more working memory (Barrouillet & Lecas, 1999).
These two theories of reasoning are not incompatible. It may be the case that
thinkers shift between rules and models depending on the situation or their
developmental phase. For instance, individuals may, at later developmental levels,
construct rules to encode their experience of reasoning with mental models and
their relations We will show in the chapters following that these two approaches
may be integrated so that both rules and mental models may be used, depending on
the developmental phase or available expertise (see Barrouillet, Grosset, & Lecas,
2000; Oberauer, 2006).
Language of Thought: inference and modularity

In cognitive science there is a long and unresolved debate about the architecture of
the human mind. In its current stage, the debate focuses on the relative importance
of a central Language of Thought as the means of understanding the world versus
the importance of modules specializing in the understanding of specific aspects
of the world.
Language of Thought (LoT)

LoT is thought to involve fundamental elements or units of meaning and a set of
rules enabling the combination of these elements to form into streams of meaning
about the world. Thus cognitive science studies the nature, origin, and possible
development of these elements and rules (Schneider & Katz, 2011). In its classic
version (Fodor, 1975, 2008), the basic units of LoT are atomic symbols that stand
for representations bearing meaning for the thinker (e.g., “cat”, “dog”, “animal”,
“life” may all be valid symbols, though their meanings may vary across thinkers).
These symbols may be combined by a combinatorial syntax that yields infinitely
compounded representations, whose meaning is defined by the symbols involved
and the rules of syntax used (e.g., “If Max is a dog, he barks”). Preservation of truth
is a basic property of LoT in that the transformation of true premises (symbols)
always results in further true premises: any set of combined representations can be
translated into any other set once initial truths are carried across sets. For many, the
rules of this syntax are the rules of logical reasoning, whatever they are (Johnson-
Laird & Khemlani, 2014; Rips, 1994).
Where do the rules of thought come from? In cognitive and developmental
science answers vary extensively. At the one extreme, the Whorfian (Whorf, 1956)
hypothesis posits the complete determination of thought by language: LoT is the
rules of syntax in language. At the other extreme, the complete independence
hypothesis posits that language is a conduit for the communication of thoughts
without any other important effect on it (Hurlburt, 1993): LoT involves its own
rules and the rules of syntax are a reflection of the rules of LoT. In developmental
psychology, Vygotsky’s and Piaget’s theories may be taken as exemplars of the
polarity about the origins of the rules of thought. At the one extreme, Vygotsky
(1986) claimed that social scaffolding and language shape thought. Specifically, inner
speech (silent language addressed to oneself) expresses and controls thought and
operates as the medium for the interiorization of the formative influences of culture.
At the other extreme, Piaget (1970) claimed that the rules of thought emerge
gradually from the coordination of mental operations. This coordination becomes
increasingly abstract, obeying logical rules able to express the relations that can be
handled at different developmental phases. Language and other representational

functions, such as perception and imagery, are subservient to this coordination.
Therefore the state and complexity of language reflect the state and complexity of
the current mental structure. We will summarize Piaget’s theory in Chapter 3.
Interestingly, the psychometric answer is very similar to the Piagetian answer:
language is thought to express knowledge already possessed rather than shape the
mechanisms of building up and using the knowledge base (Carroll, 1993).
Recently, Carruthers (2002, 2009) took an intermediate position. He postulated
that language is a general-purpose tool for bridging domain-specific information and
representations into general streams of thought, as we see in the various forms of
inference. Specifically he suggested that the capacity of thought to integrate
“different content-bearing items into a single thought” (p. 668) is patterned on the
fundamental properties of syntax: that is, recursivity (e.g., the dog chased the fox
which chased the cat which chased the rabbit), compositionality (reciting words one
after the other to covey information), generativity (words can be combined in a large
number of ways to convey the same meaning), and hierarchical organization (e.g.,
words and sentences may be embedded into different levels to convey the complexity
of realities described).
Modularity
In many theories a general LoT exists together with several modules. A module is a
system of processes serving a particular biological or psychological function more or
less independently of other modules. In their strictest expression, as introduced by
Fodor (1975, 2008), modules are automatic and informationally encapsulated. That
is they operate by especially dedicated neural networks and so they are activated by
information that is relevant to them and nothing else. Once activated they yield an
interpretation (representation) of the world that is the product of the particular type
of information involved and the dedicated network activated. Perception is an
example. For instance, colour perception is what it is: different wavelengths of light
cause the perception of different colours, such as green, red, etc., and there is nothing
that can change this other than affecting the neural basis of visual perception in the
eye, the brain, or both. This may extend to higher levels, as in the perception of
small numbers. Subitization, the perception of numerocity of small sets up to 3–4
objects lying close to each other, is automatic and comes very early in life, if not at
birth. When the input related to a module is present, the module-specific behaviours
are released to meet the needs of the moment. The assumption about modularity
posits that the human mind is an aggregate of many modules operating independently
of each other. That is, modules are domain specific in that they involve information
processing mechanisms that specialize in the processing of information specific to a
domain and they are not interchangeable across domains.
Modularity was highly influenced by Chomsky’s (1986) view of language as a
modular system. In his view, the fact that humans are the only species to acquire
language within a particular time window (at about 2 years of age) is strong evidence
for the operation of a module that frames when and how language is acquired and
used. The concept of modularity attracted considerable interest in psychology as an
explanation of often impressive variation of cognitive achievements within and
across individuals. For instance, idiot savants may be severely impaired mentally but
demonstrate advanced abilities in particular domains, such as being able to calculate
days of the week far in the past or future. Other individuals show an unusual ability
to retain information in working memory (Luria, 1968), despite serious mental
problems. Proponents of modularity argue that this bizarre pattern of accomplishments
suggests that modules do not need any common processes to operate, as assumed by
those arguing that there is a LoT. Modularity came in several versions, varying from
the theory of massive modularity (Carruthers, 2006), assuming that most cognitive
processes are modular, to approaches assuming that differences between processes
are softer and are caused by differences in learning. The issue of modularity will arise
many times throughout this book.
Awareness and consciousness

Are humans aware of mental processes? Considerations about consciousness and
self-awareness have been central in the study of the human mind. These
considerations faded in the years of behaviourism, from about 1920 to 1950, when
internal processes were not accepted as scientifically legitimate objects of study.
However, the study of consciousness and awareness has become a major focus of
research after the cognitive revolution of the 1950s. The questions of focus here are:
does the mind know itself or about itself?; does it hold representations for its own
mental states?; can it intentionally change its own functioning?; is self-knowledge
changing with age or experience?; and how does self-knowledge and consciousness
interact with inferential and processing functions? In fact, nowadays, it is a major
theme of research in brain science. The main objective is to understand how
brain functioning generates consciousness and how this, in turn, may influence brain
functioning (Dehaene, 2014; Koch, 2012).
Kant, in the mid-eighteenth century, believed that intelligence exists only by
virtue of awareness of the very ability to combine representations for the sake of
understanding and judgment: “I exist as an intelligence, which is conscious of its power
of combination” (Kant, 1902, B 158–59, see Kitcher, 1999, p. 375, emphasis added).
That is, intelligence exists only in a cognizer who is cognizant of himself or herself.
This cognizer is aware of the judgmental process generating objective judgments by
combining representations which refer to real objects, entities in the world or
mental constructions, which are causally connected. A cognizer of this process is
self-conscious with a sense of self.
For cognitive science, consciousness is an integral component of the human
mind. We noted above that, to meet their goals, humans must be able to focus
attention and process goal-relevant information efficiently. It is now believed that
consciousness evolved to enable humans to monitor, register, regulate, organize, and
reorganize their representational and inferential processes to tune them with current
making-meaning and problem-solving needs. The input to consciousness is in-

formation arising from the functioning of all other systems; the output is choices
among stimuli one is currently aware of and actions to be taken. Therefore
consciousness is a kind of spotlight that allows a small part of mental content to enter
awareness and allow systematic manipulation. Obviously, both attention and working
memory are integral components of consciousness (Dehaene & Naccache, 2001). A
general representation of consciousness is shown in Figure 1.4 (Baars, 1997).
Needless to say, not everything we attend to reaches awareness or is fully
registered. Often cognitive processes and their products, such as new knowledge or
changes in concepts already possessed, never reach awareness explicitly. However,
FIGURE 1.4 Baars’s theatre metaphor for conscious experience (adapted from Baars,
1997)
consciousness is always in operation, receiving signals from cognitive functioning

running subconsciously. For instance, individuals saw, in an experiment, a series of
characters organized according to a rule, without knowing that this was the case.
They then judged which sequences among a new set of letter sequences obeyed the
rule and specified their confidence in their judgment. Interestingly, the level of
awareness of the rule, varying from complete absence to full awareness that there
was a rule, was reflected in both the metacognitive judgments of confidence and
the speed of responding. The stronger the awareness that a rule was present the
faster the recognition of exemplars of the rule (Mealor & Dienes, 2013).
Therefore the consciousness system is sensitive to processes going on below the
threshold of awareness; this system generates a kind of blind insight reflected in
the operation of attention and attention inhibition mechanisms. Blind insight
emerges from interactions among predictive processes which yield top-down
predictions about the causes of information coming to the senses (e.g., a particular
colour or face is expected) and bottom-up projections signalling mismatches between
expected and observed stimuli (e.g., this was not the colour or the face I expected
to see). These interactions are registered and metarepresented allowing systematic
self-directed improvement of action in situations where attention is needed (Scott,
Dienes, Barrett, Bor, & Seth, 2014). Along the same lines, Tran and Pashler (2017)
showed recently that implicit learning of skills involving fast action involves
awareness. An example would be acquiring top performance in tennis by predicting
where an opponent would send the ball based on hidden signs, such as the opponent’s
gaze or head position. Only individuals who explicitly verbalize the rules linking
cues to locations learned such predictive rules. Several of our experiments, to be
presented in the following chapters, capitalize on this phenomenon.
Clearly awareness is activated when novel information is encountered or when
choices or decisions need to be made about the meaning of information or about
actions to be taken. Awareness is also activated when mental effort is needed to
construct a first representation of the problem, spot and filter out irrelevant
information, and activate past knowledge in order to evaluate current information
that is not relevant. Therefore the central executive in Baddeley’s model is by
definition the main spot of awareness. Also, mental models in reasoning and their
evaluation are, by definition, objects of awareness. Importantly, Anderson and
Fincham (2014) found recently that a metacognitive module may be located in the
brain which “creates, modifies, and rehearses declarative representations of cognitive
procedures” (p. 25). Specifically they trained 12–14 year olds and young adults
(19–35 years) to solve arithmetic problems based on a specific rule and related
feedback that would allow participants to abstract and use the rule for solving
new problems. There were four stages in problem-solving: encoding, planning,
solving, and responding. The metacognitive module can hold these steps, modify
them, create new declarative steps, and rehearse them. We will return to these
issues later, when we will present research about the relations between awareness
of these processes and intelligence and for improvements in their operation with
development.
Conclusions
We noted in the introduction that all traditions studying the human mind strive to
answer a number of questions about its organization, functioning, and development.
Of the various questions stated in the introduction, the cognitive experimental
tradition answers the first two of the queries related to the processes involved and
their relative generality. It also answers the question related to the brain bases of
mental processes. The answer to the question with respect to mental processes is
nowadays widely accepted: the human mind is represented as an information
processing system that is designed for efficiency. This system involves a suite of
mental processes carrying different tasks in the representation and integration of in-
formation for the sake of understanding and problem-solving. The efficiency of any
system that processes information is a function of its capability to complete a task
within predisposed representational limits (time, load, complexity, etc.). As an
information processing system, the mind usually operates under intense time and
representational constraints. Information (light waves, sound waves, etc.) reach the
senses in fast succession. At any point in time, any picture, sound or other sensory
input must be interpreted before it goes away. Registering and encoding information
for processing is also limited by construction. When eyes look at this they cannot
look at that. When ears hear this they cannot hear that, etc. This very reality requires
representation that would protract information until it is integrated across time or
senses. Thus processing is defined in reference to four parameters: speed, attention,
synchronization, and retention.
Speed is important because fast processing augments the amount of information
that may be attended to in a standard time unit. Attention is important because it
ensures focusing on and flexibly shifting between relevant stimuli, ignoring irrelevant
ones; the more focus there is on what is important or relevant for a given time unit
the better, because resources are not wasted. Synchronization refers to fine tuning
of information encoding within and across senses. It is important because it allows
the construction of proper mental representations of objects or episodes by properly
“packaging” slightly varying incoming information: for instance, synchronization
of information from the eyes—matching the images coming from them (i.e.,
binocular vision)—allows perception of depth; synchronization of information
coming from the two ears—by computing the minimal differences in the time at
which the sound reaches each of the ears (i.e., interaural timing difference)—allows
location of sound in space; changing perspectives of the same person under the same
visual image, or across senses—for example, “coupling” visual images and sounds
under the right face or person—ensures proper recording. Retention refers to
protraction of information in time after the stimulus has gone so that it may be
combined with current incoming information; it is important because limited or
inaccurate retention would trap the person in a constant flow of irrelevance.
Retention is the basis of working memory.
A complementary model to that of efficient information processing is needed
where the focus of analysis shifts from recognition, representation, and retention of
information, to integration and meaning-making. At this level, truth and validity

may be more important than efficiency. In other words, a logicality model is needed.
It has been suggested that there are different mechanisms to serve truth and validity:
compositionality, recursivity, generativity, and hierarchical organization are four fundamental
mechanisms for integrating information according to standards and criteria for truth
and validity. The rules governing inferential processes in inductive and deductive
reasoning implement these general mechanisms according to the type of represe-
ntations and the type of relations between representations involved. For instance,
general inferential schemes in conditional reasoning—such as MP, MT, AC, and
DA—are truth-making rules that build on the general mechanisms above and derive
truth and validity from a given series of representations and relations between
representations (or their symbols). The type of information involved—such as
visual, verbal, or acoustic information—may cause differences in the operation of
integrative processes: for example, visual representations are more transparent than
verbal representations in showing some logical relations at some points in time.
Often efficiency here comes in the skill or disposition to deal with one type of
information more easily than with another.
On top of the logicality model, the model of the decision-maker is needed.
There is, of course, a vast literature on decision making spanning several disciplines,
including mathematics, economics, and psychology (Evans, 2003; Kahneman, 2011;
Osman, 2004; Stanovich & West, 2000), that is beyond the concerns of the present
book. Suffice it to note here that, by definition, decisions involve choices. Choices,
also by definition, involve awareness. Obviously there is much in mental processing
that never reaches awareness, such as depth perception or sound localization.
However, when we come to choosing between interpretations or conclusions in
reasoning, or correcting mistakes, some awareness is needed that may vary between
tasks or phases in learning or development. On top of the decision-maker is a
human subject representing himself or herself. Thus a model of subjectivity is
needed. We will suggest ways of letting these models integrate in the chapters
following. We now turn to the differential tradition that focused on individual
differences.
2
THE DIFFERENTIAL TRADITION
The differential, or psychometric, tradition focused on individual differences in

mental abilities. Abilities are underlying latent dimensions or traits running through
several behaviours, causing their interrelations. For instance, scores on different
tests—such as mathematics, language, and science in school—tend to be related:
high scores in one school subject go with high scores in another subject. It is
assumed that there is something common running through all related scores which
channels the ability of individuals to learn and perform in these subjects. Scores in
all three school subjects above reflect reasoning and memory processes employed
for learning and problem-solving in all three domains. The emphasis in the
differential tradition is on how efficiently each process is used by different individuals
in understanding, learning, and problem-solving—rather than in the operations or
elements they involve.
Research in the psychometric tradition wavered between a single factor and a
multi-factor conception of intelligence. At the one extreme, Spearman’s (1904,
1927) two-factor theory posits that performance on any test involves only two
factors, general intelligence (or g), that is common to all tests, and specific variation,
that is skill or noise specific to this particular test. Only the first factor is of interest.
Jensen (1998) and Gottfredson (2016) are nowadays strong proponents of the notion
of general intelligence.
Technically speaking, g reflects the so-called positive manifold, the fact that all
cognitive tests correlate positively with each other. Mathematically, g stands for
a common factor related to all cognitive tasks. For Spearman, g primarily stands for a
powerful inferential competence underlying the eduction of relations and correlates.
Eduction of relations is inductive thought abstracting relations between objects or
events based on their similarities or their functional relations. For instance, Max is
a dog and barks; Rex is a dog and barks; dogs bark; or 2, 4, 6 . . . (the next number
is 8). Eduction of correlates is higher-order inductive thought abstracting relations
The differential tradition 23
between relations. For instance, understanding that “Athens is for Greece what
London is for the UK” involves understanding the relation “the capital of”, which
is recognized to be the same for both cities relative to their countries. Obviously
eduction of relations and correlates draws on the inferential mechanisms involved
in inductive and analogical reasoning discussed above. It underlies the construction
of networks of concepts about the world and grasping meaning in all domains, such
as metaphor in language, humour in social relations, mathematics and causal relations
in science, etc. Spearman also suggested that individual differences in the use of this
mechanism relate to biological processes reflecting available brain energy in
representing and processing information.
A modern differentiation of Spearman’s theory is the theory proposed by Cattell
and Horn (1978). According to this theory, intelligence involves two general factors—
fluid intelligence, or Gf, and crystallized intelligence, or Gc. Gf is basically relational
reasoning and is more or less identical to Spearman’s g. Gc is basically knowledge
about the world that is constructed by investing Gf in learning from and storing
knowledge about experience that may be used later, as needed.
At the other extreme there have been several theories arguing for specialized
abilities. Thurstone’s (1935) model of primary ability was an early version of a
modular theory of intelligence emphasizing the importance of specific abilities over
general intelligence. Thurstone specified seven primary abilities: verbal
comprehension, word fluency, number facility, spatial visualization, associative
memory, perceptual speed, and reasoning. Gardner’s (1983) theory of multiple
intelligences is the modern version of a modularized conception of human
intelligence. Some of Gardner’s intelligences are very similar to Thurstone’s primary
abilities: visual-spatial (i.e., spatial judgment and the ability to visualize with the
mind’s eye); verbal-linguistic (i.e., facility with words and languages, reading,
writing, telling stories and memorizing words); and logical-mathematical (i.e., logic,
abstractions, reasoning, numbers, understanding causal relations, and critical
thinking). Gardner added several other intelligences to this list, which originated
from research and practice in cognitive psychology and neuroscience. They are as
follows: musical intelligence (i.e., sensitivity to sounds and music—rhythm, pitch,
meter, tone, melody or timbre, singing, playing musical instruments and composing
music); bodily-kinaesthetic intelligence (i.e., control of bodily motions and skilful
handling of objects; sense of timing and goal of a physical action); interpersonal
intelligence (i.e., sensitivity to the feelings, emotions, and motivations of others, and
ability to cooperate with them); intrapersonal intelligence (i.e., ability for
introspection, self-reflection, self-knowledge, and self-control); naturalistic
intelligence (i.e., understanding natural surroundings, classifying natural forms, e.g.,
animal, plant species, rocks); and existential intelligence (i.e., spiritual and religious
understanding).
However, these theories did not support the empirical test well. Thurstone
(1938) himself admitted that his model was based on data collected from high-ability
students. It is well known that, when variation is restricted in a sample, especially
at the high end, g is concealed under specialized factors, reflecting the fact that these
individuals acquire a high level of performance in the domains of their interest but
do not operate as well in other domains. In the same vein, empirical research
showed that most of Gardner’s “intelligences” are highly dependent on g; namely
the verbal, the spatial, the logico-mathematical, the interpersonal, and the naturalistic
intelligence. Some do not correlate highly with g (i.e., bodily and musical
intelligence), but these are not generally regarded as part of what is traditionally
considered intelligence (Visser, Ashton, & Vernon, 2006; Waterhouse, 2006).
In this tradition nowadays there is growing agreement that intelligence is a three-
level hierarchical system. In current psychometric theory the dominant model of
the architecture of the human mind (see McGrew, 2009) is a model integrating the
Cattell and Horn (1978) Gf-Gc theory, already summarized, with Carroll’s (1993)
three-stratum model. This common model is often referred to as the Cattell-Horn-
Carroll model (CHC) model of intelligence. This model postulates that the human
mind is organized in three hierarchical levels.
The first level involves many specific abilities or skills, such as problem-solving
skills in various domains, information and knowledge about the world, verbal and
language skills, skills related to storing and recalling information from memory, etc.
These are organized, at the second level, into eight broad abilities. Each of these
eight abilities is identified by a few underlying mental processes shared by all first-
level specific abilities. For instance, inductive and deductive reasoning underlie fluid
intelligence (Gf); organization of knowledge and information and its applications in
culture-relevant conditions underlie crystallized knowledge (Gc); retention and
storage of information underlie general memory and learning ability (Gy);
visualization, spatial relations, closure speed and spatial scanning underlie broad
visual perception (Gv); speech, sound discrimination, general sound discrimination,
and musical ability underlie broad auditory perception (Gu); creativity, ideational
fluency and naming facility underlie broad retrieval ability (Gr); the rate of test
taking, numerical facility, and perceptual speed underlie broad cognitive speediness
(Gs); and simple reaction time, choice reaction time, semantic processing speed,
and mental comparison speed underlie processing speed (Gt). These in turn are
constrained by general intelligence, or g, at the third level. “g” is closely reflected
in measures of intelligence, captured by various intelligence tests, such as IQ tests
(Carroll, 1993; Cattell & Horn, 1978; Jensen, 1998; Gustafsson & Undheim, 1996).
Figure 2.1 illustrates this hierarchical structure of the human mind.
The current definitions of g emphasize relational thought and are still very
close to Spearman’s eduction mechanism. Many empirical studies operationalized
Spearman’s eduction competence in reference to tests of various forms of inductive
and analogical reasoning, such as Raven’s Progressive Matrices, which in turn was
identified with Gf. Raven himself stated that the abilities that “Raven sought to
measure directly were those identified by Spearman in 1923 [Spearman, 1927].
These are, respectively: (a) eductive ability (from the Latin educere, meaning ‘to
draw out’), the ability to make meaning out of confusion, the ability to generate
high-level, usually non-verbal, schemata which make it easy to handle complexity;
and (b) reproductive ability—the ability to absorb, recall, and reproduce information
FIGURE 2.1 The three stratum Cattell-Horn-Carroll model of intelligence
that has been made explicit and communicated from one person to another.”
(Raven, 2000, p. 2). Examples from Raven’s Standard Progressive Matrices are
shown in Figure 2.2. Matrices in different sets are clearly different in complexity:
simpler matrices in set A require to abstract one single rule based on a clearly visible
perceptual pattern. Matrices in sets B and C require to abstract two or three
dimensions and specify their inter-relations. Finally, matrices in the most difficult
sets D and E require abstracting multiple rules on the basis of underlying implied
but not explicitly specified principles that interlink dimensions on the basis of
conceived rather than perceived dimensions. It is noted that these patterns are
induced from standard performance of different age populations on SPM as a whole
(see Raven, 2000).
It is noteworthy that there have been several other tests focusing on a supposedly
broader definition of general intelligence. Naglieri’s test of non-verbal ability
(NNAT; Naglieri, 1997; Naglieri & Ronning, 2000) was designed to implement
Naglieri’s PASS theory (Naglieri & Kaufman, 2001). This theory defines intelligence
as a system involving planning (specification, implementation, and evaluation of
FIGURE 2.2 Examples of matrices of increasing complexity similar to those included in

Raven’s Standard Progressive Matrices
problem-solving goals), attention (stimulus selection and inhibition), simultaneous

(integration of stimuli based on their relations), and successive cognitive processes
(encoding of relations into rules that may be extrapolated into new situations).
Obviously PASS is heavily based on fluid intelligence together with executive
processes, to be discussed below. Thus the NNAT test includes items requiring
spatial visualization, pattern completion, serial reasoning, and reasoning by analogy.
Naturally this test correlates highly with Raven’s Standard Progressive Matrices
(.62). There have been several other tests supposed to address relational thought
(Dumas & Alexander, 2016) which also related highly with Raven’s APM (.49)
(Alexander et al., 2015).
Research bridging the experimental and the psychometric

tradition
Empirical evidence provided strong support to the CHC model. Specifically, the
three-level hierarchical structure is clearly valid (McGrew, 2009). All of the second
stratum abilities described by Carroll (1993) do emerge as factors of performance
organizing more specific abilities. At the third stratum, g always emerges as a
powerful construct highly related with the second stratum broad abilities. This
research showed that g is practically identical with Gf, because their relation in
structural models is equal to 1 (e.g., Gustafsson, 1984; Keith et al., 2006). Therefore
g is representational and inferential power par excellence. Research on the predictive
power of g strongly suggests that it “is a worldwide phenomenon; is highly heritable;
provides the common spine for all cognitive tests, complex or elementary, seemingly
different or not; and has pervasive correlates throughout the body, brain and
behavior” (Gottfredson, 2016, p. 120). Many life outcomes, including educational
and occupational attainment, workplace success, criminality, involvement in traffic
accidents, having extra-marital children, etc., are related to IQ. Based on these
findings, the psychometric tradition produced valid and reliable tests that can be
used to measure the relative standing of each individual on each of the dimensions.
Tests of intelligence, such as IQ or fluid reasoning tests such as Raven’s SPM, are
good examples. Overall, general IQ or performance on the Raven test is considered
a good measure of g (Herrnstein & Murray, 1994; Jensen, 1998).
Attention is drawn to the fact that the abilities specified by the psychometric
tradition overlap extensively with the processes studied by the cognitive tradition:
attention, working and long-term memory, reasoning, and specialized processes
involved in language and visualization are all present in the CHC model. The
difference between the two traditions lies in how these mental processes are
approached. In the cognitive tradition, research and theory tries to map the
composition of each process and its real-time functioning. In the psychometric
tradition, these processes are dimensions of individual differences. That is, they are
stable dimensions of ability that can be used to compare individuals and specify their
relative standing; in other words, how much they “possess” of each ability and how
well they can use each ability in learning, problem-solving, and decision making.
Science is reductionist. It seeks to reduce complex phenomena to simpler

mechanisms or principles, helping to make prediction and control of the
phenomenon of interest easier. In the present case, our understanding and control
of intelligence and learning would be very efficient if, for instance, differences
between individuals in g or IQ could be reduced to a simpler process—such as the
speed needed to recognize a simple stimulus, focus attention on a relevant stimulus,
inhibiting the processing of irrelevant stimuli, or working memory. If this were the
case, manipulating this simpler process would cause changes in the more complex
phenomenon of interest—intelligence. In this context there has been extensive
research which investigated the relations between the processes specified in the
experimental tradition as important components of mental functioning underlying
the various abilities uncovered by psychometric research.
Jensen (1998, 2006) stressed the importance of information-processing speed: the
minimum time required to identify a simple stimulus, (e.g., a colour), choose one
of two stimuli according to a goal (e.g., press the right or left button when a red
light turns green or red, respectively) or execute a mental act (e.g., reach a conclusion
based on a chain of inference). In fact this approach is implicitly present in Carroll’s
(1993) theory itself, as he associated many of his Stratum-2 factors with speed of
executing the main processes involved. According to Jensen, processing speed is an
index of the quality of information processing in the brain. Jensen (1998) reported
a correlation of about .5 between g and information-processing speed. Based on 172
studies conducted over a period of 50 years, Sheppard and Vernon (2008) found
that the correlations between information-processing speed and g, although always
present and systematic, were moderate, ranging circa .3. Moreover, it was found
that correlations tend to increase with increases in the complexity of reaction time
tasks: for instance, reaction times to tasks which require choosing between stimuli
to respond to increase with the number of stimuli one has to choose from.
Correlations also increase when tasks require some kind of attention control and
inhibition, as in the Stroop paradigm described above. Under these conditions,
correlations may be as high as .5. Obviously intelligence and simple information
processing and control processes are related. However, this relation is much weaker
than would be expected if intelligence were identified with information-processing
processes.
Other research suggested that working memory capacity (WMC) is a key
component of g. Obviously the major component shared between working memory
and reasoning is the representational efficiency in representing and mentally
manipulating information in order to consider alternative assumptions about
relations between information, induce relations, and evaluate conclusions by
reasoning. Starting with a well-cited study conducted by Kyllonen and Christal
(1990), many studies showed that g or each of its component competences, such as
inductive and deductive reasoning, are highly related with working memory.
Indeed, relations between g and working memory capacity are higher than relations
with information-processing speed, but not as high as would be expected if g and
working memory capacity were one and the same thing. Based on meta-analysis of
a large number of studies, Ackerman Beier, & Boyle (2005) found that the average
true correlation between working memory and g is .48. However, it might be the
case that the relations between working memory capacity and g are moderated by
information-processing speed. Ackerman also found that the speed-WMC relations
are higher than with g (.57). Thus a combination of these processes might raise the
variance in g that may be accounted for. Indeed, Chuderski (2013) showed that
the relation between working memory capacity and g varies as a function of the
demands of the problem-solving situations: the more one performs under conditions
of fast decision making the higher this relation, varying from .62 (performance on
Raven and analogy tests without time constraints) to 1 (performance on the same
tests under high-speed conditions). What might be common between working
memory capacity, Gf, and choice RT tasks? “Working memory capacity is needed
to establish and maintain bindings between stimulus and response representations.
Such bindings are particularly important when the mapping between stimuli and
responses is arbitrary, so that the cognitive system cannot use preexisting associations
to translate a stimulus into its corresponding response” (Wilhem & Oberauer, 2006,
p. 43; also Oberauer, Farrell, Jarrold, & Lewandowsky, 2016).
Scholars took these patterns as an index of another crucial component of g:
executive control (Blair, 2006). This is a central attention system underlying the
ability to focus processing on goal and to flexibly deploy a plan for attaining it in
spite of possible interference. Thus executive control is considered to involve
attention focusing, inhibition, flexibility in shifting, and monitoring and updating
of information to attain the current mental goal (Garon, Bryson, & Smith, 2008;
Miyake et al., 2000). However, the relations between each of these functions and
g or IQ, although significant and systematic, are also moderate, varying circa .3
(Arffa, 2007). Therefore, executive control, like information-processing speed and
WMC, did not emerge as a privileged representative of g.
In response to this state of affairs, several scholars stripped g of any distinct
psychological process: in the words of Kovacs and Conway, “There is no psychological
process that corresponds to psychometric g” (2016, p. 171). Rather, g is an algebraic
consequence of the interaction between specific processes. This interactionist
approach comes in two versions. According to van der Maas et al. (2006), the positive
manifold underlying g is not caused by any of the processes above, per se. Rather it
emerges purely by their interactions during development. That is, the correlations
between processes underlying g reflect their interactions as they are jointly brought
to bear on problems rather than any single process alone. In its current form the
model postulates that intelligence may be specified by four types of forces: (1) mutual
interactions between mental processes, such as various aspects of Gf, including
reasoning and working memory, and various aspects of Gc, such as knowledge in
various domains; (2) the centrality of a particular process in the system at a given
moment, such as attention control or working memory at a given age; (3) the effects
of external factors, such as environmental factors that may multiply the role of some
mental process more than others, including specific learning designed to boost a
specific process (i.e., working memory); and (4) sampling of specific scores that are
used in the measurement of performance, such as the intelligence test used which
samples different aspects of g compared to another test. This model is illustrated in
Figure 2.3.
This is an interesting approach, which will be discussed in the next chapter as it
has both structural and developmental implications. For instance, it suggests that the
role of g may vary with age or domain of activity involved (Demetriou et al., 2002,
2008; Gignac, 2014; van der Maas et al., 2017).
Alternatively, Kovacs and Conway (2016) suggested that g emerges as a result of
process overlap of many processes sharing the same process—executive control—for
the sake of their own functioning. This does not necessarily reflect actual common
elements between different processes but the state of the process called on, which
acts like a bottleneck masking “individual differences in specific abilities. Even if
someone were, in theory, capable of successful performance on the domain-specific
aspect of a mental test item, he or she might be unable to arrive at a correct answer
FIGURE 2.3 The van der Maas unified model of general intelligence assuming mutual
interactions between mental processes (e.g., Xf variables standing for Gf measures and
Xc measures standing for Gc measures), centrality in the power of effects of some
processes compared to others (e.g., working memory, i.e., Xf1), the effects of external
factors such as environmental factors that may multiply the role of some mental
process more than others (e.g., Ec, a specific learning environment designed to boost a
specific process, such as working memory) and sampling of specific scores that are
used in measurement of performance (e.g., this or that intelligence test). The ci and fi
test results for crystalized and fluid cognitive abilities, respectively, yield IQ. The
g-factor extracted when using factor analysis on f (and c) tests reflects these interactions
(van der Maas et al., 2017)
because of failing to meet its executive attention demands” (p.163). It is noted,

however, that placing executive control in the centre of process overlap would
imply that the relation between g and executive control would be much higher than
found by the studies summarized above.
Conclusions
This tradition focused on individual differences. Naturally, research in the tradition
provided constructive answers to those questions posed at the beginning of this
book that are concerned with the architecture of the mind and the individual
differences concerning each of the various processes involved. Specifically, research
in this tradition clearly suggests that all specific processes share common processes
with other specific abilities at several levels and are constrained by them. “g” is
everywhere and domain-specific broad mental processes delimit types of mental
processes and knowledge. Some scholars identify g with one particular process more
than another, such as executive control, or working memory, or inference. Other
scholars define g as a dynamic system where relations and coordination between
processes count more than any central privileged processes, even if privileged
processes do exist. All processes, general, broad, and specific, as well as their
interaction, may be independent sources of differences between individuals.
Interestingly, by and large the processes discussed by this tradition are the same pro-
cesses discussed by the cognitive tradition. In a sense the differences between the two
traditions are like the two aspects of the same coin. The cognitive tradition studies effi-
ciency in the operation of these processes and the psychometric tradition studies
differences between individuals in efficiency. When differences in efficiency are
systematically expressed along a dimension or scale of efficiency that is used to rank
order individuals, mental processes are transformed into abilities. Thus the unification
of the two traditions is possible. However, this tradition, as much as the cognitive
tradition, underestimated development. Even when researchers in this tradition speak
about mental age, they just refer to patterns of successes and failures at different ages
rather than specify what is the state and functioning of mental processes at different
phases of life. Thus integration would be wanting, given that development may not
be left out of the grant theory to come.
We should also note that there is an important construct completely missing from
theories in this tradition. Contrary to the cognitive tradition, consciousness and
awareness were never systematically investigated as a component of intelligence in
psychometric research. It is instructive to note that the terms “awareness” and
“consciousness” do not even appear in the subject index of three important books
which still influence research on human intelligence, namely the books by Carroll
(1993), Hunt (2011), and Jensen (1998). It is also notable that not one of the major
tests of intelligence includes any items directly addressed to any aspect of consciousness
and awareness. We will turn to this development in several chapters.
3
PIAGET’S THEORY
Developmental psychology, by definition, adopts a developmental perspective to deal

with several questions: what is the state of understanding, reasoning, problem-solving,
and knowledge acquisition at successive phases of life; what does it change in the state
of each of them; how do changes occur; and why do they occur? James Baldwin
(1894) was the first to advance a complete theory of cognitive development aiming
to account for the development of thought. Basic premises of this theory were taken
over by Jean Piaget and were integrated into his theory of cognitive development.
However, in the developmental tradition, Piaget is considered the major figure and
founding father of the discipline of cognitive development in the study of intelli-
gence. This is mainly due to the solid empirical basis of Piaget’s theory (Piaget, 1970;
see Cahan, 1984). In this chapter we will outline and evaluate this theory, while
in the next two chapters we will focus on neo-Piagetian and post-Piagetian research
and theory.
Piaget’s theory
Piaget was a biologist with epistemological interests who was trained in Binet’s
psychometric laboratory in Paris, and all three scientific disciplines converge in his
theory. As a biologist he viewed intelligence as a biological adaptation; as an
epistemologist he set himself the task to advance a theory about the origins, nature,
and adaptive functions of knowledge and intelligence. He defined intelligence as
“the state of equilibrium towards which tend all the successive adaptations of a
sensorimotor and cognitive nature, as well as all assimilatory and accommodatory
interactions between the organism and the environment” (Piaget, 1968, p. 11).
Thus he focused on concepts that were important in the philosophy of knowledge:
reasoning and logic and the basic categories of reason, such as quality, quantity,
causality, space, and time were the objects of his research for decades. The reader
may see here that Kant’s categories of reason (see Kitcher, 2011) were major objects
in Piaget’s research programme. The individual clinical examination of the child,
with systematic probing of the reasons underlying answers to tasks, both correct and
wrong, were inspired by psychometric testing.
Piaget was aware of research in the cognitive and the differential tradition of his
time but he explicitly stated that he was interested in the “epistemic subject” rather
than with individual differences: that is, the general mechanisms of thought that are
so general as to apply to all individuals rather than the factors causing variations
within an individual, related to cognitive mechanisms, or the factors causing
variations between individuals, related to individual differences. In the chapters
following we will examine if we can have an epistemic subject separate from a
psychological subject.
First, we outline Piaget’s model of the nature of intelligence. Then we summarize
the stages that he proposed as descriptions of the forms of intelligence from birth to
maturity and summarize his model of cognitive change. Finally, we attempt an
evaluation of the theory (Piaget, 1952, 1968). Our aim is to highlight the aspects of
the theory that influenced later research and stood up to the test of time, so requiring
them to be integrated into a modern comprehensive theory of intellectual
development. Readers may consult some of the many excellent books on Piaget’s
theory (e.g., Flavell, 1963; Ginsburg & Opper, 1988), and shorter summaries
(Demetriou, 1998), if they want to study this theory in more detail.
The nature and the adaptive functions of intelligence

For Piaget, reality is changing continuously. Any object or person, at any time, is
in a particular state caused by some kind of transformation. Transformations are all
kinds of change that objects may undergo—move in place, reshape or reform, get
older with age, etc. States are the forms in which things or persons can be found
between transformations: for example, the same object in different places, in
different colours, in different arrangements of its parts, so that it appears flatter,
longer and thinner than before, etc. To be adaptive, human intelligence must be
able to represent both the transformational and the static aspects of reality and
understand how they interrelate. Thus Piaget proposed that intelligence involves
two aspects, an operative and a figurative aspect (Piaget & Inhelder, 1974).
Operative intelligence is the active aspect of intelligence. It involves all actions,
overt or covert, undertaken in order to follow, recover, or anticipate the trans-
formations of the objects or persons of interest. Figurative intelligence is the more
or less static aspect of intelligence, involving all means of representation used to
retain in mind the states (i.e., successive forms, shapes, or locations) that intervene
between transformations. That is, it involves perception, imitation, mental imagery,
drawing, and language. Therefore the figurative aspects of intelligence derive their
meaning from the operative aspects of intelligence, because states cannot exist
independently of the transformations that interconnect them. Piaget believed that
the figurative or the representational aspects of intelligence are subservient to its
Piaget’s theory 33
operative and dynamic aspects, therefore understanding derives from the operative
aspect of intelligence: “intelligence consists of structuring and relating” (Piaget,
1968, p. 40). In other words, operative intelligence structures and relates figurative
representations.
At any time, operative intelligence frames how the world is understood and it
changes if understanding is not successful. Piaget believed that this process of
understanding and change involves two functions: assimilation and accommodation.
Assimilation involves active transformation of information so as to be integrated into
the mental schemes already available. Its analog at the biological level might be the
transformation of food by chewing and digestion to fit in with the structural and
bio-chemical characteristics of the human body. Accommodation involves the
active transformation of these schemes so as to take into account the particularities
of the objects, persons, or events the thinker is interacting with. Its analog at the
biological level might be the adaptation of eating and digestion to the particulars of
the different kinds of food we eat.
For Piaget, none of these functions can exist without the other. To assimilate an
object into an existing mental scheme one first needs to take into account or
accommodate to the particularities of this object; for instance, to recognize
(assimilate) an apple as an apple one needs first to focus (accommodate) on the
contour of this object. To do this one needs to roughly recognize the size of
the object. Development increases the balance or equilibration between these two
functions. When in balance with each other they generate mental schemes of the
operative intelligence. When assimilation dominates, understanding may not be
accurate and aspects of reality are ignored in favour of the individual’s personal
viewpoint. When accommodation dominates, understanding may be incomplete
although there may be an accurate representation of a situation or an object.
Following from this conception, Piaget theorized that intelligence is active and
constructive. In fact it is active even in the literary sense of the term as it depends
on the actions, overt or covert, assimilatory or accommodatory, which the thinker
executes in order to build and rebuild his models of the world. And it is constructive
because actions, particularly mental actions, are coordinated into more inclusive and
cohesive systems and thus they are raised to ever-more stable and effective levels
of functioning. Piaget believed that this process of construction leads to systems of
mental operations better able to resist the illusions of perceptual appearances and
thus less prone to error. In other words, the gradual construction of the system of
mental operations involved in the operative aspect of intelligence enables the
developing person to grasp ever-more hidden and complex aspects of the world.
Below we will summarize the development of operative intelligence.
The stages of operative development

Cognitive development evolves along several distinct but interrelated dimensions.
The three major dimensions are representation, reasoning, and perspective.
Representation moves from externally based representations, such as the perceptions
of objects and the actions on them, to internally based representations standing for
perceptions and actions, such as mental images, language, and abstract symbols such
as numbers. Reasoning moves from often faulty inference based on privileged or
dominant appearances of reality to increasingly valid and true inference based on
the integration of mental operations into reversible structures where successive states
are logically derived from each other by recovering the transformations connecting
them. According to the definition of intelligence given above, intelligence becomes
increasingly error-free because conditions of nature may be anticipated by reasoning;
when conditions are not expected, intelligence can fully restore their origins and
mentally explain them. Perspective widens and becomes increasingly flexible, taking
into account alternative points in addition to one’s own point of view. For instance,
in reasoning on tasks such as those described in earlier chapters, one may consider
alternative conclusions to ensure that the conclusion best following from the
premises is drawn and that no counter-example can be found. In social interaction
one can decentre from his or her own point of view and see a problem from the
point of view of another person.
Representational development is divided into two major periods: sensorimotor
intelligence, which lasts from birth to the end of the second year, and representational
or symbolic intelligence, which appears thereafter and develops through childhood and
adolescence. Representational intelligence is also divided into two key periods, with
the first, from 2–7 years, preoperational. That is, it is based on mental actions on
representations which are not yet coordinated into reversible operative structures.
For Piaget (1952) the foremost criterion of operational coordination is reversibility,
which involves understanding that a given action (actual or mental) can be cancelled
or undone by another action. At the age of 7 mental operations are integrated into
reversible structures, transforming preoperational intelligence into operational
intelligence. Operational intelligence is concrete until approximately age 11–12
years, when it becomes formal. Formal intelligence develops until the end of
adolescence.
Sensorimotor intelligence
Presentation here will focus on Piaget’s views concerning the development of object
permanence. Object permanence is the belief that objects exist on their own,
regardless of the infant’s perceptions or actions on them. According to Piaget, at the
beginning of the sensorimotor period, when the senses are still not well coordinated
with each other or with action, infants believe that objects cease to exist when they
are not seen, heard, or touched. By integrating perceptions and actions on objects
and persons, infants credit more stability to the existence of objects at the end of the
first year.
However, objects still do not exist independently of the infant’s actions. This is
suggested by the so-called A-not-B error. Imagine a baby rolling a ball under sofa
A. At this age the infant can recover the ball because she believes that there is a ball
under the sofa. If a little later, however, the ball rolls under sofa B in front of her
eyes, the baby will again look for the ball under sofa A, where she has successfully
recovered it earlier. This behaviour indicates that objects are not completely
independent of the infant’s actions. This is achieved by the end of the next phase,
at about age 15–16 months, when objects can be recovered from where they were
last seen to disappear. In fact, a few months later infants reconstruct hidden changes
of locations in the object’s movement. This indicates that actions have been
interiorized and integrated into plans of mental actions that can be executed
independently of their behavioural execution. Thus a new and long journey begins:
the journey towards the development of representational intelligence.
Representational intelligence I: from preoperational to concrete

thought
Internalizing sensorimotor schemes generates mental schemes, which are blueprints
of actions that can be executed mentally. Although mental schemes are present at the
early stages of representational intelligence, they are not coordinated into reversible
structures. As a result, concepts at this phase are dominated by impressive or familiar
aspects of objects or experiences rather than a balanced interpretation integrating
various aspects or points of view. Also children’s reasoning is prone to errors since
things are judged by appearances. These weaknesses are illustrated below with the
examples of some of Piaget’s best-known tasks.
Piaget (1970) believed that operative intelligence is based on grasping the logic
of classes and relations. The class inclusion task examined the development of the
logic of classes (Inhelder & Piaget, 1958). In this problem (see Figure 3.1) children
are shown objects belonging to two complementary classes, for instance four roses
(class A) and three daisies (class A’), both belonging to a higher-order class including
both of them, flowers (class B). After describing everything present, a child is asked
the class inclusion question: “Are there more flowers (i.e., Bs) or more roses (i.e.,
As) on the table?” That is, the child is asked to compare the superordinate class B
FIGURE 3.1 The class inclusion task: what is there more of, the roses or the flowers?
with one of the subordinate complementary classes, which in the standard task is
always the most numerous class.
Children in the preoperational stage say that “there are more roses because the
daisies are only 3 and the roses are 4”, suggesting that the superordinate class
“flowers” does not exist in their minds. Piaget ascribed this weakness to a lack of
reversibility in their mental operations. This would enable them to construct the
superordinate class B by focusing on the properties common to the specific classes
and then recover the subclasses by focusing on their specific properties. It is only
when composition and inversion can be applied simultaneously that the child will
be able to move between the subordinate and the superordinate classes and grasp
their quantitative relation of inclusion.
Children at the intuitive stage generally give one of two answers: they either give
the right answer (flowers are more than roses) but cannot explain why, or they say
that “they are the same”. These answers indicate an intuitive grasp of the super-
ordinate class. However, mental actions operations are not yet integrated into a fully
reversible structure of mental operations. As a result children are unable to move
up and down the class hierarchy with a clear understanding of both the general and
the particular classes. These difficulties are removed when children enter the stage
of concrete operations. At this stage they can give both the right answer and
explanations indicating that the structure of concrete operations has been established:
“There are more flowers than roses because daisies are flowers, too” (i.e., flowers
= roses + daisies); or “there are more flowers because roses are not the only flowers
on the table” (i.e., flowers - roses = daisies).
Piaget (Inhelder & Piaget, 1969) maintained that understanding transitivity of
relationships is equivalent to understanding class inclusion. In the simpler version
of the transitivity task, three sticks are involved such that, for instance, Rod A is
longer than Rod B and Rod B is longer than Rod C. Children are first shown
Rods A and B together and then Rods B and C together; Rods A and C are never
shown together. The children are then asked to infer their relation. As in the case
of the class inclusion problem, preoperational children cannot solve this problem;
intuitive children can find the right answer but they cannot explain; finally,
concrete-stage children answer and explain correctly, indicating that they can
integrate the two premises by means of reversible mental operations—once B is
longer than A and C is longer than B, C must be longer than A. According to
Piaget, this understanding indicates another form of reversibility. Specifically, the
inverse of an operation here does not cancel out the effects of the previous operation;
it is only its reciprocal. Each object in a series of increasing size is both bigger than
those coming before and smaller than those following. Being both bigger and
smaller than other objects does not affect the identity of an object, but only the
perspective from which it is seen.
Piaget believed that the understanding of every domain of reality results from
a kind of synthesis of the logic of classes, reflected in the class inclusion task, and
the logic of relations, reflected in the seriation task. For instance, in the classic
conservation of number task the experimenter shows a child a row of five coins
arranged about two centimetres apart. He then invites the child to make another
row having the same number. Children in the preoperational stage make a series
whose ends coincide with the model series, with little attention to the number of
coins. Children in the intuitive stage are able to replicate the model series, indicating
a global grasp of number concept. Thus when one of the rows is elongated so that
the distance between coins is longer, intuitive children believe that “the longer row
has more coins because it is longer”. Children in the concrete stage are not deceived
by transformations. They say that “the number is still the same, because nothing has
been added or taken away”, “this row seems to have more because it is longer but
there is also more empty space between the coins” (which is an argument for
reversibility as reciprocity), and “you can see that they are still the same if you bring
them back to their original position” (which is an argument for reversibility as
inversion).
Testing on the other conservations (quantity, substance, size, area, length,
weight, volume) produced very similar results: children first see two equal and
similar examples of the concept in question (e.g., two similar glasses containing the
same amount of water), the one is then transformed in some way without change
in the property in question (e.g., the water from one of the glasses is transferred to
another glass, which is taller and thinner), and the child is asked if the quantity is
still the same. Answers are as above: not the same for preoperational children, the
same without explanation for intuitive children, and the same for concrete
operational children (“It still is the same because nothing has been added or taken
away and it may be longer but it is also thinner and if you pour it back it will again
appear the same”).
Representational intelligence II: from concrete to formal thought

Piaget argued that the foremost structural attainment of formal thought is the
integration of the two forms of reversibility into a single system (Inhelder & Piaget,
1958). He considered this as the basis for the transition from the stage of concrete
to the stage of formal thought. Therefore the structure and the workings of this
system will be described first, followed by discussion of the more common
characteristics of this structure.
While the attainment of the structure of concrete operations was sufficient to
ensure the actual or mental manipulation of observable properties of reality,
separation of the two forms of reversibility at this stage does not permit manipulations
that go beyond or contra reality. An example is the isolation of variables underlying
hypothesis testing in scientific reasoning. For instance, children are shown several
rods of different length, cross-section, and material, and they are asked to propose
a pair of rods that would constitute a fair test of the hypothesis “Long rods bend
more than short rods”. Obviously any pair that would involve two rods similar in
every respect but length is a fair test of this hypothesis, because their possible
difference in flexibility can only be ascribed to their length. Piaget believed that this
type of reasoning can only be achieved if inversion and reciprocity are integrated
such that the thinker can understand their equivalence. In the hypothesis above, the
ideally fair test would be one which would involve rods composed of nothing but
length. Imagining such a rod is tantamount to negating or nullifying, in thought, all
those properties of rods that need to be controlled in the experiment. In reality, of
course, there are no such rods and so negation in this case is not possible. However,
any effects of the factors that need to be controlled can be cancelled out if they are
made the same, thereby affecting flexibility to the same degree. Thus any difference
between the rods in flexibility can be ascribed to their difference in length. The
reader is reminded that cancelling out or compensation is the basis of reciprocity.
According to Piaget, the understanding that the one form of reversibility can be
used in place of another signifies the construction of a structure which integrates
both of them into a single ensemble.
All characteristics of formal thought derive from this structure. The first and
more general of these characteristics is the ability to conceive of the possible. It was
explained earlier that the integration of the two forms of reversibility leads to the
mental construction of states that mimic “reality that does not exist”; that is, rods
which have no other property but length. This is obviously a possibility that can
only be constructed mentally by combining mental operations transcending
actualities.
A number of more specific abilities derive from the ability to conceive of the
possible. Combinatorial thought is a tool for specifying possibilities. It uses a
systematic method to generate all possible combinations between properties and
even to combine the combinations themselves. For instance, imagine all possible
combinations in which four differently coloured balls (red, green, blue, and yellow)
may be drawn from a box. A second ability is propositional thinking: possibilities
or combinations are expressed into verbal propositions which are then combined
into propositional arguments where emphasis is placed on the logical relationships
between the propositions rather than on their content. For instance, full conditional
reasoning is possible at this stage because all four logical arguments mentioned in
Chapter 1 (MP, MT, AC, and DA) are integrated into a common structure where
a single rule specifies how relations between propositions into each argument can
be transformed to each other. Thus, third, thought becomes hypothetico-deductive;
that is, it conceives all possibilities at the beginning and then specifies their
implications by logically manipulating the propositions in which the possibilities are
cast. Finally, formal thought has the ability to understand complex dynamic systems
whose functioning is governed by multiple interactive forces, such as mechanical
equilibrium in real or actual balances or complex mathematical relationships, such
as proportional relationships.
To conclude, then, Piaget believed that each stage endows the developing mind
with some kind of conceptual stability in understanding the world. The organization
of sensorimotor actions into a coherent structure and the subsequent internaliza-
tion of this structure enable the infant to attain ontological stability: understanding
that things have an existence of their own beyond the senses. The structure of con-
crete operations enables the school child to attain conceptual stability: understanding
that the basic properties of things, such as identity, size, etc., remain invariant despite
misleading perceptual cues. Finally, the structure of formal operations enables the
adolescent to attain ideational stability: understanding that conclusions are necessary
if they derive from a valid logical argument, even if they can never be verified
empirically. Thus formal thought is a perfectly equilibrated system of thought
towards which all development tends since birth. Once attained, it provides stability
in the relations between the mind and the environment because anything can be
reconstructed mentally, freeing the mind from any possible disturbance.
The mechanisms of change in Piaget’s theory

Piaget’s theory is often taken as a stage theory. However, it is more a theory of
cognitive change than a stage theory. To Piaget, stages were nothing more than
transient states of the advancement of an ever-changing system towards equilibration,
a condition enabling the person to mentally cope with any change in the environment
(e.g., Piaget, 1976, 1977, 2001). Why does cognitive development occur? Piaget
argued that change is necessary when the present cognitive structure yields an
incomplete or faulty understanding which generates inconsistencies and conflicts
between what the individual understands or does and the present situation. All
examples of preoperational answers above cause conflicts between the child’s
understanding and reality.
How are cognitive conflicts resolved? Piaget proposed that “reflective abstraction”
is the mechanism that removes conflicts by generating new structures integrating
conflicting views. In his later work he identified reflective abstraction as a two-phase
process. At the beginning of a stage (or dealing with a new problem) it functions as
reflecting abstraction: it takes successive actions and their results and amalgamates
them into a new system which is free of the conflicts and inconsistencies tantalizing
the previous system. Let’s take the example of a child who believes that longer,
in the number conservation task, implies more, but she realizes that the actual count-
ing of different arrangements of the same set of objects results in the same number
name; obviously this contradicts her belief that longer is more. Reflecting abstraction
is the process which compares successive counting actions, notices their common
element (for instance, that all three rounds of counting yielded 5), takes it out of
content (for instance, in the first round the coins were arranged in a long row, next
in a circle, and finally in a pile, and no coins were added or taken away), and reduces
it (Piaget would say projects or reflects it) to a new concept or scheme: “Aha! It is
always 5, it is always the same!” In conclusion, reflective abstraction is the mental
mechanism that amalgamates a series of actions and their results into a new structure
that removes conflicts and inconsistencies generated earlier. Once constructed, at a
subsequent phase, the new concept can itself become the object of reflection. For
example, the concept of number can be related to the concept of length to structure
one’s representation of space in a more accurate way. This is reflective abstraction.
In this sense, reflective abstraction is equivalent to reflection: thinking about
thinking itself.
The current status of Piaget’s theory

Piaget’s theory attracted massive interest in the 1960s and the 1970s; Piaget’s name
in Google Scholar gives 183,000 hits (June, 2017). As a result there is a solid body
of empirical evidence on most aspects of Piaget’s theory. This evidence enables us
to draw certain conclusions regarding the validity of this theory, the permanent
contributions that it has made to our understanding of human thought, and about
its potential further contributions.
The studies which evaluated Piaget’s theory can be classified into two broad
categories. One group tested the validity of the phenomena Piaget discovered. For
instance, is it true that children cannot solve the various conservation tasks before
the age of 7? This research can be summarized quite easily in a single statement:
whenever a replication study remained close to the original Piagetian conditions
the same phenomena were observed as described by Piaget. In the sensorimotor
stage, children at about the age of 8 months commit the A-not-B error; in the next
phase children fail the class inclusion, transitivity, and conservation tasks until about
the age of 7; children become able to grasp these concepts during their primary
school years; however, they do not take an “if-then” stance to problems and they
cannot reason propositionally, design experiments by systematically controlling
variables, or understand proportionality before adolescence even when they are
highly intelligent (according to intelligence test scores). Thus it is safe to conclude
that Piaget was right in both the phenomena he discovered and their approximate age of
occurrence (see the studies summarized in Brainerd, 1978; Dasen, 1977).
But is the theory right? The answer to this question is neither simple nor
straightforward. In fact the question itself must be formulated more specifically for
a fair and accurate evaluation of the theory to be made. Here only two issues will
be raised, one for mental processes presumably reflected by responses to Piaget’s
tasks (e.g., reversibility) and another for the organization of these processes into
the structures he proposed. The question on processes asks if performances on
Piaget’s tasks really reflect the processes Piaget claimed that they reflect. For
instance, does a child’s failure to solve the transitivity task really indicate that this
child cannot reason logically? Bryant and Trabasso’s (1971) study on transitivity
was one of the first and most influential studies which undermined Piaget’s position
that his tasks tap only logical reasoning. Remember that in this task children are
shown the A and B and the B and C terms together and are then asked to infer the
relationship between the A and C terms. Piaget asserted that failure to make
the correct inference implied a lack of the logical structure (to see Rod B as both
taller than A and shorter than C) to produce the inference. What would happen,
however, if children do not remember the A and B and B and C comparisons or
if they do not represent them correctly? Obviously they would fail the task. In this
case, however, their failure would not indicate a lack of logical reasoning but
a memory lapse or representational inefficiency. This is precisely what Bryant
and Trabasso found: preschool children either forget the comparisons or they
misremember them.
Hundreds of such studies demonstrated that when extra-logical factors such as

memory, language and communication, familiarity with tasks and testing procedures,
and interest are controlled, supposedly preoperational children can (or can be trained
to) classify and grasp class inclusion, seriate and grasp transitivity, conserve number
and quantity, understand another’s perspective and decipher causal relationships (see
Donaldson, 1978; Gelman & Gallistel, 1978).
Other studies focused on the organization of the processes involved in the
understanding of different domains, rather than on their age of first attainment. These
studies aimed to test whether Piaget was right to assume that each developmental
stage had an overarching structure governing understanding over different domains.
Thus, in these studies, children were tested by large numbers of tasks designed to
represent different domains, such as space, number, causality, etc. The notion of the
structure of the whole suggests synchronous development in different domains and
therefore that tasks would be performed at equivalent developmental levels, regardless
of domain, especially at the stage of formal operations. Admittedly, Piaget recognized
that there are developmental gaps (decalage) in the age of acquisition of logical similar
concepts, such as number at 6–7 years, weight at 8–9 years, and volume at 10–11
years. However, this expectation was not confirmed even for concepts supposedly
identical, such as various aspects of the conservation of number or substance. On the
contrary, these studies showed systematically that individuals may operate on different
levels in different domains. This finding suggested that there may be factors governing
the organization of cognitive processes that go beyond Piaget’s notion of the structure
of the whole (Demetriou & Efklides, 1985; Fischer & Bidell, 1998; Shayer, Demetriou
& Pervez, 1988).
Conclusions
Piaget’s work refreshed the study of the human mind throughout the twentieth
century. On the one hand, some of the phenomena Piaget studied were the
phenomena studied in the cognitive and the differential tradition, reasoning par
excellence. In fact Piaget himself compared his analysis of the development of
analogical reasoning with Spearman’s eduction of correlates mechanism (Piaget,
2001). He suggested that eduction of relations starts in preoperational thought but
it is established in the stage of concrete thought and that eduction of correlates (i.e.,
analogical thought) is attained only with formal thought. Thus he viewed his work
as a developmentally adapted mapping of the processes involved in Spearman’s g.
Perhaps Piaget and Spearman might be pleased to know research showed that
performance on the classic Piagetian tasks is highly related to performance on classic
tests of intelligence (correlations in the range of .7). In fact Carroll (1993) recognized
that these tasks capture a special type of reasoning, which he named “Piagetian
reasoning”. In his three-stratum model this is a second stratum broad ability which
goes with fluid intelligence and is subsumed under g. This reflects the fact that
Piaget enriched the study of thought with concepts drawn from philosophy and
epistemology, such as causality, number, substance, weight, volume, etc., and even
scientific method and reasoning, that were not in the focus of cognitive or differential
research.
The fundamental causal mechanisms in Piaget’s theory also bear some deep
similarities to mechanisms in both the cognitive and the differential tradition, despite
their superficial differences. For instance, there are similarities between current
definitions of fluid intelligence in terms of executive processes and Piaget’s structures
of reversible mental operations. Piagetian progression to fully reversible and thus
equilibrated thought may be taken as an ideal for perfectly fluid intelligence. Specifically,
mental reversibility may be seen as shifting between two stimuli or mental objects (e.g.,
the length and the width of glasses in the conservation task, the act of transferring liquid
from one glass to another and the act of pouring it back to the original glass) and their
integration under an overarching concept. This very process of shifting involves
inhibition of a response (e.g., “the longer glass has more water”) that would reflect the
perceptually dominant aspect of the problem (i.e., “longer glasses have more”). One
might even see a process of updating, especially at transition phases before the correct
answer is automated. That is, the conflicting interpretations (i.e., more because it is
longer versus same because nothing was added or taken away) are updated and
alternatively considered until a final judgment is made. Therefore, the close relations
between Piagetian and cognitive or differential tasks should not come as a surprise.
It is also safe to suggest that Piaget’s theory assumed that language of thought
exists long before this assumption was made by modern cognitive science. Piaget’s
LoT is based on a general mental logic in the same fashion that Fodor’s LoT assumes
a general mental logic. Specifically, Piaget’s analysis of the general processes
underlying the logic of concrete and formal operations explicitly involves all four
basic properties of a general LoT: compositionality (integration of mental operations);
recursivity (e.g., integrated mental operations may be taken as many times as
required in class inclusion or seriation tasks); hierarchical integration (e.g.,
conservation judgments based on the reduction of perspectives or actions into a
general interpretation); and generativity (i.e., new mental constructions are
generated out of old ones).
Successive developmental stages are increasingly robust implementations of this
general LoT. That is, different types of logic implement LoT at successive
developmental stages: the logic of functions (simple relations) models the language of
thought in the preoperational stage; the logic of classes and relations (rules specifying
intensive and extensive relations into concepts and conceptual hierarchies) models
the language of thought in the stage of concrete thought; and symbolic reasoning
(conditional and propositional reasoning) models the language of thought in the
formal stage. In this system, mental operations comprising the logic of thought
dominate over specific forms of representation, such as language, mental imagery, etc.
Finally, Piaget’s conception of the role of consciousness is not unlike the role
ascribed to it in the cognitive tradition. It enables the resolution of conflicts and the
creation of new mental content. In terms of the current discussion, Piaget’s reflecting
abstraction is the key mechanism for transition across the successive implementations
of LoT across developmental stages.
The similarities above (emphasis on reasoning, assumptions about a general LoT,

and consciousness) bring Piaget’s theory closer to the cognitive rather than the
psychometric tradition. In psychometric theory, reasoning is of course present
under fluid intelligence. One might even say that psychometric theory did assume
the operation of a LoT, g, but it remained largely ignorant about its nature. For the
sake of fairness, however, we note that recently fluid intelligence was analysed in
terms of the main components of LoT, such as compositionality. Duncan et al.
(2017) proposed that compositionality is fundamental to the functioning of the
mind. They found that the very process of splitting a complex whole into simple,
separately attended parts is distinct from integration, working memory, and pro-
cessing speed, and it is important for higher performance in traditional intelligence
tests, such as Raven-like matrix tests. Finally, we note that psychometric theory is
still ignorant about the role of consciousness.
The common ground is a strong basis for further development of our understanding
of the developing mind. However, further development requires noting the
shortcomings and problems as well. Thus it was noted above that empirical research
showed that Piaget’s theory is inadequate in several important respects. Additionally
we note here that Piaget remained a qualitative theorist through to the end. He never
considered the possible role of representational or processing mechanisms, such as
working memory or attention. He did consider the role of representation and, in
fact, his theory of stages may be seen as a theory of successive representational
changes. However, his approach to the role of representations was one-sided.
Changes in them (i.e., in figurative intelligence) were always considered a by-product
of changes in inferential processes (i.e., operative intelligence) and never the other
way around.
Therefore we need a theory that would accommodate these shortcomings.
Specifically, even if present, the types of thought described by Piaget are not
organized as he assumed. For instance, Piaget’s dependence on particular kinds of
logic, many of them self-tailored, as a model of the organization of mental processes
at each stage falls short of the diversity of mental networks underlying thought at
different stages. Thus Piaget’s model of the language of thought is not acceptable.
In any case, a theory of intellectual development needs to model, in addition to
inferential processes, the representational and individual differences processes
considered by the other traditions.
Piaget’s model of change, although promising, was also too general. Specifically,
his reflective abstraction may eventually be part of a theory of cognitive develop-
ment, but this theory would have to specify how reflective abstraction operates at
different phases of development, different domains, and different learning contexts.
It is in this regard that the nature of representation may itself be part of the
language of thought. We will show later that awareness of mental processes is
always present but is strongly influenced by the type of representation available in
each developmental phase. Piaget disregarded these questions. In the following
sections we will present several alternative theories that attempted to clarify these
issues.
4
NEO-PIAGETIAN THEORIES
To solve the problems faced by Piaget’s theory, several scholars drew concepts and
methods from the cognitive and the differential traditions. There have been five
lines of research in this direction, each focusing on a different aspect of Piaget’s
theory. All five lines of research generated valuable findings and hypotheses which
would have to be integrated into a new, comprehensive theory that would overcome
the limitations of each tradition. The first focused on the role of speed of processing
and inhibition in the development of thinking and problem-solving. The second,
known as the neo-Piagetian theories, focused on the role of working memory in
cognitive development. The third focused on the development of reasoning itself
and the fourth on the nature of, and development in, various domains of thought,
such as causal and categorical reasoning. Finally there was research which examined
the development of awareness about the nature of the mind itself and its role in
behaviour, which is known as theory of mind (ToM), and awareness about cognitive
processes, known as metacognition. The first two areas of research looked for a
relatively simple transition mechanism that might explain changes in higher levels
of the mind. The other approaches emphasized developmental changes in these
higher levels themselves. These five lines of research are discussed below.
The role of speed of processing and inhibition

There is abundant evidence that speed of processing increases systematically with
age. Kail (1991, 2000, 2007) showed that reaction times to a wide range of tasks,
including motor, perceptual, and cognitive tasks (such as mental addition, mental
rotation, and memory search) decrease exponentially with age, levelling off at about
the age of 17–18 (Figure 4.1 illustrates how speed changes with age). That is, Kail
found that the change in speed of processing is constrained by a common factor,
which operates across all types of tasks. This suggests the operation of a common
Neo-Piagetian theories 45
FIGURE 4.1 Development of processing speed as a function of age (Kail & Ferrer, 2007)
underlying mechanism, which may be correlated to age-related change in the rate

of neural communication or other parameters related to the representation and
processing of information in the brain. Kail (2000) suggested that changes in speed
of processing result in more efficient use of working memory, which, in turn,
enhances reasoning and thinking. Kail and Salthouse (Kail, 2000, 2007; Kail &
Salthouse, 1994; Salthouse, 1996, 2000) extended this model to account for
cognitive changes that occur in the latter years of life. Specifically, they argued that
impairment in cognitive performance, which occurs after middle age (Baltes, 1991;
Schaie, Willis, Jay, & Chipuer, 1989), is caused by a slowing in speed of processing
that begins at about the age of 40 and continues systematically until death.
Salthouse (1996) ascribed this effect of cognitive slowing to two mechanisms, the
limited time mechanism and the simultaneity mechanism. According to the first
mechanism, when processing speed is slower than the demand of a given task (e.g.,
adding 3-digit numbers), performance is degraded because there is competition
between the currently executed operations and the operations of the immediate
past (e.g., the addition of tens interferes with the addition of hundreds). That is,
“the time to perform later operations is greatly restricted when large proportions
of the available time are occupied by the execution of early operations” (Salthouse,
1996, p. 404), resulting in a processing that always lags behind current needs.
According to the second mechanism, “products of early processing (the result of
adding tens and units in a task of adding three-digit numbers) may be lost by the
time that later processing is completed (e.g., adding the hundreds digits in a task
of adding three-digit numbers). To the extent this is the case, relevant information
may no longer be available when it is needed” (Salthouse, 1996, p. 405). Thus the
operation of higher-level mental functions, such as working memory or reasoning,
may be impaired due to lack of critical information.
Inhibition, on the other hand, is the gate-keeper of information processing. It
refers to active suppression processes that protect processing from the interference of
irrelevant information, remove task-irrelevant information from the field or space

of processing, and block mental or overt actions that may divert processing from
the current goal (Bjorklund & Harnishfeger, 1995; Dempster, 1991, 1992, 1993;
Harnishfeger, 1995). Empirical research showed that inhibition changes systematically
with age, following a pattern similar to that observed in the development of speed
of processing; that is, it improves from early childhood to late adolescence, remains
stable until middle age, and declines thereafter. This pattern was observed in the
context of various task conditions, including the Stroop phenomenon (Comalli,
Wapner, & Werner, 1962; Demetriou et al., 1993, 2002, 2013; Harnishfeger, 1995;
see also MacLeod, 1991) and working memory tests. In the case of working memory,
children become more able with development to retain in memory storage relevant
information and ignore distracting information. In the latter years the tendency to
distribute attention over both relevant and distracting information surges again,
impairing working memory performance. Figure 4.2 shows changes in tasks
addressed to inhibition control. It can be seen that the pattern of change is very
similar to changes in processing speed (see Demetriou et al., 2013, 2017).
In Chapter 1 we defined information-processing efficiency in terms of speed,
attention, synchronization, and retention. The research summarized here suggested
that speed and attention may be causally related. That is, on the one hand,
improvement in speed of processing may result in an improvement in the ability to
FIGURE 4.2 Change in attention control from 4 to 17 years of age (Demetriou et al.,
2017)
work on and finish a task before distractors can interfere, thereby resulting in
improved attention. On the other hand, improvement in attention may conserve
resources that are used for the sake of the current goal, thereby reducing the time
needed to work on the task. We will return to this question later, after we present
evidence on the role of retention.
The role of working memory
Mental power: Pascual-Leone

In the developmental tradition, neo-Piagetian theorists ascribed progression across
Piagetian stages to increases in invoking working memory capacity. Although
several researchers in the early 1960s envisaged this possibility (e.g., McLaughlin,
1963), Pascual-Leone (Pascual-Leone, 1970; Pascual-Leone & Goodman, 1979)
was the first to propose a model of cognitive development integrating the
fundamental assumptions of information-processing theory with the fundamental
assumptions of Piagetian theory. He argued that mental power (Mp) involves three
constituents: the M-operator, which reflects the mental energy available at a given
moment; the I-operator, which reflects central inhibition processes that enable the
person to stay focused on a goal; and the currently dominant set of executive
schemes, which specify the current goal. Working memory involves, in addition to
all of these operators, the various content schemes that need to be held in memory
when working on tasks. Mp refers to the maximum number of independent
information units or mental schemes that the person can hold simultaneously in
mind at a given moment. It is quantitatively defined by equation 1.1:
Mp = e + k (1.1)
where e stands for the mental energy required to hold the current goal (or executive)
active and k stands for the number of independent schemes that can be represented
and operated on. Thus Mp is very close to Baddeley’s working memory analysed in
the previous chapter, as both include executive processes and storage of information.
According to Pascual-Leone, e grows during the period of sensorimotor development
until it stabilizes at the age of 2–3; practically, it is the representational capacity
needed to mentally represent a goal until attaining what it suggests. “k” is equal to
1 scheme or unit of information at the age of 3 and it increases by one unit every
second year until reaching its maximum of 7 units at age 15. Thus Pascual-Leone’s
model takes Miller’s magical number 7 for granted and specifies how it varies with
development.
Pascual-Leone attempted systematically to show that the increase in Mp is the
cause of the transition from one Piagetian stage or substage to the next (Johnson,
Pascual-Leone, & Agostino, 2001; Pascual-Leone & Baillargeon, 1994; Pascual-
Leone & Morra, 1991). It can be seen in Table 4.1 that he maintained that the
classical Piagetian tasks that may be solved from preoperational through late formal
TABLE 4.1 Mental power demand of Piagetian stages according to Pascual-Leone’s theory
of constructive operators
Piagetian stages Mental power
Sensorimotor stage (0–2 yrs) e

Pre-logical thought (2–3 yrs) 1
Intuitive thought (4–6 yrs) 2
Early concrete thought (7–8 yrs) 3
Late concrete thought (9–11 yrs) 4
Transition to formal thought (11–13 yrs) 5
Early formal thought (13–14 yrs) 6
Late formal thought (15–16 yrs) 7
thought require an Mp of 1–7 mental schemes, respectively. A lower Mp than

required by a task makes the solution of this task impossible. Thus each increase
in the capacity of Mp opens the way for the construction of concepts and skills up
to the new level of capacity. In a series of studies, Pascual-Leone and his colleagues
(Johnson, Im-Bolter, & Pascual-Leone, 2003; Pascual-Leone, 1988; Rocadin,
Pascual-Leone, Rich, & Dennis, 2007) showed that Mp is related to both speed of
information processing and effortful inhibition and task complexity (Johnson,
Im-Bolter & Pascual-Leone, 2003; Rocadin et al., 2007).
Several empirical studies examined the presumed relation between Mp
development and development along Piagetian stages (de Ribaupierre & Pascual-
Leone, 1984; Pascual-Leone & Goodman, 1979). While this research did show that
these two forms of development are related, it is a truism in science that correlation
does not necessarily imply causality. Thus these studies did not settle the question
of the direction of causality; that is, is the direction indicating that development of
reasoning is caused by Mp development or vice versa? Or is it possible that both are
caused by a third factor?
Several other studies also compared Pascual-Leone’s model with Baddeley’s
model already described (de Ribaupierre & Bailleux, 1994, 1995; Kemps, De
Rammelaere, & Desmet, 2000; Morra, 2000). These studies suggested that Pascual-
Leone’s conception of Mp captures the development of executive processes in
working memory but underestimates the operation of domain-specific processes
related to Baddeley slave systems, such as visual and verbal working memory (de
Ribaupierre & Bailleux, 1994, 1995; Kemps, De Rammelaere, & Desmet, 2000;
Morra, 2000).
Executive control structures and recycling: Robbie Case

Robbie Case (1985, 1992) widened the perspective for viewing cognitive
development. First, he dropped Piaget’s emphasis on reasoning and logical
development, shifting to executive control. In fact he defined successive developmental
levels in terms of the executive control structures that children may command, rather
than in terms of underlying logical relations. In his own words:
An executive control structure is an internal mental blueprint, which

represents a subject’s habitual way of construing a particular problem
situation, together with his or her habitual procedure for dealing with it. All
executive control structures are supposed to contain (1) a representation of
the problem situation, (2) a representation of their most common objectives
in such a situation, and (3) a representation of the strategy needed to go from
the problem situation to the objectives in as efficient a manner as possible.
Case, 1985, pp. 68–69
Second, Case argued that executive control structures in each successive developmental
level require different types of representations. He proposed that there are four types
of executive control structures: sensorimotor (e.g., seeing or grasping; 1–18 months),
interrelational (e.g., words or mental images; 18 months to 5 years), dimensional
(e.g., numbers; 5–11 years), and vectorial (ratios or analogical relations; 11–19 years).
These names draw attention to the representational unit that is used as a building
block for the construction of executive control structures in level.
Third, he promoted the notion of recycling in the command of each new type of
executive control structure. Specifically, he argued that each new executive control
structure emerges from the previous one so that the last step in the development of
each executive control structure is the first step of the new level. Moreover,
development in each level recycles through the same steps of complexity. Specifically,
Case maintained that development within each of these four main stages evolves
along the same sequence of four levels: (1) operational consolidation, (2) unifocal
coordination, (3) bifocal coordination, and (4) elaborated coordination. As implied
by their names, structures of increasing complexity can be understood or assembled
at each of the four levels. When the structures of a given stage reach their last step
of complexity (which corresponds to the level of elaborated coordination) a new
mental unit is created that is representationally richer and the cycle starts up from
the beginning.
Finally, he specified the information-processing demands of this sequence in
terms of both processing efficiency and working memory. He used the term total
processing space (TPS) to refer to processing capacity. He defined TPS as the sum of
operating space (OS) and short-term storage space (STSS):
TPS = OS + STSS (1.2)
The operating space refers to the operations that need to be performed by the
thinker in order to attain the goal. At each of the four major stages of develop-
ment the operating space is occupied by sensorimotor, relational, dimensional, and
vectorial operations, respectively. STSS refers to the maximum number of mental
schemes that the thinker can focus on at a single centration of attention. An example
is when one has to count how many elements are involved in several groups
of objects and at the end recall all values found. In this example, the operation of
counting occupies the operating space component of total processing space, and the
values found as a result of counting occupy the STSS. Obviously OS is very similar
to executive control in Baddeley’s model or e in Pascual-Leone’s model. However,
it differs from them in an important respect: Case’s OS changes with development,
which is not the case in the other models.
Unlike Pascual-Leone, Case maintained that TPS does not change with
development. Only the relations between OS and STSS change. Case asserted that,
with development, the quantity of mental resources required by OS decreases due to
increasing processing efficiency. The space left free because of these changes is used
by STSS. Thus STSS increases as processing efficiency increases. Case maintained
that the capacity of STSS is 1, 2, 3, and 4 schemes at the levels of operational con-
solidation, unifocal coordination, bifocal coordination, and elaborated coordination,
respectively. Thus Case is closer to Cowan’s definition of working memory as equal
to about 4 chunks of information, rather than to the older Miller’s conception of the
magical number 7.
In a series of experiments, Case (1992) tried to show that increases in STSS are
indeed related to increases in operational efficiency. In these experiments, operational
efficiency was defined as the speed of execution of the required operation. For
instance, to measure operational efficiency in the dimensional stage children were
asked to count the elements of different sets of objects as fast as possible. The
children were also tested for their STSS of the sets involved. It was found that
the faster the children executed the counting operation the more items they were
able to store in STSS and the further they were along the sequence of dimensional
thought levels. This relation is nicely shown in Figure 4.3.
FIGURE 4.3 Changes in STSS counting span as a function of counting speed

Case’s theory was criticized on several grounds. First, it might be the case that
increased processing speed does not free processing space that may be used to
represent more components of an executive control structure (e.g., information
about objects, planning steps, actions to be taken, etc.). Alternatively, faster
processing may enable one to process more items in the same time window, making
the representational span appear, rather than actually become, larger (Baddeley &
Hitch, 2000). This interpretation is consistent with Salthouse’s (1996) limited time
mechanism, which ascribes increases in working memory capacity to increasing
processing speed. Second, the claim that total processing space is stable throughout
development is not tenable. We now have evidence that actual working memory
capacity does increase with age (Cowan, 2016; Halford, Wilson, & Phillips, 1998).
Third, STSS may not be the workspace of thinking because a person may hold one
type of information in storage and still work on a problem of a different type. This
implies that the central executive is involved in current problem-solving while
specific STSS systems carry on a different job (Baddeley, 1990; Halford, Maybery,
O’Hare, & Grant, 1994). Probably, domain-specific storage systems are used to store
information that will be used at later steps in problem-solving (Halford et al., 1998).
In any case, however, developmental changes in executive control structures cannot
be ascribed to changes in STSS. This is an interesting point to follow up on later,
because it opens up the discussion to other factors that bring changes in executive
control, such as awareness and reflection (these were discussed by Piaget but never
systematically examined). Finally, Halford (Halford et al., 1998) suggested that
Case’s definition of complexity in terms of processing steps is flawed because the
same executive goal may be analysed in several equally successful ways.
Relational complexity: Graeme Halford

Halford proposed an alternative way to analyse the processing demands of problems
that is supposed to explain the most crucial component of problem-solving:
understanding what the problem is about. According to Halford, this understanding
is built through structure mapping. Structure mapping is a process for assigning
elements of one structure to elements of another, so that any functions or relations
between elements of the first structure are also assigned to corresponding functions
or relations in the second structure” (Halford et al., 1998). In other words, structure
mapping is analogical reasoning used to translate the givens of a problem into a
representation or mental model that they already have. Specifying what the problem
is about depends on the match between the necessary relations of the problem (for
instance, “John is taller than Michael”) and the availability of a mental template that
would be called on to assimilate and give meaning to these relations (for instance,
understanding the concept “taller” requires representing two entities A and B and
their relation such that A > B). If the necessary relations (binary here) exceed the
mental template available (for instance, the child can only represent one entity on
its own), the problem will either not be understood or be degraded to the level of
the template available: here there would be two absolute judgments represented side
by side without any integration—John is tall; Michael is short. Therefore the

processing load of a task corresponds to the number of dimensions, which must be
simultaneously represented if their relations are to be understood.
Halford identified four levels of dimensionality. The first is the level of unary
relations or element mappings. Mappings at this level are based on a single attribute.
For instance, the mental image of a cat, or the word “cat”, is a valid representation
of the animal CAT because it is similar to it. The second is the level of binary
relations or relational mappings. At this level, two-dimensional concepts of the type
“larger than” can be constructed. Thus two elements connected by a given relation
may be considered at this level. The next is the level of system mappings, which
requires that three elements or two relations be considered simultaneously. At this
level, ternary relations or binary operations can be represented. The example of
transitivity, which can be understood at this level, has already been explained above.
The ability to solve simple arithmetic problems where one term is missing, such as
“3 + ? = 8” or “4 ? 2 = 8”, also depends on system mapping, because all three
known factors given must be considered simultaneously if the missing element or
operation is to be specified. At the final level, multiple system mappings can be
constructed. At this level quaternary relations or relations between binary operations
can be constructed. For example, problems with two unknowns (e.g., 2 ? 2 ? 4 =
4), or problems of proportionality, may be solved. In these problems, the a : b pair
is mapped onto the c : d pair so that the terms c and d in the second pair are taken
to correspond to the terms a and b in the first pair, respectively, and thus the two
relations are reduced to one relation integrating both pairs. That is, at this level four
dimensions can be considered at once.
Halford’s theory is related to other cognitive, psychometric, and developmental
theories discussed above. Specifically, Halford’s analysis of dimensional complexity
and structure mapping are intended as a model of the mechanisms underlying
inductive and analogical inference. By definition, then, it may be seen as a cognitive
science analysis of Spearman’s eduction of relations and correlates mechanism. In
fact Halford’s analysis of relational complexity may be viewed as an analysis of the
developmental changes underlying the improvement of the mechanism with age.
Halford suggested that the four levels of structure mapping above correspond to
Piaget’s sensorimotor, intuitive, concrete, and formal stage, or Case’s sensorimotor,
interrelational, dimensional, and vectorial stage, and are thought to be attainable at
the ages of 1, 3, 5, and 10, respectively. In conclusion, Halford’s theory was intended
as a “hard capacity theory” in that it supposedly explains the mechanism underlying
the transition across the main cycles of cognitive development. In other words,
processing capacity as conceived by Halford may be equivalent to Baddeley’s central
executive or Case’s operating space.
Halford criticized Case for his notions of operating space and maintained that
STSS failed to explicate how external structures of relations may be turned into
internal meaningful concepts. To tackle the problem, he offered the notions of
relational complexity and structure mapping. These notions are, in turn, weak in a
crucial respect: they do not explicate how structure mapping is implemented. We
suggest that the ability to handle problems of increasing relational complexity comes
from changes in executive strategies for handling information, rather than from
sheer increases in representational capacity (Makris, Tahmatzidis, Demetriou, &
Spanoudis, 2017). These changes come, in turn, from changes in awareness of
mental processes and related self-regulation that enable the thinker to prioritize
search and organize information to build the relations of interest. Therefore the
direction of causality may go either way. These changes will be discussed in the
chapters following.
Evaluation of neo-Piagetian theories

Are any of the factors invoked by the theories summarized above able to explain
cognitive development? Luckily there is now evidence allowing an answer to this
question in regard to each of the factors. It is a simple one: none of these factors
alone is sufficient to account for intellectual development. In regard to speed,
research suggested that speed is a developmental factor. However, its relation with
various intelligence measures early in life (3–6 years, r = .43), or from childhood to
adolescence (8–15 years, r =.40), is not strong enough to support considering
changes in speed as the major driver of changes in intelligence (Carlozzi, Tulsky,
Kail, & Beaumont, 2013). In fact recent longitudinal research suggested that relations
between speed and Gf over time (in the age span 6–13 years) may even be weaker
than above (about .2–.3) (Kail, Lervåg, & Hulme, 2015).
In regard to working memory, it is clear that relations between progressions
along cognitive developmental sequences and increases in working memory do
exist (Cowan, 2016; Demetriou et al., 2002). However, working memory is not
the transition mechanism as assumed by neo-Piagetians. For this to be the case, the
cognitive level of children in reasoning would have to be at the level expected from
the condition of working memory rather than sheer age. For instance, two children
of different age but the same working memory would have to operate on the same
cognitive developmental level. This is clearly not the case. We showed (Demetriou
et al., 2013) that reasoning attainment matches age rather than working memory
level. We allocated children in each of the years 4 through to 16 to three groups
according to their performance on various working memory tasks: low (0–2 items),
medium (2–5 items) and high (5–7 items) and compared them on their performance
in several reasoning tasks. The main findings are summarized in Figure 4.4. It can
be seen that the reasoning performance of children with high working memory
capacity was closer to the performance of their age mates with low working memory
than to the performance of older individuals.
Executive control as such was also found to fall short of the main driving force
of intellectual development. Specifically, relations between executive control and
intellectual development are strong in preschool age but they drop later on. The
relation with age was very high (~.8) in the 3–6 years period but it dropped
drastically (~.5) in the 8–15 years period. Correlations with cognitive measures were
also high at preschool (~.6–.7 with block design and verbal ability) but dropped
FIGURE 4.4 Reasoning attainment as a function of age and verbal working memory
level
Note: 1=low, 0–2; 2=medium, 2–5; 3=high, 5–7. (Based on Fig. 5, Demetriou et al., 2013)
drastically from 8–15 years (.3 with block design and .5 with verbal ability) (Arffa,
2007; Zelazo et al., 2013).
It might be the case, however, that a cascade rather than a fan model is able to
account for the relations between these factors and intellectual development. A fan
model has been considered in this chapter; that is, a model that assumes each of these
fundamental processes is directly connected with developmental changes in fluid
intelligence. A hierarchical cascade would assume a chain of relations, such that the
simplest process (e.g., sheer speed) resides at the one end of the chain and the most
complex process (e.g., reasoning) resides at the other end, with the other processes
residing between and mediating between neighbouring processes. This model
postulates that each process is embedded into the next, more complex, process
residing higher in the hierarchy (Fry & Hale, 1996; Kail, 2007; Kail & Ferrer, 2007;
Kail, Lervåg, & Hulme, 2015). This model is illustrated in Figure 4.5.
Attention control lies at the bottom of the hierarchy because it is very basic:
keeping mental focus on target against salient but irrelevant object characteristics
(Diamond, 2013; Rothbart & Posner, 2015). Flexibility in shifting across stimuli or
responses according to complementary goals is the next level in the hierarchy,
because this brings mental focus under the executive control of the thinker and
allows deployment of mental or behavioural plans (Deak & Wiseheart, 2015).
Working memory is partially an attention control task, because it enables one to
maintain and execute goals despite interference from other information present in
the senses or other actions one might be doing (Meir, Smeekers, Silvia, Kwapil, &
Kane, 2017). However, in addition to these executive processes it involves storage
FIGURE 4.5 The cascade model developmentally adjusted

Note: attention is drawn to the changes of relative importance of processing according to the
developmental phase
and recall processes enabling the handling of the information one is currently
working on (Baddeley, 2012; Cowan, 2016; Kane et al., 2001). Reasoning and
problem-solving in different domains reside higher because these involve,
additionally, inferential processes interrelating representations for the sake of valid
conclusions (Johnson-Laird & Khemlani, 2014; Rips, 1994; Markovits et al., 2015).
Figure 4.5 outlines this cascade.
However promising the cascade model appeared to be, it is weak. We showed
(Makris et al., 2017) that the cascade model as a hierarchy of simpler processes em-
bedded into more complex processes (attention control (–.64) à flexibility (.58) à
working memory (–.89) à reasoning (.87) à language (.98) à awareness (.60))
cannot be discriminated from its inverse model, where more complex processes are
embedded into simpler processes (awareness (.50) à language (.67) à reasoning
(.97) à working memory (.84) à flexibility (–.84) à attention control (.48)).
Conclusions
Neo-Piagetian theories sought to explain the development of thought in reference
to factors of processing and representational efficiency from the cognitive tradition—
in the fashion that the psychometric tradition sought to explain individual differences
in intelligence. Specifically, each of these factors (i.e., speed, inhibition, executive
control, and working memory) was thought at some point to be the holy grail of
intellectual development. The findings presented in this chapter suggest that none
of them fulfilled this promise. In the fashion of psychometric research relating these
factors to individual differences in intelligence, each factor does have a role and is
definitely part of cognitive development. However, it is correct to assume that
changes in thought cannot be reduced to changes in any one of these factors alone.
It might be the case that embedding all of them together into a hierarchical
cascade does the job. Reality, though, is more complicated than developmental
researchers would have hoped. The findings above suggested that relations do not
build bottom-up from simpler to more complex, as assumed by the cascade model;
they go top-down as well, which is conceptually not acceptable. This finding lends
support to a truism: complex thought processes are more than the sum of their parts,
if the parts are those discussed here. What is missing may be the mutual interaction
between processes per se, as suggested by van der Maas (van der Maas et al., 2017),
or probably another factor driving and orchestrating this interaction and not
considered by research so far. It might even be the case that the relative contribution
of each process and their relations vary with development. We will return to these
questions later, when we will map them in development.
5
THE DEVELOPMENT OF
REASONING
Piaget’s theory associated successive stages with different kinds of logic: the logic of
functions, classes, and propositions at preoperational, concrete, and formal thought,
respectively (Piaget, 1976, 2001). However, Piaget’s theory is not an account of the
development of reasoning as such, because these types of reasoning do not really
capture each phase’s reasoning possibilities. Preoperational children may use reason-
ing supposedly acquired at later phases, and formal adolescents may fail on reasoning
tasks supposedly mastered at earlier phases (Moshman 2011). This is also true for the
neo-Piagetian theories. These theories never actually addressed the development of
reasoning as such. In this chapter we summarize research relevant to the nature and
development of inductive and deductive reasoning.
The development of inductive and analogical reasoning

Perceptual similarity is the most readily apparent form of similarity: we see things
of the same colour, same shape, same size, etc. Thus perceptual similarity is the basis
for the earliest forms of inductive reasoning. Young infants can make judgments
based on perceptual similarity from very early in life. In fact the very methods used
to study infants’ cognitive abilities, such as the habituation method or the visual
preference method (see Butterworth, 1998b), assume that infants recognize
similarities and differences between various stimuli. That is, they habituate to an
object and loose interest in it if they see it repeatedly. This implies they somehow
formed a representation of this object, recognizing it as similar after seeing it a
certain number of times.
Conceptual similarity was also found to function as a basis of inductive inferences
in infancy. Baillargeon (1995) showed that infants at 10–11 months may choose to
make inferences on the basis of functional rather than perceptual similarity. They
showed infants a cylindrical container into which they first poured salt, and from
which they then poured it out. Subsequently they poured salt either into a container
identical to the first in all respects except that it had no bottom and it thus could
not function as a container, or into another object which was perceptually different
from the original container but which could function as a container because it had
a bottom. The infants showed surprise when the perceptually similar but functionally
inappropriate object appeared to hold the salt, indicating that they were able to
formulate a category based on functional rather than perceptual properties and
generate inductive inferences accordingly.
Contrary to the popular belief that analogical reasoning is a late attainment,
recent research suggests that very young infants are sensitive to analogical
relationships. For example, Wagner, Winner, Cicchetti, & Gardner (1981) showed
that 9-month-old infants preferred to look at an arrow pointing up when hearing
an ascending tone and at an arrow pointing down when hearing a descending tone.
That is, they appeared to be able to decode the analogy [ascending tone : ↑ ::
descending tone : ↓]. Based on this and other similar evidence, Goswami (1992)
argued “that the ability to recognize relational similarity may not develop at all”
(p. 13) because it “could be an inherent quality of human reasoning” (p. 15).
The work of Susan Gelman (2003) and Keil (1989) showed that, by the age of
3 to 4, children organize their knowledge of the world into elaborate categories
which they use as a basis for very powerful inferences. For instance, Gelman and
Coley (1990) showed children a particular bird and told them that “it lives in a
nest”. Children at this age readily inferred that another bird which was virtually
identical to the target would also live in a nest; they were unsure about other birds
which differed from the target bird in some respects; and they concluded that other
creatures, like a stegosaurus, which differed markedly from the target, would not
live in a nest. Moreover, other experiments showed that associating an object
with a novel name (for example “this is a dax” or “this is a diffle”) led 3.5-year-old
children to expect that other objects of the same shape would also be “dax” or
“diffle” (Becker & Ward, 1991; Landau, Smith, & Jones, 1988).
Goswami and Brown (1989) showed that children can solve classic analogies of
the [a : b :: c : d] type once they involve familiar objects and transformations. For
example, children were presented with pictures organized in an analogy as follows:
[chocolate : melted chocolate :: snowman : ?]. Their task was to choose the missing
term d from among five pictures. Many 3- and almost all 4- and 5-year-olds were
able to solve this kind of task, choosing melted snow.
This work suggests that, with increasing knowledge and experience, inductive
inference tends to rely increasingly on conceptual rather than perceptual similarity,
although perceptual similarity is never abandoned as a basis of inference. Thus, with
increasing expertise, inference tends to rely more on attributes indicating category
membership than on simple perceptual similarity. It is also noted that the precocious
presence of analogical reasoning is largely dependent on the surface or perceptual
similarity of the terms. Even school-aged children are prone to fail on analogies
which involve very common objects and functional relationships, if there is no
obvious perceptual similarity. For example, Case (1985) found that the analogy
The development of reasoning 59
[ink : pen :: paint : brush] is understood at about the age of 10. Second-order
relations which involve abstract relationships, such as [food : body :: water : ground],
are not understood until well into adolescence. Finally, Sternberg & Downing
(1982) showed that third-order analogies requiring to understand the relation
between two abstract analogies are understood only at college age. For instance,
what is the relation between the analogy [sand : beach :: star : galaxy] and the
analogy [water : ocean :: air : sky]?
The development of deductive reasoning

According to Braine (1990), deductive reasoning develops from primary reasoning to
formal or secondary reasoning. Primary reasoning involves elementary information
integration processes enabling children to make automatic inferences in verbal
interactions and integrate information coming from different sources or at different
times. In everyday conversations the comprehension process uses all information
available to draw an interpretation (i.e., knowledge about the speaker, knowledge
about the world, conventions about speaking, grammar and syntax in language).
This kind of comprehension is fast because it takes place at conversation and reading
speeds (Braine & Rumain, 1983, p. 267).
However, there are problems which call for formal or secondary reasoning. An
example of these problems is given below:
Birds fly.
Elephants are birds.
________________________
Elephants fly.
Obviously the conclusion “Elephants fly” is logically valid, although factually false.
That is, although we know that elephants are not birds and they do not fly, we have
to accept that it necessarily follows from the premises. In other words we assume
that if it were true that elephants are birds then, given that we accept that birds fly,
we would also have to accept that elephants fly.
Note two differences in the approach required to solve this kind of problem as
contrasted to problems solved by primary reasoning. First, analytic rather than
ordinary comprehension is required: one must focus on the meaning of each
sentence as given in the argument and ignore any other previous knowledge or
information related to the words in the sentences. Second, the reasoner must
understand that an argument involves a network of relations systematically arranged
which can be used as a basis for decoding the relations. Thus, in order to grasp the
logical relations implied by a logical argument, one must be able to break down
the argument into the premises involved and focus on their logical or formal relation
independently of content. The formal relations implied by the premises in the
argument are normally indicated by particular words of natural language, such as

the conditionals “if . . . then”, the disjunctives “either . . . or”, etc. When this
approach is taken, one is, in principle, ready for formal or secondary reasoning.
Moshman (1990, 1994, 2011) proposed that the development of deductive
reasoning evolves through four stages leading from primary to formal reasoning. The
crucial factor in this course of development is an increasing awareness about the in-
ferential processes themselves, their underlying logical properties, and their relations.
Additionally, increased awareness is usually accompanied by an increasing will to
control and improve inferential processes. Moshman described the development of
reasoning as more a matter of “metareasoning” than simply inference and logic.
Metareasoning strategies are strategies of reasoning that go beyond simply

assimilating premises to unconscious inference schemata. They involve an
explicit distinction between premises and conclusions and a purposeful use
of inference to deduce the later from the former. Such strategies are typically
conscious or at least accessible to consciousness. They include, for example,
strategies for systematically generating multiple possibilities consistent with
premises, actively seeking counterexamples to potential conclusions, or
coordinating several inference schemata to construct a line of argument.
Moshman, 1990, p. 208
Thus the stages of reasoning development proposed by Moshman describe the

course from the automatic use of inference schemata to explicit, self-guided logical
reasoning. The general characteristics of these stages are summarized in Table 5.1.
Stage 1: Explicit content-implicit inference. This stage appears during the preschool
years, when children reason logically using inferences but do not think about inference.
This is obvious in both children’s ordinary comprehension and in their everyday
speech production. That is, when they speak, 2–3-year-old children correctly use
most of the connectives and conditionals involved in inference schemas, such as and,
but, or, because, and if. Moreover, at this age, children respond to commands requiring
a grasp of the relations expressed by the conditionals: “Do A and do B” (at 2 years)
or “Do A or do B” (at 4 years) (Beilin & Lust, 1975; Johansson & Sjolin, 1975). At
this age children also understand the meaning of the conditional if when it is
embedded in permission schemas: “If you want to play outside you must put your
coat on.” Harris and Núñez (1996) showed that 3- and 4-year-olds can recognize
breaches of permission rules by specifying when the child acts differently than
allowed. Understanding of the subtle and more strictly logical meaning of at least
some of the conditionals follows a little later. One such example is the directionality
signified by if; that is, understanding that events paired by if are necessarily ordered
in a particular time sequence. According to Emerson (1980), it is not until age
5 that children can recognize which of the four following sentences is “silly”: If it
starts to rain I put up my umbrella; I put up my umbrella if it starts to rain; If I put
up my umbrella it starts to rain; and It starts to rain if I put up my umbrella. At age
6 children may correct the silly sentences to make them sensible. However, children
TABLE 5.1 The general characteristics of deductive reasoning at different phases
Age Stage Explicit object of Knowledge implicit Inference schemas Comprehension Reasoning
understanding in reasoning (subject)
2–5 Explicit Content Inference: There is a cat; There is an apple/... Automatic; Primary
content Conclusions There is a cat and an apple. metalinguistic
Implicit deduced and thus There is a grape, and there is a lemon awareness is
inference distinct from or an egg/... There is a grape and a implicit
premises lemon, or there is a grape and an egg.
6–10 Explicit Inference: conclusions Logic: Form of There is a dog and a tiger; There is Ordinary; basic Primary
inference deduced from and, argument distinct not a dog and there is not a tiger / metalinguistic
Implicit logic thus, related to from empirical INCOMPATIBLE awareness
premises truth of premises If there is either a cow or goat, then
and conclusions there is a pear; There is a cow/...
(necessity) There is a pear.
There is a strawberry or a blackberry;
There is not a strawberry /... There
is a blackberry.
11–18 Explicit logic Logic: Relation of Metalogic: Formal Schemes of suppositional reasoning: Analytic; complex Secondary
Implicit argument form and logical system given a chain of reasoning of the form metalinguistic
metalogic empirical truth of distinct from “suppose p and q” one can conclude awareness
premises and natural language that if p is present q will also be present.
conclusions (validity) Subjects also start to understand that
when q is present, however, it does not
follow that p would have to be present.
19–24 Explicit Metalogic: Interrelations Metametalogic: All of the above as means for the Reflective; Metasystematic
metalogic of logical systems and Differentiation and creation and formation of logical metametalinguistic
natural languages reconstruction of systems that formalize inferential awareness
metalogic processes or systems about the mind.
at this stage are not aware that the premises constrain the conclusion. Thus they fail
to solve problems where this awareness is needed in order to analyze what is
involved in an argument.
Stage 2: Explicit inference-implicit logic. This awareness appears for the first time
by about age 5 or 6, marking the transition to the second stage. With this type of
awareness, stage 2 children realize that no conclusion can be drawn from the
argument below:
Sprognoids are animals or plants or machines.

Sprognoids are not animals.
_____________________________
Therefore, sprognoids are plants.
They realize that because the first premise involves three alternatives and the second
premise cancels only one of them, it is uncertain which of the two remaining
alternatives is valid. This understanding indicates, according to Moshman, that
children at this stage are able to view the premises and conclusions as different parts
of an argument which are connected by inference. In other words, children at this
stage are explicitly aware of the inferential process that connects premises and
conclusions into coherent arguments, and are sensitive to logical necessity. However,
logic as such is still implicit in their reasoning and does not function as a frame to
explicitly guide reasoning. Thus they fail on problems in which the logical form of
the argument must be explicitly differentiated from its content. The previous
argument on elephants that can fly is an example of the problems which cannot be
solved by children at this stage, because they focus on the content of the argument
and do not recognize that this argument is formally identical to many of the
arguments and inferences they spontaneously use in their everyday interactions.
Stage 3: Explicit logic-implicit metalogic. Preadolescents, at 11–12 years, are able to
solve these problems, indicating an understanding of the distinction between logical
form and empirical truth. In other words, they understand that “an argument is valid
if, regardless of the empirical truth of its premises and conclusions, it has a logical
form such that, if the premises were true, the conclusion would have to be true as
well” (Moshman, 1990, p. 212). Therefore, at this stage children are explicitly aware
of logic as a system of rule-bound relationships between arguments and conclusions,
which implies differentiation between formal logical relationships and the language
in which they can be expressed.
This differentiation was clearly demonstrated by Osherson and Markman (1975).
In their study, children and adults were asked to evaluate the truth of statements
such as “The chip in my hand is white and it is not white”, or “The chip in my
hand is not red or it is red”. These statements are nonempirical: their truth does not
depend on the external world but on the consistency between the assertions
involved in the statements. The colours mentioned are irrelevant. Only the syntactic
and the logical relation between the statements are relevant. When attained, this
understanding indicates that children represent logical relationships as something
different from empirical reality.
Even at this stage, adolescents still lack an explicit metalogic. This system would
explicitly specify logical systems and define their formal relations, mapping their
similarities and differences. As a result they still fail on tasks which require this
metalogic. The fallacies of affirming the consequent or denying the antecedent are
good examples of tasks requiring this metalogic. Thus Wason’s “selection task”,
which was described in the first chapter, cannot be solved at this stage.
Stage 4: Explicit metalogic. These tasks are solved at the stage of explicit metalogic,
when individuals are able to consider the characteristics of different types of
inference, or even those characteristics of different logical systems, and specify their
similarities and differences. Work on metasystematic reasoning is relevant here.
According to Commons and Rodriguez (1990), metasystematic reasoning enables
thinkers to specify the formal characteristics of each of a number of different sys-
tems and then specify the higher-order similarities and differences between these
systems. Our studies showed that only a small minority—a meagre 10%—of college
students, who were advanced formal thinkers according to their performance on
Piagetian formal tasks, demonstrated metasystematic reasoning. Moreover, these
subjects exhibited a high level of awareness of their own mental processes
(Demetriou, 1990; Demetriou et al., 2017).
Origins of reasoning: innate or constructed?

Where does reasoning come from and how is it elevated from the state of implicit
inference about the world to the state of explicit inference, logical necessity, and
explicit reflection on inference and the rules underlying inference? Several scholars
believe that inductive (Goswami, 1992) and deductive reasoning are part of the
human makeup (Fodor, 1975; Macnamara, 1986). Macnamara argued that the basic
logical notions of truth and falsity, the foundations of deductive reasoning, are
available to children from very early in life. In his opinion, the notion of truth is
implicit in the assertions that children make very early on, because making assertions
about something implies ascribing a truth value to it. Braine (1990) argued for
ready-made inference schemas corresponding to particular patterns of events and
occurrences in the environment. These are joint iteration, which refers to repetitions
of actions, events, or objects; iteration of alternatives, which applies to a situation
wherein one event or object is present and another event or object may or may not
be present; and contingencies, which refer to time-dependent sequences, which may
or may not be causal (examples of these schemes are given in Table 5.1). Obviously
these authors posit a language of thought which involves the basic logical relations
underlying conditional reasoning and which is available before children learn their
mother language. Learning spoken language, according to these theorists, is
equivalent to learning a translation into the language of thought. In other words,
learning the connectives and, or, and if, which correspond to the three patterns
mentioned above, is equivalent to mapping them onto the corresponding connectives

of the language of thought.
Arguably the fact that children at the first stage of reasoning development
outlined are able to draw implicit inferences matching all of these schemes may be
in line with this position. Of course, one might object that kindergarten children
are already old enough to have experience and practice in using these inference
schemes. In any case, the inferential possibilities of preschool children suggest that
the pillars of reasoning are present from very early in life. In psychometric terms, g
is always present as a means of processing relations. Together with awareness and
metarepresentational possibilities, g expands into reasoning allowing increasingly
refined relational processing. Therefore, assuming the operation of inductive and
deductive reasoning primitives implies a specific assumption about the development
of reasoning. It is building an increasingly differentiated code that specifies the
constraints of application of each inferential pattern. Specifically, the code prescribes
that if a specific MP relation holds, the corresponding MT also holds; however, this
is not the case for AC or DA. These remain open because the effect they prescribe
may come from factors not specified in the MP and MT relations. This code
functions as a subjective truth-table and it is thus valid for all domains. Obviously, grasp-
ing the code changes the type of representations one is attending to. Before grasping
the code, thinkers attend to content knowledge specified in the premises; when
possessing the code they attend to the relations between the premises specified
by the connectives. In the first case, building mental models is an associative
process connecting representations in memory. In the second case, it becomes a
kind of reverse representational engineering where new, even counter-intuitive
representations may be generated from the relational code.
Language learning and education are very influential in the transformation of
primary into secondary reasoning. Falmagne (1990) suggested that logical knowledge
emerges from an apprehension of the structure of language and from the
correspondence between the arrangements of propositions in language and empirical
states of affairs: for instance, the conditional connectives “if . . . then” are always
ordered in language and coupled to events also ordered in time. The assumption is
that noting the correspondence between the linguistic schemas and the actual patterns
of events may be the starting point for elaborations that will lead from primary or
automated reasoning to more self-directed or formal reasoning. In line with this
assumption, permission schema theory (Cheng & Holyoak, 1985) posits that these
inductions occur in the context of permission rules that specify the conditions under
which a given action can or must occur (Harris & Núñez, 1996). Piaget would not
be very pleased with this approach but he would not be embarrassed either. He
would suggest that language comes to sculpt on a logical operational background to
deliver the schemes of reasoning handled by middle childhood onwards.
Of course, analytic comprehension is a major force in building this subjective
truth-table: it enables the reasoner to differentiate between the various kinds of
meaning implicated in verbal statements to allow a focus on what is logically crucial.
For instance, it enables children to differentiate between literal, implied, factual
meaning, and logical relations, and prioritize the latter over all other types of mean-
ing. Obviously education contributes extensively in the development of analytic
comprehension, because it leads students to recognize the alternative readings of a
given reality that are possible. We will elaborate on the development of analytic
comprehension as a factor of logical development in the third part of the book.
However important they may be as facilitators, neither language learning nor
education would suffice to explicate the grasp of “logical insight” about what is right
and its strengthening in reasoning development. This is an internal affair carried out
by a sole mind striving to differentiate between noise (or irrelevance) and “true
reality”. The reasoner’s own awareness of the inferential process that may direct
internal attention to the proper level of analysis and the proper relations is the crucial
factor. This is important in both inductive and deductive reasoning. In inductive
reasoning there is evidence that the progression from judgments based on surface
similarity to the grasp of second-order analogical relations requires an awareness of
the analogy per se. Obviously the terms involved in many analogies may be related
by several alternative relationships (e.g., content, metaphorical meaning, etc.). To
grasp the proper relation (i.e., educe the correlates in Spearman’s terms) requires that
the thinker is aware that relations at this level do exist and she must search for them;
this would allow her to withhold judgment so that she can search for alternative
relations until finding one that can bridge the two first-order relations (DeLoache
et al., 1998).
This awareness enables reasoners to differentiate between types of reasoning, such
as inductive and deductive reasoning, and tune the inferential process accordingly.
There is evidence that this awareness does not exist before the age of 6; it emerges in
early primary school years, around 7–8 years, and continues in adolescence. Galotti,
Komatsu, and Voelz (1997) examined if children from kindergarten to late primary
school can distinguish between deductive and inductive syllogisms. Children were
asked to solve deductive (e.g., “all daxlets are squishy; all squishy animals like to yell;
do all daxlets like to yell?”) and inductive (e.g., “all squishy animals like to yell; all
daxlets like to yell; are all daxlets squishy?”) reasoning tasks, rate their confidence in
their solution, and explain their reasoning. Tasks were identical in content and logical
relation, differing only in reasoning type. Children at kindergarten, up to age 6, show
little awareness of differences between inductive and deductive reasoning, if any. The
beginning of recognition of the differences appears at second grade, when children
start to show clear differences in confidence about conclusions between the two types
of reasoning; they are more confident to make deductive conclusions, which is what
is expected provided that these conclusions are necessary. Fourth and sixth graders,
i.e., from age 9 to 11, show clear sensitivity to the differences; this is expressed in
their confidence ratings, their explanations specifying differences, and their faster time
to respond to deductive syllogisms.
This awareness is also required to differentiate between premises and conclusions
in deductive reasoning, which is necessary if one is to be able to decide whether a
conclusion can be derived from the premises. This awareness also enables the thinker
to differentiate between inference schemata, which allows him or her to formulate
a framework that can suggest what is and what is not valid in reference to a particular
type of relation. Reverberi and his colleagues conducted several studies to examine
if deductive reasoning may operate automatically and unconsciously (Reverberi,
Pischedda, Burigo, & Cherubini, 2012). They found that modus ponens is indeed
unconscious in young college students. For example, Reverberi et al. (2012) showed
that students are primed to a modus conclusion even if a part of it is unconsciously
activated. For instance, students saw and explicitly noticed the primary premise (e.g.,
“if there is a 3 then there is an 8”) but they were exposed to the secondary premise
(“8”) at very high speed, rendering it impossible to consciously see or notice it.
Surprisingly, in a series of subsequent evaluation tasks, participants pre-activated the
number “8” if the unnoticed second premise corresponded to the antecedent of
the conditional (“3”). According to these researchers, this finding suggests there is a
primitive modus ponens scheme that is activated as an ensemble (p and q, p, q) if its
two main premises are minimally associated, even unconsciously. Notably these
researchers also found that disjunctive syllogism and the affirmation of the consequent
fallacy are not automatic; the priming effect above did not apply to them, suggesting
that explicit awareness and processing of the relations are needed if they are to
be applied. College students are, of course, old enough to automate some pivotal
components of conditional reasoning. In any case, modus ponens operates as the
background on which the rest of conditional reasoning deploys its other operations,
also drawing on a minimal amount of awareness that will allow the thinker to match
an argument with the internal mental truth-table mentioned above.
Reasoning and processing efficiency

The relation between changes in reasoning and changes in indexes of processing
efficiency, such as executive control and working memory, is a crucial test of the
assumption put forward in earlier chapters that g or Gf depend on processes for
representing and manipulating information. There is a vast body of research in this
area. This research clearly suggests, on the one hand, that these relations do exist.
For instance, Chuderska and Chuderski (2009) showed that goal monitoring and
response inhibition influence updating in working memory, switching, and dual
tasking, which in turn influence analogical reasoning. Barrouillet and colleagues
(Barrouillet, Gavens, Vergauwe, Gaillard, & Camos, 2009) showed, in a large series
of studies, that working memory is highly important in the construction and
evaluation of mental models in deductive reasoning.
In Barrouillet’s model, conditional reasoning develops in three levels: the conjuctive,
which allows the handling of “p and q” instances; the biconditional, which allows the
handling of “p and q”, and “not p and not q” instances; and the conditional, which
allows the handling of “p and q”, “not p and not q” and “p and not q” instances.
Obviously these three stages correspond to the first three stages of reasoning
development discussed above. At the age of 9, children were about equally divided
between the conjuctive and the biconditional stage; at the age of 11 the biconditional
stage dominated; and at 14 the conditional dominated. It is stressed that Taplin,
Staudenmayer, & Taddonio (1974) found, many years ago, that these three levels of
conditional reasoning are attained at the ages specified in Barrouillet’s model. Mean
counting span at three age groups was about 3, 4, and 5 chunks, respectively. Of the
whole sample, children with a counting span of 3 or less operated on the conjuctive
or the biconditional stage; those having a medium span of 3–5 units operated mainly
on the biconditional stage; those having a span higher than 5 operated on the conditional
stage. Therefore “the interpretation of conditional rules develops as a function of the
number of models the child is able to produce and coordinate. The older the subject,
the higher the number of models he or she can produce to represent the state of affairs
described by the if then rules” (Barrouillet & Lecas, 1999, pp. 297–298).
Conclusions
Three conclusions may be drawn from the research summarized in this chapter. First,
inference is always present. At the beginning of life, infants are capable of inductions
that transfer meaning across representations. They are even capable of analogical
interlinking of representations, which allows the coupling of episodic ensembles with
each other on the basis of correspondence between their parts. One might even see
the seeds of deductive reasoning from as early as when language appears, in the form
of basic schemes bridging representations so that when one representation is present
the other may be derived as well.
However, second, advanced inductive and analogical reasoning and deductive
reasoning proper are not present before a certain level of awareness about the
inferential process is present. This would allow the child, at about the age of 8–10,
to place truth values on specific relations between representations, look for them,
and accept them as such, regardless of content. This process culminates in adolescence,
when full conditional reasoning may be mastered. Overall, inferential awareness
involves the following: (i) awareness that a search-match-choose-evaluate process is
required (this may be cyclical and it may thus be repeated several times in the attempt
to solve a task); (ii) awareness that there is a subjective logicality table against which
alternative choices may be placed (this table specifies prescriptions about analogies
and truth-like standards about deductive reasoning schemes); and (iii) executive self-
restraint allowing resistance to a plausible solution until the best solution is found.
Needless to say, third, building mental models is useful in reasoning because it
strengthens the likelihood that the best conclusion may be selected. Increased
working memory is also useful because it gives the representational operating
ground for the implementation of processes above. However, neither mental models
nor working memory yield the solution. This comes from the “logical insight”,
which will bring the inferential process to a closure when it lights up. The basis of
this insight is considered in the following chapters.
6
AWARENESS AND KNOWLEDGE
ABOUT THE MIND
What do children know about the mind?

The mind is the bedrock of the psychological world. We might go as far as to say
that the psychological worlds we live in exist because of the human mind. Our
actions and feelings towards each other depend on how we register and represent
the world. Our thoughts and ideas about each other guide our actions towards each
other. Our personal experience, behaviour, and interpersonal relationships depend
on the mental states we create as persons. In a sense our life is a dialogue between
minds. Thus, according to some scholars, awareness of the mind is so important that
it “is part of our social instinct”: this is what channels children to learn about
invisible, intangible, abstract states such as thoughts, beliefs, and desires (Leslie,
Friedman, & German, 2004).
The study of awareness about the human mind may involve everything
discussed in this book. Broadly speaking, it may involve awareness about: (i) the
content of the mind, such as desires, beliefs, and concepts one holds or knows (or
thinks) that others hold, and its impact on action; (ii) the processes generating
knowledge and understanding, such as thinking, reasoning, imagination, learning, and
memory; and (iii) processes of consciousness, reflection, and self-control that humans
direct to their own mind in order to know it and change it, if needed. It is notable
that research on the development of children’s knowledge about the mind has
been highly fragmented and even runs under different names. The study of
knowledge about the content of mind and its role in human action is known as
the study of the theory of mind (Perner, 1991; Wellman, 2014). The study of
knowledge about mental processes is known as the study of metacognition (Efklides,
2008; Flavell, 1979). The study of consciousness comes under various names,
including the study of self-awareness and self-representation in various domains,
such as cognition, emotions, and personality. Below we summarize these fields of
Awareness and knowledge about the mind 69
research in order to highlight how awareness of the mind develops. Researchers

ask four interrelated questions:
1. Do children understand the mind as something different from reality? That

is, do they understand that the thoughts or ideas that they may have about an
object or person cannot be identified with this object or person?
2. Do they understand the representational nature of the mind? That is, do they
understand that thoughts, ideas and beliefs stand for objects, events, or mental
states that express a particular aspect or perspective of persons?
3. Do they understand the causal role of mental activity and its products? That
is, do they realize that what people do and how they do it depends on their
thoughts, ideas, guesses, fantasies, beliefs, desires, or wishes?
4. Do they understand how mental activity is organized and functions? That is, do
they have any understanding of the mind as a complex and diversified system
comprising different functions, such as attention, memory, and reasoning, which
are responsible for different mental jobs?
In short, the four sets of questions refer to the understanding of the ontological
status, the representational nature, the causal role or agency, and the nature and
functioning of the mind, respectively. We will summarize the findings about the
four aspects of mind below.
Knowing about the mind or a theory of mind?

The study of the child’s theory of mind has been one of the most active fields of
research in developmental psychology in the last 30 years. A Google Scholar search
under the term “theory of mind” gives an impressive 4,470,000 results (May 22,
2017), far higher than cognitive development (3,420,000), learning theory
(3,250,000), psychoanalytic theory (695,000), psychometric theory (622,000) and
Piaget’s theory (101,000). Obviously theory of mind has dominated psychology for
many years.
Understanding the ontological status of the mind

There is evidence showing that children at 3–4 years understand that thinking
about an object is different from the object itself. For example, 4-year-old children
understand that a rabbit and a monster they were asked to imagine are not real.
However, when told that the researcher would leave the room, many children
were frightened to imagine a monster. Even many 6-year-olds said they were afraid
there might be a monster in the box (Harris et al., 1991). These findings confirm
our experience that children continue to be frightened by their thoughts well into
the school years. It is well known that some mentally ill patients, such as schizo-
phrenics, cannot distinguish clearly the boundaries between the real and imaginal
(Lysaker, Dimaggio, & Brüne, 2014). This evidence suggests that the imaginary
may be distinguished from the real at quite an early age. However, this distinction
continues to develop for many years and under certain conditions it may break
down at any age.
Understanding the representational and causal role of the mind

The experimental paradigm used to study children’s understanding of the mind as a
causal agent is rather simple. In the famous Sally task, the researcher places a candy
in Box A in front of both the child and an assistant, the protagonist in the experiment.
The protagonist leaves the room and the researcher moves the candy from Box A to
Box B in his absence. The protagonist returns and the child is asked to indicate where
he will look for the candy: in Box A (corresponding to the protagonist’s representation
of the candy’s location) or Box B (corresponding to the child’s representation of the
present place of the candy). Tasks designed according to this paradigm have come to
be known as the false belief tasks: the false belief ascribed to the protagonist that the
candy is in Box A. Children who indicate location A are obviously able to understand
that the representation of a given situation (a belief about it) depends on available
information and that a person’s behaviour originates from his representation (looking
for the object following the belief). Children who indicate position B are obviously
unwilling to differentiate their own representation from that of others, in effect
projecting their representation on others. Many studies showed that 3-year-old
children cannot solve this task but that 4-year-olds can. Based on this evidence,
theorists concluded that 3-year-olds do not have a theory of mind but 4-year-olds
do. Moreover, it was assumed that 3-year-olds may have a representational deficit
which does not allow them to differentiate their own mind from another’s mind or
recognize that different persons may have different beliefs which can lead to different
behaviour (Wellman, 1990).
What do standard false beliefs tasks, such as the Sally task, measure? It might
be the case that these tasks only capture a very simple dichotomous understanding
of the mind: that children at this age understand that others may not know what
they themselves know and they may therefore act differently based on what their
own knowledge is. However advanced this understanding may be, it deviates from
a complete metarepresentational understanding of the mind. This would require a
propositional attitude to representations which would direct a search for intensional
relations between them; this would allow inference about actions of different
persons associated to each person’s perspective (Rakoczy et al., 2015). Examples are
described below. In these tasks there is an overlap of properties, such that the same
object possesses two properties (e.g., a pencil that rattles, boy Peter who is also a
firefighter). In the set-up of the experiment the child tested is aware that one
property is associated with Box 1 and the other property is associated with Box 2.
However, the protagonist whose action the child must anticipate is not aware of the
overlap: he saw the first property placed in Box 1 and the second in Box 2 but did
not see the transformation of the object from the first to the second property. The
logic of these experiments is shown below and demonstrated in Figure 6.1.
FIGURE 6.1 Examples of theory-of-mind tasks involving appearance transformations

preserving the identity of stimuli involved (Reprinted with permission from Rakoczy
et al., 2015)
(1) There is an A in Box 1.

(2) There is a B in Box 2.
(3) The B in Box 2 is also an A.
(4) The protagonist knows that (1) and (2), but does not know that (3).
Test question: the protagonist is looking for an A. Where will he go to find an

A? (Correct answer: Box 1).
These tasks are solved at the age of 4–5 together with standard false belief tasks,
such as the Sally task. This finding implies that children at this age are capable
of a unified understanding of propositional attitudes and their implications. That
is, they can align specific representations with specific perspectives, thereby
accurately interlinking action with each actor’s perspective: I know that object A
(appearance-representation #1) and object B (appearance-representation #2) are the

same (transformation-representation #3) but look different because they appear
under a different appearance in boxes 1 and 2 (e.g., neutral dress and firefighter
uniform, i.e., identity-representation #4). The child possesses all four representations
and the metarepresentational awareness that actions derive from the representations
one possesses; thus knowing that the protagonist possesses representations #1, #2,
and #4 but not #3 that interlinks the rest, the child anticipates the protagonist’s
actions accordingly. The revolutionary accomplishment of this developmental phase
allows the building of reasoning abilities and the executive possibilities discussed in
the previous chapter.
Halford, Cowan, and Andrews (2007) maintained that this representational
revolution is possible at this age because the relational complexity of false belief tasks
is equivalent to ternary relations that are mastered at this age. Specifically, Halford
and colleagues argued that false belief tasks require to (1) represent the object
location (2) the actual movement of the object that (i) was seen by the child (ii) but
not by the protagonist, and (3) the representations themselves, i.e., what is
represented in one’s own (2i) and the protagonist’s mind (2ii). These researchers
showed that performance on false belief tasks was related to performance on various
other tasks requiring ternary relations, such as grasping the cardinality of number,
transitivity, class inclusion, appearance-reality distinction, and executive control.
Moreover, they showed that 80% of age-related changes in theory-of-mind tasks
were related to the ability to process increasingly complex relations (Andrews,
Halford, Bunch, Bowden, & Jones, 2003).
However, this accomplishment is only a step along a long road of developing
awareness from birth to adolescence. On the one hand, there is evidence that even
15-month-old infants have some grasp of the perception-belief-behaviour connection.
Onishi and Baillargeon (2005) presented the following sequence of events to
15-month-old infants. First an actor saw an object placed in a green box; then the
actor’s view was blocked and the object was moved to a yellow box; then the actor
appears again and looks for the object in the yellow or the green box; infants showed
surprise when the actor looked for the object in the yellow box, against his belief.
This was interpreted to imply that young infants have an intuitive grasp of the
essentials of theory of mind. There is also evidence that children adapt their behaviour
(e.g., to name and point to the place where an object was placed before) depending
on their understanding of the knowledge state of the person they are addressed to
(e.g., if object placement occurred in front of this person or not) (O’Neill, 1996).
Obviously these findings suggest that there is awareness about mental states and their
role in behaviour before the advent of language.
This is reflected in the fact that 2- and 3-year-olds who fail the false belief tasks
are quite capable of deception. From a cognitive point of view, deception implies
that the deceiver recognizes there may be alternative representations of the same
reality and that it is possible to create in the other’s mind a representation which is
different from the representation that she herself holds. In their experiments,
Chandler, Fritz, and Hala (1989) showed that, by the age of 2, children understand
that withholding or destroying evidence can deceive someone and that, by the age
of 3, they understand the role of lying. In fact there is evidence indicating that
3-year-olds can pass false beliefs tasks if they are embedded in a context of deception.
Deception is not the only context in which children demonstrate an understanding
of the other’s mind. Wellman (1990) carried out extensive research to show that
children younger than 4 are much more sensitive to desires than beliefs as mental
states which can produce a response. In his experiments he showed that 3-year-olds
can solve problems like the following: “Sam wants to find his puppy. It might be
hiding in the garage or under the porch. Where will Sam look for his puppy (garage
or porch)?” Three-year-olds were able to correctly predict Sam’s behaviour even
when the representations seemingly changed, as indicated in the following story:
“Before Sam can look for his puppy, Sam’s mother comes out of the house. Sam’s
mom says she saw his puppy in the garage. Where will Sam look for the puppy?”
According to Wellman, these findings indicate that 3-year-olds have a theory of
mind, and he argued further that, as the child’s theory of mind develops, the
importance of desire as a causal agent of behaviour lessens in favour of belief. This
seems to imply that the theory of mind is originally geared to mental states associated
with the dynamic aspects of people’s behaviour (i.e., states which are related to
emotion and motivation) and it then extends to include those states relevant to the
cognitive aspects (i.e., representations).
On the other hand, higher-order theory-of-mind tasks are grasped much later
than false belief tasks. In higher-order theory-of-mind tasks, representations about
knowledge and beliefs are embedded into one another, as often happens in real life.
For instance: {John thinks that [Mary knows that (Michael wanted)] to have an
ice-cream}. Higher-order theory-of-mind tasks may vary in complexity from the
first order, such as the Sally task, to the second order, the third, as in the example
above, or an even higher order. Second-order theory-of-mind tasks are solved in
the early primary school years but third or fourth tasks are solved later, at the end
of primary school (Rakoczy et al., 2015; Liddle & Nettle, 2006). Obviously grasping
multiple-order theory-of-mind tasks indicates the kind of compositionality,
recursivity, and hierarchical integration that is ascribed to language of thought by
cognitive scientists.
Carpendale and Chandler (1996) also showed that understanding the interpretative
nature of mind is attained at the age of 7–8. For instance, preschoolers do not
understand, but primary school children do, that different characters may interpret
the phrase “wait for a ring” (i.e., a phone call or a diamond ring) differently,
depending on the information they have. Obviously understanding of interpretations
requires a more complex understanding of the nature of the mind. This involves
understanding the role of initial premises in the chain of an argument (e.g., wait for
a professional message versus wait for a wedding proposal) and also the inferential
processes that link premises into a sequence leading to a conclusion. It is noted that
Wellman showed, in a series of longitudinal and meta-analytical studies, that the
sequences above reflect genuine changes in the representational and conceptual
abilities of children (Wellman, Cross, & Watson, 2001; Wellman, Fang, & Peterson,
2011). It is notable that the various states of the theory of mind acquired at different
age phases are longitudinally related. Brooks and Meltzoff (2015) showed that infants
who were better at gaze-following at 10.5 months possessed more mental-state
words at 2.5 years; in turn, children who knew more mental-state words at 2.5 years
were better in dealing with theory-of-mind tasks at 4.5 years. These results suggest
that gaze-following in early infancy reflects a broader capability to tune one’s own
behaviour with the mentally bound behaviour of others as reflected in their gaze.
This capability provides the framework for learning mentally rich aspects of language,
such as mental verbs. In their turn, mental verbs provide the representational
framework needed to build the intensional attitude underlying false belief and other
theory-of-mind tasks. Next we will summarize research related to the child’s
understanding of the nature and functioning of different mental functions.
Understanding the organization and functioning of the mind

Research on the development of the child’s understanding of the organization
and functioning of the mind sought to highlight how, if at all, different cognitive
functions and processes are understood at different ages. Flavell and his colleagues
presented a series of ingenious studies about the development of children’s
knowledge about thinking, which they “broadly and minimally defined as
mentally attending to something” (Flavell, Green, & Flavell, 1995, p. v). According
to these studies, the development of even this simple understanding is a process
that evolves over many years. Specifically, preschoolers seem to “have at least a
minimal grasp of the bare-bones essentials of thinking: namely that it is some sort
of internal, mental activity that people engage in that refers to real or imaginary
objects or events” (p. 78). Preschoolers also realize that thinking is different from
perceiving and that it is different from other cognitive processes such as knowing.
In one of their experiments, Flavell and colleagues showed that 3-year-olds
understand that a person who is blindfolded and has her ears closed cannot see nor
hear an object but she can think about this object. Another experiment showed
that 3- and 4-year-olds equally understand that a person is thinking when she is
in the process of choosing one out of a number of available objects or when she
tries to understand how a curious thing happened, such as how a large pear fitted
into a bottle with a narrow neck. Another study showed that young preschoolers
understand that a person can have knowledge of things she is not currently
thinking about.
In line with these findings, Paulus, Proust, and Sodian (2013) showed that
children have some awareness of their own mental states from about the age of 3.
These scholars trained 3.5-year-old children to associate individual animals with
specific objects. They showed them short videos presenting an animal doing
something (e.g., an elephant who likes watching TV). Sometime later they showed
the probe animal (e.g., the elephant) and they tested if children remembered the
object associated with it (a TV). They also asked the children to indicate how
confident they were of their judgment. Confidence ratings for correctly remembered
items were higher than ratings for incorrectly noted items, suggesting an awareness
of representations stored earlier in memory.
However, there are important aspects of thinking that preschoolers do not
understand. Specifically, there is compelling evidence that they do not understand
what William James called the “stream of consciousness”, i.e., they do not realize
that thinking is a process which goes on continuously in people’s minds, even when
they sit quietly and do nothing. In one of Flavell et al.’s studies, preschoolers ignored
very clear cues about the ever-presence of thought activity. For instance, the large
majority of preschoolers refused to agree with the statement “something is always
going on in people’s minds, so there must be something going on”.
Preschoolers also do not realize that cognitive activities such as looking, listening,
reading, and talking necessarily entail thinking. Even when they attribute mental
activity to a person, preschoolers seem unable to specify the content of the person’s
thinking despite very clear and indicative signs. Flavell and colleagues conducted an
experiment confirming this: with a preschool child as the subject, one researcher
(A) asked another (B) a thought-provoking question about an object in the room.
B said to A, “That’s a hard question. Give me a minute”, and she turned to one
side, giving non-verbal cues that she was trying to find an answer to the question.
Preschoolers were not able to indicate that researcher B was thinking about the
object named in the question and many continued to have difficulty with this
seemingly simple problem, even when researcher B stared at and touched the object
while he was thinking about it. In fact preschoolers seem to have difficulty specifying
the content of their own thoughts. For example, when asked to name the room in
their house where they keep their toothbrush they did not mention either a
toothbrush or a bathroom when asked what they had been thinking about.
Because they cannot identify the content of their thought, they are unaware of
cognitive cueing, the associative nature of the mind. That is, they do not realize that
one idea or thought triggers another, which triggers another, and so on. For
example, when told a story about a child who thinks of beautiful flowers while on
the beach, they cannot explain why that child thinks of the beach when he later sees
some beautiful flowers. Finally, preschoolers do not seem to understand that thought
is partly controllable and partly uncontrollable, i.e., that you can start thinking about
something if you decide to but you cannot always stop thinking about something
just because you want to. All of these difficulties diminish considerably or are
removed by the age of 7–8.
The studies reviewed above suggest that preschoolers differentiate thinking from
other cognitive (i.e., perception) and non-cognitive (e.g., movement) activities but
that they do not yet understand how thinking is activated or how it works. Fabricius
and his colleague (Fabricius & Schwanenflugel, 1994) reported a series of studies
concerned with a complementary question in which they examined whether children
understand the similarities and differences between different cognitive functions such
as memory, reasoning, and comprehension. Their studies involved adults and 8- and
10-year-old children. These participants were given simple descriptions of list
memory (e.g., getting all the things at the store that your mother asked you),
prospective memory (e.g., saying happy birthday on the right day to your friend who
told you her birthday a long time ago), comprehension (e.g., learning a new board
game from the instructions on the box), attention (e.g., listening to what your friend
is saying to you in a noisy classroom), and inference (e.g., figuring out what your
friend wants when he says, “Boy, that cookie looks good!”). The participants were
asked to contrast each sentence with all other sentences and indicate the degree
of similarity among the processes referred to in each pair of sentences. It was found
that, from the age of 8, children can distinguish between memory and inference. For
adults and 10-year-olds, but not 8-year-olds, the involvement of memory in tasks is
taken as an indication of similarity between the processes supposedly involved. Unlike
adults, however, neither 8- nor 10-year olds could distinguish between comprehension
and attention or between different kinds of memory. Thus it seems that by late
childhood children begin to distinguish between different cognitive processes. This
differentiation is very global, however, and limited to processes which have clear
experiential differences. In addition, children aged 6–8 do not prepare sufficiently to
cope with a forthcoming task because they are not explicitly aware that different tasks
require relevant preparation. This is attained at about the age of 10 (Chevalier &
Blaye 2016). Children in this phase understand that more difficult items require more
study time if they are to be successfully stored and recalled (Tsalas et al., 2017).
Know yourself
The knowledge about the mind discussed above focuses on specific processes and
states. However, ever since the time of the Greek philosophers, knowing oneself
has been of major concern to our understanding of human thought and action. Kant
and other philosophers noted that intelligence only exists as a part of a knowing self.
In psychology, James (1890) established the self as a central construct that generates
knowledge about one’s own attributes and characteristics and gives meaning to
experience. We will show later that the construct of the self is pivotal in our
understanding of the relations between intellectual development and personality.
In the classical theory of James (1890), the self is a central construct that organizes
and gives meaning to experience. In this theory, the self involves two dimensions:
the “I-self” and the “Me-self”. The I-self comprises self-observation and self-
recording processes. The Me-self includes the knowledge generated by the I-self
about mental, social, personality, and bodily characteristics. James’s distinction
between a knowing (the I-self) and a known self (the Me-self) is present in modern
theories of the self (Brown, 1998; Hattie, 1992; Markus & Wurf, 1987). For
example, in Markus’s model (Markus & Wurf, 1987) the working self-concept is
differentiated from the collection of self-representations possessed by the individual.
The working self-concept involves all presently accessible self-representations and
it is directly involved in the formation and control of behaviour at both the intra-
and the interpersonal level. Therefore, in this model, the working self-concept
assumes the functions of the Jamesian I-self, which generates self-descriptions,
which belong to the Jamesian Me-self. The Me-self is a hierarchical system involving
various sub-systems, such as academic self-concept, social self-concept, etc. In turn,

each of these sub-systems involves more local components, such as, for instance,
self-concept in mathematics, science, language, etc. Obviously the I-self includes
cognizance as specified in our theory. The Me-self includes knowledge and beliefs
systems about the self, as specified in various theories.
The I-self and the Me-self are molar constructs leaning on the processes
generating and modifying a person’s theory of mind and the content that it produces
for oneself. The I-self may be seen as the mechanism that generates the person’s
knowledge of the organization and functioning of the mind. To the extent this is
applied on other persons, then, the I-self becomes the source of the person’s theory
of mind. The Me-self is the crystallized aspect of the functioning of the I-self so
defined. What might the molecular mechanism be?
Theory of mind or self-awareness and mindfulness?

The research summarized above on theory of mind and the understanding of the
organization and functioning of the mind suggests a rather radical conclusion:
human understanding of the mind emerges from a broad, very comprehensive self-
centred and mind-centred monitoring system that attends to, registers, and stores
information about mental functioning and states, but also about other aspects of
behaviour and functioning. This system is associated with self-control and self-
regulation of one’s own actions and interactions with other persons. Therefore the
term “theory of mind” is very limited as a description of the changes occurring in
the child’s understanding of the mind. Even the false belief tasks, so widely used, do
not really tap children’s knowledge of the mind but only one aspect of children’s
understanding of the mind of others (Bloom & German, 2000; Stone & Gerrans,
2006). Figure 6.2 shows a general model that generates understanding and predictions
of mental states without assuming a specialized module for theory of mind.
The various aspects of this system are systematically interrelated. A core
component of the system is attention and attention control. Leslie suggested that
theory of mind is based on a selective attention-inhibition mechanism which directs
the attention of the infant very early in life to attend to mental states such as belief,
desire, and pretending, and to learn about them (Leslie, Friedman, & German,
2004). For instance, infants are attracted by the direction of another individual’s gaze
to an object or location relative to their own knowledge about this object or
location, because this helps them predict the other’s behaviour and interact with her
effectively. Others suggested that this mechanism also calls on other general
mechanisms such as recursion and metarepresentation, which allow inferences about
mental states in the same way that they allow inferences about other states of the
world, such as the biological and the physical world (to be discussed in the next
chapter). Thus, in regard to the mind of others, this mechanism implements the
human social instinct to interact with each other. Through the years it generates
increasingly refined knowledge about one’s own and others’ mental states and their
role in human actions and interactions. In fact a theory of mind that involves refined
FIGURE 6.2 A model generating inferences about mental states without assuming a
ToM module. A general metarepresentational capacity is assumed which uses
representations delivered by lower-level mechanisms to generate inferences about
states in the world, social (e.g., ToM), biological, or physical (Stone & Gerrans, 2006)
representations and attributions about other aspects of human existence, such as

personality and emotionality, is the product of the functioning of this mechanism
rather than its cause. We will discuss these relations later.
It is also noticed that there are large individual differences in the frequency of use
of different types of inner experiences related to reflection, such as inner speech,
inner seeing of visual images, and unsymbolized thinking, such as thinking a specific
thought without the awareness that this thought is conveyed by words, images, or
any other form of symbol. Some individuals never use these forms of “fixing” mental
activities and others do so most of the time (Heavey & Hurlburt, 2008). There is
very little developmental research on these phenomena despite their possible
importance in intellectual development. However, there is evidence that differences
between people in the ability for introspection relate to both self-consciousness and
the ability to attribute mental states to other persons (Frith & Happe, 1999).
Self-reflection is necessary to have attention elevated to explicit awareness. It is
clear that self-reflective awareness is present from late infancy. However, self-
reflective awareness improves with age in both frequency and accuracy. Although
initially overconfident and optimistic, with age it becomes more realistic as children
gradually become better able to record their experiences and feelings of functioning.
In the primary school years, children become increasingly able to evaluate if they have
learned what they are supposed to learn; for instance, to recognize if they learned the
meaning of kanji characters from the Japanese writing system. They can also become
increasingly able to judge if they would be able to use what they learned now in new
situations, or that their self-evaluation of learning is accurate vis-à-vis an independent
evaluator, etc. (see Destan & Roebers, 2015).
Also, improvements in self-reflective awareness relate to the ability to control
thoughts and actions (Lyons & Zelazo, 2011). Specifically, the increasing ability to
accurately record cognitive functioning and experiences and feelings generated by

cognitive functioning enables children to increasingly revisit specific cognitive
procedures, such as learning the meaning of symbols, words, or skills, in order to
modify, adjust, and tune them to a pre-specified mental goal that is also explicitly
represented. There is research showing that varying the representational requirements
or the executive selection processes of false belief tasks influences the performance
of college students and explains the difficulties that elderly persons have in dealing
with theory-of-mind tasks (German & Hehman, 2006).
This line of research assumes that reflection and awareness drive executive
control, which, in turn, drives the development of more complex processes, such
as working memory, theory of mind, cognitive flexibility, and reasoning (Diamond,
2013; Zelazo, 2015). According to Zelazo (2015), the development of executive
control is made possible, in part, by increases in the efficiency of reflective
reprocessing which allow for increases in the hierarchical complexity of the rules
that can be used to solve problems. Specifically, according to Zelazo’s levels of
consciousness (LOC) model, cognitive change comes from self-reflection which
generates increasingly higher levels of awareness. These “. . . are brought about by
a type of reflection or re-entrant processing that permits the contents of consciousness
(i.e., our representations) at one level to be considered in relation to other contents
at that same level, resulting in a more complex conscious experience” (Zelazo,
2004, p. 13). Lyons and Zelazo (2011) argued that these changes underlie changes
in executive control and metacognition. However, Zelazo did not specify how his
LOC relate to reasoning and other aspects of mental processing, such as working
memory and intelligence.
Finally, changes in the self-awareness system also relate to self-concept. There
is research showing that self-evaluation and global self-concept become increasingly
accurate and refined with development (Harter, 2012). Recent evidence suggests
that 4- and 5-year-old children already possess a representation of global self-
worth that is defined in abstract terms and is differentiated from self-representations
about specific characteristics, such as a specific school or sports-related activity. As
a result, a failure in a specific activity may be justified in reference to situational
variables and leave the general self-concept unaffected (Cimpian, Hammond,
Mazza, & Corry, 2017). In fact, by middle childhood the self-system differentially
relates to different realms of experience.
Along the same lines, a recent study showed that the difficulties of autistic
children in shared attention and theory of mind relate to difficulties in self-
categorization. Specifically, this study examined the ability of children to categorize
themselves in reference to personality characteristics related to the Big Five factors
of personality discussed in Chapter 18. They found that children who were high in
autistic characteristics were low in the accuracy of self-categorization and shared
attention. Accurate self-categorizations were related to high shared attention
(Skoritch, Gash, Stalker, & Zheng, 2017).
A recent study involving 7- to 9-year-old children showed that, by the end
of second grade, metacognitive control relates to executive functioning and
metacognitive monitoring relates to self-concept. Of these four constructs, executive

functioning was found to relate to mathematics and both executive functioning and
metacognitive control were found to relate to language. Moreover, over the span
of one year from first to second grade, executive functioning at first grade predicted
the state of metacognitive control at second grade. Self-concept at first grade
predicted metacognitive monitoring at second grade (Roebers, Cimeli,
Röthlisberger, & Neuenschwander, 2012). All in all, a central self-system generates
awareness about the self which is also related to others via what was called the theory
of mind.
Conclusions
The research reviewed above suggests some clear answers to the questions asked
at the beginning of the chapter. Do children understand the mind as something of its
own that differs from reality? They clearly do from a very young age, although this
understanding develops and becomes more refined with age. They also understand
that the mind is representational, generating representations for reality, depending on
various sources, such as perception and learning from others. These representations
are interpreted as causal origins of desires, beliefs, actions, and other knowledge. This
understanding also develops throughout infancy, childhood, and adolescence. Finally,
with age, individuals acquire an increasingly refined and differentiated knowledge of
the composition and organization of the mind.
How and why does understanding of the mind develop? There have been several
hypotheses explaining why awareness of the mind changes as children grow older;
these are complementary rather than incompatible. The first ascribes development
to the increased activation and functioning of one’s own mind. That is, as they grow
older children engage in activities and problem-solving which require them to
activate different mental functions, often unsuccessfully. For example, when an
unpleasant thought pops into their mind that they want to stop, children may realize
that this is not always possible as the thought comes over and over again. Or, when
asked to explain something to somebody, they may realize that they do not have all
the information and skill necessary to do so (Flavell et al., 1995).
Later, in elementary school, children engage in problem-solving activities in
different domains. For instance, they read, they do mathematics, they write stories,
etc. These activities drive children to realize that each domain requires different
mental operations, such as attention in reading, calculation in mathematics, memory
in making up a story. On these occasions children gradually come to “see”, so to
speak, their actual mental processes as processes rather than just as products of the
functioning of these processes. Thus they become sensitive to the presence of
different functions and purposefully act to make them work efficiently. This implies
that the development of theories and problem-solving about other domains of the
world is conducive to the development of the theory of mind itself.
The second hypothesis stresses the social dimension of the discovery of the mind.
According to this hypothesis, problem-solving in humans frequently occurs in
groups. Thus people have the opportunity to observe others trying to solve the same
problem. This is especially the case in the world of the school, where children see
each other trying to learn and solve problems in various domains. Of course, what
is going on in another person’s mind is completely private. However, in environments
targeted to problem-solving, such as the school, children exchange experiences and
they may check each other’s representations and procedures. These experiences
generate information, concepts, hypotheses, and models which gradually become
more refined, focused, differentiated, and accurate (Demetriou & Efklides, 1985;
Demetriou & Kyriakides, 2006). Thus awareness of the mind gradually gears on
three assumptions: that the mind is (1) private but disclosable at will and in shades
needed to obtain specific results; (2) complex, thus involving many different
functions; and (3) constructive, and thus part of the reality one is dealing with.
The third hypothesis builds on and integrates the two hypotheses above. This
relates to the role of awareness in the development of other processes. Specifically,
as awareness of the organization and functioning of the mind grows because of the
factors above, children become more proficient in using the processes they become
aware of. For instance, knowing that controlling attention helps them to read better,
commanding arithmetic operations helps them to calculate without errors, and
controlling recall helps them to write better stories, children intentionally turn to
these processes for the gains they offer. This becomes a self-development loop that
drives the development of self-awareness, self-regulation, and the various domain-
specific processes involved. By definition, then, knowing and controlling the mind
becomes a domain-free process underlying intellectual development and individual
differences in mental functioning and intelligence. We will return to these questions
in the following chapters.
7
CORE DOMAINS
Domain-specificity is of concern to developmental research as much as it is to

cognitive and psychometric research. In fact domains were present even in theories
which stressed the dominance of general cognitive mechanisms, such as Piaget’s
theory itself. In Piaget’s theory, all of Kant’s categories of reason—such as quality
(related to categories), quantity, space, causality, and time—figure prominently as
domains of understanding. However, Piaget postulated that the same mental
structures of reasoning run through all of the domains, driving them to develop in
concert. Neo-Piagetian theories replaced structures of reasoning by representational
and processing constraints but preserved the fundamental assumption of a central
developmental driving mechanism.
We noted in our evaluation of these theories that research showed there are
considerable differences within and across individuals in developmental rate and final
state across domains. Some individuals develop quickly and attain high-end states
in some domains but not in others. In fact high achievements in some domains may
sometimes co-exist with developmental disabilities in others. For instance, autistic
individuals are extremely handicapped in the social domain but they may be relatively
intact in other domains, such as memory or numerical reasoning (Rinaldi &
Karmiloff-Smith, 2017). In addition, learning in one domain does not always transfer
to other domains. We will see that learning research aspiring to increase intelligence
ended in disappointment: learning gains fade rapidly soon after interventions end
and do not transfer to other domains (Protzko, 2015; Salomon & Perkins, 1989). In
response to this state of affairs, a line of developmental research assumed that domains
of understanding are functionally and structurally autonomous of each other. The
implication is: if you want progress of understanding in a domain, teach the mental
processes specific to this domain and do not expect transfer to other domains. This
assumption imports in developmental theory postulates from the other disciplines
discussed in this book. It is worth remembering that the strong modularity theory
Core domains 83
in the cognitive tradition assumes that modules are informationally encapsulated and
impenetrable by each other. The “special ability” theory in the differential tradition
assumes that there are independent mental abilities, or “multiple intelligences”.
There may be different criteria for differentiating between domains. These range
from the assumption that domains are biologically hardwired in the genetic make-up
and the brain, to the assumption that domains are learned systems of processes
imposed by the environment. According to the first assumption, domains are
biological adaptations that evolved over evolutionary time as responses to adaptive
pressures. That is, each domain evolved to recognize patterns of information which
are biologically important for a species and efficiently deal with them without
depending on learning, which may not always be provided. The domains discussed
under this assumption only partly overlap with domains discussed in more traditional
disciplines. In the present context, domains co-extend with broad realms of the
world, such as animate and inanimate entities. Under this assumption, thought
domains entail the automatic differentiation between corresponding domains in the
world and grasp of the basic principles underlying their operation: for instance,
automatic recognition of biological entities as contrasted to physical entities;
recognition of members of the same species versus everything else; and understanding
of the basic principles underlying the psychological world of humans and possibly
other animals that are important for human functioning.
This line of research postulates that specific patterns of information in the input,
such as self-initiated movement in animals, the eye-nose-mouth pattern of the
human face, eye gaze, etc., are enough to trigger the relevant processing mechanism
in the brain which ascribes the “meaning” programmed in it. For instance, self-
initiated movement renders an entity a living being; a specific eye-nose-mouth
pattern results in automatic recognition that this creature is a human. It was argued
in the previous chapter that following eye gaze by an infant indicates an attention
mechanism specializing in registering and interpreting intentions for action by
others. The assumption is that domains evolved as specialized adaptations to ensure
efficient functioning, given the living conditions of the animal.
Another line of research maintains that domains are fields of knowledge that
evolved over the years in human civilization in general or in different cultures (Na
et al., 2010). Thus civilization and culture function as general frames where
individual development or individual differences occur. For instance, a reflective
style which often makes an individual relatively slow in decision making, as opposed
to a swift processing of information, may be channelled by different cultures into
what is a preferable way to deal with problems. Broad cultural productions, such as
the sciences, are also examples of broad domains of knowing that may frame how
legitimate knowledge is produced at various levels. For instance, interaction with
other humans, mathematics, physics, biology, etc., are all complex realms of
knowledge that need to be learned as such. Making meaning in each domain,
drawing inferences, and solving problems require mastering the relevant languages,
rules, and constraints. Differences between domains in each of these aspects explain
why transfer between them is limited.
A third line of research straddles these two extremes. This line views domains as
functional specializations for dealing with special types of relations in the environment
(Barrett & Kurzban, 2006), and argues that domains may emerge from specific core
recognition-processing mechanisms that have a strong biological-evolutionary origin.
For instance, the recognition of conspecifics, the automatic recognition of small
numbers, and categorical perception based on colour are all examples of core processes
that may involve a mechanism that automatically responds to a specific pattern of
information in the environment. Later, however, developmental pacing and diversi-
fication may be the result of learning, because these core mechanisms are embedded
in culturally rich domains of knowledge, which individuals need to master. For
instance, recognition of conspecifics must be embedded in social norms underlying
interaction with different groups of people in a culture. Automatic number recognition
must develop into more elaborate mathematical knowledge, as in algebra. Automatic
colour perception must be integrated into colour-related categories dominating in a
culture. In the course of mastering a domain, general cognitive mechanisms may be
as important as the initial core mechanisms because they underlie the naturalization
of core mechanisms in complex, information-rich domains, such as mathematics,
science, etc. We will now outline research conducted along these three lines.
A special version of the conception of domains as functional specialization is
known as the “theory-theory” interpretation of the developing mind. According to
the theory-theory interpretation, children’s understanding may originate from core
processes but it develops like scientific theories. In science, the term “theory” refers
to an organized body of knowledge and ideas about a particular aspect of the world
which coherently describe and explain the phenomena of interest. However,
scientific theories are modified or dropped if evidence systematically contradicts
their postulates. According to this view, cognitive development is like theory
change in science. Thus, in this approach, domains are thought to originate as
biological adaptations and develop as gradually modified scientific theories. These
include the animate-inanimate distinction, and understanding of the biological, the
psychological, and the natural world. In this approach the theory of mind examined
in the previous chapter is a special case of theory that focuses on the human mind.
Obviously these domains deviate considerably from the traditional definition of
domains, which is based on the nature of information and its perceptual basis, such
as visuo-spatial and acoustic-verbal information.
In any case, these differences suggest that rendering domain the primary unit of
analysis of intellectual functioning and development may be a slippery ground for
science, as it makes our understanding of the human mind unstable and captive to
the variations between domains. A more constructive approach is to focus on robust
psychological processes that give stability and cohesion in the developing person’s
dealing with the world. Under this approach, domain-free mechanisms are important
for dealing with domains and domain-specific mechanisms are important for the
enhancement and refinement of domain-free mechanisms. Thus in this chapter we
will present research on the domains as specified above; this will be useful for our
understanding of the general mechanisms to be discussed in the following chapters.
Core domains 85
Physical, biological, and psychological worlds

What is the ontological status of the physical, biological, and psychological worlds
in the minds of children? Are these three aspects of the world correctly discriminated
from an early age? For instance, do young children realize that a chair (a physical
thing) is not the same as thinking about this chair (a psychological thing); or do they
realize that “eating food” (a biological function) and “wanting not to get fat” (a
desire that is a psychological function) are not the same? Do they understand that
the statue of a person or animal (a physical object) is not the same as the person
herself or the animal itself (a living entity)? (Gelman, 2005). Research on these
questions showed that by the age of 3, if not earlier, children do discriminate
between the physical and the psychological worlds. In a study by Harris, Brown,
Marriott, Whittall, and Harmer (1991) children were told of a boy who has a dog
and another boy who is thinking about a dog; the children were then asked to judge
which dog could be seen, touched, and petted. Three-year-olds realized that only
the first boy’s dog could be available for these activities.
Gelman (1990) examined the animate-inanimate distinction. Specifically, she asked
children to report the contents of the insides of various animate creatures and inanimate
things. She found that even 3-year-old children report that living beings have blood,
bones, and muscles inside them, whereas inanimate things have materials such as
cotton, paper, hair, or “hard stuff” inside. Likewise, Massey and Gelman (1988)
showed that, from the age of 3, children discriminate between animates and inanimates
on the basis of life-specific characteristics, such as self-initiated movement and the
capacity to grow. They showed, for instance, that children classify very realistic statues
of animals with inanimate objects (for instance, they were judged unable to move up
a hill) and they group highly atypical animals (e.g., porcupines) with other animals.
However, some aspects of the animate-inanimate distinction are understood very late
in development. For example, it is only after the age of 10 that children understand
that plants are “alive” (Carey, 1985). This may indicate that early distinctions come
from primary information activating a core meaning-making process. For instance,
moving according to a specific pattern indicates living creatures. However, more subtle
differences (e.g., reproduction or energy exchange with the environment) need to be
learned and this is not possible before abstract principles are grasped (what energy is).
Children also discriminate between the psychological and the biological world
from an early age. Inagaki and Hatano (2002) asked children to predict who would
become fatter, a girl who wants to get fat but eats less, or a girl who wants to get
slim but eats more. They found that, at the age of 4, children predict correctly,
indicating they understand that a bodily process is affected by biologically relevant
behaviour rather than by will, which is a psychological function. This same study
showed, however, that, although able to make the distinction, children do not
understand in detail many biological processes which go on inside the body, such
as perspiration and digestion.
Despite their precocious sensitivity to quite complex aspects of the world,
children’s representations lack adult-like stability and they are quite simplistic and
rigid. Evidence indicates that a child’s ability to differentiate among physical,

biological, and psychological entities does not in itself guarantee that they think of
objects and persons as having an identity that is largely independent of superficial
changes in appearance. That is, research on what is known as the appearance/reality
distinction strongly indicates that changes in the appearance of objects or living beings
make young children believe that they also change in their very identity. For example,
Flavell and his colleagues transferred milk from a regular glass into a red glass and the
children were asked two questions: “How does it look to your eyes right now?” and
“How is it really and truly?” Children younger than 4 insisted that the milk transferred
to the red glass “really and truly” was red. In fact children continued to commit this
phenomenism error even after they were trained to make the appearance/reality
distinction (Flavell, Green, Wahl, & Flavell, 1986). Children are also prone to another
type of error, the intellectual realism error, which is complementary to the phenomenism
error. Children mistakenly believe that the object they see (for example, a sponge)
looks like the object they know it to be (a sponge), although the object has been
made to look like something else (a rock). That is, they impose what they know on
what they see.
Causal distinctions
The three aspects of the world under discussion (physical, biological, and
psychological) are distinguished by certain mutually exclusive characteristics.
Physical causality. Butterworth (1998b) suggested that very young infants are
sensitive to patterns of interaction between physical bodies, grasping specific types
of causal relationships, such as the transmission and direction of movement, physical
support or occlusion relations, etc. For instance, if a particular object starts moving
and hits another object, it is considered the cause of the movement of the second
object. This evidence suggests that our perceptual systems, vision in particular, are
able to abstract automatically certain types of causal relationships.
Research with toddlers suggests that the ability to represent causal relationships
appears at about the age of 3. According to Shultz (1982), the first representation of
causality is dynamic or generative. That is, children think there is a causal relation
if there is transmission of energy or power from one physical body to another.
Interestingly this understanding of causality as a generative relationship overrides its
understanding as a covariation relationship, a similarity relationship, or a relationship
defined by temporal and spatial contiguity. In one of his experiments, Shultz (1982)
used a candle, two air blowers, and a shield located between the candle and the
blowers. The researcher lit the candle in front of the child, turned one of the blowers
on, and then removed the shield so that the candle went out. Children as young as
2 were able to identify which blower made the candle go out. They also invoked
generative transmission to explain what happened: “The white one because it blew
it. The green one didn’t because it didn’t go.”
Despite this precocious sensitivity to causal relations, understanding of the causal
structure of the world is a slow and cumbersome process. Research into the
Core domains 87
understanding of common aspects of the world—such as force and motion and the
day/night cycle—indicates that misconceptions may persist in adulthood. For
example, many adults have difficulty integrating into their model of motion the
Newtonian principle that there is no motion without a cause (Bliss & Ogborn,
1994). Many adults also believe that the Sun and the Moon move up and down at
the opposite sides of the Earth. Misconceptions frequently coexist side by side with
scientific models of the world. For instance, to reconcile their experienced-based
intuition that the Earth is flat with the scientific model that the Earth is round, some
adults believe that the Earth is a hollow sphere with people living on a flat surface
in the middle of it (Vosniadou, 1994).
Biological causality. Biological causality refers to the transfer of effects which are
limited to animates and related to their living identity, as distinct from other
characteristics which they may have as physical bodies. For example, living beings
inherit structural or functional characteristics; this is not true of inanimate objects.
Springer and Keil (1989) showed that toddlers believed a baby animal whose parents
had a peculiarly coloured pink head was more likely to have a pink head than a baby
animal whose parents had a normally coloured head. However, genetic mechanisms
are not fully understood, even by educated adults (Caravita & Hallden, 1994).
Psychological causality. Psychological causality involves understanding the causes
of human behaviour. Obviously theory of mind is a form of psychological causality
as it is concerned with the mental causes of human behaviours and interactions.
The reader is reminded that the theory of mind was extensively discussed in the
previous chapter as part of the self-awareness system. Suffice it to highlight here
those aspects of the theory of mind that are common with every other form of
causal thought, namely: (i) decoupling or isolation of factors involved in a causal
sequence, such as perception causing representations which cause beliefs which
cause actions; (ii) recursion, such as beliefs embedded in a number of persons involved
in an action sequence of the type {A thinks that [B thinks that (C thinks)] . . .};
and (iii) prediction, because theory of mind, like all causal models, is useful as it
allows predictions of future events based on events already known (Schaafsma,
Pfaff, Spunt, & Adolphs, 2015).
In conclusion, the evidence summarized above suggests that children’s repre-
sentation of the world is organized in a way that honours the three broad domains
it involves. This indicates that children abstract the specific ontological and dynamic
characteristics of the physical, biological, and psychological aspects of the world and
organize their knowledge and behaviour accordingly. At the same time, however,
we need not overestimate these early achievements because they coexist with
misconceptions that persist for a long time.
Central conceptual structures

Case (1992; Case & Okamoto, 1996) suggested that domains exist but they are
semantic structures mapping different realms of knowledge and skills that need to
be learned as such in development. Specifically, Case identified several central
conceptual structures which coincide, by and large, with Gardner’s intelligences:

quantities, space, social behaviour, narrative, music, and motor behaviour. Case
defined central conceptual structures as networks of semantic nodes and relations
organized around a set of core processes and principles running through a broad
array of situations. For example, the concept of “more and less” for quantities,
“adjacency and inclusion relationships” for space, and “actions and intentions” for
social behaviour are core processes for the first three central conceptual structures,
respectively.
Case suggested that executive control structures in each central conceptual
structure are specialized action plans built around the core processes of the structure
concerned with representations and concepts related to a specific sub-domain. For
example, there are executive control structures for the solution of arithmetic
problems, the use of balance beams, the representation of home locations according
to their street address, etc. All of them involve the core concept of “more or less”
that defines a quantitative dimension expressed in the sub-domain of numbers,
distances on a balance, and odd and even numbers specifying addresses for each of
the two sides of a road, respectively (Case, 1992; Case et al., 1996).
Grasping the core elements of a central conceptual structure opens the way for
swift acquisition of a wide array of executive control structures in the domain
concerned. However, special knowledge structures need to be learned as such. For
instance, the use of odd numbers to designate home addresses on one side of the
street and even numbers for the other side is a special convention that needs to
be learned, and it does not come from the number sequence as such: this is a single
continuous dimension including odd and even numbers alternating to infinity the
one after the other. Thus learning in one conceptual structure does not necessarily
generalize to other conceptual structures, indicating that there may be variations
both within and across individuals in the executive control structures that can be
constructed within each central conceptual structure. These variations depend on
the environmental support provided to each structure and on an individual’s
particular preferences and involvement. However, there is an upper ceiling to the
complexity of executive control structures that can be constructed at a given age,
regardless of learning within each central conceptual structure. This is total operating
space and processing efficiency. Thus, following Piaget, the neo-Piagetians replaced
general reasoning constraints running through the domains with processing
efficiency, in effect playing down the functional autonomy of the domains.
Explaining the development of core theories

Theories about development of children’s minds may change under three conditions.
First, increasing experience with the phenomena concerned shows the theory to be
inadequate to describe or explain these phenomena. For instance, predictions based
on the theory that numerosity alters with changes in the physical arrangements of
objects are contradicted by counting, which shows that the same number comes up
regardless of arrangement if nothing is taken away or added. Thus children’s actions
Core domains 89
generate information suggesting that theory must change in specific directions. This
condition is very similar to theory change in science and is probably responsible for
the adoption of the conceptual change model of science as a model for mental
development.
Second, the mind should possess the minimum representational capabilities that
are required for an adequate representation of these phenomena and the evidence
contracting the theory involved. For instance, Flavell and his colleagues (1986)
attributed children’s difficulty to conceive of the appearance reality distinction to
their inability, up to 3–4 years of age, to keep in mind dual encodings for the same
object (i.e., to simultaneously represent an object in two different ways). Their
judgment is based on the only representation available to them, which is generally
the most salient feature of the object or situation. Along this line one may inter-
pret the difficulty of children to deal with theory-of-mind tasks. These tasks, by
definition, require at least dual encodings or representations to be solved as indi-
cated by a theory of mind: to have a theory of mind one must recognize that the
same reality can be represented in at least two different representations, either by
the same person or by two different people. In mathematics learning, teachers are
very aware of the difficulty young primary students have in understanding the
mechanism of applying the four numerical operations on fractions. This is because
it requires the capacity to construct complementary representations of numbers (i.e.,
integers and fractions), understand that numerical operations apply differently in
each class of numbers, and construct the skill to properly apply them in each class
(Braithwaite, Pyke, & Siegler, in press). However important this condition is for
mental development, it is basically irrelevant in theory change in science. Obviously,
in science, everyone involved in theory development possesses the representational
capabilities to deal with the constructs involved.
Third, meta-conceptual awareness is needed; that is, understanding that theories
are just complex representations about the world and not the world itself and thus
they are amenable to falsification and improvement. This awareness builds up in
mental development; in fact it is part and parcel of development. The research of
DeLoache is relevant here (DeLoache, 2000). She showed that a metarepresentational
understanding of the role of symbols builds up in the period from 2–4 years. In
addition, she and her colleagues showed that by 2.5 years of age children can use a
picture to find where a toy is hidden. This indicates they understand the relation
between the picture and its referent. However, at this age children cannot use a scale
model of a room to retrieve an object from the room. That is, although they are
asked to examine the scale model in order to see where to look for the object in the
room represented by the model, children younger than 3 fail to use this seemingly
realistic information to retrieve the object. According to DeLoache, this difficulty
stems from the fact that scale models require dual representation. That is, there is a
need to understand that scale models have a concrete aspect which makes them
what they are (i.e., objects which have an identity of their own) and an abstract
function, which makes them symbols of something else. To be able to look for the
correspondence between the scale model and the room, children must be able to
differentiate between the concrete aspect and the abstract function and focus on
the second. Children younger than 3 do not represent the abstract function of the
model, thus they deal with it as though it is an object to be used on its own.
DeLoache has conducted an ingenious experiment to show that, when the need
for dual representation is eliminated, 2.5-year-olds can use scale models as a source
of information for the room. Specifically, she led children to believe that the real
room was put into a “shrinking machine” that reduced the room into the scale
model. Under this condition the model is not a symbol of the room anymore; it is
the room itself. Thus no assumptions about the representational nature of the model
are required to retrieve the information about the room. The ability for dual
representation is established by the age of 4. As a result, children become able to use
various types of symbol systems, such as maps, to guide their actions in actual
environments.
Obviously metarepresentational development is irrelevant for theory change in
science. Every scientist knows that scientific theories are models of an aspect of the
world that may be falsified and thus dropped under certain conditions. In conclusion,
intellectual development may indeed involve modifications of concepts and knowledge
in the minds of developing persons that do have similarities with modifications of
constructs and theories in science. However, we draw attention to a huge difference
between developmental changes in individual minds and epistemological changes in
scientific fields: developmental changes in the minds of individuals are constrained by
genetic, brain, and social/cultural forces related to the adaptation of an individual to
his or her age and environment. Epistemological changes in scientific theories are
constrained by formally specified rules and standards about how science is done which
relate to historical, societal, and cultural forces that operate on collective organi-
zations rather than on individuals. The time-scale is also very different. In individual
development it is the period from birth to maturity in some cases, and the human
lifespan in others. In scientific development it is historical time, which may span
centuries. Thus it is no wonder that eventually the strong domain specificity approach,
regardless of the flavour, faces problems as significant as the solutions it arrives at. It
cannot account for the operation of across-domain constraints in development and
the fact that there are eventually across-domain transfers of learning, if learning is
properly designed. We will return to these issues in the last chapters of the book,
which are dedicated to learning and education. We will show how “theories” as
mental constructs changing during individual development may differ from scientific
theories changing over historical time.
Conclusions
Assuming the operation of domains is less parsimonious than assuming the operation
of a general omnipresent mechanism of understanding, such as g in differential
psychology or the central structures assumed by Piaget or the neo-Piagetians.
Domains impose more complexity than a central mechanism in modelling and
predicting learning and development. However, domains were postulated in all
Core domains 91
traditions as a solution to a problem that a central mechanism cannot solve: variation

of performance across domains. In the developmental research summarized in this
chapter, some theorists threw the baby out with the bath water and abandoned the
assumption of central mechanism altogether.
It is a credit to research in the developmental traditions that they mapped domains
operating from the very early days of life. The domains reflect types of knowledge
related to the organization of the world, such as the animate and the inanimate
sphere, or the physical, social, and biological world. The research reviewed here
makes it clear that children can make distinctions between these domains and possess
concepts about them in infancy and early preschool age. The early emergence of
these domains of understanding was taken to imply that these domains are innate
and autonomous in their functioning, free of any dependence on a general central
mechanism of thought.
There is, of course, no definitive evidence to prove the existence of domains
once and for all. One might argue that, however early these domain-specific
distinctions appear, they are never early enough to preclude a central understanding
mechanism that abstracts domain-specific patterns and builds concepts about them.
Interestingly there is recent research to suggest that conceptual domains of the
nature studied here emerge as a result of a primitive mechanism of induction based
on probabilistic learning; this abstracts categories based on dominant or recurrent
characteristics in the objects standing for domain or a particular type of relation,
such as causality (Hupp & Sloutsky, 2011; Lake, Salakhutdinov, & Tenenbaum,
2015). Obviously these findings suggest that the time is ripe for an overarching
theory that would accommodate both the operation of a central mechanism and the
operation of domains. One such theory is presented in the following chapters.
PART II
An overarching theory
of the growing mind
8
THE ORGANIZATION OF THE
HUMAN MIND
Despite their differences, the three traditions converge in several assumptions about
the architecture of the human mind. They all agree that the human mind is a
complex universe of different systems of mental processes that carry out different
tasks during real-time problem-solving. These systems are as follows:
1. All traditions recognize domain-specific systems interfacing the mind with

different aspects of the environment, enabling people to make sense of their
ongoing relation with the world. Baddeley’s perceptually based visual and
acoustic STSS in the experimental tradition, Carroll’s broad abilities in the
psychometric tradition, and ontological categories in the developmental
tradition are examples of domain-specific systems which overlap to a large
extent across traditions. From now on we will use the term specialized capacity
system (SCS) to refer to these systems.
2. They also all recognize a central relational mechanism enabling thinkers to
combine and integrate information. Reasoning in its various manifestations
(inductive, deductive, analogical, etc.) is the basic tool of this mechanism in all three
traditions. This mechanism integrates across stimuli and representations, checks
for consistency, evaluates relevance of interpretations, and builds concepts that
may be called on in the future for the sake of mental economy and efficiency in
understanding and dealing with the world.
3. In the cognitive and the developmental tradition, but not in the psychometric
tradition, researchers recognize a mechanism of awareness. This is responsible
for the explicit representation of information or one’s own interpretations,
and for monitoring and regulation of the processes activated at a given moment.
Executive control and consciousness in the experimental tradition and executive
96 An overarching theory of the growing mind
control, metacognition, theory of mind, and reflecting abstraction in the

developmental tradition are variations of the same self-awareness and self-
control mechanism.
4. Finally, all traditions recognize a representational capacity that protracts
information not currently present in the senses for the sake of rendering it
available to all of the mechanisms specified above. This is STSS (included in
working memory) in all three traditions.
This architecture is illustrated in Figure 8.1 and this chapter will summarize research
focusing on the architecture. In the first section we summarize research substantiating
the four-fold architecture. The aim here is to substantiate two important assumptions:
first, each of the four types of processes emerge as an autonomous construct
underlying cognitive performance; second, the various constructs from the other
three constellations, (i.e., the specialized domains, inference, and representational
and processing efficiency) are projected onto awareness, rendering them available
to management and metarepresentation.
The second section will focus on research exploring domain-specific systems,
while the third section will focus on g. The aim here is to show that the structure
of the human mind reminds the structure of matter in physics. In Democritus’ times,
the “atom” (in Greek “something which cannot be partitioned”) is decomposed
into elementary particles such as the proton and the electrons, which are further
decomposed into quarks, etc. In this fashion, constructs at each level of analysis can
be decomposed into more basic units so that specifying the factors holding these
units together is vitally important for understanding the functioning of the human
mind and individual differences in how efficiently it is brought to bear on problems.
FIGURE 8.1 The four-fold model of the architecture of the mind

The organization of the human mind 97
Empirical mapping of the four-fold architecture

Substantiating the four-fold architecture is empirically very demanding. It requires
that many individuals are examined by tasks addressed to all of the systems involved
in this architecture. Participants would have to be examined on several specific
processes in several SCSs so that the domain-specific systems may be identified as
autonomous entities. The model assumes that these systems are projected onto the
awareness system; otherwise consciousness would not be able to exert its evaluative
and directive functions in reconstructing current representations and processes into
better ones. Therefore participants would also have to be examined on various
aspects of their awareness about specific processes. Examples of appropriate tests of
these functions include direct examination of their grasp of similarities and differences
between tasks addressed to the various SCSs, evaluations of their own performance
on the various tasks, and evaluations of the cognitive demands of the tasks. The
model presumes that the SCSs would emerge as separate entities from self-awareness
measures in the fashion they emerge from actual performance. Third, representational
capacity and efficiency must also be independently examined. Including measures
of working memory, attention control, and flexibility would highlight if represent-
ational capacity emerges as an autonomous entity. Including measures of executive
control would also allow testing into how representational capacity interacts with
consciousness. Finally, reasoning itself must also be independently examined. That
is, participants would have to solve problems directly addressing deductive and
inductive reasoning to separate these processes from their use in the context of the
various SCSs. This would highlight how integrative processes are implemented by
each SCS, constrained by representational capacity, and directed by consciousness.
Figure 8.2 shows how the general four-fold model would appear in the conventions
of structural equation modelling.
We modelled the results of several studies according to the specifications above.
One of these studies involved children aged from 4–7 and living in Greece and
China. This study allows one to test if the four-fold architecture is valid across age
phases and cultural contexts. Children in this study were examined by various tasks
addressed to two SCSs (quantitative and spatial), deductive and inductive reasoning,
representational capacity and efficiency (attention control, working memory), and
awareness about the procedural similarities and relative task demands of tasks addressed
to each of the SCSs and types of reasoning. The model tested and the main results of
this study are summarized in Figure 8.2. It is stressed that the four-fold architecture
was very powerful across cultures and age phases. There were some differences in the
relations between g and SCS in the 4–5 years phase as compared to the 6–7 years
phase, but these differences will be discussed in depth in the next chapter, which
focuses on development (Kazi, Demetriou, Spanoudis, Zhang, & Wang, 2012).
The second study involved children from 9–15 years. In the same fashion, these
children were examined on three SCSs (quantitative, spatial, and causal), inductive
and deductive reasoning, attention control, working memory and flexibility, and
evaluated on the success and demands of the tasks addressed to the three SCSs
FIGURE 8.2 Confirmatory factor analysis model for reasoning domains, their
self-representation in cognizance, and processing efficiency and capacity,
and self-representation of the cognitive processes
and reasoning. In the fashion above, the four-fold model was tested in a two-group
set-up involving children aged 9–11 and 12–15. As above, the four-fold architecture
was very powerful in both age phases. Some differences between the phases indicated
that the transition from one developmental cycle to the relations between processes
and their contribution to the formation of g varies according to the representational
needs of each phase (Makris, Tahmatzidis, Demetriou, & Spanoudis, 2017).
Specialized domains of thought

A domain of thought is a system of mental processes specializing in the representation
and processing of particular types of objects and relations in the environment. As
such, the domains differ from each other in the mental processes used to deal with
relations in each domain and the representations that stand for these relations. We
will specify these differences below. Our research identified six domains of thought:
spatial, categorical, quantitative, causal, social, and verbal. Through the years we used
several names to denote the domains, such as “specialized structural systems” (SSS)
(Demetriou & Efklides, 1981) or “specialized capacity systems” (SCSs) (Demetriou
et al., 2002). The terms reflected the change in our views about the nature of the
domains. The name SSS indicates that the domains were regarded as structurally
autonomous systems of thought without any specific reference to their origin; that
is, whether they originated from learning from different domains of relations in the
environment or from some kind of hard-wired organization of dedicated neural
networks serving each domain. The term SCS reflected this second possibility; that
is, that they have a hard-wired origin which renders them distinct capacities rather
than just the products of learning. It will be shown below that currently we favour
the second interpretation. Thus we prefer to call them SCS, although we recognize
that they have a strong learning component as they develop in interaction with the
central systems of thought. We will return to this question in the next section.
Each SCS is a hierarchical organization involving three types of processes: (i) core
processes, (ii) mental operations (or rules), and (iii) knowledge and beliefs about
objects and persons. The processes associated with each SCS are summarized in
Table 8.1.
Core processes are innate predispositions to grasp specific types of relations in
the environment that are important for normal functioning. These relations are
so important for survival and routine functioning in the environment that they
are seeded into the perceptual systems themselves and the brain. These processes are
obviously the result of our evolution as a species (Cosmides & Tooby, 1994). An
TABLE 8.1 The three levels of organization of each specialized system of thought
Domain Core Processes Mental Operations Knowledge

and Beliefs
Categorical Perception according Specification of the Conceptions and

to perceptual semantic and logical misconceptions about
similarity; inductive relations between the world
inferences based on properties, classification;
similarity-difference transformation of
relations properties into mental
objects; construction of
conceptual systems
Quantitative Subitization; counting, Monitoring, Factual knowledge
pointing, bringing in, reconstruction, execution about the quantitative
removing, sharing and control of aspects of the world,
quantitative algebraic and statistical
transformations, the four inference rules
arithmetic operations
Spatial Perception of size, Mental rotation, image Stored mental images,
depth, and integration, image mental maps, and
orientation; formation reconstruction, scripts about objects,
of mental images location and direction locations, scenes, or
tracking and reckoning layouts maintained in
the mind
(continued)
TABLE 8.1 The three levels of organization of each specialized system of thought
(continued)
Domain Core Processes Mental Operations Knowledge

and Beliefs
Causal Perception of overt Trial and error; Knowledge,

and covert causal combinatorial operations; attributions and
relations hypothesis formation; understanding of the
systematic reasons underlying
experimentation (isolation physical and social
of variables); model events and the
construction dynamic aspects of the
world
Social Recognition of Deciphering the mental Systems of social
conspecifics, and emotional states and attributions about
recognition of intentions of others; other persons, their
emotionally laden organization of actions culture and their
facial expressions accordingly; imitation; society
decentring and taking the
other’s perspective
Verbal Use of the Identifying truth in Knowledge about
grammatical and information; abstraction grammar, syntax and
syntactical structures of information in logical reasoning;
of language goal-relevant ways; metalogical
differentiation of the knowledge about
contextual from the nature and
formal elements; justifiability of logical
elimination of biases from inferences;
inferential process; metacognitive
securing validity of awareness,
inference knowledge, and
control of inferential
processes
example here would be colour perception which is given in visual perception. That
is, our visual perception responds to different wavelengths of light in a way
that corresponds to what we recognize as the colours we know. Other species may
see other parts of the spectrum of light and thus they see different “colours” that
humans cannot see. In any case, colour perception has a clear biological background
that is the basis of similarity-dissimilarity relations underlying categorical thought.
Core processes are present from the very first few of weeks of life, if not at birth
(Cosmides & Tooby, 1994; Gelman, 2003), and they function as inferential traps
that impose their ready-made meaning on the aspect of the environment concerned
once a minimum set of conditions is present in the input information. For instance,
all humans with normal vision see colours or depth in basically the same way and
they base categorizations on them. Categorizations may be culturally determined,
such as categories related to one’s national colours, football team colours, political
party colours, etc., Obviously these categories emerge as a result of the interaction
between core processes and the other levels in the organization of the SCS to be
discussed below. Core processes are fundamental because they ground each domain
into its respective environmental realm and they are the springboards for the
development of mental operations, knowledge, and beliefs coming with time
(Demetriou & Efklides, 1981, 1985; Demetriou, Efklides, & Platsidou, 1993).
Operations arise as a result of the dynamic interactions between core processes,
the informational structures of the environment, and the central inferential and
control processes to be discussed below. That is, the core processes gradually give
rise to mental programmes of action vis-à-vis the problems posed by each domain,
which are increasingly self-guided in tune to the relations involved. For instance,
colour perception forms the basis for colour-based categories which are then
expanded by sorting and classification actions; these, when explicitly encoded, may
allow the understanding of relations between classes at various hierarchical levels or
multiple classifications, as examined in tests such as the Raven test discussed earlier.
Finally, each system involves knowledge and beliefs accumulating over the years
as a result of its interactions with the respective domain. That is, our stored knowledge
about the world is the product of the functioning of the specialized domains.
Conceptual and belief systems pertaining to the physical, the biological, the psycho-
logical, and the social world are found at this level of the organization of the various
systems. Knowledge and beliefs systems in each SCS may be seen as theories that
may be modified in the fashion of scientific theories, as discussed in the context of
the theory-theory approach detailed earlier. We describe below the core processes,
mental operations and processes, and knowledge and beliefs involved in each SCS.
The domains
Spatial thought deals with orientation in space and the iconic representation of the
environment. Arrangement in space is the basis of spatial cognition: relations such
as “close and far away”, “in front of”; “left-right”, “below and above”, and “inside
of” are reflected into core processes which register and recognize these relations
from birth. For instance, perception of depth encodes “close and far away”, and “in
front of” and is present in the first days of life (Hermer & Spelke, 1996). Integrating
variations in these relations in the environment is projected into operations specific
to the spatial SCS, such as mental scanning of visual images, mental rotation of
objects in a mental image, change of perspective in viewing a mental image etc.
Mental images, mental maps, locations, scenes, or layouts stored in memory come
from the functioning of the spatial system.
Categorical thought deals with similarity-difference relations. Physical resemblance
(e.g., colour, shape, function, use, etc.) is the basis for these relations. Core processes
here are based on the perceptual recognition and processing of similarities which
provide the primary material for building conceptual categories. As already noted,
colour perception is a very powerful primary categorization process (Carey, 2009).
Similarities in shape, size, sound, texture, etc., are primary dimensions which are the
basis for similarity judgments if needed. Mental operations of categorization involve

actions, actual or mental, which expand on these core processes. For instance,
sorting and classification according to various criteria are basic categorical operations
functioning from infancy. Their representation into integrated mental operations on
class relations, such as Piaget’s class inclusion and the processes needed to solve
Raven’s matrices, comprise categorical thought. The conceptual systems humans
hold about their world, such as natural kinds of animate and inanimate beings, types
of things we use, types of people such as good and bad people, etc., emerge from
an interaction between categorical core and operational processes and form the
knowledge and belief systems guiding people to relate with the world.
Quantitative thought deals with quantitative variations and relations in the
environment. Aggregation and distribution of objects in the environment (e.g., few
or many, more or less), and their increase or decrease (i.e., addition, subtraction)
form the basis of quantitative-mathematical cognition. Subitization, the automatic
recognition of the number of small sets involving up to 3–4 objects, is a fundamental
core operator of quantitative thought. In fact there is evidence that even addition
and subtraction within the subitization limit is present from the very first days of
life, forming the springboard of mathematical thought. Notably, both subitization
and addition-subtraction within the subitization limit are present in many other
animals (Dehaene, 2011). These core processes together with counting acts, such as
pointing, bringing in, removing, and sharing, form the basis of the four arithmetic
operations, which are the foundations of this SCS. Extensions of these operations
involve the various rules underlying the operation on quantities, such as rules in
algebra and geometry, and their applications for measurement. Factual knowledge
about the quantitative aspects of the world, such as time reading, money values,
street addresses, and rules underlying everyday transactions emanates from the
functioning of the quantitative system.
Causal thought deals with cause-effect relations. Effective interactions (e.g.,
changes in an object or event associated with changes in another object or event in
space or time) are the basis of causal cognition. Core operators enable a first grasp
of these interactions. For instance, infants understand from the first weeks of life that
solid objects cannot pass through each other; they recognize that when a moving
object hits a stationary object it will make it move in the direction of its movement
(Carey, 2009; Saxe & Carey, 2006). Trial-and-error actions aiming to specify
the causal effect of an object, such as playing with light switches, is the basis of the
development of the operations of causal thought. These actions result in the mental
operations of the causal SCS: systematic combinatorial thought allowing specification
of all possible combinations between objects; systematic experimentation based on
combinatorial thought in order to isolate variables; and prediction and hypothesis
testing based on a combination between experimentation and initial assumptions
about causal relations. More refined rules allowing deciphering of causal relations
in different sciences, such as physics, biology, or psychology, emerge as a result of
an interaction between these operations and knowledge in each discipline. Our
causal attributions about the behaviour of objects and persons reside at the third
level of this domain.
Social thought deals with the understanding of social relationships and interactions.
Recognition of species-specific important information, such as the face of
conspecifics, or basic emotions such as joy or anger, or mental states such as intention
and desire, is the basis for core operators in the social SCS (Rumbaugh & Washburn,
2003; Simion, Macchi Cassia, Turati, & Valenza, 2001). Mechanisms for monitoring
non-verbal and verbal communication or skills for manipulating social interactions
belong to this system and provide the basis for the development of the operational
repertoire in this system. It has been argued, for instance, that reasoning initially
appeared as a system to detect cheating, enabling humans to evaluate exchange of
information (Cosmides & Tooby, 1994). Even theory of mind was regarded by
several researchers to have evolved as a fundamental operation that would allow
interaction between humans or understanding the behaviour of other animals, given
that mental states cause behaviour. This system also includes understanding the
general moral principles specifying what is acceptable and what is unacceptable in
human relations (Kohlberg, Levine, & Nucci, 1983).
Linguistic ability and verbal though deals with information processing through
language. It is beyond the aims of this book to embark on the huge domain of verbal
ability. Suffice it to say here, for the sake of completion, that it comprises the
processes underlying the grasp and use of the rules of language in sake of social
interaction and mental processing. Obviously, it interacts closely with the various
other domains, social thought and inference in particular.
Empirical substantiation of the SCSs

We conducted a large number of studies to map the operational and procedural
status of the SCSs. These studies involved individuals from 4-years-old to the age of
70. These individuals were examined by large batteries of tasks which addressed most
or all the SCSs above. These tasks were designed to vary in difficulty so they could
be solved by children and adolescents at different ages. One example of the tests used
is presented in Demetriou and Kyriakides (2006). Examination of participants satis-
fied several psychometric standards about sampling of people and items, allowing
them to be seen as representative of the population and the processes involved.
It is a commonplace in psychometric research that factors reflect individual
differences. That is, factors emerge because differences between individuals in their
performance on the various tasks cluster around domains, if each domain is represented
by a sufficient number of tasks. In this case, correlations between tasks addressed to
the same domain are higher than correlations between tasks addressed to different
domains, letting domains emerge as distinct dimensions of ability. In confirmatory
research, where tasks are designed to represent different cognitive processes, it is
justified to assume that the dimensions of individual differences also represent different
cognitive processes, if statistically robust clusters correspond to the groups of processes
supposedly related to each domain. The cognitive basis of this clustering is the fact
that mental operations are not interchangeable across SCSs. That is, operations in
each domain gear on the objects and relations of the domain concerned and therefore
they are irrelevant to other domains. Thus functional specificity in the cognitive
processes associated to each SCS is reflected in their psychometric structure. Compare,
for instance, arithmetic operations in the quantitative SCS with mental rotation in
the spatial SCS or isolation of variables in the causal SCS. Arithmetic operations relate
and transform quantities. Mental rotation transforms the placement of objects in
space. Quantitative relations are irrelevant. Hypothesis testing relates possible causes
with their effects. Neither number nor rotation is relevant to experimentation as such
unless they are related to the process of isolating variables.
It is noted that a large number of studies have been conducted in different
countries, such as Australia, China, Cyprus, Greece, and India, (Demetriou et al.,
1993, 2005, 2013; Shayer, Demetriou, Prevez, 1998). We found that the SCSs
always emerge as independent factors of performance. Here we mention one of
these studies because it included a widely representative array of tasks which came
from different research traditions. This study, which we designed with Case (Case,
Demetriou, Platsidou, & Kazi, 2001), addressed the following SCSs: (i) categorical,
(ii) quantitative, (iii) spatial, (iv) causal and (v) social reasoning. Tasks addressed to
analogical and deductive reasoning were also included. It is notable that each domain
was addressed by three types of tasks: tasks drawn from our own earlier research,
tasks drawn from Case’s research, and tasks drawn from WISC-III, a well-known
test of intelligence. Therefore each domain was represented by a wide array of tasks
addressed to different processes related to it. Confirmatory factor analysis showed
that each SCS emerged as an independent first-order factor related to all SCS-
specific tasks, regardless of their origin. There was also a second-order general factor
highly related to all five SCSs which stands for g. This general model is shown in
Figure 8.3. In fact the model fits very well, even after taking out the influence of
age on the relations between the SCSs. This strongly suggests that this architecture
is present regardless of age. Many other studies resulted in similar findings (Case,
Demetriou, Platsidou, & Kazi, 2001; Demetriou & Bakrasevic, 2009; Demetriou
et al., 1993; Demetriou & Kazi, 2001, 2006; Shayer, Demetriou, & Pervez, 1988).
The reader might be interested to know the relation between performance
on the SCS and the WISC test. This information would show how much the SCSs
draw on processes addressed by a classical IQ test. We estimated this relation in two
ways. First we estimated the correlation between a common factor underlying all
five SCSs and a common factor underlying performance IQ and verbal IQ. It is
noted that these scores are standardized for age so that the test reflects within age
differences. This correlation was high but not impressive, i.e., .52. Second, we
estimated this correlation using raw rather than standardized IQ scores. In this case
this relation rose considerably to .8 for both performance and verbal IQ. Obviously
SCS and IQ tests draw on common processes to a large extent; more than 60% of
the persons who took the two tests occupy the same position, if ordered from
lowest to highest on each test. However, the standardization of scores to specify IQ
masks the developmental dimension of performance considerably. In fact the
SCS-IQ relation rose to .75 when the variance in SCS due to age was statistically
taken away.
FIGURE 8.3 The hierarchical model involving SCS-specific factors and a general factor
(Based on the model shown in Case et al., 2001)
Note: the names shown in boxes indicate the origins of the task concerned, namely the
WISC-III test, the work of Case, and our work
Interestingly, the specificity of the SCSs is also reflected in their logical and
semantic structure. We showed that each SCS (i.e., categorical, quantitative, causal,
and spatial) involves a core element that cannot be reduced to any other SCS or
standard logic, which may underlie conditional reasoning and thus relate to the
central processes associated with g. The core element for the categorical SCS gears
on the essential properties of objects which determine their reduction to a specific

category. The core element for the quantitative SCS gears on an intuitive
understanding of membership of an element in a denumerable set. The core element
in the causal system is a grasp of a necessary sequence of two events, A and B, such
that B always follows A, and there is a necessary connection such that B would not
have happened if A had not happened. Finally, the uniqueness of the spatial SCS
resides in the very nature of mental images, which are immediate, they represent
every part of their structure, and they have a physical similarity to the depicted. This
is to be contrasted to the arbitrariness of the symbols and representations needed to
represent the core elements of all other SCSs, at least after a certain phase in
development (Kargopoulos & Demetriou, 1998).
Differences in the foundational and operational elements of each SCS relate to
the fact that each of them is symbolically biased to symbolic systems that are
conducive to the representation of their own elements, properties, and relations
(Demetriou & Efklides, 1988; Demetriou & Raftopoulos, 1999). For example,
mathematical notations are more appropriate than images or words for the operation
of the quantitative system, because they unequivocally stand for quantitative
properties and relations; mental images are more appropriate than numbers or words
to represent object characteristic and relations in space, because they directly depict
these relations; words are more appropriate than numbers to represent syntactical
relations because they were evolved to stand for these relations. Of course, symbol
systems can be translated into one another, but some information may be lost at the
expense of other information when shifting from one system to the other. We
showed, for instance, that proportional relations from the quantitative SCS and
causal relations from the causal SCS may be expressed in both numerical and visual-
imaginal information. In this case, both the two SCSs and the two symbolic systems
emerge as autonomous factors interacting to co-determine performance on each
combination of SCS-specific mental operations and symbolic representations
(Demetriou, Efklides, & Platsidou, 1993). We will return to the interaction between
SCSs later when we focus on development and their relations to the general factors
of the four-fold architecture. These differences explain why learning which focuses
on the main operations within each of the domains does not transfer to other
domains (Demetriou, Efklides, & Gustafsson, 1992; Demetriou, Efklides, &
Platsidou, 1993; Efklides, Demetriou, & Gustafsson, 1992).
Convergence of the psychometric, cognitive, and logical

dimensions of the SCSs
The reader may have noted that the central conceptual structures specified by Case
(1992) are, by and large, similar to our SCSs. In fact three of our SCSs (spatial,
quantitative, and social thought) are identical to three of Case’s central conceptual
structures and Gardner’s intelligences. It is impressive that some of these domains
have been with us for a very long time. In Kant’s Critique of Pure Reason the concepts
of space and time exist in the mind before experience, as the envelopes in which
experience happens. In this context, some of Kant’s categories of reason (judgment)

are impressively similar to these domains. Kant speaks of four classes with three
categories in each. Quality is very similar to the domain of categorical thought and
includes reality, negation, and limitation. In fact reality and negation stand for the
recognition of similarities between the objects defined by a quality and negation to
their differentiation. These two properties allow the delimitation of objects’ placement
(or denial) in conceptual hierarchies. Quantity includes unity, plurality, and totality
and stands for the representation of unique elements, their belonging to sets, and their
discrimination based on their position in the set. Relation refers to causal thought. It
involves substance and accident, causality and dependence, and reciprocity between
agent and patient and is governed by the grasp of necessity and causal sequences.
Finally, modality, which includes possibility-impossibility, existence-non-existence,
and necessity-contingency, refers to inferential processes (Kant, 1902).
We note similarities between Kant’s analysis and our specification of the logical
core of each SCS. Specifically, our representation of essential characteristics in the
categorical SCS captures Kant’s reality and its use in the delimitation of objects’
placement (or denial) in conceptual hierarchies. Our grasp of membership traverses
all three of Kant’s quantitative categories. The grasp of necessity and causal sequences
in our causal SCS is the basis of Kant’s relation. Modality corresponds to inferential
processes (Kant, 1902).
In one way or another, all traditions recognized all of these domains as autonomous
constellations of mental processes. All domains were systematically studied by Piaget
himself. The famous tasks he used, such as the conservations, the classification, the
imagery, the formal thought, and the moral reasoning tasks relate to specific processes
in each of these domains. However, his search for general underlying logical
mechanisms drove him away from the procedural and developmental differences
between the domains. Later developmental research restored the domains to a large
extent. In cognitive psychology, all of these domains exist as autonomous domains
of research. Research on mental imagery and spatial thought (Kosslyn, 1980),
mathematical thought (Dehaene, 2011; Siegler, 2016), verbal and syllogistic reasoning
(Johnson-Laird, 2012; Moshman, 2011; Rips, 1994), categorical thought (Gelman,
2005; Mandler, 1992), causal thought (Kuhn, 2005; Moshman & Tarricone, 2016),
and social thought (Kohlberg et al., 1983) are different fields of research. Admittedly
the domains of categorical and causal thought are not directly mentioned in
psychometric research; interestingly, however, they are recognized as part of fluid
intelligence under the name of “Piagetian reasoning” (Carroll, 1993).
Reasoning: inference-based, rule-based, or model-based?

The four-fold model assumes that reasoning is separate from the SCS. There is
recent evidence that deductive reasoning emerges as a distinct factor when examined
together with other intelligence tasks (Kaufman, DeYoung, Reis, & Gray, 2011).
In this study there was an abstract deductive reasoning factor which stood for
performance on a large number of tasks based on Wason’s card selection task,
which we presented in the first chapter. This factor was separate from factors
representing verbal reasoning, mental rotation (spatial reasoning), associative
learning, and working memory. This reasoning factor was related to g, like the
other factors, but it was not related to the speed of reaching a conclusion on any
of the reasoning tasks. This study lends strong support to the assumption of the
four-fold model that reasoning is an autonomous dimension of mind that does not
identify with domain-specific inference and problem-solving or processing and
representational capacity.
It is important for our model to detail the status of reasoning in the mind and
specify its relations with the other systems, SCS in particular. After all, in both
problem-solving and developmental literature, the boundaries between them are
not very clear. We remind readers that in the cognitive tradition it is still debated if
reasoning is indeed organized into two major types of inferential process—inductive
and deductive reasoning (Rips, 2001). It is also debated if the inferential process is
based on logical rules framing what valid inference can be made, given the premises,
or on mental models using visual images or other types of iconic representations
enabling thinkers to visualize the logical relations involved. We ran several studies
to obtain evidence related to these debates. In one of them we constructed a battery
of 24 reasoning arguments. These arguments addressed, in sets of 6, four logical
schemes: modus tollens (MT), constructive dilemma (CD), affirming the consequent
(AC), and denying the antecedent (DA). In pairs, these arguments involved relations
from three SCSs: causal (e.g., A causes B), quantitative (e.g., number x is bigger
than number y), and spatial relations (e.g., object A resides in object B). One of the
two SCS-specific arguments was abstract, stated in the fashion of the examples
above; the other involved concrete content, where As and Bs or x and y were
replaced with real objects. These arguments included two premises and the
conclusion to be selected among four choices. From the point of view of logic, MT
and CD are deductive arguments because their conclusion is necessary, given the
premises. AC and DA are inductive because their conclusion is only likely. Examples
of tasks for each of the four logical schemes are shown in Box 8.1.
BOX 8.1
EXAMPLES OF TASKS ADDRESSING
REASONING.
Modus Tollens
If A causes B, then A comes before B
A comes after B
------------------------
A is impossible to cause B (score 3)
A causes B (score 0)
A may cause B (score 1)
A rather does not cause B (score 2)
Dilemma
If A is valid then B is valid too
If A is not valid then C is valid
-----------------------------
If B is valid then C is valid too (score 1)
If C is valid then B is not valid (score 2)
Either B or C is valid (score 3)
Neither B neither C is valid (score 0)
Affirming the Consequent

If A causes B, then B follows A
B follows A
------------------------
A certainly causes B (score 1)
A rather causes B (score 2)
A may cause B (score 3)
A certainly does not cause B (score 0)
Denying the Antecedent

If A is valid then B is valid
A is not valid
-----------------------------
Certainly B is not valid (score 1)
B is rather not valid (score 2)
Maybe B is not valid (score 3)
B is certainly valid (score 0)
In tasks with concrete content, A and B were substituted by concrete content

(e.g., if Costas has pneumonia then he has high fever; Costas does not have
high fever). Spatial tasks referred to spatial relations (e.g., if X is in Y then it is
also in Z; X is not in Z). Quantitative tasks (e.g., if a number divides another
number then it is smaller; X is not smaller than Y).
To specify the relations between reasoning applied on different types of relations

and problem-solving focusing on these very same types of relations, tasks addressed
to problem-solving in the same three SCSs were used, namely the causal, the
quantitative, and the spatial SCSs. In each SCS, different components were examined
at different developmental levels. In the causal SCS, combinatorial ability, isola-
tion variables, and grasping causal relations were examined. In the quantitative
SCS, reasoning with arithmetic operations, algebraic, and proportional rela-

tions were examined. In the spatial SCS, coordination of perspectives and mental
rotation were examined. The tasks are shown in Box 8.2.
BOX 8.2
CAUSAL SCS
Combinatorial thought: specify all possible sequences in which you can draw several
balls from a box:
1. Red, red, one green.

2. Red, red, green, green.
3. Red, red, green, blue.
4. Red, red, red, green, green.
Hypothesis testing—isolation of variables: design an experiment to examine if

increasing irrigation increases plant productivity, using plants A and B which can
be irrigated 2 or 4 times per month.
1. Hypothesis: increasing irrigation increases plant productivity. Use plants A

and B and two levels of irrigation, 2 or 4 times per month.
2. Hypothesis: irrigation increases productivity of plant A but it does
not affect plant B. Use plants A and B, irrigation 2 and 4 times per month
each.
3. Hypothesis: irrigation increases productivity of plant in area 1 but not in
area 2; it does not increases productivity of plant B in area 1 but it does in
area 2.
Specify the causal relations. The results of a series of experiments are presented.
Participants are asked to choose the patterns of results matching the following
causal relations.
1. Necessary and sufficient.

2. Necessary and not sufficient.
3. Sufficient but not necessary.
4. Neither necessary and not sufficient.
5. Incompatible with the result.
Quantitative SCS
1. Specify numerical operations in simple numerical expressions: One (e.g., 5 *
3 = 8), two (e.g., {4 # 2} * 2 = 6), three (e.g., {3 * 2 # 4} @ 5= 7), and four
operations (e.g., {5 @ 2} o 4 = {12 $ 1} * 2) were missing from the items
of each level.
2. Grasp algebraic relations: These require specifying one or more unknowns

in an equation (e.g. a + 5 = 8, specify a; u = f + 3; f = 1; specify u; if (r = s +
t) and (r + s + t = 30), specify r; when is true that {L + M + N} = {L + P + N}?
3. Specify proportional relations: the four levels required to grasp relations
between the following: (i) fully symmetrical and equivalent ratios (e.g., ½
to 3/6), (ii) equivalent but not obviously symmetrical ratios (e.g., 2/6 to
3/9), (iii) ordered pairs with two corresponding terms multiple of one
another (e.g., 2/5 to 3/7); (iv) pairs without corresponding terms (e.g.,
5/12 to 3/8).
Spatial SCS
Coordination of perspectives.
1. Draw the water line of a half-full bottle tilted at different degrees (e.g., 75 and
35 degrees).
2. Mental rotation. Choose the cube that would come out of appropriately
folding a paper, complete a puzzle requiring rotation of parts by varying
degrees.
We ran several models to test the organization of reasoning and its relation with the
SCS (see Figure 8.3). In regard to reasoning, the best model of performance on
the 24 tasks involved the following constructs: one for each logical scheme (MT,
CD, AC, and DA); one for each of the two types of inference (deductive and
inductive); and g. Neither the domain-specific relations nor the abstract-concrete
aspects of the tasks emerged as important constructs in the model. These findings
suggest that reasoning is primarily rule-based and inference-based rather than model-
based. Specifically, when reasoning about relations, the inferential process rests on
the logical relations as such, rather than on their type (e.g., causal, quantitative,
or spatial). However, reasoning is strongly related to performance in the SCS.
Specifically, we found that reasoning accounted for 63% of the variance on the
SCS. Although very strong, this relation leaves room for specialized skills related to
each SCS which need to be built on top of whatever reasoning processes are
employed to explore and specify SCS-specific relations.
The four schemes are mastered at different ages. It can be seen in Figure 8.4A
that modus tollens is fully mastered at age 12–13. Practically everyone has already
mastered this scheme by age 12 if it is stated in concrete terms. In fact, modus
tollens, as already discussed in Chapter 5, is within reach at 9–10 years if properly
stated. However, 12-year-old adolescents still fail to solve it when stated in abstract
terms; they master it at age 13. The dilemmas are mastered by the age of 13–14 (see
Figure 8.4B), although performance here is affected by both the type of relation
involved and the level of abstraction. People grasp these schemes by age 13–14 in
the causal or the spatial domain, especially when supported by concrete information;
FIGURE 8.4 Attainment of the four logical schemes as a function of age and abstraction
(concrete versus abstract)
however, when stated in abstract terms they are attained at age 16–17. The fallacies
are very difficult to grasp in any relational domain at any level of abstraction (see
Figure 8.4C). No more than about a third of the population mastered them even
at the age of 19–20, when at college. In fact, in some cases, concrete information
interfered negatively, lowering then increasing performance. Obviously the fallacies
require a clear focus on a principle that can be used to evaluate the logicality of
alternative implementations of the relation judged; concrete content may deceive
the thinker into implementing the principle by imposing knowledge possessed
on the logical relation involved. This implies that, when thought can grasp
indeterminate relations, reasoning rules do not need model-based assistance to be
implemented.
To specify the relation between each logical scheme- or SCS-specific factor and
g, a different approach to modelling was adopted. Specifically, we built one first-order
factor for each logical scheme (MT, CD, AC, and DA) and each SCS. The SCS
factors were based on performance on the causal, quantitative, and spatial problem-
solving tasks. All but one of these factors were regressed on g. “g” was regressed on the
factor left out. Thus this factor is, so to speak, a reference factor or a proxy of g.
Obviously a high relation between g and the proxy factor would indicate that g carries
the constituent properties of the proxy factor. The model was tested seven times so
that each scheme- or each SCS-specific factor was taken as a proxy of g.
The relations found are shown in Figure 8.5. It can be seen that the relations
varied across processes. It was low and negative for MT, reflecting the fact that
participants operated at ceiling on this factor while g varied extensively in the other
components. Interestingly, the two fallacies, AC (1.0) and DA (.97), predicted g
almost at unity. This is an impressive finding suggesting that once a thinker can handle
the fallacies, this thinker commands all other reasoning schemes; also this thinker is completely
able to master the skills required to solve problems in any SCS. This is not reciprocal,
however. Operating high on an SCS accounts for only about half of the variance
on g, suggesting that g is heavily marked by sheer reasoning processes. In other
words, commanding the inferential processes as such at their highest level will
automatically open the way for mastering domain-specific problem-solving skills;
however, this would not automatically yield command of the inferential processes
at the highest level. We will return to this later.
Empirical mapping of the dimensions in g

To capture the substance of g, a study would have to satisfy three requirements:
(i) psychometric g would have to be abstracted from a wide array of mental
competences, such as the broad abilities involved in the second level of the Cattell-
Horn-Carroll model; (ii) the relation between g and each of these broad competences
would have to be independently specified; and (iii) the relation between each of the
broad competences and general information processing or representational processes
(i.e., attention control, shifting, working memory) supposedly shared by the broad
processes would also have to specified.
We conducted several studies according to these requirements. One of these

studies involved children and adolescents aged 9–15 who were examined on four
types of competences. First, several aspects of reasoning: deductive (transitivity and
conditional reasoning in the fashion shown in Box 8.1), inductive (verbal analogies),
quantitative (algebraic reasoning and numerical analogies), causal (combinatorial
reasoning and hypothesis testing), and spatial reasoning (various aspects of mental
rotation). Examples of these tasks are shown in Box 8.2. Second, several aspects
of language competence: syntax (construct syntactically correct sentences out of
scrambled words); semantics (arrange scrambled sentences into a meaningful story
or understand ready-made stories), and vocabulary (define words or specify the
meaning of words). Third, several aspects of processing efficiency and executive
control: attention control (Stroop-like inhibition tasks, see Box 8.3), flexibility in
shifting (dimensional change sorting tasks), and working memory (forward verbal
and digit span and visuo-spatial working memory).
Finally, this study involved several measures of self-awareness and self-evaluation
related to performance on each of the reasoning domains mentioned above. After
solving each of the reasoning tasks, participants evaluated their own success in, and
the difficulty of, each task. To make these evaluations comparable to performance
scores, these scores were transformed into evaluation accuracy scores reflecting
concordance of evaluations to actual performance on the respective tasks (see Makris
BOX 8.3 COMPATIBLE STIMULI ADDRESS PROCESSING

SPEED AND INCOMPATIBLE (CIRCLED)
ADDRESS PERCEPTUAL CONTROL.
Examples of Stroop-like tasks addressing processing speed and inhibition

control. For speed, persons responded to the dominant dimension of compatible
stimuli. For instance, reading red written in red or recognizing a triangle made
up of triangles. For inhibition, persons responded to the weaker dimension of
incompatible stimuli. For instance, recognize the ink colour (green in reality)
of a word denoting a different colour (red in reality) or recognizing the circles
making up a square.
et al., 2017). The closer one’s self-evaluation of performance was to one’s actual
performance on a task the higher was one’s self-evaluation accuracy score. This
manipulation allows researchers to examine how differences in cognitive ability
relate to differences in self-awareness. Performance on these batteries was modelled
by a series of structural equations.
We ran several models designed to satisfy the requirements specified above.
Specifically, we created a first-order factor for each of the domains outlined above.
To satisfy the first requirement, we created a second-order factor that was related to
all domain-specific language and reasoning factors but one; this factor stands for g. To satisfy
the second requirement in a series of models, this second-order g factor was regressed on
the domain-specific factor left out of it. Therefore, in the fashion already explicated above,
the domain-specific factor is lifted up to the status of a proxy that may speak about
the identity of the common factor. Finally, to satisfy the third requirement the
reference factor was regressed on attention control, cognitive flexibility, and working
memory. This manipulation may show if any of the reference factors is a privileged
mediator between g and the supposedly shared processes.
Each of the theories summarized above leads to different predictions about the
pattern of relations expected. The theories assuming that some specific processes are
involved in g more than others would predict that the reference factors standing for
these processes would have a higher relation with g than the other processes. For
instance, psychometric theory would predict that fluid intelligence, as captured by
inductive and deductive reasoning, would be a stronger proxy for g than other processes
(Gustafsson, 1984; Spearman, 1904). Alternatively, cognitive science theory—assuming
that syntax in language contributes to the formation of the language of thought because
combinativity, recursivity, and hierarchical organization of language transfer to the
FIGURE 8.5 Relations between reasoning and SCS reference factors and g
operations of thought—would predict that language would emerge as the best proxy
of g (Carruthers, 2002, 2009). Interactive or mutualist models would predict that
the relations between g and reference factors would vary with the com-plexity of the
interactions involved in each reference factor: the higher a factor’s complexity,
the higher its relation with g would have to be. For instance, in the present study some
of the domains involved are highly specific and some are very broad. In language,
syntax is more specific than semantics; the first depends on language-specific modular
process and the second involves inferential processes needed to grasp implied meaning.
In reasoning, spatial reasoning is simpler than causal reasoning; the first depends on
highly specific processes such as mental rotation and the second requires both inferential
processes and also domain-specific hypothesis formation and testing processes. Finally,
theories assuming a ubiquitous common core (whatever this might be) would predict
that the relations between g and the reference factors would be similar across processes,
because the same core is involved in each of them.
The results of these models are summarized in Figure 8.6. It can be seen that the
relation between all reference factors and g was always very high (all > .8). Contrary
to the privileged process theory, there was no privileged proxy factor. Contrary to
the mutualist models, the very small (and non-significant) differences between
proxy factors and g cannot differentiate any of the factors in regard to complexity.
However, these results align with common core theory because they were all very
high and very close to each other. The relations of the proxy factor to each of the
three executive processes came in the same direction. They were all in the same
range (.4–.6) and very similar across reference factors.
To map the processes in g, we tested a model aiming to decompose g into its
various executive, cognizance, and inferential components. Specifically, all reasoning
and language factors were regressed on g and g was regressed on the three executive
processes, and also on cognizance and reasoning. This manipulation allows an
examination of how each of these processes differentially contributes to g. The five
factors accounted for the variance of g as follows: attention control=27%;
flexibility=18%; working memory=27%; cognizance=7%; reasoning=19%. This
amounts to 98%, very high indeed. In other words, attention control, flexibility,
working memory, cognizance, and Gf (i.e., deductive and inductive reasoning) are
all strong and distinct building blocks of the common core identified with g.
Attention is drawn to the fact that g so decomposed fully exhausted variation in all
domain-specific factors of thought and language. This finding implies that measuring
these five central processes allows one to fully predict how one would perform in
various thought domains such as mathematics, scientific reasoning, spatial reasoning
and various aspects of language.
It needs to be noted here that the relative contribution of each of these five
processes to g, although always present, changes with development. Specifically,
we found that the relations between processes residing at the lower end of the
cascade (age, attention control, flexibility, and working memory) decreased
systematically across age phases, whereas the relations between processes residing
at the higher end of the hierarchy (working memory, reasoning, language, and
cognizance) remained stable or increased. Figure 8.7 illustrates this shift in the
FIGURE 8.6 An idealized model of the structural relations between g and each of the
reference factors and between each of the reference factors with attention control (AC),
cognitive flexibility (Flex), working memory (WM), cognizance (Cogn), and inference
(Infer)
Note: the figure summarizes eight models in which the first-order factors, each standing for a
domain, were regressed on g, one of them was taken as a reference factor so that g was regressed
on it, and the reference factor was regressed on the factors standing for aspects of executive
control, plus cognizance and inference
relative importance of processes with development. These patterns suggest a shift

from executive processes related to control of attentional and mental focus in the
preschool years to processes directly related to reasoning and explicit awareness in
the primary school years. This change is fully established in adolescence, when both
attention control and flexibility faded out almost completely and self-guided
reasoning and awareness dominated as factors of problem-solving and understanding.
Therefore, the relations between processes vary as a function of developmental
FIGURE 8.7 The cascade model at four age phases i.e., at 4–6, 6–8, 8–11, and
11–14 years, respectively
phase, reflecting differences in the representational and procedural composition of

g at successive developmental phases.
Our findings here explain why reductionist models prioritizing any of these
processes as a privileged dimension of g at the expense of the others did not stand
up to expectations: g is not speed and/or working memory because they both only
reflect the state of other competences rather than being intelligence itself; g is not
just executive control because understanding requires much more; g is not just
awareness because intelligence requires applying specific processes in domain-
relevant ways; g is not just reasoning, because this is not enough for understanding
and problem-solving. Going back to Piaget, the grand processes of assimilation,
accommodation, and equilibration are not explanatory because they are everything.
What is holding these processes together? To integrate over the cognitive, the
psychometric, and the developmental tradition we need a mechanism that would do
justice to the procedural, the representational, and the generative aspects of
understanding at one and the same time. This mechanism would capture the
interactive aspect of the five processes comprising g. We suggest that this mechanism
involves three interdependent processes (Demetriou et al., 2013): (i) abstraction;
(ii) representational alignment; and (iii) cognizance—the AACog mechanism.
Abstraction spots or induces similarities between patterns of information. Although
always present, abstraction may vary in development, depending on the current state
of the organism. For instance, early in development (or in learning) a probabilistic
inference mechanism sampling over statistical regularities in the environment may
dominate (Tenenbaum et al., 2011). Later, similarities may be systematically searched
for and conceptually constructed by reasoning. Alignment is a “search, vary, and
compare” mechanism that interlinks stimuli and/or representations together according
to current understanding or learning goals; perceived current similarity and semantic
relevance provide direction and criteria for alignment. Thus alignment is an executive
mechanism of representational integration. It involves several processes traditionally
associated with executive control, such as shifting between representations or shifting
between representations and responses (Miyake & Friedman, 2012; von Bastian &
Druey, 2017). It feeds inferential processes with raw material that may lead to
inductions and deductions according to specific binding and evaluation rules activated
at a given moment (Demetriou et al., 2013). Cognizance is the process of becoming
conscious of mental content (e.g., “I know that I am thinking about numbers”) and
cognitive processes (e.g., “I know that I am looking for the bigger number in a
series”; “I knew this information”, etc.). Abstraction and alignment may be driven
by the stimuli to a large extent; however, sooner or later in the process they generate
content or activate mental processes that are registered and call for notice and attention
for the sake of further action or future use. The experiments of Zoltan Dienes (Mealor
& Dienes, 2013; Scott et al., 2014), described before, suggest that even when the
search for and the abstraction of rules are unconscious, they generate experiences that
are registered and exert metacognitive influences on further processing and reasoning.
The role of cognizance is special in the operation of abstraction and alignment.
Cognizance is a unifying force. It may arise as a side-effect of cognitive functioning
at any time abstraction and alignment fail or go gropingly. Needing to choose

between stimuli or possible actions turns the “mind’s eye” to them, thereby bringing
them into the focus of awareness. So defined, cognizance allows feedback loops
where cycles of abstraction and alignment can become the object of further
abstraction and alignment. Thus executive control, self-monitoring, reflection, and
metarepresentation, which are discussed extensively in the developmental and the
cognitive literature (Demetriou, 2000; Piaget, 1976, 2001; Schneider & Lockl, 2007;
Zelazo, 2004), express cognizance in various stages of these feedback loops. Executive
control allows the individual to choose between stimuli to respond to and between
responses to emit. Self-monitoring is an introspective process allowing mental
self-observation. Reflection is self-directed examination of alternative abstractions
and alignments. Metarepresentation is encoding of the products of abstractions and
alignments into new representations (Demetriou & Raftopoulos, 1999; Demetriou,
Spanoudis, & Mouyi, 2011). There is evidence that the consciousness needed for the
functions is present at the age of 12–15 months (Perner & Dienes, 2003).
In conclusion, AACog integrates Spearman’s (1927) eduction of relations
(abstraction) and correlates (alignment) underlying g, Piaget’s reflective abstraction
and equilibration, and reasoning and consciousness that dominated in post-Piagetian
developmental research. This mechanism must be operational in its entirety by the
end of the first year of life (Demetriou & Raftopoulos, 1999; Demetriou et al.,
2011; Leslie et al., 2004; Perner & Dienes, 2003; Stone & Gerrans, 2006).
Conclusions
This chapter summarized several studies which examined the organization of
the mind by bridging the three traditions. All studies involved tasks drawn from the
cognitive tradition (i.e., attention control, working memory, reasoning and even
language), the psychometric (i.e., Raven-like tasks and many other tasks included
in classical tests of intelligence, such as the WISC test), and, of course, the
developmental tradition (i.e., theory of mind and awareness and SCS-specific tasks).
Many of these tasks, especially those addressed to reasoning, were organized to meet
psychometric requirements, such as variation in difficulty; moreover, they were
addressed to many individuals of various ages who were representative of the total
population. Performance on these batteries was analysed by several modern
confirmatory modelling methods allowing a test of precise predictions about various
aspects of mental organization. The findings may be summarized as follows:
First, the mind is a four-fold universe involving domain-specific systems,
representational systems allowing the representation and mental processing of
information, inferential relational systems allowing the integration and evaluation
of information and interpretations, and cognizance systems, allowing self-
monitoring, self-representation, and self-regulation and self-modification, if needed.
Second, all three traditions assumed that one dimension of the four-fold model,
inferential systems, was taken as the major dimension to be understood. In the psycho-
metric tradition, inferential systems actually stand for intelligence, largely identified
with Gf or g. Likewise, in the developmental tradition, explicating changes in

inferential systems was their major objective. In all traditions, inferential systems
were identified with some kind of reasoning, primarily inductive and deductive
reasoning.
However, third, all types of reasoning, expressed in specific reasoning schemes,
such as the schemes of conditional reasoning, or implemented in different
domains, such as the causal or the mathematical domain or even syntax in language,
are interchangeable as proxies or reference factors for g. Thus we proposed a higher-
order system to capture the procedural composition of g, involving three very
general processes: abstraction, alignment, and cognizance (AACog). This may be
the background for a language of thought. Defined in reference to compositionality,
recursivity, generativity, and hierarchical integration that were assumed to comprise
LoT, the AACog processes would operate as follows: abstraction ensures com-
positionality; alignment ensures recursivity and hierarchical integration; and
cognizance ensures generativity. Specific types of inference or specific skills in each
SCS are languages to be learned and practised in special domains by adding domain-
specific qualifiers. This occurs, to a large extent, in development and it is of course
influenced by special learning opportunities, such as those offered by education.
Even types of reasoning which appear very general, such as deductive reasoning,
emerge as specializations of translating the central AACog core into a representational
ensemble of a specific set of relations in the environment that may be expressed into
a suitable syntax in language. This is the reason why even very advanced reasoners
may err in implementing deductive reasoning schemes when explicit rules for their
implementation are not present. Obviously differences in both developmental and
learning opportunities explain individual differences in attainment.
Fourth, other systems were examined as a possible fundamental explanatory
factor to which this privileged central dimension may be reduced. The studies
summarized here suggested clearly that inferential systems can indeed be reduced to
the representational and cognizance systems but none of them is a privileged factor; all
of them together are needed. In fact, at a general level, the contribution of each is more
or less the same. However, when taken developmentally, their relative contribution
varies enormously with age. Their relative contribution and interactions during
development will be further explored in the chapters following. In conclusion, this
model spans and integrates all three traditions. The studies to be presented below
will fully deploy this model.
9
CYCLES AND PHASES OF
DEVELOPMENT
Strictly speaking, psychometric g is not a developmental construct and it is not

included in any developmental theory. Moreover, as an index of individual
differences it is considered to be generally stable from early childhood to middle age.
That is, individuals who score at the top at one age will stay there at subsequent
ages; individuals who score lower than their age mates at one age will continue to
score lower at later ages (Jensen, 1998). In classical developmental theory, intelligence
changes qualitatively. As a result, individuals at successive stages of cognitive
development build different kinds of concepts about the world and solve different
kinds of problems. However, there is a central mechanism in developmental theory
as well. This is the mechanism underlying understanding in each stage and also
causing transition across stages. This is the AACog mechanism, discussed in the
previous chapter. This mechanism is the central coordinating agency that holds
together the systems associated with each of the four major dimensions of the four-
fold model.
In this chapter we will draw the big picture of the developing mind as suggested
by current research. Specifically, we will first map the developmental trends of
representational capacity and efficiency, cognizance, reasoning, and the SCS. The
aim is to show how the various processes integrate to generate a particular kind
of understanding in each developmental phase that looks qualitatively special
from a phenomenological point of view. This involves both an objective dimension
(how children in each phase solve specific problems) and a subjective dimension (what
children are aware of in each phase and how this awareness underlies their insight
into their present limitations, which drives reflection and metarepresentation leading
to the next phase). We will conclude this chapter with a discussion of the similarities
and differences between this sequence of developmental phases and other sequences,
such as the Piagetian or the neo-Piagetian stages. In the two following chapters we
will zoom in on the relations between the various processes of the four-fold model
in each developmental phase.
Cycles in the development of g

The research summarized above suggests that cognizance, reasoning, and various
aspects of executive control interact with each other in AACog. Reasoning
requires awareness of gaps in information (yielding a reasoning goal), of the
representations to be integrated, and of truth standards needed to evaluate
conclusions. Executive control requires awareness of a goal, the steps needed to
obtain it, and an inference-based decision system allowing evaluation of goal
attainment. Cognizance is a context-manipulating mechanism allowing interfacing
and holding executive control and reasoning together to collaborate. The studies
summarized also suggest that the relations between these processes vary with
development. The central assumption, from the point of view of development, is
that control of action provides the primary material one may become aware of.
Becoming aware of one’s own action allows consideration of goals, plans, and
alternative actions for the sake of success and efficiency. Reasoning is a means to
this end; it allows the evaluation of odds, truth, and validity for the sake of efficient
action. In any cycle, awareness of control experiences is transformed into models
of the mind that may be called on for the sake of reasoning. Thus changes in
control at the beginning of cycles are expressed into explicit executive control
plans in the middle of the cycle and these are then expressed into complex
reasoning-inferential systems at the end of the cycle. When metarepresented into
new inferential schemes, the next cycle begins. Figure 9.1 outlines this interplay
between processes in development.
FIGURE 9.1 The general model of changes across cycles and phases
Cycles and phases of development 123
Research suggests that the relations between these processes are transformed over
four major developmental cycles, with two phases in each. New representations
emerge early in each cycle and their alignment dominates later. In succession, the
four cycles operate with episodic representations from birth to the age of 2 (remembrances
of actions and experiences preserving their spatial and time properties), realistic mental
representations from age 2–6 (blueprints of episodic representations where spatial and
time properties are reduced, associated with symbols, such as words and mental
images), generic rules organizing representations into conceptual/action systems from age
6–11 (e.g., concepts about categories of things, exploring causal relations), and
overarching principles integrating rules into systems, where truth and multiple relations
can be evaluated, from age 11–18 (i.e., principles specifying how rules may be
integrated). Changes within cycles occur at about 4 years, 8 years, and 14 years,
when representations become explicitly cognized so that their relations can be
worked out, gradually resulting in representations of the next cycle (Demetriou &
Spanoudis, 2017; Demetriou et al., in press).
Obviously the reader recognizes that the four cycles resemble the descriptions of
cognitive development offered by several theories (Case, 1985; Fischer, 1980;
Halford et al., 1998, 2014; Pascual-Leone, 1970) with Piaget (1970) at the head.
We will elaborate on the relations with other theories later, when the four cycles
will be fully presented. In this chapter we will summarize research highlighting the
four cycles. In the next chapters we will focus on research illuminating shifts across
phases, how cognizance mediates between executive control and reasoning, and
how the relations between these processes are reworked to form a new g (AACog)
in each cycle.
Episodic thought
Infants are mentalistic creatures (Baillargeon, Scott, & Bian, 2016; Carey, 2009;
Pillow, 2008): they represent themselves and others as representational beings.
Infants differentiate themselves from objects by the age of 5–6 months (Rochat,
1998) and they recognize themselves in the mirror by 15 months (Gallup, 1982;
Povinelli, 2001), suggesting that they compare what they see with representations
of their invisible body parts. Infants talk to themselves about earlier experiences,
suggesting they reflect on them before they are 2 years old (Vallotton, 2008). For
example, they repeat instructions given to them earlier by an adult.
By 15–18 months infants show awareness of blocks of action, including an
executive sequence where past actions are intertwined with perceptions and current
actions: when encountering a familiar object set they intentionally restore the
sequence which involves representation of past experiences (e.g., insert objects of
various shapes in a toy turtle through same-shape holes) and projection into an
action plan (e.g., grasp objects and look for same-shape holes, testing by trial and
error if they do not get through). This is pre-sorting episodic representation where
perceptions, remembered representations, and actions reflected on are intertwined.
Infants also infer that someone who saw where an object was hidden will look for
it at that place (Onishi & Baillargeon, 2005), indicating some intuitive awareness of
the relations between mental states and action; this ability lies within a long sequence
of abilities forming mental awareness.
Infants are sensitive to statistical information in speech patterns. For instance, at
the age of 8 months infants differentiate between transitional probabilities of words
following one another in sentences; when they hear the sentences “The boy loves
apples. The boy loves oranges” in sequence, infants recognize that the transitional
probability between the words “the” and “boy” is 1 but the transitional probability
of the words “loves” and “apples” is .5 (Saffran et al., 1996). Noticeably, infants are
also able to abstract the algebraic rules governing patterns of stimuli which do not
obey any statistical regularity, as in the examples above. Marcus et al. (1999) showed
that 7-month-old infants can learn algebraic rules governing grammatical sequences
in language. For instance, they learn rules of the type ABA (e.g., ta li ta) or ABB
(e.g., ta li li) and apply them systematically to distinguish new rule-based sequences
(e.g., ko fe ko or ko fe fe, respectively) from sequences that are not consistent with
the rule (e.g., ko ko fe) (Marcus et al., 1999). It is notable that this ability to abstract
algebraic rules also applies to other types of stimuli, such as musical tones, animal
sounds, and varying timbres if these patterns are instantiated in sequences of speech
(Marcus, Fernandes, & Johnson, 2007). It seems that speech at this age facilitates the
activation of the rule-induction system of infants, yielding the frame in which
reasoning development may initiate. One might assume that these two forms of
early learning—abstraction of statistical regularities and fundamental algebraic
rules—provide the background for inductive and deductive reasoning.
This is obvious in the second year of life, when episodic reasoning involves
ordering episodic representations and explicitly stating an underlying relation (e.g.,
“I put this, and this, and this, all of them”, preparing for conjunction: all = this +
this + this), or reading them forward (e.g., “dad came, mom is coming too”,
preparing for implication; if A à B follows), or abstracting what runs through them.
When it concerns behavioural sequences related to a person (e.g., “Dad is going
upstairs; he is going to get dressed”), the episode may appear as a belief understanding.
The belief, however, is actually a reading forward projection of the episode for
another person rather than an explicit representation of this person’s mental states.
The reader may recognize here the fundamental schemes of mental logic discussed
in Chapter 5 devoted to reasoning development.
Realistic representational thought

Representations at age 2–3 are reduced mental projections of episodic representations
with a component of implicit awareness. There is evidence accruing in recent years
that children have some awareness of their own mental states from about 3 years of
age. Paulus, Proust, and Sodian (2013) trained 3.5-year-old children to associate
individual animals with specific objects. They showed them short videos of an animal
doing something (e.g., an elephant who likes watching TV). Sometime later they
showed the probe animal (e.g., the elephant) and they tested if children remembered
the object associated with it (a TV). They also asked the children to indicate how
confident they were of their judgment. Confidence ratings for correctly remembered
items were higher than ratings for incorrectly noted items, suggesting an awareness
of representations stored earlier in memory. Children at this age are also aware that
when one sees or hears an object one knows about it, suggesting awareness of the
perceptual origins of knowledge (i.e., I know because I see, hear, etc.) (Flavell,
Green, & Flavell, 1995). This makes theory of mind possible at the age of 4, enabling
preschool children to understand that one’s actions relate to one’s representations
(Wellman, 2014). The theory-of-mind research discussed in Chapter 6 suggests that,
at this age, children understand that different persons may have different representations
of reality depending on differences in their access to information (e.g., they saw
different parts of an episode). Emerging insight into the nature of representations
eventually brings them into focus, allowing their comparison and alignment.
At this age executive control is guided by a “scan-choose-focus-respond”
programme allowing preschoolers to set up action plans involving several steps to
be implemented in succession and shift between stimuli and responses, according
to a goal (e.g., say day when they see the Moon and night when they see the Sun)
(Vendetti, Kamawar, Podjarny, & Astle, 2015). Compared to the task where infants’
sorting of objects is guided by the match between object shape and hole shape, this
task involves a priori awareness of representations one may focus on and choose
from, organizing action beforehand.
At this early phase, representations have a transparent relation to objects or events
and they function as ensembles as a source of inference. As a result relations at this
early phase of development are intuitively “read out”, so to speak, from the
representational ensemble. However, 2-year-old children do draw inductive
inferences when perceptual patterns are clear enough so that missing components
may be integrated, based on similarity or extrapolation of characteristics across
objects (Gelman, 1988, 2003). For instance, toddlers are able to assemble simple
puzzles by matching or integrating shapes based on the similarity or complementarity
of form, colour, or other patterns. This is also evident in language learning, as noted
in Chapter 5. Inferences are also intuitively “read out”, so to speak, from the
representational ensembles (“It rains, so we need our umbrella”).
Deductive inference does not exist at this phase. Representations activated by an
incidence function as a block yielding inferences based on the episodic flow of
events: a single perceived state (e.g., [cat-on-the-edge]) is enough to generate
“plausible inductions”, often contradictory (“the cat will fall” versus “the cat will
jump”), depending on the block activated (e.g., “I fall when on the edge” versus
“cats jump when on the edge”, respectively). Plausible inductions complete
activated experiential episodes without constraining each other, if not aligned.
Thus, in this phase, the boundaries between categories are flexible, depending on
current dominant inductions. Even natural categories, such as “boy” and “girl”, may
not have fixed boundaries: Athina, at 34 months of age, wondered when Nicolas,
her cousin and 31 months old, the first author’s grandchildren, would grow up like
her to become a girl.
At about age 3–4 children start to differentiate between representations or be

able to zoom in on their components. As a result they can intentionally search
for, scan, and align them. For instance, they can solve simple Raven-like matrices
where patterns vary along a single dimension. However, they still face difficulty in
aligning patterns across two dimensions. Benoit, Lehalle, Molina, Tijus, and Jouen
(2013) conducted an interesting study of the alignment between representations of
quantities from age 3–5 (i.e., mapping dot arrays from 1–6 with number names and
number digits). They showed that 3-year-old children can only map number words
on arrays of up to 3 dots. They cannot map number words on arrays of 4–6 dots,
dots with digits or number words with digits. Obviously they have a global
representation of quantities within the subitization limit associated with
corresponding number words as an ensemble. Representations from the three
representational spaces become accessible as distinct mental entities that can be
aligned at 4 years. Four-year-olds map both number words and number digits with
arrays of up to 6 elements but they do not map number words on digits. At 5 years,
children map all representations with each other for all sizes.
When this is possible, children start to build concepts in the various domains:
there must be at least two representations to conceive of a class (e.g. “our cat is an
animal”), a quantity (e.g., “Anna has 3 and I have 2; she has more than me”), a
cause-effect relation (e.g., “Mary spilled the milk”), a spatial relation (e.g., “the toy
car is ‘on top of’ the book”), or make an inference. Alignment of representational
ensembles in this phase optimizes inductive choices and allows deals based on
pragmatic reasoning: “We agreed I can play outside if I eat my food; I ate my food;
so I go to play outside” (Kazi, Demetriou, Spanoudis, Zhang, & Wang, 2012). This
will be raised later into deductive inference. This sequence, which mimics modus
ponens (if p then q; p; thus q), is basically an induction that locks two representations
(“A occurs” and “B occurs”) together into an inductive rule (i.e., “When A occurs,
B also occurs”). Children may consider inductive options (i.e., “no eating-no play”
and “eating-play”) because their executive control programme allows them to
envisage alternative choices.
Research presented in Chapter 5 devoted to reasoning suggested that modus
ponens inference occurs automatically and unconsciously in college students
(Reverberi et al., 2012). Obviously this is natural. Time and experience, spanning
at least 15 years, is enough to consolidate and automate this basic scheme of
deductive reasoning so that it can be used as a reference scheme for the development
of other, more demanding, schemes, such as the modus tollens and the fallacies
which never automate completely, even in the minds of trained thinkers.
Rule-based thought
At 6–8 years children are explicitly aware of mental representations and their
relations with their own actions. For instance, they differentiate between easy and
difficult memorization tasks, suggesting awareness of the relation between complexity
of representations and learning. In this phase children also recognize that knowledge
may be constructed by inferential extrapolation as well (i.e., I know because I can

reason on what I saw, heard, etc., and project). Thus, in this phase, cognizance of
the inferential aspects of knowledge take over as the mediator between attention
control and working memory, on the one hand, and reasoning, on the other hand
(Spanoudis et al., 2015). However, children at this age do not yet explicitly
differentiate between mental functions, such as memory and reasoning, nor do they
explicitly associate each with specific processes (rehearsal versus inference). This is
possible at 8–10 years (Paulus, Tsalas, Proust, & Sodian, 2014), when there is an
explosion of awareness of the mental world. Children in this phase differentiate
between the metaphorical and literal meaning of verbal statements (e.g., “You are
as fast as lightning”; Olson & Astington, 2013), master second-order theory of mind
(e.g., “I know that George knows that Mary knows that. . .”; Wellman, 2014), and
recognize that lags in knowledge may be compensated for by inference (e.g., “He
sorted by colour, so blue objects would be in the blue box”; Spanoudis et al., 2015).
Children at 6–8 years do not prepare sufficiently to cope with a forthcoming task
because they are not explicitly aware that different tasks require relevant preparation
(Chevalier & Blaye, 2016). However, in the next phase, at 8–9 years, awareness of
different mental processes allows children to shift flexibly between them (e.g., to
remember you need to observe carefully and rehearse; to sort you need to follow a
sorting rule; Demetriou et al., 2014; Kazali, 2016). Interestingly, in this phase,
attention control and shifting emerge as strong predictors of reasoning. This is
expressed in the upgrading of executive control from inhibition control into a
conceptual fluency programme allowing children to shift between mental processes (e.g.,
memory versus inference) or conceptual domains (e.g., they recall words belonging
to different categories—fruits, animals, furniture—following a probe; Brydges,
Reid, Fox, & Anderson, 2012). Compared to the previous “focus-recognize-
respond” executive programme, the current programme involves analytic
representations of conceptual spaces and flexibility in variably running across them.
One might argue that Piaget’s (1970) reversibility is an index of this executive
programme. However, the emphasis here is placed on the origins (awareness of
representations and their relations) rather than on a product of mental flexibility
(reversibility).
In fact early in this cycle, at 6–7 years, there is a shift from “realistic” representations
that are visible to the “mind’s eye” to the inferential threads interlinking them. At
the beginning these may function as semantic blocks defining generic concepts, such
as object classes, number, causal attributions. The integration of various conceptual
spaces related to number, such as object arrays, number words, counting, digits, etc.,
into a common mental number line is a good example of an underlying mental
construct in the domain of quantitative reasoning (Dehaene, 2011). Thus, in this
phase, children can solve two-dimensional Raven-like matrices which require
integration of two familiar and obvious dimensions (e.g., shape, size, background,
etc.). Piagetian concrete operations in various domains (such as classes, quantities,
length, weight, area, number, etc.) and their interrelations are a strong sign of this
shift of thought from representations to their underlying relations.
In the next phase, at 8–10 years, another product of this emergent awareness is
the implicit use of rules specifying how different types of inference are interrelated.
Thus children in this phase can solve Raven matrices (e.g., series C in the SPM)
which require deciphering a critical dimension through systematic search and
transformation of one or more features of the matrices involved. They require the
generalization from instances of this dimension and reduction to a general rule: for
instance, “it is the double of each last number”, “it goes by one more”, etc. Proper
deductive reasoning requires one to evaluate a sequence of statements vis-à-vis a rule
that prescribes how they must be related. For example, to evaluate the validity of a
modus ponens argument (e.g., “Birds fly; tagi is a bird; therefore tagi flies”) requires
one to interpret it as an implementation of a relation running through it, look for it,
and check if it is consistent: “all birds fly à any bird flies”. This is possible early in
this cycle, at 6–7 years. In the next phase children can align rules as such and thus
solve more complex arguments, such as modus tollens. These require one to invert
the argument structure (e.g., “Birds fly; tagi does not fly; tagi is not a bird”) and align
it with the standard modus ponens structure to check if they are consistent. Thus,
in this period, relational definitions become increasingly dominant over particular
representations or episodic relations, yielding generic concepts supervening earlier
global representations, such as natural kinds (e.g., animate versus inanimate, etc.).
Formally, if it is accepted that “A implies B” then two possibilities are necessarily
true: when A occurs then B occurs too, and when B does not occur then A did not
occur either (Christoforides et al., 2016). Thus children grasp the relation between
modus ponens and modus tollens (i.e., if p then q; q à p; not q à not p). Therefore
awareness of underlying relations allows moving across conceptual spaces and rules
that may then guide executive control and reasoning. One can see here the second
stage of deductive reasoning development discussed in Chapter 5.
In this phase the dimensions or rules defining semantic blocks can systematically
be aligned with each other. In categorical thought, two independent dimensions
(life—living versus non-living beings—and movement—moving on earth and
flying) can be operated on so that all possible cross-classifications and their logical
relations (e.g., class inclusion) can be grasped. In quantitative reasoning, children
start to handle proportional relations (e.g., 2/4 and 4/8). This is also reflected in
children’s facility in handling analogies and metaphors (e.g., “teachers are for schools
what parents are for families”, or the matrices included in the Raven test).
Emergent logical necessity in this phase is a strong sign of this awareness (e.g.,
“All balls in the box are red, so the next to be drawn out MUST be red; Miller,
Custer, & Nassau, 2000). That is, incipient recognition that taking particular
representations as granted implies specific ensuing (actual or mental) consequences
suggests a supervisory stance objectifying representational relations as sources of
logical inference. It is noted, however, that logical necessity about these relations
is still fragile in this cycle. Even children who possess these concepts may yield
to opposing views, suggesting that the logical necessity of the relations between
rules is fully established only when these rules are embedded into principle systems
of the next cycle (Lourenço, 2016). This finding gives a Gödelian dimension in
development such that relations between inferential processes remain incomplete

and unstable within their cycle of acquisition. They only acquire relative com-
pleteness and closure in the cycles following (Piaget, 1970). Thus, when these
relations are explicitly metarepresented, these relations result in the emergence of
principle-based thought.
Principle-based thought
At 11–13 years adolescents form accurate maps of mental functions and of their own
strengths and weaknesses (Demetriou et al., in press; Demetriou & Efklides, 1989;
Demetriou & Kazi, 2006; Makris et al., 2017). As a result they evaluate their own
performance on cognitive tasks with increasing accuracy. They also cognize the
constraints of different inferential processes and they can ground inference on truth
and validity rules. Mental focus shifts from representations and rules to relations
between underlying rules connecting mental spaces, encoding them into generic
principles. For instance, they can now solve the most difficult Raven matrices
which require deciphering multiple dimensions by grasping the thread underlying
several transformations of figures and integrating them into complementary general
principles. Thus emerging principles interconnecting rules allow them to cognize
the constraints of different inferential processes. For instance, they explicitly
understand that accepting certain conditions (e.g., birds fly; elephants are birds)
imposes constraints on inference (i.e., elephants fly) even if a statement is admittedly
wrong (elephants are not birds). Formally speaking, these constraints are rules of
truth and validity which allow consistency in reasoning. The reader can recognize
here stage 3 deductive reasoning according to Chapter 5.
This is obvious in all domains. For example, in the domain of quantitative
thought they reduce the various instantiations of the mental number line into an
algebraic conception of number as a variable that can take any value (e.g., they can
understand “L + M + N = L + P + N” when M = P) (Demetriou et al., 1996;
Demetriou & Kyriakides, 2006). As a result number can be explored as such, defined
in alternative ways (e.g., natural, real, imaginary number, etc.), which can then be
compared for consistency (Dehaene, 2001). In the first phase, conceptual spaces may
be explored as such in reference to one or more alternative principles. The
hypothetico-deductive stance of the young adolescent reflects this possibility.
By age 13–14 years “reasoners have a metarepresentation of logical validity that
can be used to inform them of the accuracy of their logical deductions, at least when
reasoning about abstract materials” (Markovits, Thomson, & Brisson, 2015, p. 691).
Adolescents become aware of the logical constraints underlying different types of
relations. This is expressed in their ability to discern when an argument is logically
insolvable, as in the so-called fallacies of affirming the consequent or denying the
antecedent. For instance, they understand that no conclusion can be drawn from a
modus ponens-like argument where the second proposition is affirmed. It is
recognized that, if asserted that “birds fly” and “tagi flies”, we cannot conclude that
“tagi is a bird”. This is so because complementary representations can be strung
along a validity principle and evaluated for consistency: (i) the modus ponens
structure, (ii) the information missing (i.e., it is not specified if tagi is a bird), and
(iii) the possibility that other entities, in addition to birds, may fly. Later, principled
thought culminates into a systemic approach allowing the alignment of multiple
principles (e.g., truth-validity-morality) and their reduction into grant frames, such
as an overarching life-orientation (Demetriou & Bakrasevic, 2009; Demetriou
et al., 2011). Formally adolescents understand that accepting “If A then B” does not
allow for the drawing of any conclusion about A if only knowing that B occurred,
or for the drawing of any conclusion about B if only knowing that A did not occur,
because B may be caused by something other than A. It is noted that principle-based
thinkers do not recall all cases from memory; they can activate the general principle
as such and can then derive the implications from the principle, rather than recalling
instantiations case by case. This may occur in rule-based thought, where rules and
their instantiations are activated independently of one another.
At this phase, although rarely in the general population, systems may be aligned
with each other. The use of mathematics for the sake of problem-solving in other
sciences is an example of systemic alignment. Awareness of mental processes
develops into a detailed differentiation between mental functions, such as attention,
memory, and reasoning, and their association with relevant processes, such as choice
and inhibition, recall and association, and inductive and deductive inference, for
each of these three functions, respectively. Adolescents are also accurate in evaluating
their own performance on tasks in different domains and different levels of
complexity (Demetriou & Bakracevic, 2009; Demetriou & Efklides, 1989;
Demetriou, Efklides, & Platsidou, 1993; Demetriou & Kazi, 2006). Models of
performance in this cycle reflect this differentiation.
We showed that cognizance becomes increasingly accurate and cohesive as
indicated by the tightening of the relations between different aspects of cognizance.
In fact adolescents become increasingly able to associate a problem with relevant
mental processes. For instance, if one needs to test a hypothesis, isolation of variables
is the process needed; if one needs to fix objects in the boot of a car, mental rotation
is the process needed (Demetriou et al., in press; Makris et al., 2017). Thus the in-
ferential relevance mastery programme dominating in this phase integrates the
mental flexibility of the previous cycle into an evaluation system yielding evaluations
of the relations between mental spaces vis-à-vis various types of standards. From a
Piagetian point of view this is formal thought (Lourenço, 2016). However, the
emphasis here is on the origins of formal thought, rather than on one of its products
(i.e., hypothetico-deductive reasoning modelled on the basis of symbolic logic).
Another product is control, which in this cycle is very different from the previous
cycles. It is based on using the system of principles outlined above to co-activate
conceptual spaces, such as beliefs and knowledge about study or professional options,
and evaluate them against each other in order to form long-term life plans
(Demetriou & Bakracevic, 2009; King & Kitchener, 2002). Thus self-evaluation in
this phase becomes very accurate in reflecting the emergence of general validity and
truth criteria that may be called on in judging mental outputs relative to goals.
How are developmental cycles related to IQ?

In psychology, intelligence was always associated with cognitive development. In
fact Binet was the first to speak about mental age which refers to the typical cognitive
accomplishments of successive years of life. In Binet’s theory, mental age is defined
in reference to the kind of problems solved at a given year of age. Then the
individual’s IQ is specified as the relation between the mental age attained based on
the problems solved and this individual’s actual chronological age. That is, Binet
defined intelligence as the quotient (hence IQ) of (MA/CA) x 100. Nowadays IQ
is defined as (z × 15) + 100, where z is the z score of the individual on the test and
15 is the standard deviation of the population.
It is interesting to relate the cycles in reasoning development specified above
with the psychometric conception of IQ. For the sake of this aim, we trans-
formed attainment on our battery of reasoning development into an IQ-like score.
This attainment is indicated by the reasoning curve in Figure 9.2. In a sense this
transformation aligns mental age with the levels associated with the developmental
cycles discussed above. It is noted that this battery involved tasks addressed to all
domains of reasoning specified above (i.e., categorical, causal, spatial, analogical, and
deductive). These tasks were systematically scaled in difficulty to tap all three cycles
of development spanning ages from 3–4 to 17–18. We also note that the relation
between this battery and performance on the WISC test is very high (circa .8) (Case,
Demetriou, Platsidou, & Kazi, 2001). The total score on this battery was transformed
into an IQ-like score in the fashion that the raw score on the WISC is trans-
formed into an individual’s IQ. That is, the raw score was transformed into a z score
and this was then fed into the IQ equation: IQ = (z x 15) + 100. Therefore this
transformation shows how different levels of IQ correspond to the cycles of
intellectual development outlined here. It can be seen in Figure 9.2 that an IQ
of 100 points, which is the intelligence of two thirds of the population, corresponds
FIGURE 9.2 Alignment of reasoning development with a general IQ scale

FIGURE 9.3 Theoretical curves showing expected trends in mental age scores for
children with different IQs
to the attainments of the ruled-based concepts that are attained at the age of 9–10
years. Intelligence higher than 120 IQ points would require entering the cycle of
principle-based thought. It is noted that this transformation was also applied to the
performance attained by a Croatian sample of 8–17-year-old participants on Raven’s
standard progressive matrices (Žebec, Demetriou, & Kotrla-Topić, 2015). We
obtained very similar results.
Interestingly Carroll (1997) suggested a similar relation between IQ and MA in
their relation to chronological age. This is illustrated in Figure 9.3. It can be seen
that the IQ of 100 is attained at the age of 10 and the MA of 10 years is identical.
At this age an IQ of 120 would require a MA of 13 years. An IQ of 80 would
indicate a MA of 7.5 years.
Conclusions
In this chapter we proposed a complete model of cognitive accomplishments from
birth to adulthood. These develop in four cycles, with two phases in each. The ages
of transition coincide by and large with those identified by all scholars who studied
cognitive development. We opted for the term cycle, rather than “stage” or “level”
dominating in other theories, to emphasize the recurring nature of developmental

changes.
The emphasis of the theory was on the interrelations between all systems involved
in the four-fold architecture of the mind rather than on any privileged aspect of
it, such as reasoning (e.g., Piaget) or processing efficiency (e.g., neo-Piagetians).
Changes in the relations between these processes cause deep changes in representational
and inferential possibilities; these changes make understanding and problem-solving
appear qualitatively different from cycle to cycle.
The overall organization and sequencing of processes bear properties identified
by several great students of the human mind. We explicitly specify the properties
associated with each scholar to highlight the integrative nature of the theory. Below
we present empirical research that mapped the relations between the various
processes across cycles.
10
RECYCLING AND MEDIATION
OF COGNIZANCE
Following development through the four cycles made it clear that the relations
between executive control, cognizance, and reasoning vary with developmental
phase. Each new form of representation and reasoning is predominantly acquired in
the first phase of each cycle; the interlinking of representations into more complex
systems of reasoning and understanding predominantly emerges in the second phase.
Awareness of the new form of representation is implicit in the first phase of each
cycle and becomes explicit in the second phase; explicit awareness yields insight
about underlying relations and opens the way for the construction of the next cycle’s
representations. It seems that there are three types of developmental phenomena
underlying progression through the cycles. First, the relations between indices of
mental efficiency, such as processing speed and working memory, and reasoning
would change with phase to reflect developmental differences in the mastery of
representations and their relations. Second, cognizance is a central factor in this
developmental process, because it mediates between all processes. On the one hand,
to cognize a mental process requires thinkers to focus on, monitor, record, and
compare it with other processes; on the other hand, the very process of focusing on
each of the other processes refines cognizance itself, which may then subsequently
reflect back improvement in their own functioning. Third, these changes would
transform the very nature of g (or AACog), infusing it with new possibilities for
representation and inference. As a result they alter the relations between each of the
individual processes and g (or AACog) to reflect the transformation of the overall
unit of mental functioning and understanding as specified for each cycle.
This chapter will summarize research on the first two types of developmental
phenomena. Specifically, we will first summarize research showing how the relations
between processing speed and working memory and reasoning change with
developmental phase. We will then present a series of studies highlighting the
mediating role of cognizance between executive and efficiency processes and
Recycling and mediation of cognizance 135
reasoning. The next chapter will summarize several studies which investigated how
the relations between individual processes and general ability change with phase.
Thus this chapter will answer questions about a long-standing dispute in develop-
mental and differential psychology: whether special processes differentiate from or
integrate with g with growth.
Structural relations between speed, working memory,

and reasoning within phases
A few years ago we examined the relations between speed and working memory,
on the one hand, and reasoning, on the other, in each developmental phase of all
developmental cycles but the first one. Specifically, we specified how much of the
variance in reasoning may be accounted for by each of the two factors of interest,
speed and working memory, and also age. To attain this aim, children’s performance
on our comprehensive test of reasoning, which included a large array of inductive
and deductive reasoning tasks in various domains such as spatial, categorical, and
mathematical thought, was regressed on scores standing for processing speed
and working memory in various domains. Obviously the reasoning score stands,
by and large, for psychometric g (or our AACog). The innovation of our modelling
was that we specified these relations separately for each developmental phase: that
is, late realistic representations, early and late rule-based, and early and late principle-
based thought. This approach can show how much the attainment of reasoning in
each phase may be predicted by processing speed and working memory. Using age
in the models allows for separation of what is common between each of the two
predictors and reasoning in each phase regardless of possible age differences of
children in this particular period of time. Obviously this approach may show
possible functional and representational differences in the reasoning across phases
that may differentially relate with each of the two measures of processing and
representational efficiency, namely speed and working memory (see Demetriou et
al., 2013, 2014).
We showed that, at the beginning of cycles, reasoning relates with speed and
attention control more than with working memory; at the end of each cycle this
relation weakens while the relation with working memory strengthens. This pattern
is illustrated in Figure 10.1. It can be seen that in the early phase of each cycle the
speed-reasoning relations were high and the working memory-reasoning relations
were low. This pattern was inverted in the second phase of each cycle, when the
speed-reasoning relations dropped and the working memory-reasoning relations
rose drastically (Demetriou et al., 2013).
It is stressed that these relations were also tested by modelling the results of a large
number of published studies where speed, working memory, and general intelligence
were measured in each of the age phases above. The aim of this meta-analysis was
to test if the pattern obtained from our studies is also present in the performance of
children independently examined in other laboratories in different parts of the
world. It is noted that there were data sets from Australia, Canada, the USA, and
FIGURE 10.1 Recycling in the relations between speed, working memory,

and reasoning
the Netherlands. It is emphasized that these cycles were fully replicated, indicating
that this is a robust developmental phenomenon (Demetriou et al., 2014).
Obviously this pattern of recycling reflects differences in the command of the
inferential processes involved. At the beginning of cycles, processing speed on tasks
requiring attention may increase for several reasons. For instance, individuals master
the new executive programme with increasing skill. In the first phase of realistic
representations, children become increasingly able to focus on representations,
select those which are relevant, and inhibit irrelevant ones. At the beginning
of rule-based representations, children become increasingly able to focus on under-
lying relations and encode them into rules. At the beginning of principle-based
representations, adolescents become increasingly able to deal with abstract or multi-
dimensional concepts. In short, command of the new control programme and
related representational unit improves swiftly at the beginning of cycles and thinking
in terms of it proliferates to new content.
Later in the cycle, when the control programme is transcribed in different
conceptual domains and networks of relations between representations are worked
out, working memory is a better index because alignment and interlinking of
representations both requires and facilitates working memory. It is reminded that
alignment dominates in the second phase of each cycle. We suggested that alignment
is an executive process of representational integration that involves shifting between
representations. What is interesting to note is that alignment so defined is actually
the expression of executive control in each cycle which is adapted to the nature of
representations dominating in each cycle. Thus, given the close relations between
executive control and working memory, it is not accidental that in the alignment
phase of each cycle working memory dominates as a predictor of reasoning. That is
because working memory expresses the executive possibilities available in a cycle

better than any other process, reflecting the increasing expertise in combining and
interlinking representations and concepts.
Specifying the mediating role of cognizance

The findings above show that the relations between processes vary with developmental
phase. In this section we explore how cognizance mediates between executive con-
trol and reasoning, contributing to the specific profile of developmental g in each
phase and its transformation across phases. In the description of development across
phases we showed how reasoning and executive control interact with cognizance. To
cognize a mental process, thinkers must focus on, monitor, record, and compare it
with other processes. Actually, reflecting on mental objects requires executively
varying them and reason on the variations produced and on their relations. In
developmental time, cognizance is a background process on which executive control
is formed, because it generates judgments about the current problem that may guide
further; these judgments may subsequently merge to generate new concepts and
mental operations. Each level of executive control reflects an interaction between the
dominant mode of representation and the level of awareness available.
In the cycle of reality-based representations, representations as such, together
with awareness of their perceptual origins and the beliefs they generate, may be
connected in pragmatic reasoning sequences. However, in this phase there is no
grasp of their possible mutual constraints or knowledge-based constraints. Thus
basic common assumptions may be seriously violated, giving the impression of lack
of logic. For instance, Athina, the first author’s granddaughter, wonders: “When is
Nicolas (her cousin, 30 months old and three months younger than her) going to
grow up and become a girl? I am grown up and a girl.” Obviously there is no lack
of logic here. The reasoning is sound, obeying a modus ponens structure: I grew up
and I am a girl; when one grows up she becomes a girl; Nicolas will become a girl
when he grows up. The mistake is not in the reasoning sequence but in the
recognition of a reality constraint that derives from knowledge: gender identity does
not change with age. In the cycle of rule-based representations, representation of
rules and awareness of their underlying inferential links and related concepts may
generate systematic arguments and necessary conclusions but does not allow the
envisaging of alternatives to the conclusions generated. Finally, in the cycle of
principle-based thought, explicit representation of validity and truth principles and
awareness of the differences between states of knowledge (perception-based,
information-based, logically-based, etc.) can generate “alternative worlds” with
truth and validity values attached to each based on the epistemic state of each. In
short, reasoning implements the dominant executive possibilities into specific
information integration/evaluation processes calling on cognizance when needed.
A series of studies focused on the relations between reasoning, executive
control and cognizance, the aim being to specify how cognizance mediates
between the executive processes already discussed and reasoning. Three studies
are summarized here.
The mediating role of cognizance from 4–7 years

The first study involved children aged 4–7 (see Kazi et al., 2012, for details). The
reasoning battery involved tasks addressed to the following types of reasoning:
simple arithmetic reasoning (counting from 3–9 objects and adding numbers smaller
than 10); spatial reasoning (picture assembly and mental rotation); deductive
(pragmatic) reasoning (modus ponens, conjunction, and disjunction); and analogical
reasoning (i.e., numerical, spatial, and verbal analogies). Examples of tasks are
described in Box 17.1 in Chapter 17. Cognizance was examined by tasks addressed
to awareness of mental processes and awareness of the mental demand of tasks. Six
of the cognitive tasks summarized above (two from each domain, clearly differing
in difficulty) were depicted in separate pictures. For example, for quantitative
thought there was a child adding 3 cubes and a child adding 5 cubes (see pairs used
in Box 8.2 in Chapter 8). Children were first asked to describe each picture in order
to focus on the activities concerned. They were then asked to judge if the two
children employed the same thinking and indicate the child having the easier job:
“Is the job of this child the same as the job of this child?”; “Who is doing the easier
job?” Thus 12 scores (6 similarity estimations and 6 difficulty estimations) were
obtained. The first 3 pairs addressed comparison of tasks belonging to the same
domain (quantitative, deductive, and spatial reasoning, respectively) and the rest
addressed comparison of tasks belonging to different domains (quantitative-
deductive, quantitative-spatial, and deductive-spatial, respectively). Thus increasing
scores on these tasks reflected a shift of awareness from superficial perceptual
characteristics to mental processes. For instance, focusing on perceptual characteristics
included answers such as the following: “There are the same cubes in the two
pictures”; and “the cube in this picture is similar to the square in this picture”.
Focusing on mental processes included answers such as the following: “They are
both counting”; “one is counting, the other is classifying”; “it’s easier to count few
than many cubes”; it’s easier to count than to understand a story.
BOX 10.1
TASKS USED TO EXAMINE THE MEDIATING
ROLE OF COGNIZANCE
Processing efficiency
Speed of processing: specify the location of simple geometrical figures presented
either to the left or the right side of the screen and decide if two letters (Latin,
Arabic, and Chinese ideograms) presented side by side were similar or different.
Executive control: these tasks included a learning phase and a control phase. In
the learning phase a response set was built. For example, children were trained
to press on the stimulus-response box a figure matching the figure shown on
the screen as fast as possible. In the control phase children were instructed to
choose the key not showing the item projected on the screen. Therefore the
test examined the ability to inhibit the dominant tendency to choose the key
matching the projected stimulus in favour of the weaker but relevant response
“shift to the other one”.
Working memory was examined by the Corsi task addressed to spatial working
memory and a phonological task involving regular words and pseudo-words.
Reasoning
Simple arithmetic: counting from 3–9 objects; finding the sum of 1+2, 2+3, and
3+4.
Spatial reasoning: picture assembly and mental rotation, e.g., assemble a

square, a triangle and a circle into a house.
Deductive (pragmatic) reasoning: modus ponens, conjunction, and disjunction;

e.g., point to the picture standing for the expression “If Sally wants to play
outside, she must put her coat on”.
Analogical reasoning: numerical, spatial, and verbal analogies.
Awareness of mental processes. The pairs of tasks were as follows:
Quantitative thought: (1) a child adding three cubes and (2) a child adding five
cubes.
Deductive reasoning: (3) a child hearing a story asking her to obey one rule and
(4) a child hearing a story asking her to obey two rules.
Spatial thought: (5) a child reproducing a figure consisting of three components

and (6) a child reproducing a figure consisting of five components.
Six pairs of pictures were presented to the children:
(i) The two addition tasks, (pictures 1 and 2);

(ii) the two story-hearing tasks (pictures 3 and 4);
(iii)the two figure-reproduction tasks (pictures 5 and 6);
(iv) the easy addition and the easy story-hearing tasks (pictures 1 and 3);
(v) the easy addition and the easy figure-reproduction tasks (pictures 1
and 5);
(vi) the easy story-hearing and the easy figure-reproduction tasks (pictures 3
and 5).
Children described the pictures and then answered two questions:
1. Is the job of this child the same as the job of this child? Why do you
think so?
2. Who of the two children is doing the easier job? Why do you think so.
Processing efficiency and executive control were examined by several tasks.

Speed of processing tasks required a response to the location of simple geometrical
figures (i.e., the left or the right part of the screen) or a decision as to whether two
letters presented side by side were similar or different (Latin, Arabic, and Chinese
ideograms). Executive control was examined by a task addressed to inhibition and
shifting. Specifically, a response set was first built and then children had to inhibit
the dominant response to the relevant stimulus and exhibit a different response (see
Box 8.3 in Chapter 8). Working memory was examined by the Corsi task addressed
to spatial working memory and two phonological tasks, one involving regular words
and another one involving pseudo-words.
We tested several structural equation models aiming to examine if cognizance
mediates between executive and inferential processes and if mediation operates
bottom up, from executive to inferential processes, or top-down, from inferential
to executive processes. In these models there were factors for the following general
processes: simple speed, letter recognition speed, attention control, and working
memory; cognizance, related to awareness of similarities and differences between
mental processes and their relative difficulty; and reasoning, standing for Gf (or
AACog) and related to spatial, deductive and inductive reasoning. In the bottom-up
model the following relations were built. First, the four executive control factors
(i.e., speed, letter recognition speed, attention control, and working memory) were
taken as the background factors which were directly related to age. Second, the
cognizance factor, standing for performance on the various awareness tasks, was
related to all four executive factors. Thus this model specified how each of the four
executive processes contributes to the state of awareness reflected by the cognizance
factor. Third, the general reasoning factor was regressed on cognizance. Thus
cognizance was thereby upgraded into a mediating factor carrying the effects of the
executive factors to the reasoning factors.
In the top-down model the flow of effects was inverted. Specifically, the three
reasoning factors (i.e., inductive, deductive, and spatial reasoning) were taken as the
background factors which were directly related to age. The four executive factors
(speed, learning, attentional control, and working memory) were related to a
common factor that stands for executive control. This executive control factor was
regressed on cognizance. Thus this model specifies how much each of the reasoning
factors contributes to cognizance; this is then upgraded to a mediating factor carrying
the effects of reasoning onto the executive factors. These models are depicted in
Figure 10.2.
These two models were first tested in a two-group analysis where the first group
involved 4–5-year-olds and the second 6–7-year-olds. Thus we can map the
relations dominating in the second phase of reality-based representations and
the relations dominating in the first phase of rule-based representations. In the
bottom-up model the cognizance-reasoning relation was very high and significant
in both the younger (.76) and the older (.66) age group. In the top-down model,
the relations between cognizance and executive control were low and significantly
weaker than in the bottom-up model (.27 and .10 in the younger and the older age
FIGURE 10.2 The models of the mediation of cognizance between executive control
and reasoning. Black arrows represent relations in the top-down models, carrying
effects from reasoning to executive control; grey arrows represent relations in the
bottom-up models, carrying effects from executive control to reasoning
group, respectively). Therefore the various executive functions, primarily working

memory and attention control, do contribute, in both age groups, to the emergence
of awareness which is then used in handling reasoning. Reasoning, primarily
deductive reasoning, also contributes to the emergence of awareness. However,
awareness is weakly used in handling executive processes in the period from 4–7
years. The width of the various arrows symbolizes these differences.
To further explore the involvement of attention control in the emergence and
use of awareness we tested the models above in groups of children clearly differing
in their ability to focus attention and respond accordingly in speeded performance
tasks. Specifically, the total sample was split into two groups based on accuracy of
responses to the various speeded performance tasks: children succeeding on at least
80% of the reaction time tasks were allocated to the high-attention control group.
The rest were allocated to the low-attention control group. The simplicity of stimuli
presented in these tasks suggests that accuracy in responding reflects control of the
focus of attention and matching rather than more complex information processing.
Highly accurate persons are obviously able to efficiently focus attention on the
stimulus shown, encode it and respond as required, regardless of speed. In less
accurate individuals, focusing or matching may go astray in the process. As is usual
in research using speeded reactions, only reaction times of accurate responses were
used. This ensures that the speed of the processes of interest is analysed, rather than
irrelevant processes.
Each of the two models, the bottom-up and the top-down model, was tested
twice. During a first round the relations between processes were specified without
any manipulation of age. Thus, in these models, the relations between processes may
reflect possible across-the-board changes caused by the progression of age. During
a second round the possible influence of age was statistically removed. Thus, in these
models, the relations between processes in the two groups are purified against the
possible influence of age, reflecting their functional-operational rather than their
developmental relations. In the bottom-up model the relations between cognizance
and reasoning were very high and significant in both the less (.78) and the highly
(.98) accurate group. These relations remained strong even after the possible effect
of age was removed (.53 and .47, respectively). In the top-down model the relation
between cognizance and executive control was very low and non-significant in the
low-attention control group (.15) but considerable in the high-attention control
group (–.42). Interestingly, this difference between the two groups disappeared
when the possible influence of age was removed.
This study revealed some interesting findings about the role of cognizance as
a mediator between executive control and reasoning in the period of transition
from reality-based to rule-based thought. First, executive control systematically
contributes to the emergence of awareness about mental processes. Second, this
awareness, when acquired, is carried up, enabling children to put their processing
resources (i.e., focusing, flexibility, and representing in working memory) into
the service of information integration and reasoning. One might ask how is the
bottom-up mediating effect of cognizance distributed among the three reasoning
domains? In both groups it affected inductive reasoning (.75 and .87 for the two
groups, respectively) more than deductive (.48 and .29) and spatial reasoning (.52
and .44). Moreover, this effect came primarily from working memory (.54, .34, .38
for inductive, deductive, and spatial reasoning in the low-attention group and .42,
.14, .21 in the high-attention group, respectively). All other indirect effects were
very low and non-significant.
Third, the top-down models suggested that the experience of reasoning also
contributes to the emergence of awareness. However, carrying this awareness down
to the functioning of executive processes is weaker than carrying it from executive
processes to reasoning. In fact only individuals who are already in good command
of attention control were able to capitalize on their reasoning-based awareness to
efficiently steer their executive processes. Specifically, in these models, cognizance
significantly affected all processes in the high-attention group (.40, –.34, and –.36
for working memory, speed, and control) but not in the low-attention group (.06,
.12, and .14, respectively). These effects came from deductive reasoning (.15, –.13,
and –.14) rather than from any of the other two reasoning domains (all circa .05).
It is reminded that inductive reasoning is implemented more easily than deductive
reasoning, which is more demanding and effortful. In inductive reasoning,
conclusions are often automatically drawn. In deductive reasoning, premises must
be represented explicitly, straining working memory, and alternative models
must be considered, straining the inferential process. Therefore deductive reasoning
generates awareness of mental processes more than inductive reasoning because it
requires executive control and reflection on representations and relations. It is also
notable that children high in attention control were more able to capitalize on this
experience of deductive reasoning and use the awareness acquired top-down to
efficiently steer executive control. Obviously these children are more reflective than
their low-attention control peers. Let us examine how these relations change in later
periods of life.
Cognizance from early to late childhood

A second study explored the mediating role of cognizance by zooming in on
the processes involved in it (see Spanoudis et al., 2015, for details). This study
involved 344 children about equally drawn from each of the years 4 to 10.
Specifically, this study examined speed, attention control, conceptual control, and
working memory by tasks similar to those used in the study above. For reasoning,
this study involved a Raven-like test specifically designed for the present purposes.
This test involved matrices addressed to three levels of complexity: matrices at
the first level involved a single dimension and examined the ability to uncover the
pattern defining this dimension (e.g., same colour-same size, increasing size, same
size-alternating colour); matrices at the second level examined the ability to conceive
of the intersection between two dimensions (e.g., animal and colour, animal and
size, colour and size); while matrices at the third level examined the ability to
conceive of the intersection between three dimensions (colour, shape and size,
animal, colour, and size, and activity, colour, and size). Thus the three levels
addressed late reality-based, early rule-based, and late rule-based thought, acquired
at ages 4–6, 7–8, and 9–10, respectively.
Several tasks addressed awareness of perception and awareness of inference as
sources of knowledge. In the perceptual awareness tasks children saw a teacher
placing two toy cars in same-colour boxes (red car in red box, green car in green
box) in front of a protagonist child, John. The teacher also described what he did
in the case of the first toy car (i.e., “I put the red car in the red box”, etc.) but not
the second. Then John went away. Thus John saw and heard where the red car was
placed, saw but did not hear where the green car was placed, but he neither saw nor
heard where the blue one was placed. He was then invited to come back and was
asked to find each of the cars. The participant was asked for each of them in
succession: “Where is the red (green, blue) car? Does John know where the car is?”,
“How does he know?” After looking for the blue car, the participant was asked
about the teacher’s reason for placing the blue car in the blue box: “Why did she
place the blue car in the blue box?”
In the inferential awareness task the same cars and boxes were used. However,
in this video the protagonist child (Ann) sat across the table in front of the teacher
rather than next to her. After naming all objects as above, the teacher raised a
wooden separation between them so that the child could not see what she was
doing. She described her actions only while placing the red car in the red box (e.g.,
“I now put the red car in the red box”). She made no reference to the green or the
blue cars. Ann went away and came back, supposedly after one hour. She was asked
to find where each car was located and explain why.
Therefore, in the John task, answers about the red and green cars reflected
perceptual awareness, because John’s knowledge about the location of the cars came
from seeing or hearing where they were placed; answers about the blue car reflected
inferential awareness, because John never saw a blue car placed in a blue box. This
can only be inferred by extrapolation from perception: cars are placed in same-
colour boxes, so the blue car must go in the blue box. In the Ann task, answers about
the red car reflected perceptual awareness; answers about the green and the blue cars
reflected inferential awareness. These tasks were used together with the classic Sally
task addressed to theory of mind. Thus, in addition to mapping the development of
perceptual and inferential basis of knowledge, this study may highlight how these
two forms of awareness relate with theory of mind (see Kazali, 2016; Spanoudis
et al, 2015).
Bottom-up and top-down mediation of cognizance between processing
efficiency and reasoning was modelled for 4–6 and 7–10-year-old children in the
fashion described above. Specifically, to test bottom-up mediation, speed, attention
control, conceptual control, and working memory were regressed on age. Each of
the two forms of awareness and theory of mind were regressed on all of these
executive processes. Reasoning was regressed on all forms of awareness. In the
younger age group, reasoning was significantly related to perceptual awareness,
including theory of mind (.22) but not to inferential awareness. In this group,
perceptual awareness carried significant but weak effects from attention control
(.12) to reasoning. In the older age group, reasoning was negatively related
to perceptual awareness and theory of mind (–.81), obviously reflecting the fact that
perceptual awareness had approached ceiling in this age group. However, reasoning
was positively (and significantly) related to inferential awareness (.29). In this group,
inferential awareness carried effects from speed (.30) and attention control (.40) to
reasoning.
In the top-down model in the younger age group the awareness factors were
regressed on the factor standing for performance on the Raven-like test and the
executive factor was regressed on the awareness factors. Notably, the top-down
effects of reasoning on perceptual (.82) and inferential awareness (.73) were very
strong. The effect of perceptual awareness on the general executive control factor
(.51) was strong and significant; the effect of inferential awareness was low and
non-significant. Thus perceptual awareness carried significant indirect effects on

speed (.28), attention control (.48), flexibility (.28), and working memory (.41).
The indirect effects of inferential awareness on these four processes were low and
non-significant. In the older age group there was no effect of perceptual awareness
on executive control; however, there was a moderate but non-significant effect of
inferential awareness (.18). Thus all indirect effects carried over were negligible.
This is obviously due to the fact that the top-down effects of reasoning on perceptual
awareness (–.10) and inferential awareness (.23, significant) were weak.
To specify the mediating role of cognizance from 9–15 years we used a recent
study which involved measures of all processes required to examine mediation
between executive processes and reasoning (Makris et al., 2017). For the present
aims the mediation model applied above was tested in a three-group analysis: 9–11,
11–13, and 13–15-year-old participants. In the bottom-up model, three first-order
executive control factors (attention control, flexibility, and working memory) were
regressed on age, the cognizance factor was regressed on the three executive control
factors, and a g factor (associated with language and reasoning in the four reasoning
domains examined in this study) was regressed on the cognizance factor. In this
model the relation between cognizance and g relation was high in all three age
groups (.53, .78, .70, respectively). In the top-down model the language and the
reasoning factors were regressed on age and the three executive control factors were
regressed on a general executive control factor; this factor was regressed on cogni-
zance to capture the top-down mediation of cognizance between g and executive
control. The relation between these two factors were also significant and high in all
three age groups (.52, .93, –.54, respectively). It is noted here that the strength of
the relation increased significantly from 9–11 to 11–13 years but then dropped and
turned negative, reflecting the fact that various aspects of attention control
and shifting reached ceiling by the age of 13 years while cognizance continued to
develop.
Conclusions
The studies presented in this chapter capture three major tendencies about the role
of cognizance.
First, the mediation of cognizance between executive and reasoning processes is
cycle-specific. That is, in each cycle it is exerted through the processes underlying
the management of representation in each cycle: perception-based aspects of
representation in the representational cycle; rule-based inferential processes in the
rule-based cycle; and abstract semantic processes in the principle-based cycle. This
obviously implies that cognizance is a higher-order monitoring process that registers
the representations and the sources of knowledge available. When new higher-order
representations enter, cognizance turns onto them, often letting earlier representations
go unnoticed, as they are automated and thus in no need of supervision.
Second, bottom-up mediation is stronger than top-down mediation. This implies
that lower level executive processes—attention control and working memory in
particular—generate awareness more than inferential processes in early and middle

childhood. On the one hand, this type of mediation is difficult to scan in adolescence
because various aspects of attention control and flexibility reach ceiling. On the
other hand, top-down mediation is not attained before the second phase of rule-
based thought. Further, top-down mediation is more likely to be attained by
individuals who, when in early childhood, are proficient in control of attention; in
late childhood they are proficient in mastering inferential processes demanding
reflection, such as deductive reasoning.
Third, awareness of different modalities, such as vision and hearing, and broad
mental functions, such as perception and inference, and their role in generating
knowledge, is easier and comes earlier than analytic awareness of specific mental
processes, such as categorization, mental addition, mental rotation, etc. In fact
awareness of modalities starts in preschool years; analytic awareness starts in late
childhood and culminates in adolescence. We will focus on this developmental
process in the next chapter. Suffice it to note here that this later form of cognizance
is more sensitive in differentiating between bottom-up and top-down mediation
because it spots individuals who are high in reflection and self-monitoring.
Obviously the relation of each specific process with g varies, depending on both the
representational state of g in the various processes and the special role of each process
in each phase. These relations will be further explored in the next chapter, which
presents a series of studies designed to specify how each special process relates with
g in each developmental phase.
11
DIFFERENTIATION AND
BINDING OF MENTAL
PROCESSES THROUGH
DEVELOPMENTAL CYCLES
It is clear that patterns of change vary with process and phase. This may reflect two
related but distinct types of processes. On the one hand, it may indicate that the
strength of relations between specific abilities and general ability varies with phase,
depending on the developmental priorities in the construction of general ability in
each phase. Changes of this kind are related to a question that has been debated for
decades in both psychometric and developmental psychology: are mental processes
integrated or differentiated from each other with increasing ability or growth? On
the other hand, change in developmental patterns may demarcate when a new
ability emerges and when it reaches a level of relative stability. In turn, changes in
these patterns may indicate when this ability intertwines with g in the phase
concerned. For instance, change in a particular mental process M may accelerate
after g reaches a particular level (partly associated with age) to match this level and
it decelerates as it approaches this level.
Psychometric theory and developmental theory agree that mental possibilities
change with growth. Mental age in psychometric theory and stage in developmental
theory both capture enhancement of mental ability with age. They both indicate
that individuals deal with concepts and problems of increasing abstraction and
complexity as they grow. Developmental theory considers stages as ideal epistemic
states corresponding to successive age periods. In the present theory, the studies
presented so far showed that the profile of ability as specified by the various
representational, executive, cognizance, and reasoning processes re-morphs from
phase to phase.
Several mechanisms were invoked to account for developmental progression and
ensuing re-morphing of mental ability with age. The twin mechanism of integration/
differentiation of mental processes is a major mechanism of development. Cognitive
developmental theories postulate that increasing ability comes from increasing
integration of mental processes (Case, 1985; Fischer, 1980; Piaget, 1970). Piaget’s
equilibration is a developmental mechanism of integration generating mental

structures that become increasingly precise in their environmental implementation.
Tuning with the environment indicates some kind of differentiation because abilities
may efficiently differentiate according to the specificities of particular situations
and operate accordingly. In psychometric theory, differentiation refers to variation
in actual abilities that goes with increasing general intelligence. Specifically, psycho-
metric theory postulates that increasing g allows increasing differentiation of
cognitive abilities, because enhanced mental power may be invested into domain-
specific learning according to interests and priorities, causing domains to differentiate.
This is Spearman’s Law of Diminishing Returns for Age (SLODRage) (Jensen,
1998; Spearman, 1927). The developmental adaptation of Spearman’s differentiation
hypothesis would assume that abilities differentiate with growth because g increases
with development.
One might argue that differentiation of abilities in psychometric theory is closer
to the developmental notion of accommodation than integration. This is because it
is recognized that increasing ability allows increasingly refined learning in domains
of one’s choice. In fact the notion of integration as a mechanism of change is rather
alien to psychometric theory. For one thing, we have already shown that g is a
constraining force carried over different tasks, rather than an integration mechanism
in the fashion of developmental equilibration. For another, psychometric theory is
not basically concerned with developmental change. Even mental age, which
reflects a developmental dimension, is a statistical construct for psychometric theory,
reflecting relative successes and failures of persons of different ages on task batteries
rather than genuine developmental differences. Hopefully this chapter will highlight
that this construct may be specified in relation to cognitive developmental processes,
yielding an integration of psychometric g as an index of a ceiling of functional
constraints and developmental g as an index of the quality of representational and
understanding possibilities at a particular age phase.
Changes in factor structure with growth

Technically, the decrease of correlations between abilities with increasing g was
considered as evidence favouring differentiation. In terms of factor analysis, the
equivalent would be an increase in the number of factors needed to account for
performance of high g individuals as compared to lower g individuals. According
to Detterman’s (1987) theory of mental retardation, the malfunctioning of central
mechanisms in individuals with low intelligence causes homogeneously lower
performance across abilities: hence higher correlations and stronger g. Recent
research provided rather weak and inconsistent evidence, probably because increases
in ability and age go partly together and thereby confound each other. Specifically,
some studies did find the expected pattern of decreasing correlations or increasing
number of factors with increasing age (e.g., Deary et al., 1996; Demetriou et al.,
2013; Reynolds, 2013; Tideman & Gustafsson, 2004) but others did not (Carroll,
1993).
Differentiation and binding of processes 149
FIGURE 11.1 Factorial structure across cycles

Note: the symbols Ex, Ph, and Vi stand for executive, phonological, and visual working
memory, respectively; the symbols, Q, V, and S stand for quantitative, verbal, and spatial
reasoning, respectively; the symbols Sp, PC, and Eff stand for speed, perceptual control, and
efficiency, respectively (based on Fig. 3, Demetriou et al., 2013)
In one of our studies we examined participants aged 4–16 on several measures

of processing speed and attention control, executive, phonological and visual
working memory, two specialized capacity systems (SCS), quantitative and spatial,
and deductive and inductive reasoning (Demetriou et al., 2013). We used
confirmatory factor analysis to examine the optimum number of factors needed to
account for performance in three age phases: 4–7, 6–11, and 10–16 years.
Obviously these phases correspond to representational, rule-based, and principle-
based thought. We opted to let the three phases overlap in years so that any
possible changes across neighbouring phases emerge more clearly. It can be seen
in Figure 11.1 that in the representational cycle one single factor was sufficient to
account for performance on working memory and reasoning tasks. This factor was
related to processes indicating processing efficiency, such as speed and attention
control. This is highly interesting because in this cycle children do not yet
differentiate between different representational and inferential functions. In the
next cycle, two interrelated factors, one for working memory and one for reason-
ing, were needed to account for performance. Both of these factors were related
to the efficiency processes. It is reminded that differentiation between mental
functions in awareness emerges at about the age of 8, when children show aware-
ness of underlying mental processes connecting representations such as syntax in
language or inference in reasoning. Interestingly, in the present study repre-
sentational processes separated from inferential processes at the level of factors
underlying performance. This is an informative convergence between measures of
awareness and measures of actual cognitive performance. Finally, a three-level
hierarchical model accounted for performance in the third cycle. In this model,
three factors emerged: processing efficiency, working memory, and reasoning.

Processing efficiency resided at the fundamental level, sending influences at
working memory which resided at an interfacing level, and thus sending effects to
the reasoning factor which resided at the top. In line with our findings, Facon
(2006) found that ability differentiation is age dependent, showing up in late
childhood.
However, other researchers did not find any increase in the number of factors
with development (Hartman, 2006). Interestingly, Carroll (1993) himself analysed
several data bases in search of evidence for differentiation. He concluded “that this
is a phenomenon whose evidence is hard to demonstrate. . . . The question of
age differentiation is probably of little scientific interest except possibly at very
young ages. It is, if anything, of more scientific interest that the same factors are
found throughout the life span” (p. 681). Realistic as it might sound, this
interpretation begs the question of the mechanism underlying systematic changes
and/or differences in mental age. Our theory offers a reason for this state of affairs.
Specifically, this theory suggests that differentiation may vary according to
developmental phase and the dominant representational characteristics of g. That is,
both differentiation and strengthening of relations between specific mental processes
and g are possible. However, these changes in the relations between g and specific
abilities depend on the developmental priorities in the formation of g in each phase.
The studies presented below demonstrate this phenomenon of changing patterns of
differentiation and strengthening of relations between special processes and g with
developmental phase.
Mapping changes in structural relations and developmental

patterns
We employed two complementary methods to explore these phenomena. The first
method is appropriate to capture changes in the structural relations between specific
processes and general ability, and was recently proposed by Tucker-Drob (2009;
Cheung et al., 2015). The second method focuses on possible changes in
developmental patterns as a function of age-related ability. The two methods
together can show how changes in developmental patterns of specific abilities
reflect changes in the formation of general ability. We first explicate the two
methods below and then present their application on several data sets spanning the
lifespan.
The Tucker-Drob differentiation model

This is a structural equation model allowing the testing of possible differentiation
of abilities with increasing g and/or increasing age. This model, illustrated in
Figure 11.2, specifies how each specific ability relates with the following factors:
age, a factor standing for possible differentiation of abilities from g according to
increasing g, and g, a factor standing for a possible differentiation of abilities as a
FIGURE 11.2 The general model for testing possible differentiation of mental processes
from general intelligence (g) (based on Fig. 1, Demetriou et al., 2017)
Note: each process is regressed on age, a common factor (g), quadratic g standing for ability
differentiation, and the g x age product, standing for age differentiation
function of their increase with age. Technically, each ability is regressed on the
following factors:
1. Age as such, to separate any possible influence of age on the relations between
abilities.
2. A common factor standing for g, to separate any possible influences coming
from what is common between abilities which may always be present, regardless
of the level of ability.
3. Quadratic g (i.e., g squared); obviously squaring g magnifies differences between
possibly varying degrees of ability, allowing capture of any possible changes in
the relations of a specific ability with g at higher levels of g. This is the ability
differentiation index.
4. The product of age x g; this factor stands for developmental g and it allows capture
of any possible changes in the relations between different levels of ability as
associated with increasing age. This is the age differentiation index.
We will not overburden the presentation here by detailing the technicalities of

modelling. The interested reader is referred to a recent paper where full details are
presented (Demetriou, Spanoudis, et al., 2017). Suffice it to say that the model can
specify the relation of each specific ability with g and developmental g and specify
if they differentiate from or intertwine with it in each developmental phase. Overall,
differentiation would be indicated by a pattern of relations such that at higher levels

of g or developmental g a specific ability is lower than expected by this high
level of the general factors. An increase in a specific ability that is proportional to
the increase of the general factors would indicate that this ability gets increasingly
intertwined with the general factors.
Logistic growth
Change in the rate of change of a mental process M would alter its relations with
other processes if the rate of change across them is not the same. By implication,
this would be reflected in changes in the relations between this particular process
M and g, as g is a composite index of many other abilities. It is notable that the
model of nonlinear logistic growth is considered appropriate to describe development
of most mental abilities. This model posits that change is slow when an ability
emerges, accelerates later, attaining maximum rate of change around the middle of
its course, slowing afterwards as the ability approaches its final level, when a new
cycle will start (Grimm, Ram, & Hamagani, 2011; McArdle and Nessleroade, 2003;
van Geert, 1998, 2000). This relation is formally stated in Box 11.1 and illustrated
in Figure 11.3. One may assume that the relations between a specific process and g
BOX 11.1
THE EQUATION FOR THE LOGISTIC GROWTH
CURVE
l (1)
f ( x) =
1 + e − a ( t −t 0 )
In this equation l represents the peak of a developing ability or the final level
of the ability at the end of the growth spurt, a is a parameter stretching or
compressing time of development, quantifying the rate of change during the
growth spurt, t and t0 symbolize the age points at which the growth spurt
begins and reaches a midpoint, respectively. The curve becomes steeper as
the distance between t and t0 becomes smaller approaching 0, reflecting
increased growth rate. It is predicted here that the variations in the relations
between specific processes and g to be uncovered by the differentiation
model above can only be understood if coupled with the fundamental non-
linear logistic growth model. Specifically, the relation between a mental
process M and g strengthens with decreasing distance between t and t0.
After the crucial peak point, when development decelerates, the M-g relation
gets looser, allowing differentiation. This alternation of tightening and
loosening of M-g relations according to developmental phase is illustrated in
Figure 9.3.
in each phase strengthen in the middle of the phase to reflect that this ability is
incorporated in g, thereby coming under its control.
To capture these changes we used a special method: segmented or piecewise
linear regression. This method allows one to specify if the rate of change in the
specific process of interest changes at particular regions of change in the general
ability. To obtain a developmental index of general ability for each study we used
a weighted score of the performance attained by each person on all tasks used in a
study (the factor score of each individual on the first principal component); this
score was multiplied by each person’s age. Obviously this index is identical to
developmental g described above.
In the various models tested we always used this developmental g as the explanatory
variable and the scores standing for each specific process as variables to be specified
vis-à-vis developmental g. In each case we compared a linear model assuming that
increases in the specific ability of interest are proportional to increases in developmental
g with various segmented models. Segmented models assume that the degree of change
in a specific ability is not the same in different regions of developmental g. For instance,
change in the specific ability is faster at the lower levels of developmental g, decelerating
at its higher levels. This would appear as successive regression lines with different
slopes. Thus these models provide estimates of direction and rate of developmental
change for specific abilities as a function of general developmental ability, showing
how a specific process changes in different regions of this general ability.
FIGURE 11.3Idealized curves of the integrated integration-differentiation logistic

growth model (based on Fig. 2, Demetriou et al., 2017)
These models allow one to see if there are break points in development that match
the logistic growth model and may be taken to stand for qualitative changes in
development corresponding to the cycles and phases discussed in the previous
chapter. Obviously these developmental models may highlight the developmental
mechanism underlying the relation between specific processes and general ability
operating across developmental cycles (Crawley, 2007).
Mapping integration-differentiation patterns in structure

and development
We tested the two models on several studies to highlight how various specific
processes interact with general developmental ability to produce the patterns shown
in Figure 11.2 (Demetriou, Spanoudis, Kazi, Mouyi, Žebec, Kazali, Golino,
Bakracevic, & Shayer, 2017).
Markers of developmental g in the cycle of realistic

representations
It was shown above that mastering attention control and mental representation is
the primary developmental task in the cycle of realistic representational thought.
This may express itself in several forms: control of attention focus, control of shifting
between stimuli or responses, and linking an action sequence to a specific plan, as
in puzzle completion, etc. For the sake of the chapter’s aims, we tested the two
models above on the study presented before, which included participants aged 4–16.
It is reminded that participants were examined on processing speed, attention
control, verbal and visual-spatial working memory, and reasoning (see Demetriou
et al., 2013).
We tested the Tucker-Drob differentiation model on three age periods: 4–6,
7–11, and 12–17 years, corresponding to realistic representation-based, rule-based,
and principle-based thought, respectively. We found that in the cycle of realistic
representations, from 4–6 years, speed, attention control, and visual working
memory get strongly intertwined with developmental g, suggesting that attentional
and visual working memory processes build up in this phase, powerfully contributing
to the formation of g. To fully capture the formation of g in this phase, the Tucker-
Drob model was also applied on the studies which investigated the development of
various aspects of cognizance from 4–10 years of age. This analysis showed that in
the 4–6-year phase, perceptual awareness, including theory of mind, also strongly
intertwined with g. However, reasoning itself was not a factor in the formation of
g in this phase in any study. Obviously executive control and perceptual awareness
are major markers of the build-up of general cognitive ability in this phase. However,
reasoning as such is not a marker of this process. The reader is invited to inspect
Figure 11.4, which shows the patterns of change for all processes discussed here as
specified by segmented modelling. Attention is drawn to change in attention control
from 4–20 years. It can be seen that change in attention control accelerates greatly
with developmental g from about the age 5–7; it then starts to slow down until it
basically levels off after the age of 9. Flexibility in shifting appears to develop in two
similar phases: the first, from 4–6, years is related to attentional focus and it enables
one to shift focus across external stimuli according to goal; the second occurs in the
next cycle, from 8–10 years.
Markers of developmental g in the cycle of rule-based

representations
The primary developmental task of the next cycle of rule-based thought is inferential
control. That is, the command of the inferential process so that it can fill in gaps of
information systematically. This is primarily expressed through changes in awareness
of the inferential processes and also through improvements in the application of the
inferential processes as implicated in analogical and Raven-like tasks. Thus, in this
cycle, inductive reasoning and inferential awareness must be the major markers of
developmental g. We tested this assumption in several studies. One of them was the
study that explored the development of awareness from age 4–10 and already
described above. We found that inductive reasoning and cognizance of cognitive
processes strongly intertwined with g, with a peak of intertwining in the phase from
8–10 years. Segmented modelling showed (see Figure 11.4) that there were two
break points in the development of inductive reasoning in this period of time. One
FIGURE 11.4Idealized development of mental processes as a function of

developmental g (g x age) from 4–20 years of age
Note: this figure integrates figures 3–9 presented in Demetriou, Spanoudis, et al. (2017). All
figures were generated by segmented modelling where each process was regressed on
developmental g (g x age) because of the developmental information it conveys
was between 4–5 years and the other was at 9 years. There were also two break
points in the development of awareness, one at 5 years and the other at 9 years.
Attention is drawn to the second spurt in flexibility in shifting which occurs from
9–11 years. This change is related to semantic flexibility allowing shifting between
mental spaces, as when searching for exemplars of different concepts. For instance,
say, first, all instances of fruits that come to your mind and then all instances of
furniture. Although well established by 10 years of age, it continues to improve for
several years thereafter.
A second study focused on the development of inductive and deductive reasoning
from 7–12 years. Children were examined with many tasks addressed to processing
speed, attention control, working memory and inductive and deductive reasoning.
The inductive reasoning tasks addressed three levels: (1) identify patterns and
formulate generalizations on the basis of a single dimension or relation; (2) handle
hidden or implied relations that require to combine information present to the senses
with knowledge stored in long-term memory; (3) deal with multiple parameters and
relations simultaneously. These three levels addressed late reality-based, early rule-
based, and late rule-based thought, respectively. The deductive reasoning tasks
addressed the following three levels: (1) deal with modus ponens inferences; (2) deal
with tasks requiring to integrate modus ponens with modus tollens; (3) deal with
the fallacies. These three levels require early and late rule-based thought and early
principle-based thought.
The findings of this study were very interesting. On the one hand, processing
speed, attention control, and working memory tended to differentiate from general
ability through this age period from 7–12, reflecting their relative automation.
Interestingly level 1 and level 2 of inductive reasoning and level 1 of deductive
reasoning also tended to differentiate from g. However, level 3 of inductive reasoning
and level 2 of deductive reasoning intertwined strongly with both g and developmental
g. Thus, in the age period 7–12, advanced inductive and solid but less than optimum
deductive reasoning get integrated into g. Figure 11.4 shows how level 3 inductive
reasoning takes off at about the age of 7–8.
Markers of developmental g in the cycle of principle-based

thought
The primary developmental task in the cycle of principle-based thought is command
of cognizance and related rules so as to ensure truth and validity of inference. This
is primarily expressed through accuracy in self-representation and self-evaluation
and in explicit matching of specific processes with specific problems
Several studies focused on development from early adolescence to early adulthood
(Demetriou, Spanoudis, Kazi et al., in press, Studies 5–8). One of these studies
involved adolescents drawn from 7th to 12th secondary school grades (mean 11.5
to 17.4, respectively), university students, and secondary school teachers. These
participants were examined by a pair of tasks addressed to each of four domains:
mathematical proportionality, deductive, causal-experimental, and spatial reasoning.
The first task of each pair addressed first level (symmetrical ratios, modus tollens,
isolation of variables, mental rotation of two figures in coordination) and second
level principle-based thought (non-symmetrical ratios, matching hypothesis with
2 x 2 experiment, grasping fallacies, and coordination of shadow projections of
geometrical figures inversely varying).
Participants were asked to evaluate their success on each task on a 4-point scale
(from not satisfied at all to very satisfied). These scores were then combined with
the corresponding performance score on each task to yield the self-evaluation
accuracy score for each task. Participants were also asked to judge the procedural-
processing similarity of 22 pairs of tasks belonging to various combinations of domain
and level affiliation (e.g., tasks addressed to the same domain and level, same
domain but different level, different domain and level, etc.). Participants were
instructed to rate the similarity of the tasks in each pair “with respect to the ways of
thinking you applied when trying to solve them; take into account how your mind
worked when solving each task” (not similar, slightly, quite, very similar). They
were then asked to explain three of their similarity judgments (pairs involving two
causal, two spatial, and one causal and one quantitative task). These similarity
judgments were scored on a 4-point scale indicating full lack of any awareness of the
processes involved, content-based awareness of similarity (e.g., the two tasks involve
thinking on rods), to complete awareness of mental similarities and differences (e.g.,
these tasks are similar because they require to estimate proportional relations; these
are different because the one requires estimating quantitative relations and this
requires isolating variables).
Accuracy of success evaluation and similarity judgments capture two comple-
mentary aspects of cognizance. The first captures a system of standards enabling one
to monitor, evaluate, and adjust problem-solving accordingly; the second captures
explicit awareness of the mental processes involved. Obviously mental monitor-
ing and regulation would be easier when thinkers possess both self-evaluation
standards and explicit awareness of the processes they might have to regulate to
improve performance. Under these conditions, thinkers may be able to do both:
(i) notice possible deviations between solutions produced and the best solution
possible as indicated by the standards; (ii) explicitly select a process that is conducive
for the production of the best possible solution.
The Tucker-Drob model involved the four domain-specific mean performance
scores, the four mean self-evaluation accuracy scores, and the mean similarity
evaluation score. As expected, change in performance and self-evaluation on causal
and quantitative thought was linearly related to change in developmental g; changes
in performance and self-evaluation on spatial thought, which reached ceiling by
middle adolescence, tended to differentiate from developmental g. However,
performance and self-evaluation on deductive reasoning was positively related to g.
Also, changes in awareness of similarity between processes increasingly intertwined
with both g and developmental g. It is clear, therefore, that the build-up of g during
the formation of principle-based thought is based on the grasp of the constraints
underlying inference drawing in deductive reasoning, awareness about these
constraints, and also awareness of mental processes as such. At the same time, other
processes may increase linearly with g (e.g., causal and mathematical) or differentiate
from it (spatial reasoning).
The pattern of change of the various abilities reflected these relations nicely.
Specifically, segmented modelling (see Figure 11.4) showed that reasoning takes off
at the beginning of this period at 12 years and levels off at 16 years, which is
expected for the development of principled reasoning. Self-evaluation spurts in the
middle of this period, at around 13 years, and awareness of similarities spurts about
one year later at 14.5 years. Thus it seems that incipient grasp of principled thought
is intertwined with awareness about it before it is consolidated by the end of this
cycle.
Another study (Makris, 1995) focused on the relations between cognitive changes
in adolescence, awareness of the mental processes involved in different SCSs, and
awareness of mental effort required to decipher what the problem is about and set
up a solution plan. In the fashion of the study above, these participants— adolescents
aged 12–16—solved an early and a late principle-based thought in three SCSs: causal
(hypothesis testing and isolation of variables), quantitative (analogical and proportional
reasoning), and spatial (visual projections and mental rotations). To examine
cognitive awareness, participants were asked to rate how much several domain-
general and several domain-specific mental processes are needed when working on
each of the problems above. The general cognitive processes were as follows:
working memory (e.g., you must keep in mind many things at once to be able to
solve this problem), long-term memory (e.g., you have to remember many things
from what you know), attention (e.g., you have to focus on the problem so that
nothing else pops into your mind), understanding (e.g., you have to understand well
all relevant information and make sense of it in regard to the problem), and inference
(e.g., you have to combine information and draw conclusions from them that are
not given in the problem). The specialized processes were as follows: isolation of
variables, combinatorial ability, and hypothesis formation represented the causal SCS
(e.g., you must proceed checking factors one by one, holding everything else the
same); quantitative representation, magnitude estimation, and mathematical
analogical relations represented the quantitative SCS (e.g., you must grasp how two
magnitudes vary together); and mental rotation, coordination of perspectives, and
visual integration represented the spatial SCS (e.g., you must visualize how an object
would look if it rotates in space so you see it from a different point of view).
Specifically, a 1–2 sentence description of each process was presented vis-à-vis each
task; participants were asked to rate how much they thought that each domain-
general and each domain-specific process was used when working on each task.
The Tucker-Drob model showed that early principle-based thought levels off
at about 13 years of age. Expectedly, late principle-based thought develops through-
out the age phase investigated. Interestingly, the pattern of development of aware-
ness of the links between SCS and SCS-specific processes is very similar to the
development of late principle-based thought. Notably, an L-shape association
between developmental g and awareness of mental effort indicates that at the
beginning of the cycle of principle-based thought adolescents start to think at ease

when dealing with cognitive problems. That is, the higher their principle-based
developmental g the easier they consider the principle-based tasks they solve.
Obviously this implies a feeling of cognitive self-confidence related to the emergence
of principle-based thought and awareness of its advent.
Do adolescents differentiate between SCSs in regard to their demand on general
cognitive functions? As illustrated in Figure 11.5, they clearly do. It can be seen that
the spatial SCS was considered the least and the quantitative SCS the most
demanding of the three in all cognitive functions. This clearly implies that visual-
spatial thought is received by adolescents as a largely automatic kind of thought. In
contrast, mathematical thinking is received as highly effortful. Regarding processes,
attention and understanding are considered to be the most needed processes.
Interestingly, working memory is ascribed to quantitative thought much more than
it is ascribed to causal or spatial thought.
The relation between cognitive style and cognitive ability was also investigated by
this study. Cognitive style varies on a dimension of impulsivity-reflectivity. Impulsives
are persons who tend to respond fast before systematically examining the problem
and considering alternative solutions; those who withhold response, examine the
problem and reflect on available knowledge and solutions are considered reflective.
It is instructive that cognitive style was highly related to cognitive performance.
Reflective persons performed better on all three SCSs (see Figure 11.6).
Two studies which focused on changes from early adulthood (22 years) to old
age (85 years) will be mentioned here only in passing for the sake of completeness.
FIGURE 11.5 Association between SCS and general cognitive functions

FIGURE 11.6 Relation between cognitive style and ability on the three SCSs
Source: Makris (1995)
Specifically, the first study showed that advanced principle-based thought intertwined
with g in the period from 11 to 30 years of age. This relation vanished in the 31 to
85-year period when there was a return of rule-based thought, which intertwined
again with g. The second study, which examined persons from 16 to 45 years of
age, found that, from early to middle age, performance in all forms of reasoning
differentiated from g but cognitive self-representations intertwined with it.
Conclusions
The message of the studies presented in this chapter is clear. Differentiation/
de-differentiation of specific processes from g is a developmental rather than an
individual differences phenomenon. Thus the relation between specific processes
and g vary with developmental cycle and phase. New acquisitions in each cycle get
increasingly integrated into g, at the phase when they infuse g, impregnating it with
their properties.
Control of attentional focus is the impregnator of g in the cycle of reality-based
representations. Awareness of the perceptual origins of knowledge also contributes
in the second phase of this cycle. In the next cycle of rule-based thought, attention
control is left behind and inductive reasoning dominates as the impregnator of g.
Awareness still actively infuses g with its properties, but it mutates from perceptual
to the inferential aspects of representations. In the next cycle, inductive reasoning
recedes and deductive reasoning, in its most advanced versions of conditional
reasoning, dominates as the major source of infusion of new properties into g.

In this cycle, awareness continues to be part of g re-morphing; however, it now
comes as a refined theory of mental processes tuned to one’s own personal strengths
and weaknesses.
The findings above give an impression that intertwining dominates over
differentiation. This impression is not accurate. Differentiation is present but it
recycles with intertwining. Processes that intertwined with g when they were
integrated into it differentiate from g at subsequent cycle, when their relations with
g get loose, because they are already under command. Counting the number of
factors is a different matter. It is indeed the case that, with growth, more factors
are needed to account for performance. However, this is because the relative
power of the factor dominating in g decreases. Attention control is a sweeping
force in the reality-based representations cycle. Lapses in it cause functioning in all
other domains to falter. Later, in the rule-based cycle, representational processes
split from inferential processes in g. As a result, lapses in one can be compensated
by contributions from the other. Thus decisions are more flexible because there is
a system that can make intentional choices. In the next cycle of principle-based
thought this is further differentiated in reference to personally tailored values and pre-
ferences. All in all, psychometric differentiation comes as a result of a developmental
process underlying the transformation of g along its various constituents, where
their relative contribution varies.
12
PERSONALITY AND
EMOTIONS IN THE MIND
So far only mental processes have been discussed in this book. However,
understanding and action are often constrained by forces other than mental processes.
Personality and emotions are certainly involved. Different persons may have the
same mental ability but they may act differently because they are disposed to receive
and weigh information differently; their style of response may also be very different,
as some of them may be outgoing and easily excited and others may be reserved and
restrained. There is a huge body of literature on personality and its relation to
intelligence and it is beyond the scope of this book to discuss in detail this literature.
We will focus on three themes related to the aims of this book: first, we will
summarize research on the organization and development of personality from
infancy through to adulthood; second, we will present research on the relations
between personality and intelligence; and, third, we will present research on the
relative contribution of personality and intelligence to real-life achievements, such
as academic attainment. Our aim is to show how mental processes intertwine with
personality and emotional processes to shape action and achievement in the real
world.
The organization of personality: concepts and models

Wechsler noted that “general intelligence cannot be equated with intellectual
ability, but must be regarded as a manifestation of the personality as a whole”
(Wechsler, 1950, p. 83). Jensen (1998) concluded The g Factor, his magnum opus
on general intelligence, by urging researchers to study the intelligence-personality
nexus. Allport (1937) defined personality as “the dynamic organization within the
individual of those psychophysical systems that determine his unique adjustment
to his environment” (p. 48). In the psychology of individual differences, personality
is a system of traits distinguishing individuals (Eysenck, 1997). Traits are relatively
Personality and emotions in the mind 163
stable characteristics that can be used to describe the dominant qualities of an

individual’s behaviour and overall style of relating to the world. Currently there
are two approaches dominating the study of personality, one focusing on the early
years of life, from infancy through to childhood, and another focusing on
adulthood.
The modern study of personality starts with Freud’s psychoanalytic theory,
although theories of personality date back to the ancient Greeks (see Eysenck,
1997). For Freud, personality is a system of psychodynamic forces that combine to
define how individuals understand the world and relate to it. These forces come
from the interaction between inherited instinctual motives and socialization rules
and practices. On the one hand, there are strong hereditary tendencies to satisfy
one’s own needs for survival, reproduction, and dominance; these are associated
with pleasure and self-satisfaction and they are generally unconscious. In Freud’s
terms, these tendencies form the id. On the other hand, there are the general moral
principles of human socialization and the rules of one’s own culture and social
group which dictate what is acceptable in society. These channel the individual to
control the forces of the id so that they are attuned with societal laws and
expectations. This is Freud’s superego. These principles are initially represented in
socialization practices and education at large. By definition, these principles become
increasingly conscious: they form an external body of knowledge and skills which
must be interiorized and assimilated by the individual so that they are intentionally
and efficiently practised. With development, the id and the superego are woven
into a personal system of principles guiding interaction with the world in socially
acceptable ways. This is Freud’s ego. In Freud’s theory, the formation of the ego is
a developmental process of increasing awareness and control of action. Freud’s
defence mechanisms of the ego are actually orientation or filtering mechanisms
underlying selective attention, assimilation of information, and executive con-
trol of behaviour (S. Freud, 1927; A. Freud, 1966). We will return to defence
mechanisms later.
Temperament in infancy and childhood

Modern researchers focusing on the early years prefer to call their object of study
temperament rather than personality (Buss and Plomin, 1984; Rothbart, 2011).
Temperament reflects differences between children in their reactivity to external
stimuli and differences in their ability for self-regulation which originate from
their constitutional differences. Reactivity refers to how much a given pattern of
stimulation arouses or excites the child. For instance, when in a new environ-
ment some children tend to get excited and interested in what is new to them;
others tend to become restless and fearful; others tend to remain unexcited and
indifferent. Self-regulation refers to neural and behavioural processes that may
modulate reactivity. For instance, children in the first category tend to approach
and explore new objects whereas children in the second category tend to withdraw
close to a caregiver. Constitutional differences arise from the biological make-up
of each individual. The biological make-up at any given time is influenced by

heredity and the current maturational condition of the individual, which depends
on development and experience (Rothbart, 2011; Rothbart, Ahadi, & Evans,
2000).
The combinations of different profiles along these two dimensions are expressed
in several overarching traits in infancy through childhood. Surgency or extraversion
captures the tendency to be very active, with an uninhibited approach style and posi-
tive emotions; children with a high showing in this trait find pleasure and excitement
in social interaction. Negative emotionality captures children’s tendencies toward
sadness, fear, irritability, and frustration; also these children are not easily quieted
after high arousal. Effortful control involves the ability to orient to goals, sustain
attention towards them and organize action accordingly, inhibit competing and
irrelevant behaviour, and enjoy low-intensity situations. Orienting sensitivity refers to
direction of perception and attention to stimuli and involves sensitivity to external
stimuli reaching the senses or internal stimuli impacting on emotional or mental
processes, such as feelings of anger or association between thoughts. Affiliativeness
describes behaviours towards others which appear in adolescence (Rothbart &
Bates, 2006).
The Big Five factors of adult personality

The dominant theory of adult personality is the so-called model of the Big Five
factors of personality (Costa & McCrae, 1997; McCrae & Costa, 1999). The
dimensions of temperament above are closely related to the Big Five factors. These
are: extraversion, agreeableness, conscientiousness, neuroticism, and openness to
experience/intellect. Each of these dimensions is a continuum where each individual
may be placed. Thus a full description of an individual would require specifying
where this individual stands on each of the five dimensions. Individuals high in
extraversion enjoy being with others and actively seek their company. At the other
end, introverts are distant from others, withdrawn, shy, and self-restrained. Agreeable
individuals are oriented to others, trust them, they are warm, and they actively seek
to make good to them. At the other end, non-agreeable individuals are suspicious,
headstrong, shrewd, impatient, argumentative, and aggressive. Conscientious
individuals are goal-minded, organized, determined, and planful. At the other end,
those low in conscientiousness are careless, distractible, and lazy. Neurotic individuals
are emotionally disturbed by variations in the environment. They are thus nervous,
anxious, and moody. At the other end, emotionally stable individuals are confident,
clear-thinking, alert, and content. Finally, individuals who are open to exper-
ience are curious and have broad interests, are inventive, original, and imaginative.
Individuals who are low in openness to experience are conservative, cautious, and
mild. Box 12.1 presents characteristics associated with each of the Big Five factors
as measured for children. It is noted that the Big Five model integrated dimensions
and constructs from earlier models of personality, such as Eysenck’s (1997) and
Cattell’s (1965) models.
BOX 12.1
CHARACTERISTICS OF THE BIG FIVE FACTORS
Extraversion
Extraverts are active, lively, sociable, talkative, optimistic, pleasure-seeking, self-
confident, warm, and uninhibited.
Neuroticism
Individuals high in neuroticism are nervous, anxious, moody, tense, self-
centered, easily offended, and self-pitying.
Agreeableness
Agreeable individuals are considerate and thoughtful of others, helpful and
cooperative, soft-hearted, generous, kind, forgiving, sympathetic, warm, and
trusting.
Conscientiousness
These individuals are goal-minded, organized, determined, planful, ambitious,
energetic, efficient, determined, precise, industrious, persistent, reliable, and
responsible.
Openness to experience
Individuals open to experience are resourceful in initiating activities, curious
and exploring, open to new experiences, strongly involved in what they do,
creative in perception, thought, work, or play.
Emotional intelligence
Recently the notion of emotional intelligence was proposed to bridge traditional
theories of intelligence and personality to account for efficient functioning in the
real world. “Emotional intelligence is a type of social intelligence that involves
the ability to monitor one’s own and others’ emotions, to discriminate among them,
and to use the information to guide one’s thinking and actions. The scope of
emotional intelligence includes the verbal and nonverbal appraisal and expression
of emotion, the regulation of emotion in the self and others, and the utilization of
emotional content in problem solving” (Mayer & Salovey, 1993). Thus emotional
intelligence is supposed to involve three types of processes: (1) understanding of
one’s own and others’ emotions; (2) the ability to control one’s own emotions
according to the needs and demands of the current social interaction; and (3) the
ability to plan and organize one’s own actions according to the emotional
characteristics and needs of the other. Therefore emotional intelligence integrates
cognitive mechanisms from general theory of intelligence as they apply to the
monitoring and regulation of emotions and dispositions related to the factors of
neuroticism, extraversion, and agreeableness of personality theories. It is, of course,
relevant to ask if emotional intelligence qualifies for the status of an autonomous
intelligence because it spans several constructs studied by intelligence and personality
researchers. We will return to this question below.
All in all, mind, personality, and emotions are obviously complementary aspects
of the same living entity: the person who strives to understand, plan and act
efficiently, pleasantly, and acceptably as much as possible, capitalizing on strengths,
dispositions, and personal history. There are four recurring themes in all theories
outlined above: reactivity and activation patterns; emotions as motives driving
behaviour; control and self-regulation; and social relations. Reactivity and activation
patterns dominate in the studies of temperament in the early years of life and are
expressed via the constructs of surgency and orienting sensitivity. In the Big Five
model, emotions are expressed in the construct of neuroticism. In psychodynamic
theory they are expressed in the id. In the theory of emotional intelligence, emotions
come to the fore as the organizing forces of social behaviour. Control and regulation
are expressed in the trait of effortful control in developmental studies of temperament,
in conscientiousness in the Big Five model, in psychoticism in Eysenck’s model,
in the ego in psychodynamic theory, and in emotional self-regulation in emotional
intelligence. Social relations are expressed in the constructs of surgency and afillia-
tiveness in studies of temperament, agreeableness and extraversion in the Big Five,
superego in psychodynamic theory, and understanding and efficient management
of the emotions of others in the theory of emotional intelligence. Therefore it is
important to understand how personality, emotions, and intellectual processes
interact in the architecture of mind proposed in this book. In this chapter we will
summarize research focusing on these interactions.
Relations between personality, emotions, and cognition

The architecture of personality resembles the architecture of the mind to a large
extent. Basically, although personality dimensions and cognitive processes do not
completely match, the four-fold structure of the mind is discernible in the structure
of personality as well. Specifically, some of the Big Five factors are clearly oriented
to the environment, primarily extraversion and agreeableness. Naturally so, as these
factors are more related to the social world and they may thus interact primarily
with the SCS of social thought. Effortful control in the early years and con-
scientiousness later involve reflection and awareness needed to channel behaviour
according to goals. Therefore they are oriented, by definition, to self-knowledge in
the way cognizance operates on cognition: they both allow executive control. The
structure of personality does not directly involve a representational capacity factor
that may be analysed into a storage component and an efficiency component.
However, it involves factors that are clearly applied on representational capacity. In

addition to effortful control and conscientiousness, emotionality, orienting, and
neuroticism also set the frame for the operation of representational capacity because
they shape value weights for individual stimuli or stimuli patterns. In the same
fashion, personality does not directly involve an inference factor. However, both
conscientiousness and intellect do serve information integration functions in
personality.
It is also notable that there is a hierarchical structure in the constructs of
personality akin to the structure of mind. Specifically, in both constructs there are
core processes related to the organization of stimuli in the environment, organized
systems of action or operation emerging from the interaction of core processes with
their initial realm of application, and higher-level system-specific conceptual
knowledge and beliefs. Specifically, temperament factors in infancy and childhood
are the equivalent of the core operators in the SCS. These factors reflect dispositions
emerging from specific patterns of activity in the brain which reflect the individual’s
genetic background. They predispose the individual to receive information, act, and
relate to the world in specific ways, thereby framing behavioural tendencies
and preferences for social interaction. In cognitive terms they influence the initial
spontaneous emotional weighting of information and guide selective attention and
filtering for further processing. In cognition, for instance, a particular pattern in the
physical arrangement of a set of objects allows recognizing their number or grasping
their causal relation; in personality, a particular pattern in the emotional aspect of
objects, such as their sound or facial expression, predisposes for avoidance or
approach, pleasure or stress, underlying extraversion or emotional stability. It is
reminded that initial approach or avoidance relates to reactivity. These differences
are gradually woven into openness to or avoidance of new experience. In line with
this assumption, McIntyre and Graziano (2016) recently showed that personality
differences influence selective attention to stimuli. Specifically, individuals oriented
to other persons tend to selectively attend to social stimuli; individuals oriented to
inanimate things tend to selectively attend to objects. Along the same lines, Antinori,
Carter, and Smillie (2017) examined if individuals who are open to experience
perceive low-level stimuli differently than less open individuals. They used the
classic binocular rivalry paradigm to project stimuli to their participants. In this
paradigm, different stimuli are projected to each of the two eyes; this causes
perceptual experience to alternate between the two stimuli presented to each eye,
but occasionally an integrated mixed perceptual solution is created which encom-
passes aspects of each stimulus. These researchers showed that open individuals
“see” more possibilities in the input and they flexibly combine information from
the two eyes in a creative fashion, especially under a positive mood. Therefore
personality differences are directly linked to attention and motivational processes.
This approach concurs with Eysenck’s (1997) assumption that differences between
extraverts and introverts emerge from their difference in brain activation. In
introverts, brain arousal is high; this drives them to avoid further stimulation that
may be caused by intense social activity. As a result these persons tend to be
self-inhibited and distant from others. In extraverts, brain arousal is relatively low;
thus they seek arousal to boost brain activation.
At a higher level, the Big Five factors are middle-level constructs resembling the
operations in the SCS. According to Graziano and his colleagues (1997) the Big
Five factors contribute to the formation of self-systems such as global, social and
academic self-esteem. These systems comprise general self-representations, value
systems, and general action strategies that connect the individual with the world. At
a more specific level these general self-systems relate to more specific adaptations to
particular tasks or environments, such as academic adjustment, peer relations, class-
room behaviour, etc. For instance, extraverts have memories of their pleasure in
interacting with others, they have ready-made “social scripts” they may use to
attract the interest of other people, and they know that this adds to their personal
value in their social group as “good fellows”. Introverts may have memories of the
embarrassment they may have felt in social gatherings and they are aware they are
clumsy in starting or sustaining an interaction. Thus they choose to avoid social
activities. All in all, general self-systems related to the Big Five operate like the
level-II broad factors in the Cattell-Horn-Carroll model of intelligence discussed in
Chapter 2.
There is strong evidence that three of the Big Five factors relate to one second-
order factor and the rest relate to another (see Figure 12.1). Specifically, the first of
them, called the alpha factor (α-factor), relates to conscientiousness, emotional
stability (neuroticism), and agreeableness. This factor stands for the general trait of
stability: efficiency in organizing one’s own life, dealing with pressure, and making
oneself acceptable. The second factor, called the beta factor (β-factor), relates with
openness to experience and extraversion. This factor stands for plasticity in one’s
approach to and relations with the world. Broadly speaking, stability involves many
of the properties of crystallized intelligence as they apply to the social world and
self-management; plasticity may be seen as the expression of fluid intelligence in
personality. In a sense these two higher-order personality factors reflect the dynamic
aspects of the mind’s capacity to use available knowledge for efficient interactions
with the world (crystallized knowledge) or to cope with new knowledge and go
beyond it (fluid intelligence). These two factors relate highly to a third-order factor,
the general factor of personality (GFP). “The GFP is analogous to g and predicts
social efficiency in the way g predicts cognitive efficiency” (Rushton & Irwing,
2009, p. 564). This structure was found in many other studies and it is validly related
to actual-life indicators, such as employment performance as rated by employers
(e.g., van der Linden, te Nijenhuis, & Bakker, 2010).
What is the identity of the GFP? The GFP is highly self-representational,
reflecting a person’s self-concept and self-worth. The evidence is clear: the relation
between the GFP and self-esteem is very high (67% of the variance), even higher
than its relation to the α-factor, stability (52% of the variance), and the β-factor,
plasticity (58% of the variance) (Erdle, Irwing, Rushton, & Park, 2010). However,
there is also a strong cognitive and a strong emotional component in the GFP. We
showed in a series of studies that self-represented cognitive g, emerging from
FIGURE 12.1 The general model of the relations among general factor of personality,
fluid intelligence, and academic performance
Note: AcPerf=academic performance, Gf=fluid intelligence, Stab= stability, Pla= plasticity,
GFP=general factor of personality, EIt=emotional intelligence-trait, EIa= emotional
intelligence-ability, N=neuroticism, A=agreeableness, C=conscientiousness, E=extraversion,
I=openness to experience/intellect
self-ratings concerning processing speed, attention control, all five SCSs, and self-
monitoring and self-regulation (i.e., all processes involved in the four-fold model),
accounts for as much as 80% of variance in the GFP (Andreou, 2009; Demetriou,
Kyriakides, & Avraamidou, 2001; Demetriou, Spanoudis, Kazi, Žebec, & Andreou,
submitted). But this is actually part of general self-image which involves both
self-representations of cognition and emotions. Specifically, we examined children
and adolescents aged 10–16 by our cognitive development test already presented,
the Big Five, and on a large battery of emotional intelligence abilities, including the
cognitive ability to understand the meaning and implications of emotions and self-
representations about emotional intelligence. We found that the GFP was highly
related to both general cognitive self-image (38% of variance) and general emotional
self-image (29% of variance). Looking into each, with the exception of neuroticism,
four of the Big Five factors are highly related to self-represented g (higher than .5).
This is very close to what is called “self-assessed intelligence” in the personality
literature (Chamorro-Premuzic & Furnham, 2006). Moreover, extraversion is
related to positive emotions; agreeableness relates to management of emotions, a
positive stance to others and positive emotions. Interestingly, when both classical g
and the cognitive ability to understand emotions are involved, the later ability
dominated, accounting for a small but significant part of the variance in GFP (3%
of variance). Therefore the GFP has a strong self-worth component. Obviously
social and cognitive efficiency is reflected into how individuals esteem themselves:
the more stable, flexible, and intelligent people think they are the higher their self-esteem
(Demetriou, Spanoudis, Kazi, Žebec, & Andreou, submitted).
How does the GFP relate to g, the general factor of intelligence, and emotional
intelligence? It is stressed that, despite its very close relation with self-assessed
cognitive ability, the GFP is minimally cognitive. We found that its relation
with cognitive g is weak and it appears only in adolescence. Specifically, this relation
is very low and non-significant (less than 1% of the variance) until the age of 11 and
it becomes significant but still relatively low in adolescence (about 10% of the
variance).
The relation with emotional intelligence is equally complicated. Specifically,
“trait emotional intelligence”, that is self-represented emotional intelligence,
accounted for a large part of the variance of the GFP (28%); “ability emotional
intelligence”, that is ability to understand emotions in a problem-solving context
focusing on emotions, accounted for a smaller but significant part of its variance
(3%). Obviously these relations extend what was said above. The GFP is maximally
self-representational and minimally cognitive in both the classic cognitive g and the
processing of emotional information (Andreou, 2009; Demetriou, Spanoudis, Kazi,
Žebec, & Andreou, submitted).
It is interesting to zoom in on the relations between g or self-represented g,
each of the Big Five, and various aspects of emotional intelligence. Overall some
factors of personality have a privileged relation with some aspects of cognition or
emotional intelligence. Specifically, only two of the Big Five factors relate significantly
with g but in opposite directions: openness to experience relates positively with
cognitive g (accounting for about 5% of the variance) and conscientiousness relates
negatively (accounting for about 2% of the variance) (Demetriou et al., 2001).
Obviously, on the one hand, some cognitively able individuals are open to experience,
while, on the other, some of the not so cognitively able individuals are highly
conscientious; this may reflect that individuals of average intelligence tend to
compensate through strategies of self-organization and self-discipline (Chamorro-
Premuzic & Furnham, 2006). Along this trend, of the Big Two factors, only the
β-factor, plasticity, relates with cognitive g; interestingly, this relation comes from
attention control rather than inference. It seems, therefore, that actual cognitive
efficiency that is built in executive control rather than reasoning is projected into
self-representations of cognitive efficiency. This explains a strong proclivity of indi-
viduals who are extraverted and open to experience: they tolerate change in the
environment and they are stimulated by it (Demetriou et al., submitted).
As far as emotional intelligence is concerned, “trait emotional intelligence”
depends highly on general self-represented g: as much as 53% of the variance of trait
emotional intelligence was accounted for by general self-represented g. However,
when we turn to personality, one single factor, agreeableness, appeared to influence
three important components of emotional intelligence: management of emotions,
positive management of others, and positive emotionality. In turn, positive emo-
tionality influenced extraversion (Andreou, 2009). These findings lend support to
accruing evidence that the recent surge in interest in the concept of emotional
intelligence is not destined to last: that is, practically all the information one needs to
predict the success of individuals to deal with their emotional life and the emotional
aspect of their relations with others may be obtained via more classical and well-
established methods of testing cognitive ability and personality (Waterhouse, 2006).
Therefore the relations between personality and cognitive functioning are

indirect rather than direct, mediated by self-awareness and self-regulation processes.
In fact the relations between personality and self-representations about cognitive
functioning were close enough to justify the claim that personality, as examined by
self-report inventories, is, to a large extent, part of the general self-representation
system. Thus individuals maintain a cognitive-affective view of themselves that is
accessible to consciousness (McCrae & Costa, 1999) and preserves maps of our
personality characteristics.
In conclusion, personality and intelligence are related in multiple ways.
Cognizance is the royal road of this relation. Openness/intellect has a special role
as it is the first-hand reflection of cognitive ability. Openness/intellect carries, so to
speak, one’s feeling of mental power into actions where the mental power available
may be put in use. A major bridge over this road is executive function. This
naturalizes, so to speak, cognizance and mental ability in the realm of personality.
In this realm, conscientiousness is the agent that takes cognizance and executive
function and transforms them into plans and strategies for the long-term regulation
of behaviour. In line with this analysis there is evidence showing that executive
control is the mechanism used by conscientiousness to set action plans and organize
their implementation (Hall & Fong, 2013; Hall, Fong, & Epp, 2013). Interestingly,
of the various aspects of executive control studied in childhood (i.e., attention
focusing, mental shifting, and inhibition), it is shifting that relates to conscientious-
ness (Fleming, Heintzelman, & Bartholow, 2015). Of the rest, extraversion and
agreeableness relate mainly to the social SCS. High perceived g and high g tend to
go with higher extraversion and higher agreeableness, probably reflecting both the
self-confidence to get engaged with others and the skill to do this successfully.
The moderate size of relations between g and the GFP implies that many persons
hold self-representations about their characteristics and abilities which are not always
accurate. In fact Eysenck included a lie scale in his personality test to capture the
tendency of many individuals to be lenient about themselves, drawing a picture that
is more positive than justified. We found that scores on this lie scale decrease with
age, indicating that individuals tend to become more accurate in their self-
representations as they grow. We come to the question of the development of these
relations in the section below.
Developmental changes in personality and personality-

intelligence relations
The Big Five factors are discernible from early childhood, at around the age of
4 years, and they are generally stable throughout childhood and adolescence.
Asendorph and van Aken (2003), drawing on a 9-year longitudinal study covering
the age span from 4 to 12 years, found that the Big Five emerge from very different
measures throughout childhood and early adolescence. Moreover, they found that
each of the Big Five factors systematically correlates with related behavioural indices
as specified by parents and teachers. Specifically, they found that neuroticism and
low extraversion go together with social inhibition; low agreeableness and low
conscientiousness go together with aggressiveness; conscientiousness and/or
culture/intellect/openness go together with school achievement. We will further
explore these relations in the chapter focusing on the relations between cognition,
personality, and school performance. Drawing on a longer longitudinal study,
covering the age span from about to 2 to 15 years, Lamb, Chuang, Wessles, Broberg,
and Hwang (2002) found that the stability and reliability of various factors, including
the Big Five factors and also positive activity and irritability, increased with age.
Overall, conscientiousness, irritability, and positive activity were present and
relatively stable since early childhood; extraversion and neuroticism stabilized after
the age of 8; openness was never stable, suggesting that it may not be a meaningful
dimension of personality prior to adolescence. This study also found changes in the
prevalence of the various factors over the years. With age, children become less
extraverted and more agreeable, more conscientious, and more emotionally stable;
openness varied, increasing early in childhood and adolescence and decreasing later
(see also Roberts, Walton, & Viechtbauer, 2006). The same trends were observed
from middle adolescence (14–17 years) to middle age (older than 50). Con-
scientiousness and agreeableness increased and extraversion and neuroticism
decreased across many different nations (McCrae et al., 2000).
Self, ego, and ego-defences

We showed that there is a powerful general self-representation entity that subsumes
all aspects of self-representation, including the general factor of personality, self-
esteem, and cognitive and emotional self-representation. This entity organizes how
people think of themselves in regard to their mental, social, and emotional
characteristics and competences. What they think may not always be accurate, but
it does influence their behaviour. This entity was introduced in psychology under
several different names, depending on the tradition. Here we focus on two of them,
which were intended to account for the functioning of the person as a whole: the
self and the ego. We remind that the self as an agent of reason dates back to Kant;
the ego as an agent integrating experience dates back to Freud. We outlined the
construct of the self in the chapter dedicated to self-awareness and understanding of
the organization and functioning of the mind. Below we will focus on the construct
of the ego, which is central to the concerns of this chapter.
The ego
In its modern form the concept of the ego was introduced by Freud (1949). In
modern psychology, Loevinger formulated a theory of ego development (1976) and
Cramer (2006, 2007, 2008, 2015a) studied changes in the use of ego defence
mechanisms. We noted that in Freud’s theory the ego gradually integrates internal
drives with social rules and norms into a (relatively) self-aware and self-driven
adaptive system. Thus control of impulse is an important function of the ego.
Control of impulse is often stressful because impulses driving individual priorities

and preferences are often in conflict with social and moral rules and norms. Conflicts
of this nature may cause frustration and distress. Thus, on the one hand, ego
development broadens self-awareness as the person negotiates his or her desires and
priorities with the norms and priorities of others. In so doing, ego development
enables the individual to make sense of experience. On the other hand, ego
development broadens self-regulation. In modern terms, ego development is the
frame where the maturation of executive control and self-control happens.
In Loevinger’s theory, ego develops in nine stages (Hy & Loevinger, 1996). Here
we will only focus on the kind of awareness available to each stage, because this
is relevant to our concerns. At the first stage, the prosocial stage, the infant does not
differentiate the self from the world, focusing on gratifying immediate needs. The
second, the impulsive stage, comes with language; the sense of self emerges power-
fully and is projected to others by negativism. Toddlers are immersed in the here and
now, and divide the world between those satisfying needs and desires and those
rejecting them. In the third stage, the self-protective stage, which dominates in
preschool, children make the first step towards self-control. Children at this stage can
delay immediate advantage; they are aware of rules and they can play with them for
their advantage; they are creatures of opportunistic hedonism. At the fourth stage, in
middle childhood, children progress to conformity and view themselves and others as
acting according to accepted codes and norms. This is an organized but rather
simplistic approach to the world which is viewed in dichotomies; children are aware
of their characteristics (for instance, gender), and they identify with their group.
However, behaviours are judged externally rather than by intentions. The fifth stage
is the self-aware stage, which dominates in late childhood or early adolescence.
Individuals differentiate the “what I am” from “what I ought to be”; thus they
examine the self, and they conceptualize inner life, so that identities are associated to
both actions and feelings. The individual is aware of his or her self-identity, as
contrasted to the group, and is thus aware of options for action. At the sixth stage, the
conscientious stage, which may be acquired in adolescence, control is internal and is
based on conscience and self-constraining principles rather than rules. Identification
is a basic mechanism for internalizing principles. Conscientious persons are reflective.
At the next stage, the individualistic stage, people recognize individual differences and
they are aware of the difference between the inner and the outer self. This is obviously
a stage attained in adulthood. The last two stages are very rare. At the autonomous
stage, the need for autonomy is a central characteristic. Thus there is awareness of
the complexity and the multifaceted character of real people and real situations. At the
last, the integrated stage, self-actualization is the fundamental property.
Several studies examined the relations between ego development as captured by
the sequence above and intelligence. Cohn and Westenberg (2004) presented a
systematic meta-analysis of 52 correlations between ego level scores and intelligence
scores. The weighted average between these scores ranged between .20 and .34.
These values suggest that ego development and intelligence are related but that this
relation is, on the one hand, moderate enough to suggest that the two constructs are
separate of each other, while, on the other hand, the origin of this relation would
have to be specified. Cognizance and self-regulation, driving both ego and
intelligence to higher levels of functioning, are very strong candidates. These authors
also found that ego development distinctly predicts several other conditions, such as
aggression, when the influence of intelligence is removed. Therefore the system of
ego development above does have predictive validity that goes beyond intelligence.
Defence mechanisms
Defence mechanisms are powerful tools of ego development that have an emotional
and a cognitive component. The emotional component relates to their function:
the alleviation of stress and anxiety caused by conflicts between impulses or
individual priorities and socialization norms or practices. The cognitive component
relates to their operation: they transform information or condition received as full
of threat or stressful. It is stressed that defence mechanisms operate unconsciously.
In fact the use of defence mechanisms declines when individuals become aware of
their operation. That is, “lack of awareness is one of the reasons that defenses are
successful—that is, we are unaware that we are ‘deceiving’ ourselves. In thinking
about why children abandon certain defenses and replace them with others, it
seemed that the issue of awareness might be critical” (Cramer, 2015a, p. 117). Thus
different defence mechanisms dominate at different developmental phases. Therefore
the grasp of awareness is the major mechanism underlying the development of
defence mechanisms. In terms of James’ theory of the self, defence mechanisms are
tools of the I-self. They colour or distort what the I-self can see in the environment
or in the self, explicating why the content of the Me-self is not always accurate. This
assumption is in line with the assumption stated above that cognizance is the bridge
between intelligence, ego, and the self.
Three mechanisms were studied systematically in development: denial,
projection, and identification (Cramer, 2006, 2007, 2105a). Denial is rejection of a
stimulus or situation; when using denial, individuals fail to see, recognize, or
understand the existence or meaning of an internal or external stimulus. Projection
denies thoughts, feelings, or intentions as one’s own and attributes them to others.
Identification is the inverse. It causes a person to take thoughts, belief, values, or
behaviours of another person as one’s own. There is a clear developmental trend in
the use of these three mechanisms. Early in life, in preschool, denial is a prominent
defence. The reader is reminded of the development of awareness. Children can
map events and stimuli onto each other and become aware of representations at age
5–6. At this age the use of denial declines. Projection dominates in the primary
school years. It is interesting that this mechanism declines when children become
aware of the underlying cognitive mechanisms of thought and inference.
Identification dominates in adolescence. This is an important mechanism in this
phase because it allows the adolescent to shape his or her identity, according to the
values and priorities he or she espouses. It is reminded that this is the mechanism
that Loevinger associated with the conscientious stage.
Cramer (2008, 2015b) showed that the use of defence mechanisms relates to
intelligence in adulthood but not in childhood or adolescence. Individuals with high
IQ used identification more often than other mechanisms; individuals with low IQ
used projection and denial more often. Thus it seems that brighter individuals tend
to use more cognitively complex defences. Overall, ego development relates with
both intellectual development and changes in the use of defence mechanisms.
However, this relation is complex. On the one hand, higher levels in ego
development go with higher levels of intelligence. On the other hand, the relation
between ego development and use of defence mechanisms is U-shaped. Ego levels
with low internal impulse control use defences more often; ego stages with external
control of impulses rarely use defences; ego stages with internal control use defences
more often. In fact individuals with low IQ but a high level of ego maturity tend
to use defence mechanisms more often than high IQ individuals with a high level
of ego maturity (Cramer, 2006).
Conclusions
The research discussed above suggests several conclusions. First, it is clear that
intelligence, personality, and ego and self are related but they are distinct of each
other, because they represent different aspects of mental functioning and action in
real life. Intelligence comprehends, plans, and solves problems; personality frames
and colours comprehension, constrains plans and solutions to suit one’s predis-
positions, preferences, and values; ego and the self ensure coherence in making
meaning of the world and one’s own life and experience.
Second, there is a royal bridge connecting these realms: cognizance and the
awareness that it engenders. A plausible assumption is that the self-representation
system gradually builds pointers to different combinations of (i) problem-solving
skills and processes, (ii) dispositions to go on with a particular pattern of activity or
abandon it, and (iii) feedback received about successes and failures and the ensuing
feelings of satisfaction and dissatisfaction. These pointers are used by the person for
the sake of both self-regulation and self-representation. That is they direct the person
to choose those action patterns and environments that are appropriate and reward-
ing to him or her. Thus both action patterns and self-representations come out as
packages involving combinations of abilities, dispositions, styles, and interests. At the
beginning, when these packages are in the process of being formed, the relations
between their cognitive and personality components are strong because they may
frequently require review or reflection. Later, when they are established, these
relations weaken because action tendencies become automated. As a result personality
tends to appear less related with intelligence during the years of maturity.
Chamorro-Premuzic and Furnham (2006) suggested an interesting analogy about
the relative influence of intelligence and personality on actual life tasks:
Metaphorically, then, we may conceptualize Intelligence as the engine,

Conscientiousness as the accelerator, and Openness as the map. Neuroticism
and Extraversion, on the other hand, may be regarded as indicators of the

driver’s level of nervousness, optimism, confidence, and energy, whereas
Agreeableness (only a marginal indicator of intellectual competence) may
indicate how ruthless and competitive the driver is. Crucially, all personality
traits are somehow related to individual differences in intellectual competence
(p. 260).
Third, with development, the relations between the various processes are refined
and projected into one’s self-system. In the process the mechanisms of the ego, such
as the ego defences, ensure that this refinement and amalgamation will occur
smoothly and adaptively. The functioning of these mechanisms reflects the level of
cognizance that is possible at successive developmental phases. Children at 5–6 years
use denial because they have little understanding of it. It is reminded that up to this
age children are at the very beginning of becoming aware of representations and the
representational nature of knowledge. Obviously representations that are not stable
because they are not represented as such can easily be ignored or deleted. Thus
denial can take place for stressful or endangering representations because they do
not persist anyway. When children are aware of representations, but not of their
underlying interferential connections, they can project them on to others.
Interestingly, 11-year-olds have difficulty in understanding projection. When
children become aware of the inferential anchoring of representations, they
understand that one cannot have thoughts one did not reason about. Thus projection
dominates in middle childhood. Finally, in adolescence, awareness of mental process
allows identification with the other’s admired views and beliefs. This is not needed
when the adolescent acquires, by middle adolescence, an accurate self-representation
of characteristics, strengths, and weaknesses. It seems then that when children
become aware of a defence and its function they abandon it, adopting a more
complex one that is again not well understood. The use of defence mechanisms
increases under conditions of stress or threat to the self. However, excessive reliance
on age-inappropriate defences is associated with psychopathology (Cramer, 2008).
The research summarized in this chapter bears some clear implications for old
philosophical theories of the mind and the self. Specifically, the evidence allies with
rationalist philosophers, such as Kant and Descartes, rather than with the empiricist
philosophers, such as David Hume. It is clear that there is an overarching, implicitly
and explicitly represented, self-concept built around this core of cognizance; this
provides cohesion to the individual’s sense of uniqueness and subjectivity and to the
image projected to others. Obviously Kant would see his concept of consciousness
operating here. William James would see both his I-self and his Me-self. Hume
would be disappointed because this is against his bundle theory of mind, which
claims that the mind is “a bundle of perceptions” without any unity or cohesive
quality (Flage, 1990). Cohesion is built gradually, as the developing person strives to
assemble a well-functioning ego allowing him or her to give meaning and direction
to his or her life, correcting and repairing experiences that hurt, if necessary.
13
GENETIC, PSYCHOLOGICAL,
AND CULTURAL ASPECTS OF
THE MIND
The discussion so far may be summarized in a few state-of-the-art conclusions about

developing mind and intelligence. In short, (1) g is a multidimensional psychological
construct defined by control, representational, awareness, and inferential processes.
(2) The contribution of each of these factors, although always present, varies with
age. (3) As a result, g is re-morphed in development, with some processes
intertwining with and others differentiating from it according to developmental
phase. (4) This makes g appear qualitatively (representationally and experientially)
different across phases, rendering individuals more intelligent with growth (5).
One might argue that the architecture and development of mind described in
this book is specific to western culture, where the background studies were
conducted. For instance, schooling and habits in the West deliver a package of skills
and knowledge that requires withholding pleasure (hence executive control),
dealing with information in a demanding environment (hence attention and
flexibility), dealing with relations from various points of view (hence working mem-
ory), making decisions and choices and resisting deception to the best interpretation
at a specific moment (hence reasoning), and capitalizing on what was good in the
past (hence awareness and reflection). Building these skills requires time (hence
the developmental cycles and phases). Thus variations in specific environments
within the culture, such as social groups, families, nutrition, etc., cause the individual
differences we found. Is this the case? This chapter answers two questions:
1. What is the genetic basis of the various mental processes discussed here? Is there
specific evidence that would link specific aspects of the architecture and
development proposed here to specific genetic influences?
2. What is the generality of the model of mind presented here? Specifically, how
universal is the architecture and developmental sequence of cycles presented
here? Are there aspects of this architecture and development that are more
amenable to cultural influences than others?
Framing the debate about the origins of mental processes

There has been a long debate about differences between cultures in intelligence. The
discussion was polarized in the twentieth century between two diverging
epistemological positions. The first maintains that intelligence is a primary human
characteristic and, therefore, is the same across cultures. This view takes a second step,
assuming that tests of intelligence developed in the West do measure this universal
human characteristic. Therefore differences in attainment, such as differences in
average IQ attained by different ethnic or cultural groups, reflect differences in how
much of this characteristic is possessed by each group. Research conducted from this
perspective established a well-documented finding: western nations average about
100 IQ points, eastern nations, such as the Chinese, average about 105 IQ points,
and African nations average about 85 IQ points (Lynn, 2008). These differences were
considered equivalent to differences in what is involved in psychometric g (Jensen,
1998).
Some scholars argued that these differences are genetically determined, reflecting
selective pressures exerted on different human populations through the millennia as
a result of their selected environments when humans migrated from Africa about
100,000 years ago. The central idea was that the colder Eurasian environments
resulted in the selection of larger brains, more forward planning, family stability,
longevity, and slower individual maturation. These adaptations are associated with
the differences in average IQ mentioned above (Lynn, 2008). Naturally the debate
was heated and politicized and, as a result, was not very constructive. Research
on group differences in intelligence was politically tainted and demoralized by
opponents.
A famous book by Herrnstein and Murray (1994), The Bell Curve, further
inflamed the debate. The two authors developed their argument on the basis of
six propositions: that g exists; IQ tests measure it accurately and they are not
biased against any group; IQ is what people call intelligence; IQ scores are stable
in time and heritable. Based on these propositions they developed several
arguments with strong social and political implications: that IQ is little influenced
by the environment or education; and that social class and social mobility are
associated with intelligence. By implication, according to the critics, they asso-
ciated social class differences with genetic, and thus inherited, differences. The
book was both criticized (Devlin et al., 1997) and applauded. For instance, 52
professors, including researchers of intelligence, published a note in The Wall
Street Journal in December 1994 in defence of many of the views expressed in the
controversial book. Carroll (1997) himself reviewed the evidence for the six basic
postulates on which the book is based. He concluded, on the one hand, “that all
of them are well supported in psychometric and behavioral genetic research. On
the other hand, this says nothing, however, about the validity of the conclusions
Genetic, psychological, and cultural aspects 179
that Herrnstein and Murray draw about the consequences of variations in

intelligence” (p. 47).
The second position takes the opposite stance: it maintains that intelligence is the
expression of the worldview adopted by a culture. Therefore, according to this
position, it is not even possible to have a common measure of intelligence across
cultures with different worldviews. Thus it is just wrong to use a system of
measurement capturing a particular worldview, such as western IQ tests, to compare
different cultures; this would miss the essence of intelligence as developed and used
in all other cultures. Nisbett (2003) suggested that cultures differ along important
dimensions, such as the orientation to the individual as contrasted to the social group,
the orientation to a holistic as contrasted to an analytical approach to problems,
and the orientation to real-world functional organization of knowledge as contrasted
to an abstract rule-based organization. Differences between cultures reflect differences
in their history which, in turn, reflect long-term differences in needs and activity
imposed by differences in the environment. For example, the individualist stance of
the West was initiated in the geographically compartmentalized world of the ancient
Greeks and their ensuing localized administrative and political structures. The right
to argue, provoke, and evaluate views in an attempt to convince each other and
synthesize different views—all of which constitute the basis of a democratic state—
resulted from the fact that “polis”, the basic unit of the ancient Greek city-state, was
small enough to allow its citizens to know and interact with each other regularly.
This is contrasted, for instance, to the large, open spaces of the Chinese environment
with large populations where individuals do not know each other. Sustaining large-
scale crop production needs a collective mentality; having a functioning state
involving millions of people requires citizens to obey centralized law and respect
the lawmaker, rather than having equality of participation in law-making. Hence the
centralized administrative and political structures of these kingdoms, as contrasted to
the democratic organization of the Greek city-state. According to this position,
differences in IQ, such as those noted above, do not convey any meaningful
information other than that IQ tests capture an expression of intelligence that is
tuned to the West, favours the East, and disfavours the South (Nisbett, 2003).
The American Psychological Association set up a task-force, including some of
the most distinguished researchers of intelligence, to evaluate the knowns and
unknowns of intelligence, including the origins of individual differences in intelli-
gence. The aim was to provide a solid frame of evidence on the nature of intelligence
and the influence of genetic and environmental factors that ignited the debate above
(Neisser et al., 1996). A second paper followed recently (Nisbett et al., 2012), with
the aim to update conclusions in the light of recent research. For instance, at the
time of the original report, research on the relations between brain organization,
functioning, and development and intelligence or between genes and heredity and
intelligence were not as developed as in recent years. Several conclusions remained
strong and were well sustained over time. In the discussion below we will capitalize
on these two reports and draw on recent research specifically concerned with the
questions driving this book.
Genes, intelligence, and intellectual development

It is commonly accepted that gene action always occurs in a biochemical
environment and an ecological environment. Thus genes do influence intelligence
through their association with brain structures underlying intelligence. However,
genetic effects on the development of mental functions are influenced by the
environment (Neisser et al., 1996) and vary with age and specific social or cultural
environments. Specifically, research into the genetic bases of intelligence makes it
clear that similarity in intellectual achievement increases with genetic similarity.
That is, monozygotic twins, who are genetically identical, are more similar to each
other in intelligence than dizygotic twins, who share half of their genes; these, in
turn, are more similar than siblings born apart, due to greater similarities in the environ-
ment, and unrelated individuals. Overall, about 50% of variability in intelligence is
due to heredity. However, this varies with age. It is about 40% in childhood, 55%
in adolescence, 60% in early adulthood, and 80% later in life (Bouchard, 2004;
Bouchard & McGue, 2003; Hunt, 2011). It seems that, with age, people become
more selective of their environment and they even actively shape it according to
their own genetic propensities, leaving little room for influences that deviate from
what they feel is suitable and manageable. Therefore a large part of individual
differences in intelligence is indeed due to heredity, but the effect of environment
is also very important, particularly in early childhood.
It needs to be noted that the heritabilities for specific abilities are very similar to
the heritability of general IQ. Bouchard and McGue (2003) report the following
values based on meta-analysis of multiple sources: 48% for verbal ability, 60% for
spatial ability, 64% for perceptual speed and accuracy, and 48% for memory. Notably
the heritabilities of the Big Five factors of personality are in the same range: 54%
for extraversion, 42% for agreeableness, 49% for conscientiousness, 48% for
neuroticism, and 57% for openness (Bouchard, 2004).
Heredity expresses itself through the genome. That is, the genes are made up of
DNA which encodes the general plan of the build-up of animals’ bodies, including
the brain. There is extensive evidence showing that there is genetic g reflected in
indices of elementary cognitive processes. This g reflects physical measures, such as
qualities and quantities of neurons; physiological measures, such as plasticity of the
synapses connecting neurons to form new connections that would generate neural
networks to sustain new learning; and genetic measures, such as gene expression in
various aspects of the brain structure and functioning, as noted above. This genetic
g is related to psychometric g, which reflects measures such as processing speed,
working memory, performance on the Raven test, etc. (see Figure 13.1A). Recent
research on twins suggests that grey matter density in several brain regions, which
relates thus to genetic g, is highly heritable, substantially intercorrelated across brain
regions, and moderately correlated with g (Plomin & Spinath, 2002, p. 175). In line
with this model, Kievit et al. (2016) showed recently that various aspects of the brain
directly influence various aspects of processing speed and working memory, which
influence fluid intelligence (see Figure 13.1.B). Overall, then, genes shape, directly
FIGURE 13.1 (A) A model of the relations between genes, cognition, and
psychometric tests. It assumes that genes determine psychometric g, which determines
cognition and performance on tests (Reprinted with permission from Plomin and
Spinath, 2002). (B) The watershed model showing how various parts of the brain
influence various aspects of processing speed and working memory, which influence
fluid intelligence (Reprinted with permission from Kievit et al., 2016)
or in interaction with the environment, the formation and functioning of various

aspects of the brain underlying the functions of mind. At a functional level, very
global measures of these functions, such as speed and working memory, reflect the
efficiency of the brain in constructing integrated sequences of mental functions that
carry environment-related or action-important meaning. We will tackle these
relations in detail in the next chapter in order to show how the translation of
operations may occur from level to level in the brain.
Are there connections between genetic differences and differences in the nature
of the brain? Are there then specific genes associated with g? Bouchard and McGue
(2003) remarked, a few years ago, that no such genes had been discovered. This has
changed since then. Due to advances in the methods of studying genetic influences
on psychological characteristics and behaviour, an increasing number of studies
explored the relations between the genome and intelligence, g or IQ in particular.
These are mostly genome-wide association studies (GWAS). GWAS investigate the
entire genome, looking for genes that may be associated with a particular trait in
the population. In general in these studies, individuals are classified according to a
phenotypical characteristic of interest and they are then compared for genetic
differences. For instance, individuals are classified in groups varying in IQ, such as
low, average, and high IQ, and are then examined for possible differences in genes
across the whole genome. Technically speaking, researchers look for differences in
single-nucleotide polymorphisms (SNP); that is, for variations in single nucleotides
at specific positions (or loci) in the genome, where each of the variations is present
to some degree in the population. If there are systematic differences in the SNP of
specific genes between the groups it is assumed that the genes involved are related
to the trait of interest.
Recently, Sniekers et al. (2017), based on a GWAS meta-analysis of 78,308
individuals, identified 15 new genomic loci and 40 new genes associated with
intelligence. These accounted for 4.8% of the variance in intelligence, almost double
what had been obtained so far. Three of the genome-wide significant genes are
expressed in the brain and they are involved in neuronal function; that is, they
contribute to synapse formation, axon guidance in brain development, and
regulation of myogenic and neuronal differentiation. These genes were found to
have very high genetic correlation (.70) with educational attainment, strongly
indicating that they have a very powerful g-like effect on different aspects of
cognitive attainment. Hill and colleagues (2017) extended this study, using the same
sample together with the samples of two other similar studies. They found 107
independent associations for intelligence and increased the number of genes involved
in intelligence to 338. Specifically, 104 SNPs were implicated in expression
differences in the brain, and most of them (i.e., 100) were expressed primarily in
cortical tissue rather than other tissue types. These predicted 7% of individual
differences in intelligence and might even contribute to predicting an individual
level of intelligence. Interestingly, Zabaneh et al. (2017) showed that extremely
high intelligence (i.e., mean IQ of 170, which is .0003 of the population) is a
polygenic trait and is highly heritable. Noticeably they found that this relation was
associated with the plexin gene family. Plexins are implicated in axon guidance,
neural connectivity, and axon regeneration in the central nervous system; these
factors are related to several neuro-developmental disorders. The findings suggest
that extremely high intelligence is continuous genetically with normal-range
intelligence.
There has also been a search for specific genes associated with particular cognitive
abilities. Hansell et al. (2015) investigated the heritability of relational complexity
as specified in Halford’s theory already presented in Chapter 4. They showed that
relational complexity is highly heritable: genetic sources accounted for 67% of
individual variability in handling relational complexity. Further, they presented
preliminary evidence that four variants near or in gene DGKB and the NPS gene
appeared related to relational processing. These genes are related to insulin secretion,
whose decline is implicated in cognitive decline. Thus there is some evidence for a
direct link between specific genes and a powerful aspect of g, relational processing.
Similarly, Benyamin, Pourcain, et al. (2014) found that, although highly polygenetic,
childhood intelligence is associated the FNBP1L (formin binding protein 1-like)
gene. This gene is involved in a pathway that links cell surface signals to structural
aspects of brain cells that relate to information processing and inhibition, such as the
actin cytoskeleton.
There is also evidence for the genetic basis of domain-specific abilities. A set of
10 SNPs were found to account for a significant amount of variance of performance
in mathematics (2.7%) (Asbury, Wachs, & Plomin, 2005). Others found that
polymorphisms of the dopamine D4 receptor gene (DRD4) were associated with
performance on theory of mind and executive control tasks. Specifically, preschoolers
with shorter alleles of this gene outperformed those with one or longer alleles
(Lackner et al., 2012). Probably these receptors influence the development and
functioning of the frontal lobe, which exerts a general neuro-anatomical constraint
on the development of general metarepresentational ability. This assumption would
be consistent with the generalist genes hypothesis, which assumes that a common
set of genes constrain the operation of different cognitive abilities. In line with this
hypothesis, Kovas et al. (2007) showed that reading disability and mathematics
disability are affected by the same genetic factors to a very large extent.
Developmental research is at its very beginning in this field. Interestingly there
is research showing that some genes express themselves stably across age. However,
there is evidence that the expression of some genes related to intelligence is
developmentally paced. Ronald (2011) reviewed research showing that “to some
extent the same genes influence early and later cognitive and behavioral traits (from
infancy to adulthood)” (p. 1476). However, there was also “some evidence for
changes in genetic influence and that not all candidate gene associations replicate
across ages” (p. 1476). Thus there may be genes activating changes associated with
the various cycles discussed above. We will return to this issue in the chapter
focusing on brain development.
Is there an animal g?
Genes have existed for four billion years longer than humans (Fortey, 1997). Their
role has always been the same: guide the formation of the animal to form, grow,
survive, and reproduce so that it continues to exist. The products of this processes
evolved along many lines and took many turns and forms. One of the lines was that
which resulted in the species of homo sapiens. Even this line of evolution is longer
than humans by several million years (Boyd & Silk, 2014). Therefore it is natural to
assume that gene-based solutions for intelligence exceed humans and are present in
other animals as well. In any case, there is recent evidence, confirming the expected,
that genes related to general cognitive ability in humans did change in the last 6,500
years; specifically, genes related to dealing with novelty and abstraction gradually
dominated in the Holocene (11,700 years ago), when sedentary and agricultural life
dominated (Woodley of Menie et al., 2017). If this assumption is correct, it is
justified to expect a g-like ability in other animals as well. This would suggest that
intelligence evolved along a common dimension of gene-brain development,
despite its spurts and turns at some points in time in some species.
Indeed, recent research on animal intelligence abstracted a general factor
corresponding to human g which accounts for learning and problem-solving. This
animal g was related to attention control, working memory, and other executive
functions in the fashion found here. The similarity between human and animal g
implies an evolutionary dimension in the emergence of this structure (Burkart,
Schubinger, & van Schaik, in press; Matzel & Kolata, 2010). Perhaps this similarity
between human g and animal g suggests that research will have to search for the
equivalent of cognizance in animals.
There is strong evidence that some animals, such as dolphins, monkeys, and apes,
do have some capacity for metacognition. Does this mean that these animals are
consciously aware of their own knowledge states and mental processes? The
evidence available suggests that “some animals know when they do not know. But
do they know ‘I know’ or ‘I remember’ or ‘I believe’ or some non-linguistic
equivalent? Do they represent their knowledge as a belief state or a memory state
or do they have some more primitive way of monitoring their minds?” (Couchman
et al., 2012, p. 86). There are no definitive answers to these questions. Obviously
answering them would shed light on how human consciousness evolved over the
millennia, in liaison with changes in the size, the structure, and the functional
organization of the human brain. Aligning the brain of different animals along some
well-specified metacognition-related dimensions to the degrees and nature of
consciousness and awareness they possess would be an important step towards
answering these questions.
Culture, intelligence, and intellectual development

It was clear a long time ago that environmental factors contribute substantially to
the development of intelligence, but it was not clear how these factors work.
Attendance at school was considered important, but it was not known what aspects
of schooling are critical (Neisser et al., 1996, p. 97). Now we know that adoption
adds 12–18 points to the IQ of unrelated children, who are usually from lower
socio-economic backgrounds. Therefore a substantial fraction of the IQ advantage
is due to the environment, independent of the genes associated with them (Nisbett
et al., 2012, p. 136).
At the group level there have been large IQ gains in industrialized nations; this
is the so-called Flynn effect (Flynn, 2009). Specifically, Flynn found that fluid
intelligence had increased by 18 points and crystallized intelligence by 10 points in
western nations between the beginning of the twentieth century to the present. This
trend came to a halt in western nations but it started to take off in eastern and
African countries that embraced industrialization. Nisbett et al. (2012) suggest that
the ultimate cause is the Industrial Revolution, “which produced a need for
increased intellectual skills that modern societies somehow rose to meet” (p. 141).
This was implemented through the adoption of a scientific approach to reasoning
with an attendant emphasis on classification and logical analysis. This is implemented
in schools through a shift from fundamental mechanical skills, such as counting, to
pattern analysis and relational processing involved in relations between numbers and
other concepts. We will return to this possible underlying cause of the Flynn effect
in later chapters focusing on learning and enhancing intellectual development.
It is notable that the explanation for racial differences has clearly shifted in favour
of environmental/cultural influences. In the years of the Neisser report it was
recognized that there was not much direct evidence on black/white differences in
psychometric intelligence, but what little there was failed to support the genetic
hypothesis (Neisser et al., 1996, p. 95). Currently it is accepted that this difference
may be explained fully by environmental differences in the education-related
environment of the black family. For instance, African-Americans with European
genes who grow up in the environment of a middle-class Caucasian family exceed
African-Americans with similar genetic background by about 13 IQ points. The
differences in achievement between Asian Americans and white Americans may also
be explained on cultural grounds; east Asians espouse Confucianism, which involves
a strong belief that intelligence is primarily a matter of hard work. Thus they try
hard and they excel.
Cultures do impose a mental attitude for approaching reality and problems.
Luria (1976) showed that Asian peasants refuse to take the analytical stance and
apply inference to solve a simple syllogistic reasoning problem: “All bears in the
wood are black; Mischa met a bear today; what colour is this bear?” These peasants
responded that they could not know, as they had not seen the bear themselves!
Many other studies showed that individuals in different cultures may be very
proficient in solving locally relevant problems but inept in dealing with standard
school-like or test-like problems. Cole et al. (1971) showed that children in the
Kpelle tribe in Africa use locally realistic and functional criteria for classification
(robin is flying) rather than taxonomic criteria (robin is a kind of bird). The use of
taxonomic criteria increased with schooling. Gladwin (1970) showed that the
Puluwat in the South Pacific developed a sophisticated system of navigation that
integrates information about the weather, the wind, the ocean currents, and the
stars. This system enables them to navigate efficiently between islands. Carraher and
colleagues showed that street children in Brazil, who make their living by selling
various goods on the street, were more proficient in arithmetic calculations related
to their business than in the standard school-based formal context. Moreover, they
used strategies involving the mental manipulation of quantities in their street

calculations (e.g., they produce global quantities which they then combine) rather
than the manipulation of the symbols standing for the four arithmetic operations, as
is usual at school (Carraher, Carraher, & Schliemann, 1985).
Different cultures may have long-established institutions that facilitate the
relations and use of key systems of intelligence, leading them to excel in overall
intellectual performance compared with other cultures. We noted that schooling is
probably the most influential of these institutions in fostering intelligence. However,
there may be other, more specific, cultural institutions that relate to the overall
intellectual achievements of different cultures. A good example here is the learning
of the logographic system of the eastern cultures, such as Chinese or Japanese.
Learning this system is extremely demanding on processing efficiency and working
memory because children, from early in life, must be able to recognize and produce
thousands of complicated visual patterns. In two studies we examined Greek and
Chinese children, aged 4–14, on all dimensions involved in the mental architecture
(Demetriou et al., 2005; Kazi et al., 2012). Specifically, children were examined on
various aspects of processing efficiency requiring simple speed, attention control and
shifting. Speeded performance was examined with commonly familiar objects
and tasks related to reading (i.e., Latin, Arabic, and Chinese characters). They were
also examined on various aspects of working memory (spatial and verbal), reasoning
(inductive, deductive, and spatial), and cognizance (awareness of procedural
similarities and differences of various reasoning tasks). The aim was to investigate
possible effects that learning the Chinese logographic system had on possible
differences in intellectual development between these ethnic groups.
We found that the overall organization of cognitive processes was the same in
the two ethnic groups and was consistent with the four-fold architecture. However,
we also found that there were differences in the strength of relations between
processes and their relations with g. Specifically, in the Chinese, processing efficiency
related to reading more closely than it related to working memory; in turn, in this
culture, working memory was more closely related to reasoning. Notably, awareness
of cognitive processes was more closely related with working memory in the Greeks
but with reasoning in the Chinese. It seems that the Greeks needed the monitoring
processes involved in awareness more than the Chinese to handle working memory
tasks. However, the Chinese became more aware of cognitive processes from early
on, probably because they are forced to monitor and reflect on cognitive processes
to master the logographic system, which places a heavy demand on self-monitoring
and self-regulation.
Regarding attainment, Chinese always outperformed Greeks in speed of
responding to tasks in the logographic system from early on. Later this advantage
spread to letter recognition in general. Later still, at 7 years, Greeks caught up in the
systems in which they acquired practice. Interestingly, Greeks outperformed Chinese
on simple reaction-time tasks testing sheer processing speed, which are less related to
the early learning experiences in the logographic writing. That is, Chinese, because
of their logographic-specific experience, may have developed a more careful general
stimulus search strategy that gives an advantage when dealing with complex tasks.
However, this experience does not provide any advantage in very simple tasks where
plain stimulus identification is required. Along the same line, Chinese outperformed
Greeks on visuo/spatial and pseudo-words working memory tasks, which were
advantageous to them, but not on the regular verbal memory tasks, where they
seemed not to have any advantage. The Chinese also outperformed the Greeks in
spatial reasoning throughout the age span from 4 to 14 years. However, Greeks
exceeded the Chinese on deductive reasoning tasks. All in all, Chinese excelled in
processing efficiency, representational capacity, and reasoning only where they have
had a cultural advantage; that is, in dealing with spatial information.
Our findings that children developed along the same sequence of levels but at
a different rate, which was related to culture-specific experiences, is in line with a
large volume of cross-cultural research on cognitive development. In the heydays
of the Piagetian theory, thousands of studies tested if the Piagetian stages of cognitive
development are present in very different cultures—including that of Aboriginal
Australians and several cultures in Africa. This work was pertinently reviewed and
evaluated by Pierre Dasen, who was himself actively involved in this endeavour.
There is a very clear finding. On the one hand, the sequence of stages and sub-stages
was universal. The performance of children from all cultures could be pertinently
assigned to some Piagetian sub-stage in any task they were examined on. On the
other hand, there were large cultural differences in the relative rate of development.
In many cases children never reached the end of the stage of concrete operations,
let alone the stage of formal operations (Dasen, 1994). These differences in develop-
mental rate between cultures are in line with differences in IQ between the same
cultures noted above, and suggest that highly multidimensional and abstract
problem-solving is not demanded and thus not nurtured in some cultures.
Obviously these similarities suggest a common baseline in the development and
the organization of the developing mind that applies across cultures. Overall the
architecture of the mind is the same across cultures. In the same fashion, the cycles
and sequences of development are also the same across cultures. However, the mind
is a system for responding to and coping with variations in the environment. Hence
differences between cultures may appear in the relations between mental processes
and their development whenever there are differences between cultures that
systematically influence mental functioning and learning. Well-practised processes
would relate differently with other processes or with g than less-practised processes;
these processes would also develop faster in the culture where practice was massively
provided. These patterns are similar to the changing patterns of intertwining and
differentiation of mental processes within the same culture, depending on their
developmental state. We showed these patterns again and again in many different
cultures, such as India (Demetriou et al., 1996) and Pakistan (Shayer, Demetriou,
& Prevez, 1988).
Naturally, the discussion above focused on intelligence. It needs to be stressed,
however, that similar findings exist for personality. Specifically, there is abundant
cross-cultural research on personality which clearly suggests that the Big Five factors
of personality are reliably present everywhere on Earth. Schmitt et al. (2007)

addressed the universality of this structure by a specifically designed study that was
conducted in 56 nations speaking 28 languages. They found that, while the Big Five
factor structure was robust across all nations, there were systematic differences
between nations in the degree of dominance of each of the various dimensions. For
instance, some nations scored higher on openness than other nations; some scored
high in conscientiousness or neuroticism. Obviously differences in social practices
and priorities between nations are related to their relative standing on each of the
five dimensions in the fashion described above in regard to mental functioning.
An interesting question that might bridge cross-cultural differences in mental
functioning and personality relates to the role of self-awareness and self-regulation
in each culture. Nisbett et al. (2012) suggested that differences between groups in
intelligence may reflect differences in self-regulation practices which facilitate the
development of intelligence and reflect common influences that may originate from
personality. Thus we would argue that this approach is promising in its implication
that further progress may come from the integration of cognitive, psychometric,
developmental, and personality research, which is a major target of this book.
Conclusions
In conclusion, there is increasing evidence that specific genes are systematically
related to specific aspects of the organization and functioning of the brain, which
are then systematically related to cognitive achievements and individual differences
in the brain. We would argue that this picture of genetic influences on intelligence
aligns with our assumption that g involves some very general and ubiquitous
processes, such as the AACog processes, which are activated in any particular process
associated with g, be it executive, representational, or inferential. However, despite
the progress made, the association between genes and intelligence, or intellectual
development, is elusive and limited. So far only a small amount of variance in g is
accounted for by the genes associated to it. Even this is associated with some of the
processes involved in g in the global manner these are represented by tests, such as
the Raven test or global IQ. If the four-fold architecture is taken as a basis for
searching genes-mind-intelligence associations, very little has been discovered about
the relations between genes and the various SCSs or the various aspects of
cognizance. In the words of Nisbett and his colleagues: “It may simply be that the
number of genes involved in an outcome as complex as intelligence is very large,
and therefore the contribution of any individual locus is just as small as the number
of genes is large and thus very difficult to detect without huge samples” (Nisbett et
al., 2012, p. 135).
It has also to be borne in mind that the genome, the brain, and the mind are
three very different levels of reality (and analysis), each made up of different structural
components: the genome only involves nucleotids, themselves made up of proteins,
which are expressed in the brain, constraining and directing its construction,
functioning, and development. The brain involves neurons, connected by axons
and synapses, organized as discussed in Chapter 14. The mind involves the structures
that this book is about, such as the structures of the four-fold architecture. It should
also be borne in mind that there are no thoughts or mental processes in the genome
or in the brain. The brain is expressed via electrochemical events, measured by
various means, such as the recording of electrical activity or mapping the activation
of different brain areas (to be discussed in the next chapter). Mental processes exist
only at the cognitive level. There may be a direct causal relation between genes and
the brain and the brain and mental functioning and intelligence: that, for instance,
this and that gene cause this and that cell structure, thickness of the cortex, initial
connectivity, neuro-transmitter functioning; in turn, this or that particular brain
structure, initial connectivity, etc., cause this and that processing speed, working
memory, inferential power, etc., (as suggested in Figure 9.1 in Chapter 9).
Similar considerations apply about the influence of culture. It is generally
accepted that the genome is enveloped in a particular body guided by a brain that
acts and interacts with a particular environment. Therefore the pattern of genetic
effects observed in different social groups suggests strongly that the relations between
genes, heredity, and environment are much more complex than originally assumed.
For instance, it is well-documented that individual differences in intelligence among
children coming from lower social classes are more associated with environmental
differences; individual differences in intelligence among children coming from
higher social classes are more associated with genetic differences. This may simply
imply that children from lower social classes do not fully develop their genetic
potential, suggesting that they are in more need of social and educational support.
In any case, at present there is no way of knowing how much of the IQ advantage
that children with excellent environments have is due to the environments per se
and how much is due to the genes that parents creating those environments pass
along to their children. In addition, some of the IQ advantage of children living in
superior environments may be due to the superior genetic endowment of the child
producing a phenotype that rewards the parents for creating excellent environments
for intellectual development (Nisbett, 2012, p. 136).
The findings of cross-cultural research on intellectual organization and
development suggest a strong conclusion: one genome, one intelligence; one humanity,
one mind; many cultures, many mentalities. That is, human intelligence is the same
across human cultures because it is framed by the same genome. Even the overall
conception of the human mind seems to have many similarities across cultures
because, to the extent cultures observe it and explicitly describe it, they all see the
similarities. It is noteworthy, for instance, that the Confucian conception of human
intelligence involves many of the features involved in the Carroll three-stratum
model discussed in Chapter 2. According to Confucius, human intelligence
comprises “(a) the ability to identify areas of intelligence in others, (b) the capacity
for self-knowledge, (c) knowing how and when to solve problems, (d) verbal ability,
(e) the ability to think actively and flexibly, and (f) the ability to make healthy
personal decisions” (Pang, Esping, & Plucker, 2017, p.167). Attention is drawn to
the fact that all abilities are involved in the psychometric model except the capacity
of self-knowledge; however, this ability came full force in both cognitive and
developmental theory, as extensively discussed in this book. This theory, taken by
persons as a general life orientation rather than as a specific educational theory,
guides educational and learning practices in the Chinese culture. It is plausible to
assume that it may thus underlie the differences observed in their favour when
compared to western education, where teaching in self-knowledge was abandoned
after the dominance of behaviourism.
It is notable that the command of general processes, such as executive control
and reasoning, reflect the level of abstraction attained by a group. For example, the
analytical stance is secondary in development and it needs to be cultivated and
practised systematically before it can be used spontaneously. At the same time,
advanced knowledge and expertise reflect how general cognitive processes and
abilities interact with a cognitive domain in a cultural context. The success of
Brazilian street children in their version of mathematics does not indicate that these
children have a different mind than their middle-class contemporaries in school. It
only reflects that their transaction needs fostered the development of general
inference and quantitative reasoning in ways that were enough for the task, but, of
course, were short of the needs of formal computation skills required to deal with
abstract arithmetic tasks at school. In the same way, the sophisticated navigation
system of the Puluwat indicates that general inferential mechanisms, spatial thinking,
and other domains are all integrated to generate expertise in a highly valued culture-
relevant domain. Obviously this system is very remote from the abstract,
electronically based navigation system used to steer an aircraft. However, they both
organize space and time according to a complicated system of rules and knowledge
that needs extended training and practice to be mastered. The systems outlined here
are involved in both cases. It might be the case that the advantage of eastern nations
in general IQ noted above is related to this.
14
MAPPING MIND-BRAIN
DEVELOPMENT
It is taken for granted that the brain is the underlying biological mechanism of
the mind, because the mind emerges, in all of its expressions, from the structure
and functioning of the brain. Here we examine how these levels of analysing mental
functioning, the psychological and the biological, relate with each other. Thus we
are in search of the principles underlying the brain’s capability to generate mind and
the organization of mental structure. The frame for discussion is the four-fold model
of the architecture of the mind presented in Chapter 8 and the model of
developmental cycles discussed in Chapters 9–11. This chapter is organized into two
parts. The first focuses on the architecture and the second on the development of
the brain. In each of the sections we first remind the reader about the basic
assumptions and findings about the mind and then present evidence and theory
about the underlying brain structures and mechanisms. It is hoped that this
organization will make similarities and differences in assumptions, hypotheses, and
findings at the two levels transparent.
This chapter aims to answer the following three questions:
1. Is the architecture of mind suggested by psychological research reflected in the

organization and functioning of the brain?
2. What are the neuronal processes implementing the mental processes and
mechanisms? For instance, how are abstraction and inference carried out in the
brain?
3. How are changes in gross parameters of the brain (such as increases in neuronal
volume, myelination, neuronal networking, and neuronal pruning) reflected in
the phase and cycle changes of cognitive processes?
We caution that answering these questions is not easy. On the one hand, psycho-
logical research involves: (i) observable affective and cognitive behaviours measured
by a variety of tasks expressed in a variety of scales; and (ii) subjective experiences

expressed in various modes, such as self-descriptions, self-evaluations, certainty
evaluations, etc. On the other hand, brain research involves biological entities,
such as the nature, volume, and organization of neuronal matter and its functional
correlates, expressed through various modes (e.g., blood supply and glucose
consumption of activated brain areas and electrochemical activity). Structurally the
human brain comprises many cytoarchitecturally discrete regions (Brodmann,
1909; Glasser et al., 2016) which have both functional similarities and differences.
However, the brain develops and operates in an integrated manner as “the product
of a complex series of dynamic and adaptive processes operating within a highly
constrained, genetically organized but constantly changing context” (Stiles &
Jernigan, 2010, p. 327). The scales and precision of measuring each of these
dimensions (mind and brain) vary both within and across the two levels of
description, the structural and the functional. We do not yet fully know how to
map the different levels of analysis onto each other, and when we map them
precision falls short of the optimal.
A theoretical model seeking to link the structure and function of developing
mind with brain development must take into account the following assumptions.
First, the structure and organization of the human brain is constrained by the
genome. Human genes drive a prolonged process of brain development,
beginning in the third gestational week and lasting for decades. We showed in
the previous chapter that there is a genetic g that is related to psychometric g
and that their relation is directly expressed by elementary, content-free, cognitive
functions that reflect, on the one hand, the quality of brain functioning in
representing and processing information, and, on the other hand, the functional
efficiency of the mind to compose them into understanding or reasoning
sequences involving different types of information, such as visual, verbal,
numerical, etc. (see Figure 13.1 in the preceding chapter). This process is affected
by an incessant interaction between genes and the environment at all levels
(Gottlieb, 2007).
Second, the input of genes or the environment does not prescribe or determine
directly the outcome of mind development. Although brain structure and function
are the physical substrate of mind structure and function, their development over
time is a bidirectional complex, adaptive, and dynamic process. The research
summarized in the previous chapter suggests that our physiological structures are
not prewired in the genes; genes express themselves in an environment which
influences how they will express themselves into the structures they are related
to. Thus, although the general architecture of the brain and its physiology is
tightly constrained by genetic predispositions, the gene-environment interplay
sets a developmental course that is unique for each individual; by implication this
ensures a unique developmental course of the mind itself. Implementing the
probabilistic epigenesis framework (Gottlieb, 2007) in our analysis, we could
bridge findings coming from three traditions of research discussed throughout
this book.
Mapping mind-brain development 193
Architecture
Brain architecture may be analysed at different scales (e.g., at the level of the whole
brain, cortical or subcortical structures, synaptic connections, the neurons
themselves) with different techniques. Brain morphometry is concerned with the
mapping of brain architecture from various points of view. In fact the multiplicity
in the organization of the brain structure is an important property of its functioning;
its microscopic and macroscopic nature has been the focus of brain research for
decades. Most research and theorizing about the developmental relations between
brain structure and the mind are concerned with the macroanatomic level.
Locating cognitive functions in the brain

We argued here that the human mind is a complex universe of networks organized
into systems carrying out different tasks during understanding or problem-solving.
According to the four-fold architecture of the mind, there are four systems of
processes: domain-specific, representational, integrative, and cognizance systems. We noted
that domain-specific systems are directly related to the environment, grounding the
mind in the real world. The other three systems are increasingly detached from the
environment, organizing environment-specific information and mental constructs,
such as concepts and rules, at various levels of abstraction. Recent advances in
developmental cognitive neuroscience sketch a first draft of an integrative theory of
brain-mind function that is based on network architectures and mechanisms.
Domain-specific brain networks

Recent research on the brain suggests that each of the SCSs or their cognitive
facets may be mapped onto one or more specifically dedicated brain structures or
networks (see Figure 14.1). Object properties related to categorization, such as
colour, form, and size are processed by various parts of the visual cortex. Processing
of verbally expressed categorical information is associated with the entire superior
temporal gyrus, which is related to language understanding and it thus analyses the
“object” properties of auditory signals (Galaburda, 2002). However, implementing
already acquired expertise in object categorization, regardless of their identity,
involves the fusiform gyrus in the inferior extrastriate visual cortex (Gauthier et al.,
2000). Different aspects of visuo-spatial information, such as colour, depth, move-
ment, and form are processed by different regions in the visual cortices in the
occipital lobe (Livingstone & Hubel, 1988). However, mental rotation activates
the right posterior parietal lobe, centred on the intraparietal sulcus (Harris et al.,
2000). Interestingly, intellectually gifted persons activate frontal and anterior cingulate
cortices in addition to this region (O’Boyle et al., 2005). Quantitative information,
such as separate numbers and number relations (e.g., increase), is processed in the
inferior parietal cortex, particularly its posterior convolution called the angular gyrus
and the intraparietal sulcus. The frontal cortex is also involved (Dehaene, 2011;
FIGURE 14.1 Brain regions associated with mental processes presented in Table 14.1
Nieder & Dehaene, 2009). Perception of causality is served by visual cortices, such
as V5, but understanding of causal interactions between objects is processed in the
medial and dorsal part of the superior frontal cortex (Fonlupt, 2003). Crucial aspects
of social understanding involve several regions. Face encoding involves the right
hippocampus and left prefrontal cortex; however, face recognition involves the
fusiform gyrus in the inferior extrastriate visual cortex, showing similarities with
expert categorization (Kanwisher, McDermott, & Chun, 1997) and the right
prefrontal cortex (Haxby et al., 1996). Admittedly there is not as a yet a full map of
the networks serving each of the domains at its three levels of organization (i.e., core
processes, mental operations and belief systems, at different age phases).
It is important to note, however, that each domain may be served by a core or
hub network where different networks converge. These hub networks operate as
semantic unifiers that allow the thinker to interpret widely varying stimuli, in the
context of each domain, as just “one of the kind” in each domain. These semantic
hubs may be considered equivalent to meaning-bearing symbolic units. For instance,
the anterior temporal lobe may be the semantic hub for the categorical system
(Patterson, Nestor, & Rogers, 2007). The angular gyrus in the parietal region may
be the hub for the numerical system (Dehaene, 2011), and the claustrum for the
visuo-spatial system (Crick & Koch, 2005). Table 14.1 shows brain regions associated
with mental processes and Figure 14.1 localizes the processes in the brain.
Brain networks underlying representational capacity

Intrinsic interdependence analysis focuses on large-scale brain organization
independent of task-processing demands (Bressler & Menon, 2010). Using this
analysis on fMRI data acquired from subjects at resting state (i.e., metabolic activity
of the brain when a subject does not perform any task) offers an image of intrinsic
connectivity networks (ICNs) in the brain that is not biased to specific tasks. ICNs
suggest that the human brain is intrinsically organized into distinct functional
networks. ICNs have been found to be involved in executive control, episodic
memory, autobiographical memory, and self-related processing (Vincent et al.,
TABLE 14.1 Locations of mental functions in the brain
No. in Mental function Brain region

Fig. 14.1
Domain-specific
1 Categorical occipital, superior temporal gyrus
2 Spatial occipital
3 Quantitative inferior parietal (angular gyrus (BA 39))
4 Causal occipital, superior frontal (medial and dorsal)
5 Social (face medial prefrontal, superior temporal sulcus, temporal
recognition) poles
STS
6 Visual posterior parts of the superior frontal sulcus, entire
interparietal sulcus, amygdala, hippocampus
7 Verbal, rehearsal left-lateralized premotor-parietal
8 Verbal, anterior-prefrontal/inferior parietal
maintenance
9 Episodic right middle frontal gyrus, pre_SMA
Executive control
10 Inhibition ventral and dorsal prefrontal
11 Selection medial pre-frontal, parieto-temporal association areas
12 Switching the inferior frontal junction, premotor-intraparietal
network
Reasoning
13 Binding hippocampus, medial temporal and inferior temporal,
dorsal and/or anterior PFC
14 Inductive inferior frontal gyrus, right insular cortex
15 Deductive temporal (BA 21, 22), frontal regions (BA 44, 8, 9),
occipital (BA 18, 19), left parietal (BA 40), bilateral
dorsal frontal (BA 6), left frontal (BA 44, 8, 10), right
frontal (BA 46), right superior parietal lobule, thalamus,
right anterior cingulate
Consciousness
16 Visual claustrum
17 Theory-of-mind right and left temporo-parietal junction, medial parietal
cortex (including posterior cingulate and precuneus), and
medial prefrontal cortex (mPFC), ventral and dorsal
attentional
2006). It is clear that several structures and networks serve working memory.
Jonides, Lacey, and Nee (2005) suggested that the short-term storage component
of working memory is mediated by the same structures processing perceptual
information, and the executive parts, such as rehearsal, are mediated by brain
networks involved in attention control. Indeed, phonological storage is served by a
bilateral anterior-prefrontal/inferior parietal network that is related to language

perception. Visuo-spatial storage relies on a different bilateral brain system, which
includes the posterior parts of the superior frontal sulcus and the entire intraparietal
sulcus. Some brain regions are activated during processing of verbal rehearsal and
the visuo-spatial transformations, such as spatial sequencing and rotation: these
are the right middle frontal gyrus and the pre-SMA, as well as bilaterally in the deep
frontal opercular cortex and the cortex along the anterior and middle parts of
the intraparietal sulcus. This is in line with the assumption that central executive
processes, such as shifting, and episodic integration processes may exist (Baddeley &
Hitch, 2000; Gruber & von Cramon, 2003; Repovš & Baddeley, 2006).
The hippocampus is crucial for several aspects of working memory, especially
spatial working memory (Squire, 1992). It is involved in the explicit representation
of information rather than in implicitly represented memories related to automated
skills and habits. It temporarily binds together several distributed areas in the cortex
representing various kinds of information. In fact it is itself differentiated so that
different structures in it serve different aspects of working memory, such as: (i) retain-
ing information received from perception-based structures (e.g., from visual or
phonological cortices); (ii) specifying arrangement information, such as object
location and sequential order (Lisman, 2005), and (iii) delivering information to the
frontal cortex for further processing, such as relational integration (Friedman &
Goldman-Rakic, 1988).
Therefore working memory functions are widely distributed: the hippocampus
represents novel information, medial temporal and inferior temporal regions
represent visual and multimodal object representations, the dorsal and/or anterior
PFC capture relations between objects, and the ventral and/or posterior PFC
maintain the currently relevant items (Raganath & D’Esposito, 2005). The inferior
frontal junction (IFJ), which is situated at the junction of the precentral sulcus and
the inferior frontal sulcus, carries task switching, causing the attentional bottleneck
because it can retain only one rule at a time (Vergauwe, Hartstra, Barrouillet, &
Bras, 2015). Finally, response selection is carried out by medial and ventrolateral
pre-frontal regions (Cowey, 1996; Gruber & Goschke, 2004; Yoon, Okada, Jung,
& Kim, 2008). Thus the executive functions of working memory are carried out in
the frontal cortex rather than in the hippocampus as such. These networks which
serve executive processes in working memory are also involved in the integrative
and awareness processes described below.
Brain networks underlying integrative inferential processes

There are several structures and networks in the brain associated with the integrative
processes above. Simple inference based on Bayesian probabilistic covariation
involves the left cerebellum and adjacent visual cortex; this network seems to
integrate relations between movement and visual information (Ide et al., 2013;
Zheng & Rajapakse, 2006). Bayesian abstraction of functional covariation between
events involves, in addition to this network, the anterior cingulate and prefrontal
FIGURE 14.2 A view of the left cerebral hemisphere with the areas of cortex
numbered in accord with Brodmann’s cytoarchitectonic map
cortex, which is always present in relational thought (Ide et al., 2013; Wendelken
et al., 2015). Inductive reasoning activates the frontal gyrus (serving integration) and
the right insular cortex (serving, as noted above, salience detection and switching
between large-scale networks) to set attention and working memory in the service
of the salient representation selected (Menon & Uddin, 2010). Deductive reasoning
activates a set of networks serving different tasks at different stages of the inferential
process. Specifically, content-based propositions activate temporal (BA 21, 22,
serving language processing) and frontal regions (BA 44, 8, 9, serving integration).
Formal propositions activate occipital (BA 18, 19, suggesting the construction of
visual mental models of the relations implied by the formal propositions), left parietal
(BA 40, building associations), bilateral dorsal frontal (BA 6), left frontal (BA 44, 8,
10), and right frontal (BA 46) regions, serving integration, evaluation, and selection
(Goel, 2007; Goel, Buchel, Frith, & Dolan, 2000; Goel & Dolan, 2000, 2001). In
the same spirit, other studies suggested that deductive reasoning involves the right
superior parietal lobule (serving associations between concepts), the thalamus
(relaying information between systems), and the right anterior cingulate (serving
selection of competing responses) (Osherson, Perani, et al., 1998).
Recently Vendetti and Bunge (2014) proposed that three regions are central in
higher-order reasoning (e.g., analogical or deductive reasoning). The first region,
located in the inferior parietal lobule (IPL, BA 39/40), represents specific rather
than general relations and it scales with the number of relations to be considered.
This region makes first-order relations available to the second region, the left
rostrolateral prefrontal cortex (rlPFC, BA 10). This region abstracts second-order
relations, comparing and integrating mental representations over common relations
running through them. The third region, the right dorsolateral prefrontal cortex
(dlPFC, BA 9), provides a supporting role, enabling performance monitoring,
interference suppression, response selection, and manipulation of items in working
memory. The second region does not scale with difficulty but the third does. With
age, increases in myelination of the left rlPFC-IPL tract, but not the corresponding
right tract, is a particularly strong predictor of changes in reasoning ability.
A recent study by Schubert, Hagemann, & Frischkorn (2017) deserves special
mention. This study examined where the relation between speed of processing and
g comes from. They contrasted the traditional view that the influence of speed
reflects the quality of brain-wide properties of information processing aspects of the
brain, such as myelination, speed of neural oscillations, or white-matter tract
integrity or specific aspects of network communication, such as a kind of top-down
transmission of information from frontal attention and working memory processes
to temporal-parietal processes. They found, using measures of g, processing speed,
and EEG recordings of brain activation, that more intelligent individuals are more
efficient in this top-down transmission of information. This is consistent with the
theory advanced here that top-down selection of networks to be activated is crucial
to high intelligence.
Brain networks underlying awareness and consciousness

There is no consensus about the brain bases of consciousness. Several scholars
maintain that consciousness is associated with specifically dedicated brain structures
and others that it is an emergent dynamic condition reflecting the co-activation of
several networks. In any case, however, there are certain regions and networks that
are always involved.
Attention occupies a crucial role in consciousness because it is the gate between
unconscious and conscious mental functioning. According to Petersen and Posner
(2012), the attention system involves three fundamental components, each served by
a different network in the brain. The alerting system controls arousal and vigilance,
it is based in the brain stem system and the thalamus, and it is basically modulated by
norepinephrine. This is a bottom-up network that projects to the frontal and the
parietal cortices. The activation of this network renders one ready for awareness but
it does not by itself generate awareness. The orienting system is related to the
prioritization of input through selection of modality or location. This system is served
by the frontal eye fields (FEF) and parietal areas and frontal areas. Moreover, this
system seems to involve two discrete functions, one serving top-down visuo-spatial
selection (IPS/SPL and FEF) and one serving bottom-up reorienting (TPJ, i.e., IPL/
STG and VFC, i.e., IFg/MFg). This system is basically modulated by acetylcholine
and may be the gate to conscious awareness. Finally, there is a third, top-down,
system that handles conflicts and it is related to executive control. This system also
involves two networks, a frontoparietal network (precuneus, mCC, dlPFC, dFC,
IPS, and IPL) serving moment-to-moment task monitoring, and a cingulo-opercular
network (dACC/msFC, aPFC, al/FO) serving task maintenance. Wimmer et al.,
(2015) showed that PFC biases visual thalamic reticular nucleus (visTRN) to control
thalamic sensory gain, selecting appropriate inputs for further processing. In other
words, the TRN functions as a searchlight directing the internal spotlight of attention
to thalamo-cortical circuits that process ongoing behaviour. This is mostly modulated

by dopamine and serotonin. It is obviously the focus of awareness.
Consciousness relates to attention, in addition to whatever else it may involve.
Specifically, students of the brain bases of the theory of mind looked for a mental-
izing network serving awareness of mental states. They found that this network
includes the right and left temporo-parietal junction (TPJ), the medial parietal
cortex (including posterior cingulate and precuneus), and the medial prefrontal
cortex (mPFC) (Frith & Frith, 2003; Siegal & Varley, 2002). Other studies showed
that ventral and dorsal attentional systems are activated by the mentalizing network
to regulate the processing of self and other mental states (Abu-Akel & Shamay-
Tsoory, 2011; Mahy, Moses, & Pfeifer, 2014). Obviously ToM, attention, executive
control, working memory, and relational processing are served by overlapping
neural systems participating in the sequence of events generating consciousness
(Petersen & Posner, 2012).
How does consciousness emerge? For many, consciousness emerges from the
recurrent activation of the same network or the coordinated activation of complementary
networks (Edelman & Tononi, 2000; Lamme, 2006), rather than from the functioning
of any specific region. According to the global workspace model advanced by Baars
(1989), many modular networks operate in parallel at any time, often processing
information unconsciously. The content of processing will reach consciousness and
become the object of explicit awareness when the corresponding neural population is
mobilized by top-down attentional amplification into a self-sustained brain-scale state
of coherent activity that involves many neurons distributed throughout the brain. It is
as though this particular content seizes the whole brain for some time. As a result this
particular content is made available to various processes, such as categorization,
quantitative estimation, and long-term storage, through the long-distance connectivity
of these top-down amplified “workspace neurons” (Sergent & Naccache, 2012,
p. 102). Recent evidence suggests that two regions are causally crucial in this process:
the medial prefrontal cortex and the precuneous in the medial parietal cortex. When
these are inactive because of anaesthesia awareness breaks down (Boly et al., 2013).
It is worth noting that Crick and Koch (2005) proposed a model of visual
consciousness that involves an ad hoc specialized network. According to this model,
the spot of visual attention and consciousness is the claustrum (Cl), a subcortical
structure located in the basolateral telencephalon. Crick and Koch (2005) argued
that the Cl is richly connected to all sensory and motor regions of the cortex and
the amygdala, which is related to emotion. It receives input from all areas activated
at a particular moment and it binds them together, yielding awareness about the
mental content involved.
Parieto-frontal integration theory: an integrated model of brain

functioning underlying intelligence
Jung and Haier (2007) advanced an integrated model of brain functioning, the
parieto-frontal integration theory (P-FIT), claiming that it is the brain analogue of
fluid intelligence. Here we will map the P-FIT model onto the four-fold archi-
tecture proposed in this book. According to P-FIT, information is first registered
and processed in regions of the cortex which specialize to deal with different types
of sensory information, such as the visual (BA 18, 19) and the auditory cortex (BA
22). Therefore, the sensory areas involved in the P-FIT model may be more related
to the domain-specific processes represented by the present model, such as cate-
gorical, spatial, or quantitative core operations as noted above. From there,
information is fed forward to several regions in the parietal cortex (BA 7, 39, 40),
which primarily carry out elaboration, association with past knowledge or action,
and abstraction. Therefore the parietal areas of the P-FIT model may be related to
the abstraction and alignment processes involved in AACog. Then these regions
interact with frontal regions (BA 6, 9, 10, 45–47) in search of alternative solutions
to the problem at hand. Finally, the anterior cingulate (BA 32) is engaged to
constrain response selection by inhibiting alternative responses. The anterior
cingulate is supposed to intervene as a conductor orchestrating when each network
is to come into play in the sequence of brain events needed to reach a final decision.
The anterior cingulate is called on when there is interference caused by the fact that
the same neural networks are activated by different blocks of information (Gruber
& von Cramon, 2003; Klingberg, 1998). Therefore the frontal areas of the P-FIT
model relate to working memory, attention, and executive control; the anterior
cingulate relates to intentional planning, inhibition, and selection, which require
cognizance and metarepresentation.
Communication between brain networks

So far we have talked about interactions between brain networks related to cognitive
processes but we have not specified the code of these interactions. It is clear that the
brain responds to changes in the environment or changes in its priorities by changes
in its electrodynamic activity. For instance, at a rather global level, the preparation
of the brain for reaction to an expected stimulus (e.g., be ready to recognize the
colour of the incumbent stimulus) is expressed as a negative shift in EEG, which is
known as the contingent negative variation (CNV), and it arises bilaterally in the
anterior cingulate and adjacent structures. This reaction signifies the involvement
of executive control networks in setting the expected stimuli-relevant networks to
respond, when the stimulus appears (Petersen & Posner, 2012).
Thus networks speak to each other in a code based on various rhythms of brain
activation. Brain rhythms are periodic oscillations in excitability of groups of
neurons as reflected in EEG activity. Rhythms vary from very low (i.e., .05 Hz
frequency) to very high (200–600 Hz). Perceptual and cognitive activity is mainly
expressed into theta (4–10 Hz), beta (10–30 Hz), and gamma rhythms (30–80 Hz).
It seems that components of stimuli, such as successive letters or digits presented in
working memory experiments, are encoded by high-frequency rhythms, such as
gamma oscillations. These stand for the neural letters of thought. These letters are
combined into “neural words” and neural sentences according to a specific rule,
such as their presentation order, and are encoded by lower frequency rhythms, such
as theta oscillations. According to Buzsaki and Brendon (2012), these rhythms
constitute the basic components and syntactic rules of brain language.
Interestingly, the dialogue between regions serving working memory functions
in the hippocampus, such as place order of objects, and regions in the prefrontal
cortex serving executive functions is held as a sequence of gamma oscillations
representing items in the hippocampus and theta oscillations in the prefrontal cortex.
These latter theta oscillations organize hippocampal gamma oscillations into the
proper sequence (Jensen & Lisman, 2005). It has been suggested that working
memory capacity equals the number of gamma cycles (standing for individual
stimuli to be stored) that can go within a theta cycle. Thus integrated gamma/theta
cycles stand for a brain code for storing multiple items in working memory (Jensen
& Lisman, 2005). It has recently been suggested that theta activity is the fundamental
integrative mechanism of the brain that coordinates different types of information
expressed into other brain rhythms (Sauseng, Griesmayer, Freunberger, & Klimesch,
2010). Therefore increasing coordination between brain rhythms may underlie the
coordination between representations and inferential schemes underlying thought.
These findings are in line with recent non-invasive research investigating the
possibly causal role of different oscillatory rhythms in different types of cognitive
activity. Thut and Miniusi (2009) concluded, based on the review of a large number
of studies, that alpha (7–13 Hz) and beta oscillations (10–30 Hz) in V1/V2 of the
visual cortex are related to visual perception, conditioning the brain for forthcoming
perception; however, modulation of this band over rolandic areas of the motor
cortex are involved in the transformation of perceptions into action. Interestingly,
the dorsolateral prefrontal cortex (DLPFC) responds in other bands, mainly delta
(1.5–4 Hz) and theta (4–10 Hz). Thus there seems to be an association between type
of rhythmic activity and brain regions.
Abstraction in the brain relates to brain rhythms. The information at each stage of
processing (or in each part of a network) is physically the same (brain oscillations
of some band). When delivered to another network it may change in some physical
dimension (another band) to differentiate the stage or component of network
involved. Thus each stage encodes some other aspects of between stimulus relations
(e.g., first- or second-order). What it really differentiates between them is the
awareness that second-order relations preserve some of the first-order relations
(although there may always be real exemplars to ground each stage of processing or
level of abstraction) in reality.
Development
This section examines if the overall pattern of brain development parallels the
overall pattern of mind development already discussed. The human brain is genetic-
ally designed to grow over a developmental course that yields a unique kind of
information-processing organ that is increasingly capable of managing a multitude
of cognitive and emotional information. The functional specialization which arises
from cortical interactions is a developmental process occurring in a parallel and

synchronized way with intellectual development (Casey, Tottenham, Liston, &
Durston, 2005). As regions of the brain interact with each other over the course of
development they “fine tune” each other’s functions from general to increasingly
specific. In a similar and synchronized way, as different domain-specific intellectual
skills evolve, the human mind is transformed into an increasingly modular system,
functionally and structurally (see Barrett & Kurzban, 2006). This course—from
undifferentiated structures and global, all-purpose, functions to compartmentalized
structures and specific functions—defines the transition from a critical period of
plasticity during infancy and early childhood to a period of fixed neural circuitry
(Bardin, 2012), where central orchestrating processes become increasingly powerful.
In what follows we advance a domain-general framework for relating large
bodies of data on the postnatal development of human brain functions with specific
domains of cognition. Parting from the neo-Piagetian theorists who interpreted
emerging cognitive functions in terms of the maturation of particular regions of the
brain (e.g. frontal cortex), or gross brain processes as myelination (Case, 1992;
Fuster, 2002), we adhere to an interactive specialization approach. Two key ideas
underlie this approach. The first idea posits that activity-dependent interactions
between brain regions develop and, because of development, determine the
functions of these regions. As a result, the activity of the regions becomes increasingly
specialized to a narrower set of events; for instance, a region involved in a wide
variety of sounds becomes increasingly responsive to human voice. The second idea
posits that the computational power of the brain depends critically on how individual
processing elements at the molecular, the cellular, and the systems scales are
networked together (Sporns, 2012). In other words, the operation of each specific
cortical area is partly determined by its patterns of connectivity to other areas and
the patterns of activity of these other areas (Johnson, 2011).
Changes in mind-related brain networks

It is well established that the brain undergoes both structural and functional changes
in development. Postnatal brain development at the level of the neurons themselves
is both progressive and regressive. That is, there are expansions in various aspects of
the neurons, such as their number, the quality of their axons that transmit
information, and their interconnections at the synapses linking neurons with each
other. However, there are also decreases, such as a loss of neurons or connections
that are not in use. These changes are nonlinear over time and they vary across
regions. The most striking structural changes during brain development occur in
the relative proportions of grey and white matter: there is a peak followed by
a decline in grey matter volume and cortical thickness in early adolescence, but a
relatively continuous increase in white matter during adolescence and young
adulthood. These changes are heterochronic (occur at different age phases) and
follow a general trend of maturation from posterior to anterior and inferior to
superior areas; thus, somatosensory areas controlling perception and movement
mature earlier than association cortices controlling thought, particularly areas in the
frontal lobe (Colby, O’ Hare, Bramen, & Sowell, 2013).
Specifically, although the total brain size is about 90% of adult size by age 6, the
brain continues to undergo dynamic changes throughout adolescence and well into
young adulthood. Connections between neurons also change systematically in
development. For example, synaptic density (i.e., the number of synapses/mm3)
changes exponentially up to the first year (from about .5 x 108 at the fifth gestational
month, to 2.5 x 108 at birth, to 5.8 x 108 at the end of the first year of life) and it
then slows down (to 3.5 x 108 at 10 years and 3 x 108 at 70 years). Also, it is noted
that synaptogenesis and synaptic pruning differ across brain regions. These changes
peak at the second month after birth in the sensorimotor cortex, at the end of the
first year in the parietal and the temporal association cortices, and from 4–6 years in
the prefrontal cortex (Tau & Peterson, 2010).
A remarkable fact about brain development is that structural and functional
connectivity of brain networks are mutually related. Recently developed methods
are used to systematically specify the structure and interlinking of networks at both
the cognitive and the neuronal level. Exploratory graph analysis is a promising
approach. This analysis explores network models by specifying the links between
particular units. These may be measures on different levels of analysis. One may be
the cognitive level, involving measures of specific cognitive processes, such as
various types of (i) attention control, (ii) working memory processes, (iii) arithmetic,
(iv) proportionality, (v) algebraic reasoning, (vi) Raven matrices, etc. Another is the
brain level, including measures of different regions of the brain (e.g., the (i) frontal
lobe, (ii) the hippocampus, (iii), the parietal lobe, (iv) the occipital lobe, etc.)
activated when working on different types of cognitive problems. Exploratory graph
analysis estimates the relations between units to examine if they form clusters;
additionally it specifies the relations between clusters.
Figure 14.3A illustrates how networks are represented in graph analysis. It
can be seen that there are nodes connected with each other so that they form a
cluster. These clusters obey small-world structural relations. Small-world networks
involve nodes connected to one another so that any node can be reached from any
other node by a small number of steps. These nodes may be anything: in cognitive
research they may be measures of different types of attention control, working
memory, etc.; in brain research they may be groups of neurons co-activated to serve
a function. A cluster is a group of connected nodes regardless of the relations between
this cluster and other clusters. Technically, a small-world network, in cognitive or
brain terms, may stand for a module. The relations between clusters may vary from
a distributed set of relations connecting each unit from one cluster to several units
in the other cluster to privileged relations between specific units in each cluster
which form nodes or hubs extending links across clusters. In Figure 14.3A, hubs are
represented by dots connected by lines connecting the three local networks. Strongly
connected clusters via connections between their representative hubs form rich club
systems. In cognitive terms, for instance, there may be a cluster for visual short-term
storage, a cluster for verbal short-term storage, and a cluster of executive processes
all connected to form working memory. This method is similar to various methods
of factor analysis, such as confirmatory factor analysis, in that clusters in graph analysis
are assumed to operate as factors or latent dimensions in factor analysis. The
advantage of this method is that it allows specifying the direct relation between units
at various levels and also the role of each unit in the functioning of other clusters.
This method has been used in both cognitive developmental research and brain
research. In cognitive developmental research we applied this method on many
of the studies we have already presented in this book (Chapters 8 and 11) to specify
the organization of various processes targeted. For instance, Figure 14.3B shows the
results of this analysis applied on the data presented in the hierarchical model shown
in Figure 8.3 in Chapter 8. Figure 14.3B shows that this analysis abstracted four
clusters: one for spatial, one for mathematical, one for social thought, and one
for fluid intelligence underlying logical analysis. Attention is drawn to the fact that
there are many connections both within and between clusters. However, some
nodes within each cluster are more strongly related than others, indicating that these
nodes play a central role in the organization of the cluster. Some nodes from each
cluster also have a stronger connection with some nodes in other clusters, indicating
that these nodes operate as hubs allowing the clusters to interact (see thicker lines).
Obviously factors of various orders in confirmatory factor analysis and structural
equation modelling grossly represent the network of relations between units (first-
order factors), nodes and their relations (second-order factors), and hubs and their
relations (third-order factors, such as g).
Figure 14.3C shows how modules would be represented in the brain as clusters
of nodes in different brain regions connected with each other into broader systems
(rich clubs). There is an increasing number of studies using these methods to explore
the connectomics of the brain both structurally and functionally (Cao, Wang, Dai,
et al., 2014; Sporns & Betzel, 2016; Vértes & Bullmore, 2015; Zuo, He, Betzel, et
al., 2017). The overall current spirit of this research can be summarized as follows:
networks are topologically complex at birth; early in development clusters are based
on physical proximity and neuronal similarity of the neuronal systems involved
(Zuo et al., 2017); however, their organization gradually evolves over childhood
and adolescence from a local architecture dominated by sensory and sensorimotor
areas to a more diffuse topology, forming rich club networks, facilitating higher
level integrative functions. That is, from infancy to early adulthood the proportion
of short-distance connections decreases and long-distance connections increases
(Cao et al., 2014). These network changes are related, on the one hand, to synaptic
pruning and decreases in grey matter volume and, on the other hand, to progressive
myelination of long-range connections. These later changes are related to increasing
functional connectivity resulting in the strengthening of relations between structure
and function (Vértes & Bullmore, 2015).
These studies suggest that the synergetic function of brain areas working together
as large-scale networks might point to how the brain generates mind (Bressler &
Menon, 2010). This reciprocal interaction takes place on multiple time-scales
during development (Sporns, 2012). On a slow developmental time-scale this
FIGURE 14.3 Representation of cognitive and brain networks according to graph theory. Panel A shows an abstract representation of
modules and systems of modules. Panel B shows cognitive modules and a general system of modules standing for g. Panel C shows
brain connectomics indicating modules and between modules relations in different brain regions
Source: Vértes & Bullmore (2015).
dynamic interaction reshapes the brain’s anatomy through mechanisms such as

activity-dependent dendritic development, synaptogenesis and synaptic pruning
(Rubinov, Sporns, van Leeuwen, & Breakspear, 2009). At faster time-scales funct-
ional networks connectivity fluctuates, but gradually it becomes stable and is perhaps
constrained by emerging structural hubs. The emergence and activation of these
hubs may also relate to the fine interplay between the dissociation and integration
of functionally specialized processing during development; this is considered the
hallmark of brain connectivity and functional specialization (Anderson, Kinnison,
& Pescoa, 2014).
Understanding how connectivity emerges during development is the route to
understanding the complexity of the developing brain. According to the traditional
model, brain development includes two phases: an early genetically predetermined
phase resulting in a gross wiring of the neural networks; and a later phase during
which the extant connections are tuned because of brain-environment interactions.
This traditional model is modified nowadays into an interactive model assuming
that the organization and function of neural networks are shaped by continuous
interaction between genetic information and experience during all stages of
development.
The brain shows patterned activity from prenatal age. Numerous studies have
demonstrated that spontaneous neural activity plays an important role in forming
connectivity patterns in the developing brain (Sporns, 2012). For instance, Moreno-
Juan et al. (2017) showed that prenatal thalamic spontaneous calcium waves are key
regulators of sensory cortical area size and that prenatal alteration in inter-thalamic
sensory nuclei communication may trigger size adaptation in cortical fields. This
evidence aligns well with the finding that the foundations of resting-state networks
are already in place before birth, with rapid neural growth in the last three months
of pregnancy (Hoff, van den Heuvel, Benders, Kersbergen, & De Vries, 2013).
These networks include bilateral primary motor, primary visual and extra-striate
visual, parietal-frontal and frontal (executive control networks) insular-temporal/
anterior cingulate cortex (ACC; salience and shifting networks). It is notable that,
although in place, these networks reach maturity levels at successive age phases
which correspond to the changes in symbolic units that dominate in working
memory in each cycle (Hoff et al., 2013). Early in infancy, basic acoustic, visual,
and motor networks are active, providing the basis for episodic mental units
(Franson, Aden, Blenow, & Lagercrantz, 2011). These areas are connected, although
minimally, with the integrative superior parietal cortex. These connections are
strengthened throughout childhood, from age 3 to 7, probably explicating the spurt
in visual working memory and related mental activity. However, these connections
are still weaker than in adulthood, reaching maturity in middle adolescence.
Interestingly, in adolescence long-distance connections between prefrontal hubs and
the cerebellum are established, which may relate to fine tuning and error detection
by spotting network (and representational) inconsistencies (Hwang, Hallquist, &
Luna, 2012). Thus it seems that structural potentials of the brain for the episodic
(primarily sensorimotor), the realistic mental (primarily visual), and the rule-based
(primarily verbal) mental representations are made available at the beginning of the
respective cycles.
The topology of brain networks changes greatly over the human life. These
changes occur in the number, arrangement, and interdependence of brain structural
nodes influencing the effectiveness and spatial pattern of their dynamic interchanges
at all scales. It is worth emphasizing that network development and plasticity is one
of the most significant drivers of cognitive development. Many neuro-imaging
studies examining different age groups have started to cast light on neural substrates
of cognitive development (Johnson, 2011).
Cycles in brain development

Thompson, Gledd, Woods, MacDonald, Evans, and Toga (2000) showed that
changes in the connectivity within and between the two hemispheres are very
similar to the phases of intellectual development described here. Specifically, the
orienting network is active by the age of 6–7 months. However, executive attention
emerges as a control mechanism only at 18–20 months, rendering episodic executive
control possible. In the 3–6-year period there is a drastic change (60%–80% increase
in connecting fibres) in the frontal circuits of the corpus callosum, which sustain
vigilance and regulate action planning. Early in this period, from 2–4 years, the rates
of glucose metabolism double (Chugani, Phelps, & Mazziotta, 1987). This goes well
with the finding that in this phase, at 3–4 years, the inhibition network is function-
ing effectively, rendering the “focus-choose-respond” programme possible. Later,
at 6–7 years, there is dramatic change in the callosal isthmus, which supports
intraparietal associative and language functions. In addition, in the period from 7–11
years the anterior cingulate gets differentiated from the orienting network and
connects to the executive and the associative IP networks (Rothbart & Posner,
2015), rendering the conceptual fluency programme possible.
Obviously we are a long way from understanding how the various brain rhythms
coordinate with each other in the fashion proposed by Buzsaki and Brendon (2012)
to create the brain units corresponding to the representational units of each of the
phases and cycles outlined here. Interestingly, however, the changes above are
tractable in spurts in the amount of EEG energy found in the alpha-frequencies.
Specifically, Thatcher (1992, 1994) found that electroencephalographic coherency,
which reflects changes in connections within and between regions, develops in
growth spurts that are nearly identical to the time-frame of the developmental cycles
described above: these cortical connections develop in approximately four-year
cycles, within which there are growth spurts in EEG coherence (a measure of
cortical connectivity) that last for 6–12 months (Somsen, van’t Klooster, van der
Molen, van Leeuwen, & Licht, 1997). The three cycles extending from early
childhood onwards occur at approximately 1.5–5 years, 5–10 years, and 10–14
years, with growth spurts occurring at the ages of 6, 10, and 14 years. Along the
same line, Epstein (1986) showed that spurts in EEG activity coincide with the first
phase of the cycles presented here in most of the cases: they occur at the ages of 2–4,
6–8, 10–12, and 14–16. Additionally, Hudspeth and Pribram (1992) reported
changes in relative power indices across the brain occurring in stages that almost
perfectly match those presented by Thatcher and in the cycles proposed here (1–6
years, 6–10.5 years, 10.5–13 years, and 13–17 years). It may thus be the case that
transitions across cycles relate to the establishment of the brain networks (based on
EEG coherency and power) necessary to project representational alignments of an
earlier cycle into the more abstract networks capturing the new units that emerged
from these alignments. Changes in the second half of each brain cycle relate to the
extending and consolidating connections within regions.
It may be the case that, at the beginning of cycles, the time taken to evaluate
information is longer because the networks serving the new mental units are neither
fully established nor well-tuned to each other. This is expressed in two kinds of
evidence. Activation of the necessary anterior cingulate network in the evaluation
of the same stimulus is much slower in 4-year-old children. The respective P300
response is only 50 msec in adults but 400 msec in children. “Another important
difference between 4-year-old children and adults was the distribution of effects
over the scalp [see Figure 14.1]. In adults, the frontal effects appear to be focalized
on the midline, whereas in children the effects were observed mostly at prefrontal
sites and in a broader number of channels, including the midline and lateral areas.
In addition, the effect on the P3 appears to be left-lateralized in the adult data but
lateralized to the right side in the children. The focalization of signals in adults as
compared to children is consistent with neuro-imaging studies conducted with
older children, where children appear to activate the same network of areas as adults
when performing similar tasks, but the average volume of activation appears to be
remarkably greater in children compared to adults” (Rueda, Posner, & Rothbart,
2005, p. 586).
Interestingly, Wendelken, Ferrer, Whitaker, and Bunge (2015) found systematic
changes in networks associated with first- and second-order reasoning in the years
coinciding with the cycles of rule-based and principle-based concepts. Specifically,
in the 7–10 years cycle, the IPL-rlPFC network is activated but it processes both
first- and second-order relations, which are not well differentiated. This
differentiation occurs in two phases coinciding with the recycling phases described
above. That is, in the 6–8 year phase, reasoning development goes with weakening
of connectivity between the DLPFC and the VLPFC. Interestingly, in this period,
changes in speed mediated changes in reasoning ability, suggesting that entering the
cycle of rule-based reasoning is associated with differentiation between DLPFC and
VLPFC, causing faster processing. In the next phase, from 9–11 years, the RLPF
dominated and it was coupled with the right RLPFC. Notably, in this age phase,
changes in working memory mediated changes in reasoning, suggesting that
consolidation of rule based-reasoning is associated with the consolidation of
the relevant brain network, which is expressed in working memory expansion.
In the 11–14 years phase, left and right dlPFC as well as dorsomedial PFC were
more strongly engaged in processing second-order rather than first-order relations.
That is, in this phase the second pole takes its primary function in processing
second-order relations. In the 15–18 years phase the left rlPFC and bilateral IPL
were engaged in processing second-order relations as contrasted to first-order
relations. It seems that cortical reorganization within the IPL leads to greater
efficiency in processing first-order relations, thereby reducing relational processing
demands within rlPFC. Notably, Mackey, Miller Singley, and Bunge (2013) found
that training in reasoning strengthened the connections between the rlPFC and the
IPC but also between the IPL and the striatum. Thus the establishment of these
networks seems related to the acquisition of the inferential relevance mastery programme.
In this last period, the speed-reasoning and the working memory-relations
disappeared, suggesting the consolidation and automation of higher-order inference.
Structural equation modelling of longitudinal data found the same trend at the
behavioural level (Žebec, Demetriou, & Kotrla-Topić, 2015).
In conclusion, the general pattern of brain development seems to follow a
recycling course. In particular the human brain is genetically equipped (1) to detect
sensory regularities of the outside world, (2) to store these regularities in cortical
networks, and (3) to convert these regularities into conceptual knowledge, rendering
them representational of the realities that gave birth to them. This pattern of brain-
environment interactions appears to follow a recycling course matching the
developmental cycles of the mind proposed in Chapter 9. Thus we claim that
conceptual development starts with formation of unconnected sensory and motor
systems in conceptual tasks; at a subsequent phase it advances to the construction of
amodal conceptual representation systems; and eventually it ends up with modality-
specific conceptual features integrated into coherent systems. The evidence
summarized here indicates close links between the sensory and motor brain systems,
on the one hand, and the conceptual system, on the other (Kiefer & Pulvermüller,
2012; Machery, 2016). The neuronal recycling hypothesis was proposed by Dehaene
(2009) to explain the neural processes underlying reading. This hypothesis resembles
our conception of recycling. In our proposal, we use the term neural recycling
to mean the gradual transformation of unconnected sensory/perceptual systems to
highly networked conceptual systems.
Brain development under adverse life conditions

There is clear evidence that adverse life conditions directly affect brain development.
These conditions include poverty and ensuing nutritional deprivation and poor
sanitary conditions, and institutionalization and ensuing psychological deprivation
concerning a rich sensory environment where infants can interact normally with
physical objects and a social/parental environment where infants can build healthy
patterns of social interactions with other human beings. Specifically, growing up in
an environment of poverty, as happens in many developing countries, may exert
highly adverse influences on as many as 30%–40% of infants. Hanson et al. (2013)
showed recently that growing up under conditions of poverty negatively influences
various aspects of brain development. In their study they examined the development
of grey and white matter over different areas of the brain in children aged 5–37
FIGURE 14.4Frontal lobe grey matter volumes according to age and SES
(Reproduced with permission from Hanson et al., 2013)
months. Figure 14.4 shows the growth of grey matter in the frontal brains of
children with high, middle, and low socio-economic status (SES) over this period.
It can be seen that there are clear differences between the three SES groups which
increase with age.
In line with these findings, a recent study conducted in Dhaka, Bangladesh,
showed that children in poverty with stunted growth have smaller volumes of grey
matter than non-stunted babies. This study also “detected stronger electrical activity
among children with stunted growth, along with a range of brainwaves that reflect
problem-solving and communication between brain regions” (Storrs, 2017, p. 152).
Other studies showed that children raised in institutionalized settings show
various neuro-developmental deficits in executive processes, working memory, and
problem-solving which relate to brain development impairments caused by the
institutionalization experiences. Unfortunately many of these problems persist for
many years after these children have left the institutional environment for family
life (Pollak et al., 2010).
Conclusions
This book aims to highlight the architecture and development of the mind. The
present chapter summarized research on brain architecture and development and
explored how these two levels of description of the human mind are interrelated.
Here we evaluate the evidence in reference to the three questions stated at the
beginning of the section focusing on the brain.
Architectures of mind and brain. The first question asked if the architectures of
the mind and the brain resemble each other. The four-fold architecture des-
cribes (i) several specialized domains of thought, and central (ii) representational,
(iii) integrative, and (iv) cognizance processes. The evidence reviewed in this
chapter shows that there are several networks in the brain sub-serving each one of
these mental functions. Specifically, (i) different networks carry out the core
processes comprising each of the psychological domains (rooted in sensory cortices
but extending into various regions depending on the relations involved); (ii) other
networks serve working memory by protracting information in time, so that it can
be related to information following or past knowledge (rooted in the hippocampus
and the reticular formation); (iii) other networks, which are mainly rooted in
temporal, parietal, and prefrontal cortices, take as inputs the networks mentioned
above, align them, and abstract their common elements; (iv) finally other networks
monitor, orient, select, and regulate the networks above to optimize goal-related
abstractions (rooted in frontal and medial cortices) and handle differences. By
definition, information entering this last type of network is available to awareness
and metarepresentation.
The reader may have noted that we emphasized brain networks rather than brain
structures as bases of mental processes. The evidence is compelling that mental
functions are served by overlapping, developmentally changing, brain networks
rather than single structures. In terms of Anderson’s (2015) neural reuse model,
“individual neural elements (at multiple spatial scales) are used and reused for
multiple cognitive and behavioral ends” and they participate in multiple, overlapping
neural coalitions (p.1). Coalitions are established through a neural search process
that reminds one of the alignment processes described at the beginning of this
article. By analogy, mental entities participate in a mental reuse process such that
one entity is used for multiple cognitive tasks. For instance, mathematical estimation
is used to express class relations, causal relations, spatial coordinates, etc. A flow of
brain events as described in the P-FIT model may capture how the various processes
succeed each other in time for the sake of understanding goals. In conclusion,
functional mind structures specified by the four-fold model are served by both
physical and functional structures in the brain.
Central processes in mind and brain. The second question focused on the possible
relation between mind and brain processes. Meaning emerges from mapping
representations onto each other so that they are compared, integrated, redefined, or
re-represented into new representations. AACog is a package name standing for
these processes. The brain analogue of AACog lies in the interactive and syntactic
processes of the brain. Specifically, various oscillatory rhythms reflecting the
activation of brain units and networks stand for “letters”, “words”, and “sentences”
in the meaning-making process. At the mental level, comparison between
representations and the search for similarities and differences is implemented by
alignment processes, which are inherently executive. At the brain level, these
processes are expressed in oscillatory co-activations between the networks
representing the mental entities involved. Ideally, complete rhythm coupling would
signify representational alignment. At the mental level, abstraction occurs when the
commonalities between the representations aligned are identified. The brain
equivalent of abstraction may be the lock of the rhythms coupled above through a
rhythm of a different band, such as when several gamma oscillations are bridged by
a theta oscillation. Cognizance may emerge when this new theta oscillation is made
available into a broader theta-based network, thereby functioning as an autonomous
token of a new mental object. Focusing on this new object itself may bring it into
the focus of awareness. Ensuing brain activations may be the equivalent of this
awareness. Rostrolateral prefrontal cortex is the region primarily serving these needs
(Dumontheil, 2014).
Mind-brain development. The third question asked about the relation between
mind and brain changes in development. Developmental research suggests that
intellectual development occurs in four cycles of emergence and alignment yielding
insights into the cycle’s dominant mental unit and alignment process, thereby letting
them be re-encoded and metarepresented into mental units differing in level and
form of representation. All major development theories described four levels of
intellectual development with transitions at ages 1.5–2, 6–7, and 11–12 (e.g., Piaget,
1970; Case, 1985), associated with increasingly abstract (i.e., episodic, realistic
mental representations, rule-based concepts, principle-based systems, respectively),
differentiated, but precisely interlinked representations.
The research on brain development reviewed above strongly suggests that the
power underlying these cycles lies in the fact that brain networks change in cycles
reminiscent of intellectual cycles. Thus the cycles of intellectual development
correspond to successive expansions of neuronal networks such that earlier net-
works are integrated into the hub architecture of the networks constructed later.
The crucial aspect of these expansions lies in the addition of extra connections to
the parietal and the frontal hubs. An interesting hypothesis for future research to test
is that the changes in the relations between g and different forms of executive
processes and reasoning at successive developmental cycles are related to the changes
summarized above in the formation of small-world and rich club networks. With
development, decreases in intra-modular connectivity render the modules
increasingly refined and task-specific. Increases in inter-modular connectivity by
strengthening between-hub connections allow viewing each rich club module from
the point of other modules in the club. In these processes some hubs become
stronger and better connected to the modules involved, almost literally fleshing out
rules and principles. That is, when activated as such they allow activation to flow
through the networks involved, delivering cases and implementations of the hub
dominating. The more fluent the activation the better and faster the functioning;
developmentally speaking, establishment of a network of this nature may correspond
to a developmental phase or cycle; psychometrically speaking, differences in the
efficiency of the activation of the network may underlie relevant individual
differences.
Interestingly, the basic symbolic units of each cycle are discernible at both the
mental and the brain level. At the mental level, they are episodic in the first cycle,
representational but visually dependent in the second cycle, rule-based and language
encoded in the third cycle, and principle-based and language or arbitrary symbol
systems (e.g., mathematical) encoded in the fourth cycle. At the brain level, the
dominant networks are located in the sensory and the motor cortices, the reticular
and the parietal, the prefrontal and the frontal cortices, respectively. In other words,
epigenetic mind-brain interactions transform the mind into a powerful represent-
ational machine capable of creating and using complex abstract representations in
the service of different domains of knowledge. Thus each of the four cycles is a
dynamic state of functioning at both the mental and the brain level. At the mental
level, each state may be characterized in terms of representational priorities and
AACog (e.g., inferential) possibilities. At the brain level, each state may be defined
by the critical dynamics of the network synchronizations that are possible (Chalvo,
2014). It is notable that adverse life conditions directly influence the development
of the brain, which is then expressed into various mental deficiencies. Fortunately
some of these deficiencies are reversible when the life conditions of children
improve.
Transitions across cycles occur when already established networks (and represent-
ations) are embedded into more complex networks. In the brain, long-distance
connections between regions allow the transcription of current networks into
higher level networks that may express new relations in the input. Network
expansion may be indexed by several psychological markers. Speed is a powerful
marker of the initial phases of network expansion. For instance, when sensori-
motor networks are hooked onto a parietal association hub or when, later, this
integrated network is hooked onto a pre-frontal hub, time is needed to practise and
consolidate the new network. Thus, at an initial phase in the functioning of the
network, increases in processing speed would reflect changes in the flow of its
activation until the core of the network is consolidated. After a certain point in time
the network expands to include already available instances of the lower-level
networks. At this phase, working memory capacity would be a better psychological
marker of network expansion, because it reflects its horizontal expansion. However,
increases in working memory capacity reflect rather than cause increases in the
complexity of networks. The findings of Wendelken et al. (2015) summarized
above indicated that speed was a predictor of change in reasoning at the phase
(6–8 years) at which dlPFC was dissociated from the vlPFC as a basis of reasoning.
In the next phase, when the left and right dlPFC were connected, working memory
dominated as a predictor of reasoning change. This pattern implies that speed
suggests brain rewiring associated with a change in the kind of reasoning possible
and that working memory suggests changes in the enhancement of the networks
associated with the consolidation of the new kind.
It is notable that the fundamental components of the fronto-parietal executive
control network are in place from very early in life and they systematically expand
in the fashion of the four executive control programmes outlined in the first section
of the chapter. Attention is drawn to a broad similarity in the development of core
executive control programmes across cycles and related brain networks. It was noted
above that the core executive programme of each next cycle integrates and extends
the core programme of the previous cycle in representational scope, procedural
flexibility, and cognizance resolution. In the same fashion, underlying brain
networks of each cycle integrate and extend the brain network of the previous cycle
by laying extra paths between local networks and first- and second-order abstraction
networks. Thus further representational and inferential possibilities come within
reach because further brain lines are there to support them. It is similar to when a
telecommunications network is expanded by the addition of new lines or the
upgrading of the carrying capacity of the existing network. More clients can use
the network at the same time or each client can transfer more information.
This is a skeletal network that expands by addition of new networks or previously
unconnected regions. We view this network as a scaffold for the development
of more specialized networks, such as the various logical schemes of deductive
reasoning, problem-solving strategies in mathematics, moral principles in the social
domain, etc. That is, content-rich networks are built around the executive scaffold
of each phase via a process of translation of the scaffold network into the domain-
specific relations and constraints. Building these networks requires the activation of
specific circuitry that would carry on the representation of the specific information
and relations involved. Thus speed and working memory are good indices of both
developmental and individual differences because they reflect the functional state of
underlying skeletal networks.
We suggested that cognizance rather than speed or working memory is the causal
factor for transitions in intellectual development, because it takes mental activity as
input to further mental activity. The view of consciousness as a recursive system of
interactions between a central executive-selection network and other brain systems
may be the system generating new mental content through its continuous rewiring.
In fact a developmental version of the global workspace model would capture how
insight is built in each developmental cycle, opening the way to the next cycle.
Specifically, the global workspace model implies that, at any time in development,
the content, resolution, and precision of awareness and cognizance depend on the
state, differentiation, and synchronization of the globally synchronized networks.
Thus in each developmental phase the awareness possible is commensurate with
the network available. We showed that in each next phase increasingly more
local networks are hooked onto the global workspace network and more long-
distance connections are added. This addition goes hand in hand with awareness in
each cycle. That is, the resolution and precision of awareness in each cycle reflects
the differentiation and tuning of the brain networks involved in the consciousness
oscillation cycle that produces mental units to be availed for attention and executive
control.
Obviously there are many unanswered questions and problems to solve. For
example, there is no cognitive function whose corresponding brain structures and
networks are fully known. Moreover we still do not know what is truly general
and what is truly specific in both the brain and the mind. Specifically, how much
of each general mental function, such as speed, control, or representational capacity
is associated with general brain qualities (i.e., sheer total brain or cortical volume,
the overall physical state of neurons and neurotransmitters, connectivity, etc.) and
how much is accounted for by the fact that particular brain systems (such as the
attention or control networks) are always engaged in cognitive processing? Also, we
still do not know how each of the various networks carries on its own job (e.g., in
terms of rhythms), how the networks interact with each other (e.g., by direct
structural connections or by functional coordination), and how they are integrated
into a final solution behaviourally and subjectively. And, of course, we know very
little about how the various types of change in the brain (e.g., myelination, electrical
activity, volume, dispersion, activity of neurotransmitters, connectivity, etc.) interact
with cognitive developmental changes. Therefore the grand neuro-cognitive develop-
mental theory of intelligence that would integrate brain with functional and
subjective maps of mental functions into a common landscape is still far ahead of us.
PART III
A developmental theory
of instruction
15
SCHOOL AND INTELLECTUAL
DEVELOPMENT
Some animal species do provide some kind of training to their offspring that is
directly related to their survival and fitness in the environment (Caro & Hauser,
1992; Thornton & Raihani, 2008). However, humans are the only species to have
developed an elaborate system of education aiming to induce children into the
world of adults. All human cultures have educational systems, varying from plain
apprenticeships focusing on specific activities to the elaborate and very long
education of the modern world. Obviously education and culture are possible
because of the human mind. Social institutions, such as customs and traditions in
different civilizations, religion, collective life, occupations and professions, common
knowledge and science, are all products of the human mind. Some of the skills
and knowledge in each of these realms are acquired by individuals as they grow in
their social group; many others, however, such as reading and writing, mathematics,
professional and scientific skills and knowledge, must be learned, often over decades.
In a sense, education is a process of tuning individual minds with the collective mind
of their culture and time. For this reason many scholars suggested that this tuning
shapes individual minds on the pattern of their culture. However, individuals differ
in what they learn, how well they learn what they are taught, and how they use
it in their life. This is a function of their mental possibilities, their developmental
phase, and their potential for further development. Therefore theories of intelligence
and intellectual development are crucial for education.
Naturally, different theories about intelligence and mental development lead to
different implications for education. For some scholars, intelligence reflects con-
straints rather than possibilities. Proponents of the classic psychometric theory of
intelligence who consider IQ as a stable trait argue that learning cannot go further
than the possibilities associated with a given level of IQ (Jensen, 1998; Herrnstein &
Murray, 1994; Murray, 2009). Interestingly, classic cognitive developmental theory
espouses the same approach. Piaget, for instance, claimed that learning cannot go
220 A developmental theory of instruction
much beyond the child’s current developmental level. Along the same lines, neo-
Piagetian theories claimed that the capacity of working memory defines the upper
limit of the complexity of concepts that can be learned (Case, 1985; Halford et al.,
1998; Pascual-Leone & Goodman, 1979).
This part of the book focuses on learning and education. The present and the
next chapter will tackle the question of learning and examine relations between
various aspects of developing mind and education. In this chapter we will first
summarize research on the relations between intelligence, cognitive development,
and learning in school. Then we will summarize research that examined whether
intelligence can be increased in specifically designed learning environments. In the
third chapter we will outline a programme designed to increase intelligence in
schools.
Relations between developing mind, school performance,

and learning
How intelligence influences schooling

The relation between intelligence and education is bidirectional: intelligence affects
learning in schools and schooling affects intelligence. A large number of studies
substantiate this claim. Specifically, performance on intelligence tests and broadly
used school achievement tests relate highly. For instance, the relation between IQ
and attainment on the Scholastic Assessment Test (SAT) or the General Certificate
of Education (GCE) is about .8. Thus about 65% of variance in scholastic
achievement tests is accounted for by IQ. These tests address reasoning and
knowledge and problem-solving skills in school-related domains, such as reading
and mathematics. Performance on them is crucial for university admission in the
USA and the UK. There are similar tests in many other countries with very similar
relations with performance on IQ tests. The relation between intelligence scores
and school marks is lower but still high (circa .5). Obviously school marks are less
reliable than standardized achievement scores because they are affected by various
factors, such as teachers’ skills in evaluation, school policies, etc.
Overall, the idea is that the higher one’s intelligence the better one performs
at school. Deary, Strand, Smith, and Fernandes (2007) showed that general
intelligence at age 11 strongly predicts school achievement at the age of 16 across
25 school subjects examined (the relation between the two latent factors was .81).
Notably, of those who were at the mean level of g at age 11, 16% obtained
5 GCEs at the age of 16. However, 91% of those who were one standard deviation
above mean g obtained 5 GCEs at this age, although this relation may vary across
school subjects. Specifically, intelligence explained 59% of variance in
mathematics, 48% in English, and 18% in art and design. Rohde and Thomson
(2007) showed that performance on the Raven matrices strongly predicts
performance on the SAT. Verbal and spatial ability, working memory, and
processing speed only weakly contributed to predicting SAT. However, spatial
School and intellectual development 221
ability and processing speed did add to the prediction of SAT maths scores in
addition to general cognitive ability.
Jensen (1998) argued that the relation between test performance and school
achievement is caused by the dependence of both on g. Therefore, measures of
processing efficiency must reflect school performance because they reflect g. In line
with the theory presented in this book, reality is more complex than simple reductionist
models of intelligence. Specifically, Rindermann and Neubauer (2004) showed that
speed of processing is indirectly related to school performance through its effects on
g and creativity. That is, processing speed influences both general intelligence and
creativity. Intelligence and creativity both have an impact on school performance.
Figure 15.1 summarizes these findings. Krumm, Ziegler, Buehner (2008) showed that
reasoning is a good predictor of school achievement in science (mathematics, physics,
biology, and chemistry) and language. The executive and coordination processes of
working memory additionally contribute to performance in science; verbal storage
additionally contributes to achievement in language. Therefore it seems that different
subjects in school require a different combination of domain-specific and domain-
general processes. Schweizer and Koch (2002) suggested that general information
processing capacity as reflected in processing speed and working memory is the
fundamental mechanism that allows the investment (Cattell, 1957, 1963) of fluid
intelligence (reasoning) into crystallized intelligence and school learning.
How schooling influences cognition

The studies above suggested that school achievement does depend on intelligence
and intellectual development. Is the opposite equally true? That is, is there an effect
of school on intelligence and cognitive development? The studies to be summarized
below show that schooling does exert an effect on intelligence and intellectual
development but the size of this effect varies with several conditions.
FIGURE 15.1 Model showing the relations between processing efficiency (speed),
cognitive constructs (intelligence and creativity), and school performance based on
several school subjects (e.g., language, mathematics, science, etc.)
Hundreds of studies have examined the effect of schooling on intelligence.

Drawing on this research, Ceci (1991) concluded that each extra year of schooling
leads to an increase of 1.8 IQ points. Later, Winship and Korenman (1997)
concluded that this effect varies between 1 and 4 IQ points. Gustafsson (2008)
investigated the effect of schooling on different components of intelligence in a very
large sample of 13,903 cases. He found an effect of 2.5 IQ points for each extra year
of schooling. His breakdown of this effect into components suggested that it came
from changes in fluid intelligence (i.e., reasoning and problem-solving processes)
rather than g (where processing efficiency as such is also involved). Interestingly,
this effect came mainly from academic programmes that promote reasoning,
reflection, and consolidation of abstract knowledge, such as learning mathematics
or an ancient language such as Latin or ancient Greek, rather than from vocational
programmes that promote skills in different domains, such as mechanical skills.
Moreover, Gustafsson found that programmes oriented to science and technology
exerted a positive effect on visuo-spatial ability and crystallized intelligence.
Few studies examined the effect of schooling on cognitive development as such.
Kyriakides and Luyten (2009) studied this effect from age 12–18. They used our
comprehensive test of cognitive development (Demetriou & Kyriakides, 2006) that
examines the student’s level of reasoning in all SCSs but social thought. To specify
the effect of schooling they compared same-age students who differed by one school
grade due to the fact that there is a conventional cut-off point in birth date
determining when students enter primary school. They found that the effect of one
extra year of schooling was significant (Cohen’s d = .61) throughout the six
secondary school grades. This difference is equivalent to about one third of a
developmental cycle. The effect of schooling on mathematics (Cohen’s d = .68)
and language (Cohen’s d = .72), expectedly, was even larger. This is an interesting
finding, showing that schooling accelerates cognitive development as such in
addition to increasing crystallized intelligence related to school subjects.
At the high end of the spectrum, Lehman, Lembert, and Nisbett (1988)
investigated the effects of graduate training on various aspects of reasoning.
Specifically, they examined how graduate training in law, medicine, psychology,
and chemistry influences statistical reasoning, methodological reasoning about
confounded variables, and deductive conditional reasoning. They found that
training in psychology and medicine positively affects statistical reasoning about
everyday-life problems, methodological reasoning about everyday-life problems
enabling the handling of confounding principle, and conditional reasoning enabling
the handling of the fallacies. Training in the law affects conditional reasoning.
Training in chemistry does not seem to affect any of these kinds of reasoning.
It is highly interesting that Brod, Bunge, and Shing (2017) found the same effect
of schooling on executive control. These scholars employed the same method of
comparing same-age students who either entered first grade of primary school
because they were born before a specific cut-off date or remained at kindergarten
because they were born after this cut-off date. The mean age of both groups was
5.4 years. Children were tested on several tasks addressed to executive control,
which included attention control, inhibition of response, and mental flexibility.

Children were also brain scanned to examine if possible performance differences are
also expressed at the level of the brain. It is stressed that children in primary school
performed better on the tasks addressed to executive control. Moreover, these
children exhibited greater increase in activation of the right posterior parietal cortex;
it is reminded that increased activation in this region is related to improvements in
accuracy of responses to stimuli. Obviously the need to pay attention to teachers,
sit still in a classroom, and organize one’s own actions for the sake of meeting school
tasks, such as reading and writing, has a direct effect on executive control which is
directly projected on the relevant regions of the brain. These findings are highly
interesting in their implication that a central influence of schooling underlying the
relative advantage it offers in general intelligence and intellectual development is
due to the pressures it exerts on students to organize information and know and
manage themselves better. Thus it directly affects all three AACog processes.
A half glass may be seen as either half empty or half full. According to Brody
(2008), the size of the effects above suggests that the influence of schooling on
intelligence is limited and accounts for no more than about 6% of individual
differences. This is small when compared to other factors: for instance, parental
differences in IQ at birth—reflecting genetic differences—account for about 20%
of adult intelligence; and information-processing differences in early infancy account
for about 50% of differences in adult intelligence. However small it is, though, this
influence does affect fundamental inferential and abstraction mechanisms in g,
thereby accelerating the rate of cognitive development (Gustafsson, 2008; Kyriakides
and Luyten, 2009). It is not just an improvement in test-taking skills, as some
scholars suggested (Ceci, 1991). When taken year by year, an increase of about
2 degrees for each school year may look small; it is very large, however, if taken
cumulatively. Obviously this influence of schooling is related to the Flynn (2009)
effect already discussed in Chapter 13. It is reminded that there has been an increase
of 18 IQ points in fluid intelligence in the West over the last century.
Let us take two imaginary 4-year-old children to make the argument clear. Let
us suppose that they both have an IQ of 100 and they have both reached the second
phase of realistic mental representations. One of them never goes to school and the
other goes through all levels until college; i.e., she spends 16 years at school. Based
on the literature above, as a young adult the first imaginary individual would still
have an IQ of 100 and he will stop at the second phase of rule-based concepts. The
second imaginary individual would have an IQ of 132 and she will be operating at
the end of the principle-based cycle. This is a huge difference with immense
consequences for life opportunities.
Developmental relations between school performance and

cognitive performance
The research reviewed above may yield a wrong impression: that prediction of
performance in school may be equally valid or accurate regardless of grade. This is
not the case. We conducted several studies to examine how school performance
relates to the various mental processes discussed in the previous chapters. Actually,
in some of the studies already discussed in previous chapters we asked the teachers
of our participants to rate their students in various aspects of learning in three subjects:
mathematics, science, and Greek. Specifically, we asked the teachers to rate each of
their students on their “actual performance”, “possibility to learn complex concepts”,
and “how fast they learn” in the subjects. Rating was done along a 7-point scale.
In one of these studies, primary school children from first to sixth grade (i.e.,
from about 7 to 12 years of age, N=140) were examined by several measures of
processing and representational efficiency and reasoning. Processing efficiency was
addressed by processing speed, attention control and inhibition, and working
memory. The reasoning test addressed development of inductive and deductive
reasoning across rule-based and principle-based thought. The items in each type
of reasoning also involved verbal (e.g., verbal analogies in inductive reasoning or
verbal statements in deductive reasoning), quantitative (e.g., numerical analogies
or deductive arguments involving numbers), and spatial context (e.g., Raven-like
matrices for inductive reasoning and spatial relations in deductive reasoning
arguments). We have gone to some length in presenting this reasoning test because
we obtained some unexpected results: not all aspects of reasoning are predictive of
school performance, especially at the first grades of primary school.
Specifically, in a first model we tested the mediating model shown in Figure 15.1
above. In this model we created a latent factor standing for reasoning. This factor
was defined by mean performance on all verbal, mathematical, and spatial tasks. In
this model, attention control was regressed on age (–.79), working memory was
regressed on attention control (.84), reasoning was regressed on working memory
(.88), and each of the three school subjects—mathematics, science, and Greek—
were regressed on reasoning. Notably, all relations between mental processes (shown
in parenthesis above) were very high, as expected. However, the relations between
each of the school subjects and reasoning were strikingly low: all smaller than .07.
Inspection of relations between school performance and particular groups of
reasoning tasks revealed that the size of relations varied with tasks and grade. On
the one hand, reasoning in the verbal domain, both inductive and deductive, was
significantly related to all school subjects in the three lower school grades. On the
other hand, reasoning in the arithmetic domain, both inductive and deductive, was
related to all three school subjects in the three higher school grades. To capture
these trends, we redesigned the model above so as to include two latent reasoning
factors, one for verbal and one for arithmetic reasoning. This model was run in a
2-group analysis: the first group involved the three lower grades and the second
group involved the three higher grades. For the sake of parsimony, we created one
school performance factor associated to mean performance on each of the three
school subjects. The relations between mental processes were held as above. Based
on trends noted above, the school performance factor was regressed on the verbal
reasoning factor in the group of younger children (grades 1–3) and the arithmetic
reasoning factor in the group of older children (grades 4–6). Indeed, both relations
held well: .53 and .41, respectively. In fact, when this model was tested on the
whole sample, including all six grades in a single group, the general school
performance factor was found to relate highly not only to the arithmetic reasoning
factor (.59) but also to attention control (.67).
These findings suggest an interesting and rather novel conclusion in regard to
the prediction of school achievement from performance on cognitive tests. First,
early in school life predictability of school achievement based on specific cognitive
processes is quite unstable and unreliable. This is due to the fact that school
performance early in school depends on many factors that are not related to mental
ability: these include motivation, adjustment to the school environment, and the
acquisition of skills that require extended practice on top of wit, such as reading and
arithmetic skills. Thus, in the early school years, reasoning in the verbal domain
proved to be a good predictor because this requires reasoning in a medium that is
well under the control of the child, thereby reflecting his or her early ability that
is invested in school learning. In the later years, reasoning in the context of arithmetic
takes priority because this reflects an interaction between individual ability with
demanding learning at school.
A second study involved children and adolescents aged 10–18 (N=131). These
participants were examined by a set of processing efficiency measures (speed of
processing and attention control), our cognitive development test involving all
SCSs but social thought, the self-concept in regard to these five domains, and the
WISC-R. We also used the teachers’ evaluations of school performance on the three
school subjects mentioned before: that is, science, mathematics, and Greek. We found
that performance on the SCSs and IQ almost equally but distinctly accounted for
school performance (.55 and .53, respectively). These factors mediated between school
performance and processing efficiency (–.56) and working memory (.61). Interestingly,
cognitive self-concept was significantly influenced by school performance (.38), which
mediated the influence of actual performance on the SCSs.
Two conclusions are suggested by this study. On the one hand, reasoning in the
SCS is not identical with what is represented by performance on the WISC test of
intelligence. Obviously knowledge as reflected in crystallized intelligence is a
stronger factor in the WISC than in the SCS test: the WISC test involves many tasks
requiring knowledge, such as the information and vocabulary tests. Our test captures
developing reasoning and problem-solving mechanisms. Therefore Carroll (1993)
rightly distinguished in his 3-stratum reasoning represented by developmental tasks
(in the spirit of those days he called it Piagetian reasoning) and reasoning represented
by fluid intelligence tasks included in intelligence tests. These mostly gear on
inductive reasoning. Notably, the mental abilities represented by these two different
tests are equally strong predictors of school performance. In turn, performance at
school directly contributes to the formation of a student’s self-representation about
cognitive domains. In fact, school performance even mediates between the self-
concept and actual attainment in the domains.
A third study involved many participants (N=851) from primary school grade 5
(age 11) through senior high school grade 3 (age 18). These participants were
examined by two tests: our full SCS test (spatial, quantitative, causal, qualitative, and
inductive and deductive reasoning) and our full cognitive self-concept test. Teachers
were also asked to rate the school performance of each student in mathematics,
science, and Greek, according to the scale already described. The large sample size
allows an examination of relations in three age groups: primary school (from 11–12
years, primary grades 5 and 6), junior high school (grades 1–3, age 12–14 years), and
senior high school (grades 1–3, age 15–18 years). A rather complicated model was
tested here. Specifically, there was a first-order factor for each of the SCSs. These
factors were regressed on a second-order factor which stands for general cognitive
performance. In the spirit of the distinction proposed above between general
developmental intelligence, this factor includes fluid intelligence as traditionally
defined but also includes extra reasoning and problem-solving processes, such as
command of deductive reasoning and handling of perceptual deception. There was
also a first-order factor for each domain of cognitive self-concept that corresponds
to each SCS. Additionally, there were self-concept factors for working memory and
self-regulation. These factors were regressed on a second-order factor which stands
for general cognitive self-concept. Finally, there were three school performance
factors standing for performance on each school subject which were regressed on a
general school performance second-order factor. Both the general self-concept and
the general school performance factors were regressed on the fluid intelligence
factor. The school performance factor was regressed on the general self-concept
factor. It is notable also that inspection of relations in the model suggested that self-
concept for verbal ability was additionally related to school performance. Thus this
factor was added to the model.
The pattern of relations was very interesting in showing how the relations
between these processes vary with development. Specifically, the relation between
self-concept and cognitive ability was very weak at primary school (.03), substantial
at junior high school (.36) and low but significant (.16) in senior high school. The
relation between school performance and cognitive ability was always very high (.66,
.80, and .57 in the three educational levels, respectively). Interestingly, the relation
between general self-concept and school performance was significant, albeit low,
only in junior high school (.08, .17, and –.08, for the three levels, respectively).
However, the impact of the verbal ability self-concept was always moderate and
significant (.28, .31, and .26, respectively). All in all, cognitive ability highly predicts
school performance throughout this long age period. The predictive possibility of
cognitive self-concept is always much weaker, although it becomes significant in
adolescence. We discuss the possible implications of these differences after we specify
the role of personality in the prediction of school performance in the section below.
Personality, intelligence, and academic performance

Academic performance is an appropriate field to dissociate the relative influence of
intelligence and personality in real-life outcomes. Intelligence defines the readiness
and facility to learn; personality defines the disposition to use the available ability as
efficiently as possible for the sake of actual learning. Limited ability constrains the
range and complexity of what can be learned, whatever goodwill and motivation
one may have for learning. Limited self-discipline and persistence as reflected in
limited conscientiousness might limit the range and complexity of what can be
learned, however high one’s intelligence is. It is trivial to note that there are concepts
and skills that require hard work to be mastered, however smart one is. This is
especially the case for school learning. Concepts and skills at school evolve over
many years and mastering them requires sustained engagement and effort.
This is reflected in the relative contribution of cognition and some but not all of
the dimensions of personality in academic performance. There is a vast literature
showing that conscientiousness is related to many positive life outcomes, including
academic and job performance (see Hill & Jackson, 2016). In our research we
examined how each of the constructs involved in the four-fold model, the Big Five
factors, and emotional intelligence are distinctly related to academic performance.
Our findings clarify the picture presented above by placing cognitive effects in the
perspective of personality: g accounts for between one fourth and one third of
the variance in each of three important school subjects already discussed above: 34%
for Greek, 31% for science, and 24% mathematics. However, the α-factor of
personality, stability, accounts for much more: 65% for Greek, 68% for science,
and 37% for mathematics. Most of this effect came from conscientiousness: 27%
for Greek, 68% for science, and 25% for mathematics. Agreeableness appeared to
have a small but significant effect on science (–.11, 1%) and Greek (–.13, 2%);
however, this effect was negative. Interestingly, the effect of the general cognitive
self-representation factor was nil (see Figure 12.1 in Chapter 12).
Conclusions
The research summarized in this chapter leads to several clear conclusions about
relations between the mental processes, personality dispositions, and learning and
achievement at school.
First, cognitive ability and school performance are highly related. The relation is
bi-directional. Schooling influences cognitive ability and development substantially.
Each extra year of schooling boosts intelligence by about 2 IQ points, or a good
part of a developmental cycle. If one takes into account that in the West compulsory
education varies between 9 and 12 years, this effect may fully explain the Flynn
phenomenon of significant secular increases of general intelligence.
Second, developing cognitive ability captured by tests addressing reasoning and
problem-solving processes acquired at each of the four developmental cycles and
cognitive ability addressed by intelligence tests are complementary to, rather than
substitutes of, each other as predictors of school performance. Therefore an accurate
prediction of learning possibilities at successive developmental phases would have
to include both types of tests.
Third, the relation of cognitive ability with school achievement varies with age,
being higher in secondary than in primary school. In fact, different mental processes
may be better predictors of school performance in different school levels, depending

on their command by the student in each level.
Fourth, cognitive self-concept is tuned with actual ability in the early secondary
school years rather than earlier. In fact, it seems that school performance affects
more the formation of mental self-concept than it is affected by it. This tuning may
be explained by two factors. On the one hand, the emergence of principle-based
thought and ensuing accuracy in self-monitoring and self-evaluation renders young
adolescents more astute observers of their own mental activity. On the other hand,
the pressure exerted by increased learning needs coming with transition from
primary to secondary school feeds self-monitoring and self-evaluation with more
differentiated information about mental functioning and related successes and
failures than before. It is highly interesting in this regard that the self-concept about
verbal ability is always related to school performance. This suggests that self-
monitoring of verbal performance is more accurate than self-monitoring of other
cognitive abilities.
Fourth, personality does have a very strong influence on school performance that
is additional to the influence of cognitive ability. When personality is integrated into
the picture, cognitive self-representations pale as predictors. In other words, it
matters very little if you think you are smart but you are not. What matters is only
actual cognitive ability and the self-discipline and determination needed to invest
this ability into learning. Conscientiousness is unique among the Big Five factors
because it drives individuals to continuously invest their ability in actual life
achievements which accrue with time, thereby increasing their return to the
individual (Hill & Jackson, 2016). When these two factors work together, being
extravert or introvert, emotionally stable and calm or anxious and stressful, or
agreeable to others makes little difference, if at all. In fact, highly agreeable persons
should know that being too much oriented to helping or pleasing others may have
a negative effect on their own endeavours. Also, smart people must be aware that
they sometimes overestimate the possible effect of their high ability, because this is
not enough for many demanding long-term life tasks; conscientious individuals of
average cognitive ability must be aware that their determination and planfulness may
increase their achievements in real-life tasks but do not increase their intellectual
competence as applied to new complex tasks on the spot.
16
ENHANCING INTELLIGENCE
IN THE LABORATORY
In the context of the psychology of individual differences there is a long tradition

of research aiming to increase intelligence (Kyllonen & Kell, 2017; Protzko, 2015).
In developmental psychology there has been extensive research aiming to accelerate
cognitive development by raising children’s thought to higher stages of cognitive
development (Brainerd, 1973; Efklides, Demetriou, & Gustafsson, 1992; Shayer &
Adey, 2002). Some scholars combined the psychometric with the developmental
approach to enhance the likelihood of success (Klauer & Phye, 1994, 2008). Both
lines of research showed that increasing intelligence or elevating the developmental
level is possible. However, both lines hit deadlock and are not used in their intended
domain, education, for simple reasons. The research aimed at increasing intelligence
stumbled over the fade-out effect; that is, gains in IQ drop drastically, from about
7–8 points to about 1–2 points, in the three years after the end of intervention
(Protzko, 2015). In the research addressing acceleration of cognitive development,
the developmental boundary proved to be a barrier; improvement of thought does
occur but this is limited within the current developmental phase. That is, training
helps to consolidate the current level of thought rather than to cause transition to
the next level.
This chapter will answer three questions:
1. Can we train each of the processes involved in g? How much can each of these
processes change because of training?
2. How stable are the results of training over time?
3. Does training in one process transfer to other processes?
The ultimate aim in answering these questions is to open the road for more successful
learning interventions in education.
Enhancing intelligence
Early research aiming to train intelligence focused on the relational and inferential
processes directly tested by intelligence tests. The Head Start Program conducted in
the USA is one major programme which aimed to improve the learning skills, the
social skills, and the health conditions of poor children (Currie & Duncan, 1995;
Neisser, 1998). Another example, also conducted in the USA, is the Abecedarian
study which focused on children at risk since infancy. Training in this programme
involved skills related to school learning. Specifically, interventions were designed to
promote emergent maths interest and activities and measuring, counting, doing
arithmetic operations, etc. For language, children were exposed to new and rich
vocabulary, elaborated syntax, decontextualized language about not-present events
and objects, etc., (Campbell & Burchinal, 2008) In recent years, training research
followed the reductionist path: that is, it examined if training simple processes such
as processing speed, attention control, shifting, and working memory would
generalize to fluid processes (e.g., Jaeggi et al., 2008; Protzko, 2015). There is a
practical concern here. Training simple processing and representational efficiency
processes is easier. Thus, if there is transfer, focusing on these simpler processes would
be much more time- and cost-efficient than focusing on inferential processes which
are more cumbersome to train.
It is worth noting that examining the direction of transfer across processes is a hard
test for the theories discussed in the chapters above. Normally the direction of transfer
would have to follow the direction of causality assumed by our models. For instance,
transfer of training bottom-up from a specific process (e.g., attention control or
working memory) to g would indicate that this specific process is an integral component
of g so that any change in it automatically alters g as well. Lack of bottom-up transfer
would imply that the process involved may not be sufficient for the operation of the
higher-level g processes per se, even if part of them. A top-down transfer of gains from
a general process associated with g, such as relational thought, to a more basic process,
such as flexibility or working memory, would indicate that this general process is
spontaneously involved in activating or using the specific process. Lack of top-down
transfer would imply that the specific process is not activated when the general process
is activated. Transfer of gains from a particular process both ways—bottom-up to a
process known to be part of g, such as inductive reasoning, or top-down to a more
basic process, such as any executive control process—would imply that this specific
process is a regulator or mediator that may be called on by both specific or general
processes to help tune them to the specificities of the target at hand.
Protzko (2016) noted recently that only top-down effects are allowed by the
3-stratum Cattell-Horn-Carroll model. Inspection of Figure 2.1 in Chapter 2 shows
that arrows go from g to second-stratum broad abilities and from these abilities to
first-stratum specific abilities. Interpreting the direction of arrows literally, according
to the rationale underlying the construction of structural equation models, Protzko
pointed out that only top-down causal effects are allowed by this model. That is,
changing g might change any of the broad abilities and changing any of the broad
Enhancing intelligence in the laboratory 231
abilities might change any of the specific processes associated with this particular
broad ability. However, bottom-up effects are not allowed because they are not
assumed by the causal model. Therefore, if this model has any value, training would
transfer top-down according to the causal paths affected but it would not transfer
bottom-up because no causal relations are assumed in this direction.
Our theory, as encapsulated in the four-fold model (see Figure 8.1 in Chapter
8) suggests a simple rule for predicting transfer: follow the path of relations between
the processes involved. Various types of effects of varying magnitude are possible
under certain conditions:
1. First, transfer between processes that belong to the same type of system is more
likely because they share common components. For instance, transfer from
attention control to working memory or vice-versa is likely because they both
share focusing of attention and interference control.
2. Second, some processes are more central than others, cognizance and inference
par excellence: cognizance is a mediator that translates and transcribes processes
from specific domains to general instructions and inference is needed for the
implementation of domain-specific skills and operations in any domain, such as
the SCS. Thus transfer from these processes both bottom-up and top-down is
more probable than from any other process.
3. Third, the models shown in Chapter 8 (see Figures 8.2, 8.5, and 8.6) suggest
that, even if bottom-up effects are present, these effects are relatively limited (on
average, circa 20% of variance for each of attention control, shifting, and working
memory) and they vary with development. Actually some of these effects
decrease up to non-existence after a certain age. For instance, attention control
and shifting after the age of about 13 have minimal influence on cognitive
change; some recycle consistently, such as working memory, and some others
increase systematically, such as cognizance. Therefore if training these processes
is to transfer to inferential processes involved in g, it must be properly delivered
at the right age and also properly measured. Moreover, this effect may be
cumulative so that it must exceed a certain threshold to change g. For instance, how
much training of attention control and working memory is required at the ages
of 9–11 or the ages of 11–13 so that it transfers to g at these age phases? How
much of this training would have to be directly associated with cognizance of
the processes involved? Obviously no satisfactory answer to this exists.
4. Finally, a special note about the SCSs is required, because these are the interface
with the real world. No transfer of training would be expected across SCSs if the
mediating processes were not involved. This is because SCSs are special systems for
the processing of special types of relations.Thus acquiring the special operations
of one SCS (e.g., mental rotation in the spatial SCS) does not necessarily provide
any advantage for another SCS (e.g., proportional reasoning or isolation of
variables).This may happen only to the extent that learning the special operation
in an SCS activates the central systems. But then prediction must be based on
the central systems, not the SCS involved.
Research in the experimental and psychometric tradition

There has been extensive research examining transfer effects along all directions.
The pattern is clear by now: in regard to the first question above about trainability,
all processes can be trained. Speed in various contexts (Mackey et al., 2010), attention
control in the Stroop task (MacLeod & Dunbar, 1988), shifting in working memory
(Chein & Morrison, 2010), and reasoning in various domains (Christoforides et al.,
2016; Papageorgiou et al., 2016) do improve as a result of training at any age. In
fact even video games have this effect. A recent study which reviewed a large
number of experiments showed that video games improve reaction time, task
shifting, target and change search, time estimation, sensorimotor coordination,
timely decision making, and resistance to illusions (Boot, Blakely, & Simons, 2011).
In regard to the second question about stability of training, it is safe to conclude that
gains do weaken after a certain time but some of them remain at the end.
In regard to the third question, the transfer of training does follow the paths
specified by our model. That is, training may generalize across processes within the same
group of the four-fold model but it does not generalize to processes associated with psychometric g,
including performance on the Raven test, the WISC test, deductive reasoning, verbal,
arithmetic, or non-verbal (Boot, Blakeley, & Simons, 2011). Several review studies,
based on a large number of empirical studies, concluded recently that gains in attention
and working memory do not transfer to intelligence, real-world cognitive skills, or
academic achievement (Melby-Lervåg et al., 2016; Sala & Gobert, 2017; Shipstead
et al., 2012). Transfer, when found (see Au et al., 2015a, 2015b), comes from the fact
that attention and working memory training also included relational processes, such as
abstraction and awareness. When there is transfer across hierarchical levels, top-down
effects are more likely than bottom-up effects. For instance, Motes et al. (2014) trained
children to use comprehension and inductive reasoning processes to facilitate inferring
the essential gist or abstracted meanings from materials. Gains in these processes did
transfer top-down to inhibitory control in Go/No-Go tasks, enabling the refinement
and recruitment of domain-general inhibitory control processes.
Research accelerating cognitive development

There has been extensive research aiming to boost or accelerate cognitive
development. This research was based on classical Piagetian theory (Brainerd, 1972;
Csapó, 1992; Shayer & Adey, 2002; Shayer & Adhami, 2007; Strauss, 1972) and
neo-Piagetian theories (Efklides, Demetriou, & Gustafsson, 1992; Halford &
Boulton-Lewis, 1992). Overall, this research attempted to teach Piagetian-type
mental concepts, such as the various conservations (i.e., substance, number, weight,
and volume) or reasoning schemes, such as analogical reasoning or hypothesis
testing. Training aimed to enable children to induce specific types of mental
operations, such as reversibility in conservation or isolation of variables in hypothesis
testing. Cognitive conflict between one’s own expectations (e.g., quantity is more
when longer) and reality (e.g., quantity remains the same unless something is added
or taken away), and guided reflection on the differences, was the primary focus of
training. This method was based on the Piagetian assumption that cognitive change
results from attainment of more integrated and reversible (flexible) mental structures.
Overall this research showed that change is possible but limited. It accelerates
cognitive development by one sub-level within a major stage (Brainerd, 1972;
Strauss, 1972) and it does not generalize across domains.
Testing for transfer within and across SCSs

Along these lines, one of our earlier studies (Efklides et al., 1992) examined if training
quantitative and causal thought along the levels of principle-based thought is possible
and if transfer across these SCSs may be achieved. Additionally we examined if
training is related to fluid intelligence. To train quantitative thought, participants
learned to solve mathematical analogies at one of four levels of complexity, spanning
early and late principle-based thought. Specifically, they learned to estimate relations
in 2 x 2 (level I), 2 x 2 x 2 (level II), 2 x 2 x 2 x 2 (level III), and 2 x 2 x 2 x 2 x 2
tables (level IV). For instance, at level I, how is the productivity of two plants A and
B (3kg versus 6kg/hectare) affected by watering (2 versus 4 times/month)?; at level
II, two such tables are presented side by side showing the effect of watering on the
two plants in two areas, 1 and 2. To train causal-experimental thought, participants
learned to design experiments in order to test hypotheses. The tasks were similar in
content to the quantitative tasks and structurally equivalent in complexity. That is,
participants were asked to design an experiment to test how watering influences the
productivity plans. The aim was to test how participants may implement isolation of
variables at increasing levels of complexity. The four experiments required the
isolation of variables in a 2 x 2 (level I), 2 x 2 x 2 (level II), 2 x 2 x 2 x 2 (level III),
and 2 x 2 x 2 x 2 x 2 experiment (level IV). For instance, at level I, participants were
asked to test the hypothesis that “the increase in watering frequency increases the
productivity of plants”; use plants A and B, which you may water 2 or 4 times per
month. At level II, the hypothesis became specific to each plant: “watering increases
the productivity of plant A but it does not affect the productivity of plant B”. All
levels address principle-based thought: levels I and II address early principle-based
thought processes; levels III and IV address late principle-based thought.
The study involved 10-, 12-, 14-, and 16-year-old participants, divided between
a control group, an experimental group trained on the quantitative SCS and an
experimental group trained on the causal SCS. Participants were tested on tasks
addressing these levels both before and after training. They were also tested with
the Kid of Factor-Referenced Test which involves various measures of inductive
reasoning (Ektstrom, French, & Harman, 1976). Training was individually specific.
That is, each participant was trained at one level higher than his or her pre-test level.
Training involved solving tasks at the level concerned with specific direction and
it was quite short, lasting for one hour. Thus this study was designed to test if
training a specific operation (estimating proportional relations and isolation of
variables) can cause improvement along a developmental sequence and can transfer
across another SCS.
The results were clear. First, both SCSs may be trained: 50% of those trained on causal
thought and 70% of those trained on quantitative thought improved by at least one
level. However, second, training was constrained by initial level: it was easier to move
children from level 1 to level 2 or from level 3 to level 4 than from level 2 to level 3.
This indicates that change within a phase is easier than shifting to a higher phase.
Additionally, third, fluid intelligence was related to change because of training. The higher
one’s performance on the fluid intelligence test the more likely it proved to be that one
would positively respond to training. Finally, fourth, there was no transfer across SCSs.
Bridging Piaget with Vygotsky

Shayer & Adey (1993) developed a complete intervention programme aiming to
accelerate cognitive development along Piagetian stages with a focus on science and
mathematics concepts. These were related to the Piagetian formal operational
schemata, such as combinatorial thought, proportionality, hypothetico-deductive
reasoning, and isolation of variables. In Shayer’s words, this programme is conceived
as teaching art integrating Piaget, Vygotsky, and Feuerstein. Piaget offered the
concepts from science and mathematics, such as isolation of variables and causal
interpretation, volume, heaviness, etc., and the developmental hierarchies.
Vygotsky offered two important notions. The first is the notion of the zone of
proximal development which points to the fact that mastering concepts moves
through a zone of competence such that initially one grasps the bare bones of
concepts and it then builds them up, grasping various nuances and skills and allowing
their full use and implementation. The second is social interaction with peers and
adults. Vygotsky suggests that negotiating one’s own understanding with others,
peers and teachers, helps one move through the zone of proximal development and
reach the higher levels of a concept. Mechanisms such as internal speech help one
reflect and interiorize the others’ perspective, thereby enhancing one’s understanding.
Feuerstein offers the recognition of the fact that classrooms involve students of
varying levels and this is a reality to capitalize on for the benefit of all.
The intervention programme evolves along four acts, as illustrated in Figure 16.1.
In Act 1 (8–12 minutes) students are introduced to the subject of investigation and
learn the technical vocabulary needed. It is ensured that at least 80% of the class can
enter into a dialogue about the subject of learning. In Act 2 (10–15 minutes) working
groups work at the task within their group ZPDs. The aim is to prepare for Act 3. In
Act 3 (typically 15 minutes for CASE) groups report their ideas to the rest of the class.
The aim is to lead students to witness and internalize the concepts taught. In Act 3
students also reflect metacognitively on what has been achieved. In Act 2 the teacher
scans the class and intervenes to enhance group energy and give questions that would
induce cognitive conflict when a group operates at a level of processing. Explicit for
the teacher and implicit for the students is that the peer-peer mediation occurring in
Act 2 and the exchange of ideas in Act 3 are the major drivers of intellectual
development. It is the teacher’s job to manage the lesson so that peer-peer mediation
is maximized, a very different skill from ordinary instructional teaching. There are
FIGURE 16.1 Stages in cognitive acceleration as implemented by Shayer (2003)
two driving questions: “How do I know what I think until I hear what I say?” and
“How do I know what I think until I hear also what the others say?” In Act 4,
achievements of Acts 2 and 3 are extended in other areas of science and mathematics.
This programme formed the basis of a programme of professional development
offered by King’s College, London, for teachers and schools. Tens of schools in
various districts in England were involved. Obviously this is a top-down
intervention aimed at developing principle-based thought. It is noted that this

level of thought is fully achieved by only 10% of students aged 16, at present
(Shayer, Ginsburg, & Coe, 2007). The main findings are summarized in Figure 16.2.
FIGURE 16.2 Distribution of cognitive developmental levels in the general population

and cognitive change as a result of intervention (A) and transfer of training effects to
school performance (B)
Note: levels stand for a cognitive developmental scale based on the probability of success on a
large battery involving tasks discriminating at Piagetian early and late preoperational thought, a
sequence of levels in concrete thought and a sequence of levels in formal operational thought.
These levels correspond to the first and second phase of realistic representations, rule-based, and
principle-based thought
It can be seen in panel A how the slope of change becomes steeper as a result of
the intervention, which is marked by a black bar in the figure. It can be seen that
there is improvement because of the intervention in children of all levels. In terms
of Piaget’s stages, on average, trained children moved from late concrete thinking
to early formal thinking. However, on the one hand, only children operating
before training at late concrete thought moved to any level of formal thought and
only those operating at transitional levels between concrete and formal thought
moved to late formal thought. On the other hand, those operating on preoperational
or early concrete thought moved up to late concrete thought but not beyond.
At the end of the second year of intervention (2002) all schools had their national
tests, Key Stage 1. Performance on these tests allowed an insight into the possible
generalization of training effects from experimentally based cognitive processes to
actual school performance. This is made possible by comparing the test performance
of schools involved in the experiment with schools not involved. Panel B of Figure
16.2 shows that there was a clear difference between the CASE experimental schools
and the control schools. The mean effect-size for mathematics was .65 (.30 SD) and
for English it was .43 (.59 SD). Four years later, at the end of primary education, all
schools have the national Key Stage 2 tests. By this time it is likely that the children
will have passed through the hands of four different teachers. The effect-sizes, although
lower, were still significant (circa .24 for maths and circa .35 for English).
Thus it is clear that a large-scale intervention programme addressing complex
inferential processes in different contexts resulted in a rather large progression
along Piagetian levels, shifting from concrete to formal thought. Most importantly,
these gains in thought were expressed in actual school-related subjects, such as
mathematics and English. Comparing this study with our short intervention study,
which showed that transfer across SCSs is not possible if intervention focuses on
a SCS-specific process, suggests an important conclusion. To obtain transfer
learning must be long-lasting, occur in multiple contexts so that underlying rules
and principles may be abstracted across contexts, explicitly encoded, and reflected
on. These findings justify asking the question: what if the mediating processes
themselves are trained? This is the question to be answered by the studies
following.
Research manipulating cognizance
Improving conditional reasoning

It has been argued throughout this book that reasoning is important for intelligent
functioning, because it allows integration and evaluation of information. Conditional
reasoning, an important part of deductive reasoning, is grounded on the four logical
schemes already explicated: modus ponens (MP), modus tollens (MT), affirming
the consequent (AC) and denying the antecedent (DA). We remind that two of the
schemes, MP and MT, are decidable and rather easy to grasp, because all information
needed for a conclusion is present in the premises. The remaining two, AC and DA,
are not decidable because the conclusion depends on information not given in the
premises. These two schemes are called “logical fallacies” because they may prompt
the thinker into drawing a conclusion that is not tenable; the two logical fallacies
deceive the thinker that a conclusion can be drawn because they appear equivalent
to MP and MT, respectively. We also remind that MP and MT are attained early
in development, at 7–9 years, by practically everyone. The two fallacies are not
mastered before the age of 11–12 and then no more than about one third of adults
can handle them systematically (Gauffroy & Barrouillet, 2009; Johnson-Laird &
Wason, 1970; Moshman, 2011; Markovits, 2014; Overton, 1990; Ricco, 2010;
Wason & Evans, 1975).
Obviously events and discourse in everyday life are often patterned according to
them. A failure to recognize and resist them results in misinterpretations and wrong
decisions. For this reason the fallacies have been studied extensively in psychology
and the cognitive sciences in a search for their causes and for training methods that
would enable thinkers to cope with them (Nisbett, Fong, Lehman, & Cheng, 1987;
Ricco, 2010). The studies designed to train conditional reasoning have met with
limited success so far. This study implemented a training programme aiming to
enable children to master the logical fallacies and specify the possible contribution
of various aspects of mental processing, such as attention control and working
memory, and intelligence, such as inductive reasoning and cognitive flexibility.
Below we will first review cognitive, developmental, and learning research on
deductive reasoning and then state the predictions of the present study.
The focus of different studies varied depending on the theory espoused as a basis
of training. The studies assuming that logic is crucial in reasoning trained participants
to master the truth tables associated with each of the various logical schemes. That
is, they trained participants to recognize when a conclusion is true and when it is
false given the premises. However, success was meagre (Staudenmayer & Bourne,
1977; Müller, Overton, & Reene, 2001), suggesting that focusing on the underlying
logical relations is not enough.
Another approach was based on Johnson-Laird’s (1983, 2006) mental models
theory. Instead of teaching the logical rules as such, training here aimed to enable
children to envisage the mental models necessary to represent the relations in each
scheme and reason accordingly. This approach succeeded with participants who
were advanced in their grasp of logical relations and had working memory capacity
high enough to enable them to represent the necessary relations involved in the
critical mental models (Barrouillet, 1997; Simoneau & Markovits, 2003).
The studies assuming that reasoning emerges from pragmatic experiences that
may direct children to intuitively grasp underlying logical relations organized
training to familiarize children with examples and counterexamples related to each
scheme. The assumption was that children would abstract the implications of each
scheme and construct the necessary inferential patterns on their own. This approach
was successful among adolescents who were inferentially advanced enough to use
the examples encountered to flesh out mental schemes they were already using.
However, it was not successful with younger children with limited proficiency in
conditional reasoning (O’Brien & Overton, 1982; Overton, Byrnes, & O’Brien,
1985).
Obviously none of the factors examined by the training studies above (logic,
mental models learning, or pragmatic examples) was sufficient to generate the
change wanted. This study capitalized on the successful aspects of these studies with
an emphasis on the factor that may unite the various types of experiences provided
by each: cognizance. That is, we hoped that building awareness of the four logical
schemes, training them how to relate each rule to a mental model, and how to
transform actual examples into relations and mental models would succeed in
improving logical reasoning in children (Christoforides, Spanoudis, & Demetriou,
2016).
This study focused on two critical phases in the development of conditional
reasoning. That is, the second phase of rule-based inference, from age 8–10, when
the two determinate schemes are mastered, and the first phase of the principle-based
inference, from years 11–13, when fallacies come within reach. Third and sixth
grade primary school children represented these two phases, respectively. Thus we
could test if a relatively short training programme (about a month) simulating the
grasp of awareness and related experience that presumably unfolds spontaneously
over these two phases would be enough to enable children to master the fallacies.
Training aimed to develop an analytic approach to propositions involved in an
argument as contrasted to their everyday use, raise awareness about (i) the chain of
relations between propositions leading to a conclusion, and (ii) the four basic logical
schemes, and provide practice in the construction of mental models following from
each scheme. Training always started with an argument given to the child to solve
and evolved according to the plan related to the session concerned. There were six
sessions, each focusing on a particular aspect of the inferential process and delivered
individually. These sessions are fully described in Table 16.1. It is noted that the
order of presentation implements a systematic progression of training from general
(e.g., everyday versus analytic and formal approach to premises and conclusions) to
more specific themes (e.g., contradiction and logical necessity) and culminating in
the explicit representation of the four logical arguments. Children were introduced
to each session’s target concepts in reference to a specific problem and they were
then asked to solve sample problems, receiving feedback about their answers.
The first session aimed to raise awareness about the analytical approach to
logical arguments as contrasted to their “everyday” use in language. In the second
session children learned to differentiate between the stated and the possibly implied
meaning of propositions and focus on the first. The third session focused on logical
contradiction and truth. The fourth session focused on the notion of contradiction
and consistency by providing practice in the recognition of propositions which are
consistent with a target proposition and propositions which are in conflict with it.
The fifth session focused explicitly on the notions of logical necessity and
sufficiency. The last session focused on the explicit recognition of the four logical
schemes and the construction of alternative mental models implied by each. The
content used in the training arguments was always different from the content used
in pre-test and post-test batteries.
TABLE 16.1 Aims, instructions, and examples of interventions across sessions
Session Aim Instruction to children Examples
First Raise awareness Take the meaning The premises “John is 15 years old”
about the of words in the and “George is 16 years old” allow
analytical arguments as one to conclude that “George is older
approach to given. Accept that than John” but not that “they both go
logical in reasoning the to school”, although this is possible
arguments as conclusion derives and we may know that it is true.
contrasted to from what is stated
their “everyday” in the premises and
use in language. not from what one
Children must knows or other
grasp the possible
concept of connotations.
“conclusion” Discriminate
and discriminate between premises
between and conclusions
necessary vs. and conclusions
likely which although
conclusions. right do not agree
with reality.
Second Differentiate Analyse the Examples include same meaning
between the meaning of propositions differing in prose,
stated and the propositions, propositions contradicting each other,
possibly implied compare them, and and propositions inverting each other.
meaning of specify if meaning For example, the propositions “If you
propositions and and logical do your homework I will buy you an
focus on the first implications agree ice cream” and “If you won’t do your
or disagree. homework I won’t buy you an ice
Elaborate on cream” may sound the same in
alternative everyday language but they are not
interpretations of logically identical.
propositions and
their logical and
semantic
connotations.
Third Differentiate Specify if The propositions “When we irrigate
between logical propositions in plants they grow and become better”
contradiction agreement or in and “John irrigated his plant and it
and truth. The contradiction with dried out” are in contradiction,
aim is to each other and although the first may not and the
discriminate with reality and second may be true.
between logical judge if they are The propositions “When one is 30
contradiction logically consistent. years old one goes to kindergarten”
and agreement and “Andreas is 30 years old and he
with reality. goes to kindergarten” are both false in
reality but they are logically consistent.
Session Aim Instruction to children Examples
Fourth Grasp the notion Focus on target Target proposition: “If you go out,
of contradiction proposition and then you put your sweater on.”
and consistency. evaluate if the Choose which of the following
Several models other propositions propositions are not consistent with it:
are evaluated are in “He went out and he put his sweater
which are contradiction with on.”
derived from an it. “He didn’t go out but he put his
initial premise or sweater on.”
they are in “He didn’t go out and he didn’t put
contradiction his sweater on.”
with it. “He went out but he didn’t put his
sweater on.”
Fifth Grasp the Choose a Target proposition: Figure A is a
notions of proposition which triangle.
logical necessity ensures that the Which of the propositions following
and sufficiency. target proposition proves that the target proposition is
is true. That is, true:
which one is Figure A has a right angle.
necessary and/or Figure A has exactly three sides.
sufficient to Figure A is red.
conclude that the The area of figure A is equal to
target proposition triangle B.
is true.
Sixth Recognition of Focus on the target DA argument: “If John gets 18 at the
the four logical argument and exams he is happy; John did not get
schemes and the imagine other 18; is he happy”—yes, no, can’t say.
construction of arguments that are Children were directed to conceive of
alternative consistent with it. all possible models that are consistent
mental models Choose which of with the proposition “John did not get
implied by each three possible 18” (i.e., any mark that is higher or
of them. propositions can be lower than 18) which makes the
deduced from the argument non-decidable.
premises of the
argument.
Source: Shayer, 2003
In addition to a control group in each age group which did not receive any
training, there were two levels of training: limited (LI) and full instruction (FI)
(the terms “training” and “instruction” are used interchangeably). The LI group
was explicitly induced into the logical structure of each of the four logical schemes
and also into the notions of logical contradiction and consistency. That is, they
received training of sessions 1 and 4: they learned the logical structure of the four
schemes and the notion of logical contradiction. It is assumed, based on the analysis
above, that this is the minimal requirement for grasping the general principle
TABLE 16.2 Pre- and post-test percentage success and effect-sizes for conditional reasoning
study (Source: Yuan et al., 2017)
Age 8 years 11 years

Type LI FI LI FI
Reason AC DA AC DA AC DA AC DA
Pre-test 15 19 15 19 28 34 28 34
Post-test 42.5 48.5 58 57 43 50 71 67
d 0.93 0.89 1.33 1.0 0.99 0.90 1.35 0.96
Note: AC and DA stand for the affirming the consequent and the denying the antecedent fallacy,
respectively. LI and FI stand for limited and full instruction, respectively.
integrating all four logical schemes into a system that specifies the logical
implications of each scheme. The full instruction group, which received all
6 sessions, learned, additionally, to adopt an analytical approach to logical argu-
ments as contrasted to their “everyday” use in language; differentiate between the
stated and the possibly implied meaning of propositions; recognize logical contra-
diction and truth in propositions and reality; and grasp the notions of logical
necessity and sufficiency.
Moreover, to specify how, if at all, learning to reason depends on the various
processing and intelligence processes discussed above, all of these processes (i.e.,
processing efficiency, working memory, inductive reasoning, and cognitive
flexibility) together with the four logical schemes and awareness about them were
examined at pre-test.
The main findings of this study are summarized in Table 16.2. It can be seen
that, in terms of spontaneous developmental time, this short training programme
pulled children up by an almost full developmental phase, enabling both groups to
master the fallacies: overall effect-size for reasoning and awareness was .72 and .36
and .92 and .37 in the limited and the full instruction group, respectively. The key
to this success was awareness of the inferential identity of each scheme and the
principle of logical consistency. It can be seen in Table 16.2 that the limited
instruction group trained in these two aspects of inferential awareness performed
close to the full instruction group, which was trained in all aspects of reasoning. That
is, trained third graders handled problems at the level of principle-based reasoning
if aided by context; sixth graders moved to this level regardless of content and context.
Overall, cognizance almost fully mediated the influence of training on deductive
inference. Cognizance itself also improved but improvement was only limited in
the full instruction group. Whatever was attained, it was highly dependent on
attention control (–.47) and it strengthened further with training (–.62). So, given
the contexts, the 8-year-olds were aware of what they should do in handling fallacies
but, unlike the 11-year-olds, they did not translate that into a conscious abstract
rule, relating the two forms of inference. These rules, which require an explicit
representation of the pairwise relations between the schemes, were mastered by the
11-year-olds, who became able to consider possibilities beforehand. Obviously
these findings provided an experimental demonstration of the mediation models
above that a top-down design involving children thinking about reasoning itself
affects cognizance in such a way that those trained in one fallacy master the other
as well without any specific training. This is fully consistent with the findings
presented in Chapter 8 that commanding the logic underlying any of the fallacies
allows this to spread to others, as well as other kinds of problems. It is also consistent
with the results presented in Chapter 5 about awareness in reasoning: these types of
reasoning are based on awareness.
Enhancing mathematical thought and transferring to g

We conducted several studies to examine if changing intelligence is possible and
what is the crucial mechanism that must be targeted to attain change. One of these
studies examined if training inductive reasoning in mathematics and related
awareness would improve performance in several aspects of mathematics and if
this would generalize to other aspects of intelligence (Papageorgiou, Christou,
Spanoudis, & Demetriou, 2016). This study involved 11-year-old children; by age,
thus, the study involved children at transition between rule-based and principle-
based thought. Half of them were randomly assigned to the training group and the
rest were used as controls. All children were pre-tested on various aspects of
attention control, working memory, and reasoning (deductive, analogical, spatial,
and causal-scientific reasoning).
They were also examined on a test of mathematical reasoning and problem-
solving specifically designed for this study. This battery included tasks focusing on
processing similarities and differences between numbers and ensuing relations. Some
of these tasks were concerned with number grouping according to common
attributes and some were concerned with number seriation according to their
relations. Specifically, some problems required participants to identify common
attributes between numbers (e.g., what is common between 4, 16, 8, 32, 20, 100,
and 40), form sets based on common attributes (e.g., “select what is common
between these numbers {12, 14, 7, 56, 28, 36, 84, 54, 49, 19} and spell it out
explicitly”), extrapolate number sets by adding new objects sharing their defining
property (e.g., which one of the numbers {9, 12, 6, 7, 3, 8} belongs to the set {24,
36, 18, 15, 63, 30}), identify numbers that do not belong to a set (e.g., 9, 21, 11,
15, 12, 6, 35) because they do not share the set’s defining property, detect similarities
and differences in two-dimensional series (e.g. complete with the right number: 8,
4, 2 to 1, ½, ¼ to 1⁄8, 1⁄16 . . .). In some of the tasks children were asked to explicitly
specify the relations running in two or more items. These tasks are developmentally
patterned. Specifically, tasks requiring participants to specify an explicitly pre-
sent relation (e.g., each next number in a series is the double of the previous one)
address early rule-based reasoning. Tasks requiring mapping relations of this kind
onto each other require late rule-based reasoning. Finally, tasks requiring specifying
relations between relations and to explicitly spell them out require principle-based
reasoning.
Our training programme aimed to enable children to identify the various
dimensions underlying the various mathematical reasoning tasks described above,
explicitly conceive of their various groupings, and build the problem-solving skills
associated with each. Specifically, students were taught to look for and abstract
properties and relations, based on similarities and differences between tasks and task
types, align them according to a specific goal, conceptualize problems, and build
problem-specific problem-solving strategies. Students were instructed to identify
different problem types based on their mathematical and inferential requirements,
explicitly represent each structure, and specify similarities and differences between
problem types. Thus they were required to explicitly metarepresent both problem
structures and processes as well as their associations. The emphasis was on formative
concepts like “attributes”, “relations”, “similarity”, “dissimilarity or difference” and
their instantiation in the various problem types.
The intervention comprised 12 lessons, organized in 3 phases. In the first phase,
children were guided to search for and identify relevant attributes or relationships
involved in a problem and explicitly represent them into conceptual maps of similarities
and differences between tasks. For instance, children were instructed to specify the
relation underlying various patterns of numbers and separate patterns into those ruled
by the same relation and those differing. In the second phase, children were trained
to recognize different type of problems (e.g., increase, decrease, relations between
whole numbers, relations between fractions), solve them (e.g., first specify the relation
between the two numbers of a fraction and then specify the relations between
fractions) and practise on other problems. Children were guided to construct
procedural diagrams explicitly representing the sequence of steps involved in the
solution of a problem. Finally, in the last phase, training focused on three major
processes: first, to enable children to encode relations into rules (e.g., fractions are
relations where the number below the line denotes how an entity is divided and the
number above the line denotes how many parts of those specified by the other number
are taken); second, specify relations between rules (e.g., all fractions can be reduced into
a number specifying how an entity is divided); and, third, transfer of problem-solving
strategies to new problems, combine strategies in complex problems requiring more
than one strategy, evaluate solutions, and explicitly metarepresent them. In terms of
metarepresentation, in this phase, students were also required to recall strategies from
memory according to problem prompts standing for different problem types and
explicitly describe solution processes in detail and explicate why each was appropriate
for each problem type. The general scheme guiding actions in phases 2 and 3 involved
three steps: (i) search, specify, and classify problem; (ii) compare problem with other
problems; (iii) solve problem choosing the best strategy available. Feedback was
provided to children about the appropriateness of their answers.
The content of problems was taken from the mathematics curriculum of fifth
and sixth grades used in Cyprus. For example, activities involved concepts related
to the factorization and the divisibility of natural numbers, algebraic expressions and
generalizations about the properties of numbers and numbers’ operations (e.g.,

odd+odd=even, the sum of two consecutive triangular numbers), numerical
proportions, number sequences (e.g., Fibonacci number sequence, the sequence of
triangular numbers, etc), and attributes and properties of two-dimensional and
three-dimensional figures (e.g., different kinds of parallelograms, properties of
parallelograms, analogy tasks with figures, etc.) and geometrical patterns.
Thus this instruction programme focused on establishing and consolidating
processes primarily pertinent to the rule-based cycle. In terms of the other cognitive
domains tested in this study, instruction was related to analogical reasoning in
addition to mathematics, which was its primary aim. The other domains addressed
by the batteries above were only minimally and indirectly related.
The change in the domain of mathematical reasoning was considerable soon after
the end of the intervention (effect size =.38), although not all of it was sustained
about six months later (effect size = .20). However, the gains did transfer to domain-
free analogical reasoning tasks (effect size = .20) and, to a lesser extent, to other
domains, such as deductive reasoning (effect size = .12). Gains in deductive
reasoning were stable from second to third testing (effect size = .13), when they
dropped below significance in other domains, implying transcription of gains into
more formal inferential processes. Also there was a strong effect on working memory
(.93) and a less strong but significant effect on fluid intelligence (.38) and attention
control (.10); these effects were preserved at delayed post-test. Obviously these
effects indicate that cognizance mediated transfer of training from top-down process
involves executive processes. It is reminded that these effects resemble the effects
obtained by the mediation models presented in Chapter 10 (see Figure 10.2).
A second study (Panaoura, Gagatsis, & Demetriou, 2009) focused on the impact
of self-awareness on leaning. This study is based on an on-line programme for
educating mathematical thought. Specifically, this study was designed to lead children
to reflect on the processes they use when solving different kinds of mathematical pro-
blems. The aim was to enhance their awareness about the similarities and differences
between the problem-solving processes activated by different types of problems. The
self-representation of children about their own general cognitive efficiency, their
performance in mathematics, and their preferred problem-solving styles and strategies
were systematically recorded. The study showed clearly that the more reflective and
self-aware a person is the more this person profits from training.
A third study (Demetriou, Christou, Spanoudis, Pittalis, & Mousoulides,
forthcoming) aimed to assist learning in mathematics and specify the involvement
of the various dimensions of mind, as they were presented above. Specifically, this
study involved 9- and 11-year-old students who were first examined on processing
speed and control, working memory, and reasoning in the domains of quantitative,
spatial, and verbal thought. Moreover, they had been examined on various aspects
of self-representation, self-monitoring, and problem-solving styles. These children
were then allocated into three groups. The main experimental group received
20 one-hour sessions addressed to mathematics, aiming to develop both specific
mathematical skills for dealing with different types of problems, such as proportionality
and algebra, and general problem-solving skills. Training took place by specifically
trained teachers who used specifically designed material. A second experimental
group was simply exposed to the training material. That is, they were given the
material and they were asked to study it and solve the problems involved. Finally
there was a control group. All three groups were tested on all measures both before
and after the end of the training phase.
This study showed that children who were high in processing efficiency and
working memory were equally able to profit from plain exposition to the teaching
material or the systematic training. However, those who were low in these two
fundamental parameters of cognitive functioning were able to profit from systematic
training but not from plain exposition to the training materials. The message that this
study conveys for learning is clear: when the learning environment is well structured
and systematic, everyone can profit because structuring the learning environment
compensates for weakness in processing efficiency. In other words, on the one hand,
a well-scaffolded learning environment enables the students who are weak in their
processing and representational capabilities to learn despite their weaknesses. On the
other hand, the students who are strong in processing and representational capacity can
themselves compensate for the shortcomings in their learning environment because
they learn fast and efficiently, thereby discovering and constructing by themselves the
relations and concepts that float, so to speak, in the information provided.
Noticeably a recent study by Mackey, Park, Robinson, and Gabrieli (2017)
obtained similar results. Specifically, this study used commercially available games
to train fifth grade primary school children on several cognitive processes, including
processing speed (matching cards with a target card as fast as possible), working
memory (matching cards on a predetermined property that was played n turns ago),
and fluid reasoning (sorting cards according to one and two rules). The positive
effects of this intervention, which lasted for a school year, on the processes trained
but also on school related measures were much more pronounced for students with
lower academic performance in the previous year.
Conclusions
The studies summarized in this chapter suggest several conclusions about training
mental processes. They are as follows. First, training is possible. Any ability improves
with systematic training; some of the gains are lost with time, but much of them
remain. Second, transfer is possible, and it occurs according to the relations between
the processes involved. Top-down transfer is much more likely than bottom-up
transfer. The royal road to transfer is training cognizance and relational thought.
Training of these processes transfers to executive control processes, such as attention
control and working memory. Third, learning is phase sensitive. To succeed it must
take into account the present state of thought of the persons involved. To be
sustainable it must recycle to upgrade each phase’s core cognizance and relational
thought processes.
17
TOWARDS A THEORY OF
INSTRUCTION
This chapter outlines a theory of instruction explicitly focusing on the various

aspects of the mind discussed in this book. We discuss how learning in schools
would have to be organized to strengthen the operation and development of mental
processes involved in the architecture of the mind. Obviously education is a very
complex enterprise, aspiring to socialize children in a particular culture and time.
The present chapter will focus on the cognitive and developmental aspects of
learning in schools. We claim that the theory of instruction proposed below is
universal and is thus addressed to education in any place or culture on Earth.
It sounds trivial to argue that education as a cognitive endeavour is motivated by
three interdependent goals: (1) to enable children to read, write, and command
arithmetic. Literacy and numeracy are means to the two major goals of education
following: (2) to acquire a solid body of knowledge and understanding in various
fields of knowledge, such as the social, natural, and biological sciences; and (3) reason
soundly, efficiently, and critically to solve problems and make decisions in everyday
life and a profession. Obviously different realms of knowledge make different
assumptions about the nature of knowledge and its relation to reality. Contrast, for
instance, the humanities to mathematics or the natural sciences. The humanities
emphasize the human experience; mathematics emphasizes axiomatic truths based
on logic; science emphasizes observation and experimentation as key methods in the
search for truth. Knowledge in the humanities is largely subjective and constrained
by cultural and historical contexts. Thus it is taken for granted that truth is relative.
Mathematics searches for eternal truths whose power comes from the logic of proof
that generated them, rather than from correspondence to reality. Knowledge in the
sciences is always under check by the method of discovery: a combination of reason
with experimentation may topple any scientific theory, however long held or
inviolable it used to be.
Therefore the stance on the nature of knowledge differs drastically between these
disciplines. Still, knowledge specific to any of these realms must be represented,
understood, integrated, and critically evaluated. Old knowledge may be rejected or
modified and new knowledge may be created in the processes, calling on both
general mechanisms of representation, understanding, integration, and evaluation,
and realm-specific or discipline-specific criteria and standards. How best can a
theory of intellectual development, such as the theory proposed in this book, guide
learning in schools towards achieving the three goals specified above and naturalizing
them to work effectively and efficiently in any realm or discipline?
The architecture of developing mind outlined here suggests that domain-specific
systems always work together with general mechanisms and processes. It is reminded
that the domain-specific systems differ between each other in their core operators,
the mental processes they use to process information, and the knowledge and belief
systems associated with them. These systems operate in close liaison with general
representational, cognizance and control, and reasoning systems. All of these systems
of the developing mind are brought to bear on learning in the school in all know-
ledge and skill domains. Obviously different realms of knowledge require different
combinations of the systems of the mind. For instance, the human sciences capital-
ize more on social thought and awareness; the axiomatic sciences capitalize more
on quantitative thought and reasoning; the sciences capitalize more on causal
thought and reasoning. Figure 17.1 translates the four-fold architecture presented
in Figure 8.1 (in Chapter 8) into an educationally relevant statement of priorities
and principles of learning.
This chapter is organized into two sections. In the first section we present the
general principles derived from our theory for education. In the second section we
present the implementation of these principles for the various aspects of the
developing mind. Presentation in this section is organized according to developmental
cycles. The aim is to show the unity of educational aims and practices within each
period of development that corresponds to major levels in current education, such
as kindergarten, preschool, primary, and secondary education.
The general principles of education

The theory advanced here suggests a number of general principles that any
educational system must follow to fully capitalize on and expand the learning and
development potential of each student. Needless to say there are practices in modern
education that may be taken as implementations of some of these principles.
However, this implementation is implicit rather than explicit. One of the main aims
of this chapter is to give the frame that would allow policymakers and educators to
develop practices explicitly capitalizing on the knowledge generated in the recent
decades in cognitive developmental science. The general principles are as follows:
Principle 1. Education must cater for all four types of processes involved in the four-fold
mental architecture: these must enable children to (i) acquire, consolidate, and sharpen
problem-solving processes and skills related to every SCS and connect them to
Towards a theory of instruction 249
different domains of knowledge; (ii) handle representational and processing

possibilities as efficiently as possible; (iii) master inferential processes and counter
reasoning illusions; and (iv) sharpen awareness and master reflection so that
metarepresentational processes are fully facilitated.
Principle 2. Developmental priorities differ at different developmental phases;
thus children have different learning needs at each phase, although there may be
common learning mechanisms across phases. Children in different developmental
phases live, in a way, in different worlds; education must recognize these worlds to
meet its goals. The general educational principle in this regard is: saturate the
developmental needs of each cycle, starting from its developmental priorities. Overall, help the
episodic mind to command the core operators in each SCS and build explicit episodic
representations of each; help the realistic representational mind to properly interlink
representations in representational ensembles and acquire executive control of action;
help the rule-based mind master the inferential process vis-à-vis its underlying
representational and inferential rules; and help the principle-based mind to master
reasoning and logical principles, precisely map individual mental strengths and
weaknesses, and formulate long-term plans organizing action and decision making.
Principle 3. Learning at any phase must be planned under the perspective of learning at
earlier and later phases. In regard to the earlier phases, learning at a given phase must
enable children to rework earlier representations and skills at the level of the current
phase. This would free children from lags, shortcomings, and misconceptions
associated with the representational and inferential weaknesses of the previous phases.
In regard to later phases, once the concepts and thought patterns are consolidated at
the level of the current phase, children must start viewing their weaknesses based on
pointers to how these weaknesses may be removed. In short, education must help
the developing child consolidate the world view that prevails at each phase, use it
for the sake of better understanding and problem-solving relative to the previous
phase, and prepare for ascension to the next phase. For instance, in infancy, practising
core operators in each SCS may be more important than anything else. Later, in the
cycle of realistic representations, children must learn to recognize exemplars of
concepts but not their underlying organizing rules; thus learning would have to
demonstrate examples of concepts rather than define them. To acquire this ability,
children in this cycle would have to master the representational integration processes
that would generate generic concepts later on. Thus showing how exemplars may
relate to each other is a good learning priority for this cycle. Later, in childhood,
practising mental processes vis-à-vis their use in specific knowledge domains and
freeing inferential processes from contextual constraints may be more important. In
adolescence, coming to grips with the central premises and principles of different
knowledge domains and belief systems in one’s culture and time may be very
important. This is the substance of education for critical thinkers.
Principle 4. Although related, complexity of concepts and processes is not
identical to the representational priorities as specified above. Thus the complexity of
concepts taught at successive school grades must take into account the possibilities for handling
complexity typically associated with each grade. For instance, applying numerical
operations on multi-digit numbers exceeds the complexity that preschool or first

grade children can handle. It requires breaking each number in more than one
dimension (e.g., units and tens for 2-digit numbers), operating on each and then
recombining them. This requires a working memory span of at least 3 units and
alignment flexibilities attained at about 8–9 years of age. Therefore numerical
operations at preschool or first grade must be taught using single digit numbers. In
addition, it must be recognized that anyone may operate lower than his or her
optimum level when facing a new task. Thus teaching must always start with
examples demanding less than the students’ optimum capacity. Children must also
be educated to know and handle their representational and executive limitations;
this must also occur in phase-appropriate ways to be discussed below.
Principle 5. Educational timing. Concepts and educational materials are structured
in time according to school grade and educational level. Organization of educational
content in the curricula across grades and school levels does not always respect
developmental possibilities and constraints. This is due to several reasons. For
instance, education is much older than developmental science. Thus timing of
concepts in the curriculum were often answered independently of developmental
theories; as it is well known to all those concerned with education, education is
resistant to change. Admittedly too, as shown in this book, developmental science
has often erred in specifying developmental priorities. Piaget’s theory is an example
in this regard. This rendered educators cautious in adopting developmental models
as drivers for their curricula. Therefore the bridging of educational timing and developmental
timing is a major goal of educational science. We hope that this chapter will facilitate this
major goal.
Principle 6. Presentation of knowledge and learning opportunities is often
based on considerations related to the historical and epistemological characteristics
of different knowledge fields rather than developmental concerns such as those
noted above. This approach was strengthened in recent years by the emergence of
the theory-theory and the conceptual change approach in psychology, already
discussed in this book. It is reminded that these approaches suggested that learning
and knowledge change in the developing mind occurs in a fashion similar to
theory change in science. We showed in earlier chapters that these assumptions
are often wrong: change in developing minds obeys different principles than
change in science, despite some similarities. Most likely, it is confusing for
developing children to be involved in the controversies and changes that have
taken place in a discipline up until this point. Thus instruction must cause controlled
conceptual change in the minds of individual children following the principles of cognitive
developmental science rather than the epistemological principles of the history of knowledge
in a discipline. This change must rectify the various misconceptions that any of the
students may have about the world and generate new grade-appropriate knowledge
in the various school subjects represented in the curriculum following all other
principles above.
We now delve into the implementation of the principles above for each of
the major developmental cycles outlined in earlier chapters, initially spelling out the
FIGURE 17.1 Educational priorities according to the four-fold architecture of the mind
developmental priorities and weaknesses of the cycle and then specifying what
education would have to do to capitalize on priorities and remove the weaknesses.
Educating infants
The major developmental priority of infancy is to create a mental reality out of
perceptions and actions. Therefore revisiting episodic blocks and elaborating on the
characteristics of objects and their possible changes as a result of actions on them
allow the infant to abstract action patterns, interrelate them, and represent them in
language or other representations, such as mental images, generating the realistic
representations of the next phase. Thus the episodic mind is captive of environmental
variation, guided by it as much as it errs because of it.
Through the years education has expanded to ever-younger and older ages. In
the early twentieth century the vast majority of people went only to primary school
from ages 7–12, if at all. In our time, and for most people in the industrialized world,
education starts at kindergarten at about age 4 and proceeds up to about age 22
when the individual is at college. However, infancy is still excluded from formal
education. Education up to this age is the responsibility of the parents and the care
givers, either at home or in day-care centres. Informally, however, there are two
sources of influence on the education of young infants. One is the popular “how to
teach your infant” literature that abounds now and the other is the toy industry,
which produces all sorts of toys for infants from a few days old to school-age
children. Both the popular literature and the toy industry capitalize on the knowledge
generated by developmental science to create an educational market that goes with
the child-oriented culture of our time.
However useful these industries may be, they are not enough. We believe that
the time is ripe for a more systematic approach to education in infancy. For one
thing, the increasing demands of modern society and education coming soon after
infancy make it necessary to prepare infants for the world of the school, which
comes at about the age of 4. For another, the recent explosive expansion of our
knowledge about the mental characteristics and capabilities of young infants provides
a sound basis for the design of “educational environments” for infants.
Educating core operators in the SCS

The core operators in each of the SCSs are the starting point for interaction with
the world in different realms of knowledge. Gradually the core operators become the
cornerstones for understanding the physical, the biological, and the psychological
worlds and interacting with them. Carey (2009) suggested that the mind of infants is
organized in three realms: objects, making up the physical world; agency, comprising
the psychological and biological world; and number, which is concerned with the
quantitative relations between everything. Therefore education in infancy must enable
infants to construct (i) a basic understanding of the nature, behaviour, and relations of
objects, (ii) their own relations with objects and other persons, and (iii) the basics
of number. Below we give examples of training the core operators of each SCS.
Specifically, in the categorical SCS, infants must practise classification skills based
on various forms of physical and functional similarity: colour, shape, sound, smell,
and taste are all sources of physical similarity; the use of objects for eating,
transportation, playing, etc. are examples of functional similarity. The aim of
education at this phase must be to show how objects hang together based on their
physical characteristics or their functions and uses and how this may vary from
moment to moment according to which property or function of an object one may
focus on. For instance, a fork is an eating object but it may also be used as a can
opener. A ball is a football object but it may also be a sitting object.
In organizing activities it must be borne in mind that, in this cycle, infants do not
explicitly learn by rules and thus they do not associate key information to rules
underlying relations between objects. However, they focus on salient information
which they use as a basis of their explorations (Rivera & Sloutsky, 2016). Thus
organization must occur on the level of salient information. Specifically, to facilitate
explorations there may be special markers directing the classification of objects
according to shape and colour which provide feedback about acceptable and not
acceptable actions. For instance, particular objects (e.g. toy cars) can be associated
with specific colours (e.g., green trucks, yellow taxis, red racing cars) to indicate
that the combination of characteristics creates multiple specified categories.
In the spatial SCS, explore the form, size, shape, and relations between objects
that lead to the formation of mental representations in the visuo-spatial system.
Manipulating objects to fit in other objects or assembling them to build more

complex objects allow the construction of mental operations in this SCS, such as
mental rotation. For example, there may be special markers directing the connection
of objects according to shape and colour which provide feedback about acceptable
and not acceptable actions. Lego toys are a good example in this regard.
In the causal SCS, action on objects with the aim to make things happen would
enable infants to realize that particular actions may be connected with particular
effects. Moreover, varying actions to change the effect would allow infants to grasp
the differences between causal links that may exist between objects or actions and
objects. Particular objects or components of objects may be associated with special
effects, such as sounds or light colours that enable the infant to build causal connections
between his behaviour, objects, and effects.
In the quantitative SCS, adding objects to sets and taking them away would allow
infants to grasp the basic operations of mathematical thought; counting and number-
ing within the subitization limit would help them to associate operations with
numerical concepts, such as absolute and ordinal number.
Finally, in the social SCS, recognition of emotional expressions and deciphering
of the emotional states of others would promote social thought.
Educating representational capacity

Infancy is marked by limited representational capacity and executive control. As a
result, concentration and attention float and infants are easily attracted by new
objects in the environment. Moreover at this phase the use of language as a medium
of representation and instruction is limited. Therefore education focusing on the
core operations before age 2 must capitalize on the active discovery and exploration
activity dominating in this phase and initiate control over attention shifting according
to changing stimulation in the environment.
This can be attained by a careful organization of what is available to the infant
to play with. For instance, objects must be organized in a clear episodic sequence
which generates a specific interesting effect: for instance, in the first year, presentation
of a sequence of lights where colours alternate positions systematically, so that the
infant may represent the difference between sequences. Later, in the second year,
we may hide an object under several (2–3) places in sequence and let the infant
retrieve the object involved following the sequence. One may also use two boxes
where a number of objects (e.g., from one to several) are put into each box
according to some characteristic, such as colour, and the infant is asked to imitate
the sequence of action. Tasks such as these train both working memory and SCS-
specific operations, such as sorting.
Educating reasoning
Obviously it is not possible to educate reasoning explicitly in this cycle. However,
we may build explicit representations of basic logical schemes at the end of the
second year, when language emerges. For instance, iterative placement of objects in
a box according to a specific property, such as shape (e.g., square), but not another
(e.g., colour) and tagging the action with an appropriate word (e.g., a square, and
another square, and another square, all of them are squares) may create a representation
of conjunction and intension (i.e., the common property shared by all objects
involved). Placing the light switch at an on or off position (e.g., up or down) and
associating this with the effect (i.e., dark or light) and properly tagging the relations
(i.e., up and dark or down and light) may create the representation of disjunction.
Educating awareness
Like reasoning, awareness cannot be explicitly trained in infancy. However, the
education episodes above may be used to generate some awareness of the processes
involved. For instance, if repeated with varying content and complexity, these
episodes may make the infant realize that changes in the organization of information
cause changes in experience. For instance, the hiding sequence of an object or the
placement of objects in boxes may become as complex as to make the infant realize
that she cannot imitate them anymore. These changes in experience may be tagged
with appropriate language to show that we can speak about mental experiences
(e.g., “oh, there are too many, I see, this makes it difficult”).
Educating preschoolers
In early childhood, at 2–4 years, children master representation which transforms
their thought from episodic to representational. In this phase command of represent-
ational media, language par excellence, is the main developmental priority. Thus
explorations, imitation, and make-believe games, together with command of
language, dominate as activities. Knowing and using the representational tools is
more important than directed control. Controlled attention and working memory
are very limited in this phase, both in time and capacity. This comes in the next
phase, from 4–6 years. In this phase mastering attention control relevant to the
variation of external stimuli and mastering mental focus that would allow alignment
and integration of representations take priority. Children in this phase are able to
focus on, compare representations, and shift between stimuli according to a goal. In
this phase children can also hold in working memory 1–2 instructions, understand
the intentions of others, and reason pragmatically. The realistic representational
mind blurs boundaries between imagination and reality, enjoying the imaginary
world as much as it may be deceived by it. Thus attention and executive control
training must be the educational priority of this period.
From core operators to representations and representational

ensembles
Language appears at the edge between infancy and toddlerhood. This signifies a
major change from the episodic to the representational mind. The representational
mind is much more than any of its tools, including language. It is a stance that the
world is represented and representations may be manipulated independently of
the objects or events that gave rise to them. Thus the primary aim of education at
this initial phase of the emergence of the representational must be to raise core
operators from action to representation and transform them into the mental
operations of the specialized domains concerned.
Therefore infants must be systematically facilitated to tag core operators to words
and other symbols and use them mentally. In this phase the mental operations in
each SCS must be practised as such. For instance, sorting and classification in the
categorical SCS must be properly connected to verbs to tag the classification actions
themselves (e.g., look at me: I put these red cars together and these green cars
together; I classify them); additionally, the objects of sorting and classification must
be properly named so that categories may be tagged in reference to specific
characteristics (for instance, look at them: red cars are racing cars; yellow cars are
taxis; all of them are cars).
In the quantitative SCS, operations must be practised within the subitization limit
of 3–4 objects (e.g., look at me: I count these red cars, 1, 2, 3, there are three; and
these green cars, 1, 2, 3, 4; green cars are more than the red cars, we have 1 green
car more than the red cars). The aim would be to show that there is a quantitative
aspect of the world that can apply to all sorts of objects. Specifically, pointing to sets
of objects and naming their quantitative relation is the first step in establishing
quantitative representations: these are many (5 or more objects); these are just a few
(3 or less objects). Taking in and removing objects across sets may also show that
quantitative states may change by operations: taking 3 objects from the set of 5 and
adding them to the set of 2 objects makes the first “less” and “just a few” and the
second “more” and “many”. Counting and enumerating must be systematically
practised to show that number names correspond precisely to counting acts, absolute
values, and ordinal relations.
The spatial SCS may easily be confused with any of the others. Spatial proximity
may give rise to the perception of number; phenomenally linked changes in the
shape of objects may give the impression of a causal relation, etc. A primary aim of
education here would be to enable toddlers and preschoolers to discriminate
between operations pertaining to the spatial SCS as such from operations related to
the other SCSs. First of all, at this phase children must acquire awareness of mental
imagery as a major tool of the representational mind that may be connected to other
tools. That is, children must realize that what is represented in mental images
emerges from the senses (seeing and viewing) and may also be drawn, named,
enumerated, etc. Thus, for instance, having a visual image depends on having direct
contact with an object or event and that the precise nature of the image is a matter
of perspective. Second, they must understand that real actions correspond to mental
actions: rotating objects in reality may also be done mentally and that mental rotat-
ions can direct real actions. Adding or taking objects away from sets result in
corresponding changes in the mental images. Teaching must also enable pre-
schoolers to recognize how different words and symbols, but also the grammatical
and syntactical structures of language, are connected to realities related to each of

the core operators represented in the visual image.
In the causal SCS, education must have two priorities. The first is to show that
reality may be different to what it seems. Children must grasp that simple spatial or
time contiguity does not necessarily imply causality. For instance, a particular effect,
such as a light or sound, looks as though it was caused by a specific agent, such as
pushing button A, but actually it was button B that was the cause of the effect.
Obviously, grasping isolation of variables is not possible in this cycle. However, trial
and error explorations of these relations would enable children to differentiate the
phenomenal from the real. Second, children must form schemes of basic causal
relations. For instance, there are toys where specific actions or combinations of
them produce specific effects. For instance, button A is enough to make a toy car
move (necessary and sufficient); button B does not have any effect (neither necessary
nor sufficient); button C blocks the action of button A (incompatible); buttons D
and E may be used together to make the car move (necessary but not sufficient).
Educating executive control and representational capacity

The primary developmental task in this cycle is mastering executive control, mainly
inhibition and shifting. Practice in so-called hot executive control seems particularly
useful in preschool (Garon, 2016). This involves the ability to inhibit a present
action that will result in a pleasant reward (e.g., eat a small treat now, a small amount
of money) for a later (the time may vary from minutes to hours or even days) action
so that one receives a bigger reward (e.g., a box of smarties, more money, a new
toy, etc.). It is well known that children who are weak in other measures of
executive control discussed throughout this book are also weak in delay (Garon,
Bryson, & Smith, 2008). Training in this task may involve several strategies: for
instance, diverting thought from the present reward to something interesting until
the time required is over; thinking about the difference between the present, small
reward, and the future, bigger reward; or reflecting on the advantage of self-control
for other activities as well.
Practice in flexible shifting between actions according to changes in goals is also
important. Specifically, children need to practise checking the compatibility
between items and the current goal with ensuing deactivation or displacement of
irrelevant items away from attention. For instance, when sorting objects according
to a criterion, children must practise checking if a current move (e.g., placing a blue
car in the group of blue objects) is consistent with instruction (i.e., placing cars with
cars regardless of colour) and prioritize attention to identity rather than colour.
Additionally, practice in the various dimensions involved in a task, such as the
Dimensional Change Card Sorting task, facilitates children to shift between
dimensions in sorting later on, if needed. There is research showing that when
children are trained to recognize and store in memory several levels across dimensions
(e.g., hue variations of a colour, or size variations along the dimension of size) they
can then more easily shift between them if required. In line with our theory, this
training provides the representational resolution necessary to look for instantiations

of representations, align them, and shift between them (Perone et al., 2015). In
other words, training the primary developmental task of this cycle, executive
control, requires building advanced organizers of representational spaces so that
children can navigate in them following their current mental goal.
Representational capacity is very limited in this phase, about 2 units of new
information. Making children aware of these limits would increase personal control
of action. Increased personal control of representational capacity would require
leading children to monitor ongoing performance in order to improve sensitivity
to variations in its content. Specifying variations in performance as a result of
variations in the amount of information one is dealing with would be helpful. For
instance, children may be trained to alternate between blocks or types of information
and binding items according to type and time of presentation. Presenting visual
information on the one side of a screen and a verbal description on the other side
and asking children to integrate in order to formulate a complete representation and
story is an example. Also, practice in reorganization and re-chunking would enable
children to trade-off increasing information volume with increasing semantic
density of representations to be held in focus (e.g., by reducing several objects in a
category to a generic representation with a recall marker).
Educating reasoning
In the first phase of this cycle children may become familiar with inferential schemes.
This familiarization may be embedded in the episodes oriented to SCS as noted
above. For instance, joint iteration (i.e., and, and, and) associated with sorting in
the categorical SCS hints to conjunction and gives the basis for categorical reasoning.
That is, it demonstrates that adding cases does not change a class property, intension
(e.g., “animal”), but only its extension (e.g., “cats and dogs and mice are animals”),
opening the way for grasping class inclusion. It also may be used to point to the
difference between categorical and mathematical relations: joint iteration does
change the cardinal value of a set (e.g., a cat (1 animal) and a dog (2 animals) and a
mouse (3 animals)), despite differences in the particular objects involved. Pointing
to disjunctive relations (either . . . or) opens the road to grasp logical consistency.
Pointing to necessary causal sequences prepares for conditional reasoning underlying
modus ponens: for instance, “if it falls, it breaks”. Pragmatic deals and discussion
about their consequences put reasoning into its social perspective.
Examples for training reasoning in the second phase of this cycle are taken from
Box 17.1 (which is based on Table 16.1 in the preceding chapter, dedicated to
learning to reason). The first set of examples addresses inductive reasoning. They
refer to the very familiar condition of flying birds. Their aim is to show that
participating in a class allows the child to generalize class properties (e.g., they have
wings; they fly) from the general class (e.g., birds) to novel class members (e.g.,
imaginary nigles); also that if an organism has one class property (nappows fly) this
makes it a member of the general class (are they birds?) and this leads to other class
properties (e.g., do they also fly?). The second set of examples addresses analogical
reasoning. The emphasis of instruction here shifts from object similarity to relational
similarity. Examples horizontally involve analogical relations at the same order,
namely between specific elements (1), classes (2), and general functions (3). Thus
children may be instructed to pinpoint and elaborate on the relations within and
across pairs within and across analogies. The third set of examples addresses deductive
reasoning. The aim of the examples here is to show the difference between induction
(actual information is relevant and essential in induction), analogical reasoning
(a general property, such as motion, constrains relations between apparently differing
elements or properties), and deductive reasoning (form constrains inference,
knowledge about properties is irrelevant to inference, truth may remain an open
question). The first two examples stand for classical and easy modus tollens. Given
the premises, the conclusion is valid in both cases, although it is not true in the
second example. The two last examples are inconclusive. Nappows may or may
not be birds and cats may or may not walk, given the premises. These arguments
may be compared with each other and with the inductive and analogical problems
presented above from a number of respects. Overall, children would have to grasp
the difference between a true statement and a statement taken as true.
BOX 17.1
EXAMPLES OF TASKS THAT CAN BE USED IN
LEARNING TO REASON PROGRAMMES
Instruction aiming to enhance reasoning must adhere to

the principles following:
1. Demonstrate the difference between automatic and analytic inference.
Specifically, individuals must become aware of the dual nature of reason-
ing and acquire the necessary flexibility in moving from automated
reasoning to analytic reasoning that requires explicit representation of
premises and their integration according to specific rules.
2. Decontextualize inference. That is, they must understand that inference
obeys rules that are independent of specific information and context. Thus
they must be able to systematically activate and apply the rules of analytic
reasoning according to the relations involved rather than content of
context.
3. Differentiate between inferential processes and logical forms, such as
inductive, analogical, and deductive reasoning.
4. Make use of mental models for the sake of inference and reasoning. It is
important to use mental models in relation to the knowledge described
above about the management of representational and executive pro-
cesses in order to protect the inferential processes from representational
lapses that may be caused by excess of representational demands of the
information involved.
5. Make use of metarepresentation to automate inference and reasoning. We

will explicate how these aims may be attained in reference to the examples
presented below.
Inductive reasoning
Pigeons are birds, they have wings, and they fly
Sparrows are birds, they have wings, and they fly
Hawks are birds, they have wings, and they fly
Nigles are birds: Do they have wings? Do they fly?
Nappows fly: Are they birds? Do they have wings?
Analogical reasoning
Wings are to pigeons as feet are Wings are to airplanes as wheels are
to cats to cars
Wings are to birds as feet are Wings are to flying machines as
to animals wheels are to rolling machines
Flying is to birds as walking is Flying is to flying machines as rolling
to animals is to rolling machines
Flying, walking, and rolling enable motion, given the constrains of each living
being or vehicle
A (flying) : B (birds) :: C (walking) : D (animals) ::: E (flying) : F (airplanes) :: G
(rolling) : H (cars) à motion
Deductive reasoning
Birds fly Birds fly Birds fly Animals and birds
either walk or fly
Nigles are birds Cats are birds Nappows fly Cats are animals
-------------------- -------------------- -------------------- --------------------
Nigles fly Cats fly Nappows are birds Cats walk
P and Q P and Q P and Q R or S
P Q p1
-------------------- -------------------- --------------------
Q ?P ?R
-------------------------------------------------------------------------------------------------------------------
In regard to inductive reasoning, instructions in this phase should give priority

to elaborating on object comparison and similarity identification. That is, they
should elaborate on the reduction of individual elements into a class through a
common essential property present in each of them (e.g., they fly). Symbols standing
for the class may be used to symbolize the connection between reality and
representation (e.g., an abstract symbol of a bird). In regard to analogical reasoning,
teaching in this phase may start from studying actual animals and objects and
specifying their relations within and across pairs within each analogy (or their toy
representation). For example, they both have parts enabling them to move.
Observations may then be encoded into verbal statements with the explicit aim to
show how one kind of representation may be expressed into another kind of
representation. In this way observations or their action or visual models are
metarepresented into language.
In regard to deductive reasoning, emphasis must shift from similarities of element
properties or relations to relations imposed by the structure of the argument. For
this to be possible children must focus on the meaning of each sentence as given in
the argument and ignore any other previous knowledge or information related
to the words in the sentences. They must also understand that an argument involves
a network of relationships systematically arranged which can be used as a basis for
decoding the relationships. Thus, in order to grasp the logical relationships implied
by a logical argument, one must break down or analyse the argument into the
premises involved and focus on their logical relationships independently of content.
Attention should be drawn to the role of connectives such as “is”, “are”, “and”, “if
. . . then”, “either . . . or”, etc. as indexes of logical relations.
Educating awareness and metarepresentation

In this cycle, education must build awareness of representations and their relations
with reality. The activities described above for grasping the various aspects of SCS
are an appropriate frame. For instance, turn-taking with another person, age mate
or adult, is conducive to realizing the multiplicity of representations and their
function in causing differences in knowledge and beliefs. This is also a good context
to make children aware of multiple representations, their symbolic expressions, and
their mapping onto their reality referents. For instance, tagging photographs to the
objects they show, and then varying the perceptual similarity between an object and
its photograph until transforming the photograph into an abstract representation of
the object (e.g., showing only the shape of the object) is a method for training
toddlers in the nature of representation in the categorical domain. Tagging numbers
onto sets of objects may have the same function in the quantitative domains; tagg-
ing written words to expressions may have the same function in the domain of
language. Moreover, associating alternative representations with the same reality
may be helpful in enabling the child to dissociate representation from thought as
such. Finally, tagging photographs showing different perspectives from which an
object may be viewed and associating it with different persons is good practice in
perspective taking and theory of mind.
Representational alignment is a major mechanism of change in this cycle.
Properly organized map-reading may be used to enable preschool children to align
representations and their relations across realities, such as a room and a map, and
FIGURE 17.2 A schematic illustration of guided-alignment aiming to enable preschool

children to align maps with reality (Reproduced with permission from Yuan, Uttal, &
Gentner, 2017)
symbol systems, such as language and pictures. Figure 17.2 shows how guided
alignment may be used to enable 3–4-year-old children to align maps with the
realities they represented (Yuan, Uttal, & Gentner, 2017). Specifically, objects in
reality (the room) are explicitly named and properly pointed to in the order they
appear in the room. In the same fashion, the objects in the map are also named and
pointed to in the order they appear in the map. Their correspondence and their
relative position is also explicitly specified and pointed to in both the room and the
map. This method enables children to grasp the analogical relations involved
between reality and map.
The method above, in addition to training relational thought and representational
alignment, may also enable preschool children to become aware of the very pro-
cess of structure mapping, facilitating metarepresentation as such. To support
metarepresentation and facilitate the emergence of general reasoning patterns from
domain-specific processing, teaching must raise awareness of what may be abstracted
from any particular domain-specific learning. Children must become aware that
representations may stand for underlying relations that surpass content differences;
for instance, a name (cat) refers to an intensional quality (whatever is “catness”)
regardless of the many differences between cats (in colour, size, ferocity, etc.). Also,
the very mental processes used to produce an abstraction underlying a representation
(e.g., sorting) constrain the representation. For instance, sorting constrains an under-
lying quality (intension) that guides the actions of sorting; counting differentiates by
definition as it creates sets of different numbers.
The activities above provide an appropriate frame for educating executive control
as such. This must first enable children to understand that accepting a goal imposes
constraints on thoughts or actions to follow. Specifically, children must recognize
that adopting a goal implies focusing on goal-relevant information and a specific
course of related action. For instance, if the goal is to find the sum of two sets, one
must first count each set and then add the two results; if the goal is to find how each
button affects an effect, the various buttons must be examined one after the other.
Specifying the actions or thoughts following an accepted goal would enable the child
mentally to oversee the link between the sequence of actions and thoughts implied.
Educating primary school children

In middle childhood, explicit reasoning and explicit awareness of inferential pro-
cesses emerge as major developmental priorities. Thus concepts are defined in
reference to general rules integrating over different object characteristics. Flexibility
in moving across them also appears in this cycle. Awareness of mental processes may
be called on to help bypass weakness, as when children, for instance, recognize that
a difficult task needs more work to be completed. The rule-based mind allows a
well-organized representation of the world which often lacks cohesion and logical
validation.
Educating the SCS

In this period the mental operations underlying the SCS must be explicitly educated.
The aim of education must be to make the underlying rules explicit and readily
applicable. In categorical reasoning, for instance, sorting must be elevated to classification
in reference to explicitly defined rules in various contexts, such as basic properties
of matter (e.g., material, weight, volume), social reality, such as social activities (e.g.,
occupations, sports), etc. Also, condensing the results of classifications into general
rules and stating their implications would be a first step in the direction of their
formalization.
In the quantitative SCS, school mathematics is definitely a major source of
educational experiences into the mental operations involved. On the one hand,
associating them with everyday activities, such as transactions in the supermarket,
would highlight how they are related to real life. On the other hand, mathematics
may be the royal road to transition from rule-based to principle-based thought. One
may invoke many examples here. One of them is the understanding of fractions and
the operation of them. It is well known that children have difficulty grasping the
concept of fractions and applying the four arithmetic operations on them. One of
their main difficulties stems from the fact that, up to the age of 10–11, children do
not understand why numerical operations cannot be applied to them in the way
they are applied on integers (Braithwaite, Pyke, & Siegler, 2017). However, to be
able to understand similarities and differences children need a principled approach
that would specify how the numerical operations transform integers and how they
transform parts (divisions) of an integer. Teaching fractions at the end of primary
school may thus help, on the one hand, children grasp the general principle that
numbers are variables that may be transformed at various levels, with each level
obeying its own rules. On the other hand, it may help build the flexibility needed
to operate across rules and principles.
In spatial reasoning, fundamental spatial operations, such as mental rotation, must
be practised. The composition of three-dimensional objects through necessary
rotation of their components or the folding of objects would enable children to
grasp the preservation of analogical relations between components. The main
objective here would be to enable children to build the connections between the
real world and its projection into the visuo-spatial mental world.
In the causal SCS, classifications would have to be associated with causal relations:
that, for instance, one particular profile of characteristics, such as cloudy weather, is
associated with one type of effect, such as raining, and another type of characteristics,
such as clear skies, is associated with a different effect, such as hot weather (in some
places of the world). Moreover, associating operations across the SCS, such as
classification and measurement, may reveal more complicated causal relations than
would be seen by the naked eye: for instance, that cloud density would have to
exceed a certain limit, together with some other characteristics, to have a specific
effect, such as rain. By the end of primary school, mastering operations must be
combined with scientific theories in different domains, such as science and biology.
Teaching according to this approach may start from a phenomenon that is
familiar and is dealt with by different disciplines. The example of motion is one of
many. For the developing person, motion is self-initiated, perceived in other
persons, living beings, or objects, and experienced in many different ways from the
very first moments of life. In terms of the architecture of the mind proposed here,
concepts related to motion emerge out of the functioning of the core operators and
mental operations that define the various fundamental domains (i.e., the categorical,
the quantitative, the causal, the spatial domain, and the social domain). As such,
these concepts, at any moment in development, give meaning to motion and frame
how the person interacts with the world and solves the problems that it poses.
Children construct concepts about the fundamental aspects of the physical world
(such as force, energy, and gravity), the biological world (such as power, biological
needs such as hunger, and survival), and the psychological world (such as intention,
effort, and success in the environment) and use them to interpret the behaviour of
objects and other persons and guide their own actions. For example, when an object
is thrown at them they interpret the intention of the person who threw the object
(e.g., dad would not harm me), its appearance (the objects looks heavy though),
and speed (it is coming fast) in order to decide if they will try to catch it (it is
small and light so it cannot be dangerous) or avoid it (it is heavy and coming very
fast, so catching it may be painful). To make a decision and act accordingly,
computations are performed by all fundamental domains: the categorical domain
provides information about the identity of the object; the quantitative domain pro-
vides information about its speed; the spatial domain provides information about its
position and direction; the causal domain provides information about its possible
effects; and, finally, the social domain informs about the motives that initiated the
movement.

This is the period for educating representational capacity and its executive control
par excellence. Children must become familiar with the role of working memory
in learning and understand its limitations. Familiarization with capacity limits
requires activities where the volume of information and complexity may be
systematically manipulated. Primary school abounds with such activities. For
instance, recalling number digits or words of an increasing volume and recording
the cut-off point between success and failure externalizes what one can store and
recall: for instance, “I can remember numbers up to three but it becomes difficult
if they are more, I may forget any one of them”. Recalling information under
different organization (e.g., integrated into meaningful sentences) or recall strategies
(e.g., in presentation order or backwards) shows that different executive plans
influence how much information can be stored. Alternatively, recalling the output
of the same mental operation on information of increasing complexity—for instance,
multiply numbers of increasing complexity, such as 1 digit (e.g., 8 x 9), 2 digit (e.g.,
27 x 46), and 3 digit (e.g., 464 x 639)—demonstrates that mental processing
generates information that needs to be represented together with the initial
information and mental operations. Recalling a set of objects presented visually and
verbally demonstrates that there may be different limits according to the type of
information. Therefore more rehearsal may be needed to process and remember
verbal representations as compared to mental images.
Increased personal control of representational capacity would require training
children to monitor ongoing performance in order to improve sensitivity to
variations in its content. There are several aspects of processing that may be trained.
For example, using distractors (such as intense sound or a flashing new picture) to
divert attention from a demanding mathematical calculation would demonstrate that
attention is crucial for representing information and executing specific mental
operations at particular stages of mental processing. Managing more than one task
simultaneously by alternating between types of information that may be processed
at different steps of a multi-step problem would highlight the role of mental
flexibility. For instance, presenting visual information on the one side of a screen
and a verbal description on the other side and asking children to integrate them in
a complete story is an example of shifting between information for the sake of a
major goal. This also shows that relevance of information varies with the relation
between a particular step and the goal. Thus what is relevant now may not be
relevant later; checking the compatibility between items and the current goal and
activating and inhibiting accordingly is part of successful problem-solving and
understanding. Finally, practice in reorganization and re-chunking would enable
children to trade-off increasing information volume with increasing semantic
density of representations to be held in focus. For instance, several objects may be
reduced to a category and this may be associated with a generic representation that
may be called on later in the process. These examples highlight how executive
control of representational capacity may be enhanced.
Educating reasoning
Let’s turn again to the examples shown in Box 17.1. In regard to inductive reasoning,
emphasis must focus on class specification through multiple properties and relations
between classes (all birds have feathers and wings, a particular body structure, they
fly, they lay eggs, etc.). Any of these properties may also be present in other animals
but it is not enough to make them birds (e.g., bats fly but they are mammals), or
not present but this does not exclude them from the class (e.g., ostriches do not fly
but they are birds). In regard to analogical reasoning, at primary school the relations
may be elaborated across analogies with the aim of showing the relations between
relations (i.e., that flying and walking is motion).
In regard to deductive reasoning, children in early primary school may be trained
to realize that the information in the premises is connected by inference. Actual
models of the organisms involved and visual representations of the line of inference
going from the one to the other would be useful. Later, directed comparisons across
the various arguments would enable children to differentiate form from content and
understand that logic constrains inference. That is, when children understand that
the conclusion “cats fly” necessarily follows from the premises, given that we accept
that “cats are birds” and that “birds fly”, they already know that logical structure
underlies inference and that content is irrelevant to the conclusion.
Educating self-awareness
In the early primary school years self-awareness includes mental activity itself. This
is evidenced in a first understanding of the stream of consciousness and inner speech.
Children are now aware that they speak to themselves when they do so. Moreover
they can relate their thought with their ongoing activity. If they write, they know
that they think about what they are writing. By the middle of primary school, that
is at about age 8–9, they start to be able to differentiate between cognitive functions,
such as attention, memory, and inference. Finally, at this phase they can self-regulate
their actions through inner speech and other means of attention, motivation, and
stimulus control. That is, they know that they can give instructions to themselves
or that they can intentionally direct their attention to a particular object in order
not to think about something else. As a result of these new possibilities, during these
years children gradually grasp the constructive nature of thought. Thus they begin
to have intentional control of the thought processes. Logical necessity is a strong
sign of these possibilities.
However, primary school children still do not clearly differentiate between
cognitive functions and they do not understand the learning effects of different
cognitive functions or activities. For instance, at this age they may be deceived that
what is in their eyes now will stay in their mind later. Likewise they may be
deceived that what is in short-term memory is also in long-term memory. That is,
they think that they will remember something later on and be able to recall it simply
because they understand it when it is in front of them or when they have just
formulated a representation of it. It is to be noted that these weaknesses coincide
with the years of primary school; that is, a period of intensive formal learning of
new concepts and skills, such as reading, writing, arithmetic, science, etc.
Thus, at primary school, education must focus on building awareness of the
differences between mental functions and of their differential impact on learning.
Moreover it must focus on bridging activities with different mental functions and
aspects of learning. For instance, recalling and associating old information with what
is in front of one’s eyes help understanding but rehearsal helps storage in long-term
memory for later use: that association and relating with prior knowledge help
learning but variation and differentiation help originality in problem-solving.
Moreover, education in the primary school years must focus on revealing the
connections between concepts and operations and facilitate children to see their
own overt or mental actions vis-à-vis the concepts and operations. At this phase
children must be trained to realize that a series of representations are logically linked
with each other so that a particular series necessarily results in particular conclusions.
This will help them understand that logic underlies inference and that it is governed
by restrictions that need to be respected.
Educating adolescents
By middle adolescence core operators, operations in the SCS, all aspects of executive
control, and the major foundations of reasoning are in place. However Piagetian it
may sound, the major cognitive priority of this cycle is the integration of principle-
based relations into an overarching paradigm placing truth and value weights to
various forms of inference. This goes with an awareness priority and an executive
life priority: know yourself and choose a life course. A major limitation at this
period of life may be the mixing up of reality with the subjective and the tilting of
the balance on either side without the ability to realize that this happened and the
ability to rebalance. Obviously the major aims of education are very different from
the previous phases. We will preserve the same structure, though, to make similarities
and differences clearer.
Educating the SCS

By adolescence the basic operations in each of the SCSs are well established. Thus
education would have to enable adolescents to automate these operations, formalize
them, map them onto each other and create mental models for each that would
enable one to call on the set of operations required according to the problems at hand.
In categorical reasoning, basic operations of class logic and categorical syllogism
must be clearly grasped and applied. The aim would be to enable adolescents to
grasp the connection between real classificatory actions, actual classes and concepts
held, and underlying formal principles governing the formation and relations
between these classes and concepts. The basic principles of set theory must eventually
be grasped.
In quantitative reasoning, types of number, such as natural number (positive
integers), real number (a quantity along a line), negative number (numbers below
0), rational number (fraction of two integers), and complex numbers and operations
on them must be grasped. The basic properties of mathematical relations (i.e.,
closure, associativity, commutativity and distributivity) must also be grasped in this
cycle. The background for all is to understand number as a variable (e.g., be able to
specify when a + b + c = a + x + c or x + 2y + 3z = x + 3y +2z, when b = x and
y = z, respectively); this would enable understanding of its possible expression in
any of the other forms.
In spatial reasoning, visuo-spatial operations and spatial relations already mastered
in the previous cycle must be connected with geometrical and topological concepts
and operations. The main objective would be to let adolescents realize that mental
operations on objects, size, and their position in space have a formal aspect that may
be represented and specified according to several principles.
Finally, in causal thought the experimental method must be exhaustively trained.
Adolescents must know how to isolate variables, connect their isolation of variables
processes with specific hypothesis, and interpret results accordingly. Interpretations
must take place within the frame of causal relations where the precise type of
causality is formally understood (i.e., necessary and sufficient, not necessary but
sufficient, necessary but not sufficient, incompatible).
Training reasoning and problem-solving in each SCS must take place in an
epistemic framework where adolescents and young adults understand how each type
of problem-solving is embedded into the context of different knowledge traditions,
such as humanities and the natural sciences. This framework will be discussed in the
next chapter, in relation to the education of critical thinking.

Representational capacity reaches its limits in adolescence. The emphasis of training
here must focus on the recognition of personal strengths and weaknesses and their
management: for instance, that one is more at ease in representing visual information
as contrasted to verbal or numerical information. Moreover, in this period,
adolescents must familiarize themselves with differences in the mental demands of
different types of information, such as concrete, rule-based, or principle-based.

They should also recognize that representational capacity is sensitive to experience
and learning in a field or one’s own physical condition (e.g., in need of sleep or
under the influence of a drug).
Educating reasoning
Educating reasoning in this period should focus on developing the “if . . . then” stance
to concept construction and problem-solving and build epistemological awareness
about the characteristics, possibilities, and limitations of different knowledge domains
vis-à-vis their methods, functions, and priorities. It should also automate readiness to
embed inference drawing into the context of conditional reasoning schemes as noted
above. The aim would be to automatically resist the various logical fallacies.
As regards inductive reasoning, emphasis must shift from inductions in reference
to objects and their properties to the nature of inductive generalization as such. That
is, that it is likely but never necessary. Thus belief in inductive generalizations must
always remain open to future falsification. In analogical reasoning, relations may be
formalized in abstract representations as above. This may occur through discussions
of relations from the context of different knowledge domains, such as biology (motion
is needed for survival), physics (wings and feet make use of similar principles to ensure
animal motion), and technology (artificial parts such as wheels make use of the same
physical principles). The aim would be to move from a learning exercise in reasoning
to embedding reasoning into different knowledge domains and evaluating it from the
point of view of different theories with the aim to bridge reasoning with epistemological
awareness about the nature and limitations of both knowledge extraction and handling
mechanisms and knowledge validation itself (Gentner, 2005). Finally, in deductive
reasoning adolescents must be introduced to the conditional and suppositional nature
of reasoning and the role of form and constraining inference.
Educating awareness
In adolescence, self-awareness gradually becomes process-driven, the self-concept
becomes dimensionalized and generally accurate, and problem-solving becomes
planful and systematic. Therefore education at this phase should focus on awareness of
the differences between cognitive domains in the mental processes they involve
and how they relate to their world domains. Moreover adolescents must be driven and
practised to know where they are strong and where they are weak, and they can plan
their problem-solving activities from the start so that they can seek information when
and where it is needed and integrate it into their current problem-solving endeavours.
Assessment for learning and learning to learn

Assessment in education is crucial for learning because the information it provides
is useful both for the learner and the teacher. As regards the learner, accurate
cognitive developmental assessment may inform the learner of possible divergences

between learning and developmental goals and respective outcomes. The model
of mind proposed here offers a full framework for assessment that would be very
useful for learning in school so that students may (i) become aware of their cognitive
strengths and weaknesses and their facility with different domains of knowledge;
(ii) enhance their knowledge about the mind; and (iii) sharpen their self-monitoring,
self-representation, and self-regulation skills. In terms of the present theory,
assessment must first enable both the learner and teacher to identify the learner’s
position on the various dimensions of the four-fold mind specified earlier. Second,
it must also enable the learner and teacher to recognize the sources of difficulty
in understanding a concept or solving a problem and assist them to develop
strategies for overcoming these difficulties. The ultimate aim is to obtain information
that would allow the teacher to prescribe for the learner the right cognitive
developmental support programmes discussed in this chapter. These provisions are
in line with the aims of formative assessment or assessment for learning (e.g., Black
and Wiliam, 2009).
Outline of a cognitive developmental diagnostic tool

The theory presented in this book suggests that mental diagnosis must be cycle and
phase sensitive, focusing on the developmental priorities of each phase. In the epi-
sodic cycle, diagnosis must focus on sensory discriminations, abstraction of patterns
in stimulus sequences, sensorimotor coordination, understanding of instructions,
and signs of revisiting and reflecting on experiences. In the realistic representations
cycle, diagnosis must focus on control of attention focus, inhibition as examined by
Go/No Go tasks, flexibility in shifting between rules (as in the Dimensional Change
Card Sorting task), and understanding of other minds by theory-of-mind tasks. In
rule-based thought, diagnosis must focus on Piagetian-like tasks addressed to
integration of dimensions and to resistance to perceptual deception, inductive
reasoning as examined by Raven-like matrices requiring integration of dimensions,
and flexibility in shifting between conceptual spaces as in tasks requiring to recall
categories based on a specific characteristic. In principle-based thought, conditional
reasoning must be examined, because perhaps it is the best indicator of principle-
based reasoning.
It is reminded that the studies of reasoning summarized in Chapter 8 showed that
if an individual is in full command of conditional reasoning then he or she would
be very good on every one of the other processes. However, conditional reasoning
at lower levels is not predictive of other capabilities because the lower levels of
conditional reasoning are mastered by everyone. This finding suggests that reverse
diagnosis is important. That is, in each cycle diagnosis must look for the state of the
individual’s command of this cycle’s top abilities. If acquired, others would be
present too, at least as possibilities. Thus education would need to raise the
individual’s performance in different domains to the level of the individual’s
competence. In a sense the approach here concurs with the Vygotskyan concept of
the zone of proximal development; that is, persons operate in a zone of possibilities
that may be realized if support is properly scaffolded. This zone is precisely specified
here as the individual’s upper level of attainment of a cycle’s major developmental
priority and the deviation between this level of attainment and the individual’s
actual operation in the various processes of the four-fold architecture.
Overcoming mental difficulties

Difficulty may arise from any component of the architecture of mind proposed here.
For example, students take problems or concepts as difficult if they cannot make
sense of them in reference to prior relevant knowledge or if they are inconsistent
with each other. Assessment here must highlight what is lacking and direct the
student to search for necessary knowledge and use it to address the problem at hand.
Search must be both self-directed (e.g., “Do I know something similar?” “Have I
solved problems like this in the past?”) and directed externally, if needed (e.g.,
turning to the teacher or other knowledge sources for help). In the case of conflict
or inconsistency between concepts, assessment must direct the student to notice the
superiority of the new concepts to explicate the phenomena of interest: for example,
why is the heliocentric model of our planetary system superior to the geocentric
model?
Problems are also difficult for students if the amount or the presentation rate of
information exceed the available representational or processing resources. Assessment
here must make the students sensitive to their “personal point of command”. The
discussion above about monitoring and regulation of representational capacity is
relevant. Finally, inferential processes may be a source of difficulty. That is, if the
concept or problem to be dealt with involves relations that require inferential
schemes that are not available, the crucial relations will not be worked out.
Assessment here must highlight the inferential schemes missing and the student be
practised in their use.
Conclusions
This chapter suggested that education, at any phase, must lead the student to develop
and refine the following cognitive skills:
• focus on relevant information;

• scan, compare, and choose according to goal;
• ignore irrelevant information;
• represent what is chosen and associate with extant knowledge;
• bind into models and rehearse if necessary;
• evaluate models in reference to evidence;
• reason by deduction to evaluate truth and validity of models and conclusions;
• estimate consistency with beliefs, extant theories, dominant views, etc.;
• encode, symbolize, and embed into the system.
These general goals may be summarized as follows:

First, developmental milestones can direct the formulation of educational prior-
ities in regard to crucial epistemological questions about the origins, nature, and
change of knowledge and learning. Specialized domains may direct the training of
the various specialized mental mechanisms for the sake of better handling of different
types of relations in the environment and different types of knowledge. The ultimate
aim in each phase must be to consolidate crucial concepts and facilitate transition
to the next phase. Representational capacity may be a guide to the control of
instructional complexity children encounter at different phases of their school life.
The development of awareness must inform programmes addressed to learning to
learn, self-development, and self-guidance in knowledge-relevant life choices.
Ultimately the aim is to enable individuals to reason systematically, resisting
deception or impressions, so that they properly evaluate information or past or
incoming knowledge.
Second, teachers operate in a vast field of intra- and inter-individual differences.
Therefore curriculum, instruction methods, and moment to moment teaching must
adapt to different students and to different subject matters. It is reminded that
students who are weak in processing efficiency need more help and support to learn.
The implications of this postulate for the development and use of appropriate
diagnostic tools and the education of teachers are enormous. After all, democratic
education must lead each student as close as possible to his or her potential across
the board at successive developmental phases.
Third, in regard to general issues about educational policies and orientations, the
model suggests that plain constructivism, which has dominated in discussions about
educational priorities since the 1980s, is not enough for efficient education. In
addition to self-directed activity and discovery, guided abstraction and meta-
representation are very important for stable learning and learning transfer. This is
especially needed by weaker students. In fact the research showing that genes are
related to intelligence and educational achievement implies that the difficulties
many children have to attain high performance at school originates from genetically
determined constraints. However, our learning studies showed that students who
are cognitively initially weaker gain more from training. Therefore it might then be
educationally useful to locate students who are genetically and cognitively inclined
for lower attainment in order to integrate them into specifically designed programmes
that would compensate for their weakness.
Finally, it is clear that dominant theories in the various traditions that have
studied the mind since the late nineteenth century do not suffice to explicate or
direct education for learning to learn. In psychometric theories the assumption of a
powerful general intelligence mechanism controlling everything (Jensen, 1998) is
not enough to account for learning to learn because this mechanism involves only
a part of the necessary processes, namely the inferential part (Carroll, 1993). The
multiple intelligences (Gardner, 1983) theory is also in trouble in attempting
to account for learning to learn. Postulating another domain-specific intelligence to
account for self-awareness and self-regulation cannot explain how this intelligence
exerts its effects in a fragmented mind of autonomous intelligences. The model

summarized here keeps the balance between general and specialized processes and
relegates ability for self-development to the dynamic relations between all of the
processes, because in this system the ability to know oneself and learn to learn is part
of the evolutionary endowment underlying our ability to face the unexpected and
deal with it.
18
EDUCATING CRITICAL
THOUGHT
Critical thinking is a major goal of modern education. Often educational systems are
criticized for just delivering rote or authority-based knowledge rather than enabling
students to think critically in order to evaluate knowledge and search for better new
knowledge. The criticism reflects the assumption that any knowledge is a mental
construction that is inherently incomplete or wrong. In contrast, critical thinking
aims to impart to students a spontaneous “what if . . . ” inquisitive attitude that
would enable them to evaluate knowledge, assertions, assumptions, and problem-
solving methods for accuracy, completeness, truth, and their relevance to current
problems and questions. This is all the more relevant in our modern impression-
making world where mass media, social media, and the Internet flood citizens with
information about practically everything, often inaccurate and conflicting if not
intentionally fake. In this context, critical thinking is essential both for the individual
citizen and society. While it may appear trivial to suggest that a critical attitude is
important for efficient everyday functioning because it may protect the individual
from misjudgment that may result in wrong decisions, at the collective level this is
highly important in a democratically functioning world where one’s decisions
influence the choice of leaders and directions nationally and internationally.
There is a vast body of literature on critical thinking (see Ennis, 1962, 1996;
Sternberg, 2006; Watson & Glaser, 1980). Here we provide a developmental pers-
pective to critical thinking. We draw on research that is developmentally relevant
(e.g., Heyman, 2008; King & Kitchener, 2002; Kuhn, 1999) and integrate with the
four-fold model of the mind presented in this book. The aim is to show the priorities
of education for critical thinking suggested by the development of the various systems
specified by the four-fold model in each developmental cycle. Our approach is based
on the assumption that critical thinking requires appropriately capitalizing on all
mental systems given one’s age. Thus we view critical thinking as the ability to
embed intelligence into real-life contexts and make decisions taking into account the
information available, together with an evaluation of possible outcomes and their

possible value both for the present and the future. We also assume that becoming
critical is a long developmental process requiring that critical attitude and skills
capitalize on the strengths of each cycle and overcome its weaknesses.
What is critical thinking?

In the literature, critical thinking is considered to be both ability and disposition. As
an ability, critical thinking is self-directed thinking aiming to attain mental or
behavioural goals based on knowledge and information available, systematically
evaluated to the best-reasoned conclusion or decision. As such it involves the following
characteristics: (1) identifying central issues and assumptions in the theme of interest
or in an argument; (2) interpreting data or statements using systematic logical analysis
in order to recognize problems or unanswered questions because of lack of evidence
or coherence; (3) envisaging alternative models; (4) associating each with its own
supportive evidence and logical substantiation; (5) embedding it into its own
conceptual or belief system; (6) adopting an informed preference based on evidence
and well-reasoned argumentation, often against previously held beliefs and personal
biases; and (7) tolerating ambiguity and remaining open to the possibility of change
in conclusions and interpretations based on new evidence or analysis (e.g., Ennis,
1996; Halpern, 2006; Watson & Glaser, 1980; West, Toplak, & Stanovich, 2008).
Obviously the abilities above are based on accurate representation of the
information involved and the manipulation of information by means of the various
forms of reasoning analysed in this book. Depending on the problem or question
involved, anyone or any combination of the various SCSs may be needed to put
the questions and information in perspective of relevant mental operations and the
belief systems involved. Inductive and deductive reasoning are also crucial tools in
all of the abilities mentioned above. Finally, the very stance of being critical involves
awareness of the possibility that any knowledge or information may be short of
accuracy or completeness. It also involves awareness both of the mental processes
involved and their relative strengths and weaknesses. Therefore critical thinking is
a system-wide stance that systematically directs the mind to make interpretations
and decisions to the best interests of the thinking person or the group this person
belongs to. Ultimately, to be critical, thinking must be embedded in a context of
epistemological understanding, specifying what is evidence, what is truth, and what
is important or valuable (Siegel, 1989).
Can everyone be a critical thinker? As a disposition, critical thinking is an
interaction between mental and personality characteristics. Specifically, individuals
differ in their inclination to adopt the “critical stance” noted above. At the one
extreme, some individuals are habitually inquisitive, information-seeking, withholding
final judgment until complete evaluation is attained, flexible and ready to reconsider
if new evidence or reasoning suggests it. At the other extreme, non-critical thinkers
are inclined to accept rather than search, believe rather than question, decide rather
than leave options open, fix rather than reconsider. There is research showing that
Educating critical thought 275
critical thinking relates highly with various indices of cognitive performance, such as
scholastic achievement tests of analytical reasoning, verbal, and mathematical ability
(Halpern, 2006; West, Toplak, & Stanovich, 2008). Recent evidence also suggests
that critical thinking distinctly predicts real-life events, adding significant predictive
power over measures of intelligence. Specifically, critical thinkers were less likely to
experience negative life events than less-critical thinkers with the same level of
intelligence (Butler, Pentoney, & Bong, 2017). In addition, an elaborative and
synthetic style of processing, which is part of critical thinking, relates primarily with
openness to experience; interestingly, these characteristics relate negatively with
neuroticism (Halpern, 2006). Similar to openness is open-minded thinking (e.g.,
“People should always take into consideration evidence that goes against their
beliefs”) and need for cognition (e.g., “I would prefer a task that is intellectual,
difficult, and important to one that is somewhat important but does not require much
thought”) (West, Toplak, & Stanovich, 2008).
There is also a developmental dimension to critical thinking. For instance, lack
of a theory of mind would make children take others’ views at face value; lack of
command of the principles of conditional reasoning would render critical thought
incomplete because conditional reasoning is necessary for full critical thinking, as
specified above. Thus the reasoning training programme outlined in the previous
chapter must be part of any programme addressed to the education of critical
thinking. This is the topic of the section below.
Educating critical thinking

Educating critical thinking should lead developing persons to realize the limitations
that their own knowledge, understanding, and decision making may have. Also they
should understand the difference between reasoning and rationality. That is,
reasoning is a tool for analysis, evaluation, and selection of solutions; rationality is
integrative judgment that bridges multiple frames and points of view. Thus attending
to the formal prescriptions of reasoning ensures that valid conclusions may be
derived. At the same time, valid conclusions may be useless or even harmful if they
are not relevant to the situation at hand. For example, the statement “Peter was so
good, he was flying” cannot be induced or deduced from any of the example
arguments given in Box 17.1 in the preceding chapter. It may be induced or
deduced, however, if embedded in a metaphorical context where the meaning of
flying is symbolic of something else, such as that Peter was very successful at the
contest. Thus, to become critical, children must learn how to evaluate or choose
the relevant spaces in which logical analysis may be deployed.
Educating critical thinking in preschool

Education must capitalize on the worldview prevailing in each developmental
phase. It is reminded that, early in preschool, children are not aware of the relation
between perception and mental states, inference as a basis of knowledge or belief,
possible differences between persons in their mental states or perspectives relative

to their relation to various sources of knowledge, and they cannot grasp the re-
presentational nature of symbols and differentiate between reality and appearances.
Children also lack knowledge about many aspects of the world that would allow
them to view current information from alternative perspectives. Researchers
viewing cognitive development as equivalent to conceptual change replacing mis-
conceptions with the right concepts assume that young children are universal
novices. That is, they lack knowledge and expertise in all domains (Carey, 1985;
Vosniadou, 2013). This combination of characteristics renders them particularly
weak in critical thinking. On the one hand, they lack the epistemic background
for questioning and evaluation; as a result they tend to believe others or take their
own view as the only one available. On the other hand, children’s limitations
in critical thinking may just reflect their ignorance about the world, because to be
critical requires knowledge about the issue at stake.
Later, in the 4–6-year phase, children start to be aware of mental states, they can
reason in pragmatic contexts, and they start to realize that reality may not always
correspond to appearances. However, they still may be deceived by appearances or
their own point of view, although they start to know that individuals may differ in
credibility. That is, at the age of 4, children recognize that if a person is not accurate
in naming an object with the right name, this person must not be believed in the
information she may convey later (Koenig & Harris, 2005; Harris, 2007).
Therefore there are two major aims for education for critical thinking at preschool.
First, children must understand that there may be alternative or conflicting represent-
ations about the same reality. For example, they must realize that the knowledge of
different children about an object (e.g., that it is red, or green, or blue) may be right,
depending on each child’s perspective (i.e., because each of the three sides of the object
is differently coloured and each child has access only to one side). Second, the example
above may also be used to show that assertions are not always accurate. Sometimes
they only represent a person’s perspective; sometimes assertions are “intentional lies”
aiming to deceive. This may be achieved if two persons are asked to exchange their
perspective and talk about their past and current perspectives. This is also a good set-up
to enable children see the difference between conveying information that (i) one
thinks is accurate but is not (any person’s perspective above) and (ii) intentional
deception as one gives information other than truth. For instance, in this latter case,
children may be exposed to a person who “sees and knows” that an object is “red”
but she says that this object is “green”. Hopefully this approach would help children
overcome the absolutist stance imposed by the cognitive profile of this cycle that
knowledge is either right or wrong, and someone has it anyway. This would prepare
for the next cycle, aiming to show that reality is not directly knowable.
Educating critical thinking in primary school

Children in primary school are good observers of reality, especially after the age of
8–9. They are aware that people may differ in what they know and where their
knowledge may come from—other people, reading, films, etc. They are also aware
that knowledge may not be accurate or complete and that others may deceive
intentionally or unintentionally. However, in this period children still lack a
complete system allowing full evaluation of their own interpretations or the truth
of assertions. It is reminded that children up to the age of 11 fail conditional
reasoning fallacies. This reflects a bias towards confirmation rather than falsification.
Therefore assertions or interpretations may be taken at face value, especially if they
originate from authority.
Based on the characteristics above, education of critical thinking at primary
school must aim for the following. First, reality is not directly knowable. Children
must also realize that observations are integrated by inference and integration may
be fallible for several reasons. For instance, conclusions may be reached on the basis
of faulty or incomplete information. Thus the reasoning underlying a conclusion
might be right but the conclusion is wrong because of wrong initial assumptions.
Alternatively the information may be accurate but the line of reasoning wrong, thus
leading to an inaccurate interpretation. Moreover, children must be gradually
introduced to an epistemological approach to knowledge enabling them to
understand that the knowledge generated by different knowledge extraction
mechanisms, such as observation and experiment, may differ in accuracy and
validity, depending on how well confounding factors are controlled by each. Plain
observation is often fallible because it highlights appearances but not hidden
relations. If things are not as they appear, simply observing may lead to wrong
conclusions. For instance, humanity believed for millennia that the Earth revolves
around the Sun. Experimentation may be more sensitive to hidden relations but
there is always the risk of uncontrolled variables that may deceive one. Distance
from truth is only a matter of appropriateness and precision of methods and ingenuity
of controls, which children improve with accretion of knowledge in time (Chandler,
Hallet, & Sokol, 2002; Wildenger, Hofer, & Burr, 2010).
Educating critical thinking in adolescence

Formally speaking, principle-based thought is critical thought by definition.
Entering this cycle implies a suppositional stance enabling the adolescent to approach
problems and information from alternative points of view, to envisage alternative
possibilities, and to check for truth and validity. Thus education in early adolescence
must consolidate the suppositional-inquisitive stance and associate with habits of
reasoning directed to truth-checking and validating. Specifically, adolescents must
take knowledge as mind production that is inherently uncertain. Thus one must
question both the accuracy of assertions and the truthfulness of judgments. They
must automate the skills for “catching fallacies” in the line of an argument or proof
by both staying alert to the possibility that reasoning may be deceived if information
is not exhaustively analysed and by systematically implementing the schemes of
conditional reasoning discussed in the previous chapters or the inherent weaknesses
of any form of reasoning that was used to derive an assertion or judgment.
A major aim of education for critical thought at the level of principle-based

thought is to let principle-based thinkers stay aware that deception and wrong
inference always lurk in arguments and complexly organized information. Let us see
an apparently easy task, which the reader may answer before considering the answer
below:
Jack is looking at Anne but Anne is looking at George. Jack is married but
George is not. Is a married person looking at an unmarried person?
A. Yes
B. No
C. Cannot be determined.
This task was first used by Levesque (1989) and studied by Stanovich (Toplak &
Stanovich, 2002). The vast majority of adults chose C, when the right answer is A.
In the words of a respondent who solved the problem, “Anne could be either
married or not. If she is married then the fact that she is looking at George meets
the condition set by the question. If she is not married then the fact that Jack is
looking at her meets the condition set by the question. So, either way, some
married person is looking at an unmarried person” (see also our informal study
below). This reflects disjunctive reasoning, allowing a respondent to resist the
easiest (wrong) conclusion that the answer cannot be determined because
information is missing (if Ann is married). Interestingly, in an informal experiment,
we gave this task to 16 persons, all university graduates; 12 of them have a doctorate,
10 being professors in various fields. Only 8 solved the problem: 2 professors of
mathematics, 1 professor of science teaching, 3 with a doctorate in reasoning
development, and 2 teachers. Impressively 5 professors and 3 university graduates
failed. Obviously, to stay critical requires more than possessing reasoning skills at
the top; it requires staying alert to the possibility that reality may be different than
it looks and thus inhibiting judgment until exhaustive consideration and evaluation
of all possibilities is complete.
Education for critical thought for adolescents and young adults who entered
principle-based thought must also provide the historical and epistemological frames
where knowledge and information may be placed and evaluated. In Kagan’s (2009)
terms, in science there are three general cultures: humanities, social sciences, and
natural sciences. Each of them specializes in investigating, interpreting, and re-
presenting a different broad domain of human experience or of the world.
Humanities deal with the human experience and condition and they are primarily
descriptive rather than interpretive. The historical, cultural, or personal point of
view dominates in descriptions and interpretations, and control is limited. Social
science deals with human experience and condition as in the humanities. However,
in the social sciences control of knowledge dominates over personal or subjective
experience or point of view. Ultimately it is aspired that interpretations grasp
underlying mechanisms of how reality operates. Limitations of control here come
from two sources: the complexity and variability of the phenomena involved; and
also from the fact that humans may not always be objective in dealing with the
human condition. Finally, natural sciences deal with the natural world. Although
subjectivity cannot be fully ruled out, natural sciences are less susceptible to the
weakness of control of the other realms, because the phenomena of interest are
more stable, they are less susceptible to effects from history and culture, and they
can more easily be subjected to experimentation. Within each of these “knowledge
cultures” there are specialized domains, such as literature, philosophy, and philology
in humanities; psychology, sociology, and political science in social sciences; physics,
chemistry, and biology in natural sciences. Obviously technicalities may differ
between disciplines within the same knowledge culture.
Students must understand that knowledge and concepts about the world are
constrained by the peculiarities of the institution and the discipline that generated
them. In other words, education must develop epistemological awareness about
these realms and domains and lead the students to develop and practise knowledge
and skills within each of them. This can proceed in two ways. First, the students
must develop mental models and templates appropriate for each. Ideally they should
develop a systemic approach allowing them to see each model from the point of
view of the others. Second, they must be aware of the possibilities and constraints
of each. This goes with an understanding that cognitive processes in the SCS are
knowledge extraction and knowledge handling mechanisms that can be put in the
service of the realms and domains of knowledge specified above. Thus these must
be practised and refined with content from each of the knowledge domains,
understanding that at any time errors may occur.
For instance, although causal relations are formally identical across domains (e.g.,
to be a necessary and sufficient cause, A must always come before and make, on its
own, B happen) they are expressed differently in physics, biology, human relations,
and history. Specifically, cause-effect relations are expressed on different time-scales
(milliseconds in chemical reactions, thousands of years in evolution, billions of years
in astronomy); take place on different levels of reality (observable reality versus
hidden reality) that is not accessible to the senses; they may be masked by different
kinds of confounding factors; and they thus require different manipulations to
become obvious and established. Mental images have a different role in art (they
may symbolize realities only remotely related to the image as such), geometry
(analogical to the object of interest), natural science (closely related as representations
of the realities of interest), and human interactions (laden with personal meaning
and emotions). Quantitative thought is the mental background for mathematics but
its function and uses in physics, economics, psychology, and history differ markedly.
Thus students must understand that core operations and processes in the various
domains are expressed differently in different knowledge domains. As a result their
modelling is the same at one level of analysis (i.e., the fundamental causal relations
are the same everywhere) but different at another level (i.e., content and form of
interactions between factors).
We suggest that the construction of new knowledge and skills and the
epistemological awareness that is necessary may well be served if organized around
the big ideas that were elaborated in different disciplines through the years. For
example, ideas such as gravity, energy, or force in physics, evolution, heredity, or
homeostasis in biology, or intelligence, motivation, or intention in psychology are
good examples of ideas that can be developed in the curriculum. Each of these ideas
may be presented from several perspectives. For example, they may be presented
along a dimension that varies from their direct relevance to personal experience and
everyday life to abstract models about underlying general laws that are remote from
personal experience and everyday life. The historical perspective is also important.
That is, it may be explained how each of these ideas was conceived in different times
in the history of a discipline. This will enable the students to see that the same
phenomenon may be differently understood over time as knowledge gets refined
because of the use of new methods and controls. The domains identified by
psychological research, such as the various SCSs, do not fully coincide with subjects
of knowledge in the curriculum or science. For instance, it was noted above that
causation is expressed differently in physics, biology, human relations, and history.
Therefore each of these ideas may be analysed and explained from the point of view
of different disciplines. This would highlight the differences in knowledge
production mechanisms of different disciplines; it would also show the multiplicity
of representations that may exist for the same aspect of the world. This would
greatly enhance cognitive flexibility and metarepresentational construction of
general inferential patterns of thought.
Conclusions
The main implications of this chapter may be summarized as follows. First, critical
thought is a general stance of inquiry rather than any specific cognitive process. One
might argue that intelligence and critical thought are distinct in the literature and
the testing practices rather than in reality. That is, intelligence as studied so far
comprises the processes used to integrate and make meaning of information, solve
problems, and make decisions. Critical thought comprises the metacognitive realiza-
tion that there may always be more information to integrate, alternative meanings
to be considered, better solutions to produce, and wiser or more useful decisions
to make.
Second, becoming critical is a long process that must overcome the weaknesses
of successive developmental phases. To be critical in preschool the child must learn
to envisage reality from alternative points of view. Anything can be represented in
more ways than one. In primary school the child must understand that all knowledge
is constructed and it may thus be fallible. There may always be a better rule to
organize observations. Adolescents must realize that reasoning and rationality are
not identical and reasoning may always err if not properly and exhaustively applied.
Late in adolescence and in college, students must embed analysis in historical and
epistemological perspective and recognize relativity of approach, based on the frame
chosen. Lapses of attention and reasoning, lags of knowledge, and weaknesses of
judgment may always outsmart everyone.
19
CONCLUSIONS
Towards an overarching theory
of the growing mind
At the beginning of this book we stated several questions we wanted to answer. In

this concluding chapter we summarize our answers as they emerge from the theory
advanced in this book. Hopefully, our answers will guide further research on the
growing mind. Obviously the theory of mind and development advanced in this
book owes much to earlier theories. In answering the various questions we outline
how our theory relates to other theories, focusing on the psychological realities each
theory prioritized in research.
The composition of the growing mind

The first question asked: what are the cognitive processes underlying intelligence? In
a nutshell, the research presented in this book suggested that all processes investigated
by the three traditions are needed to account for the functioning of the human mind.
However, no one is privileged in explaining its functioning or development. Thus
the four-fold architecture described in Chapter 8 is our answer to this question. To
reiterate, the human mind involves four types of mental processes:
• several domain-specific processes specializing in the representation and

processing of specific types of information and relations in the environment,
such as similarity, quantity, space, cause, persons;
• representation processes allowing preservation and updating of incoming
information (short-term storage capacity) for the sake of its processing and
integration with past information or current action;
• reasoning processes ensuring integration of information across time; senses, and
choices according to goals, filling in gaps and evaluating for
• cognizance, which ensures choices of choices, based on considerations of accuracy,
validity, and truth; that is, that choices and evaluations will be optimized, drawing
on past experience (including punishments and rewards), the options available,

and relevant criteria (logical, moral, interests, etc.).
The reader might have noticed that highly popular processes, such as attention,
executive control, and working memory were not mentioned. This is not because
they are underestimated. They are very important, but they are interactive products
of the processes mentioned before: attention control, including inhibition, reflects
awareness of a goal and a stimulus or an action; and executive control requires
awareness of a goal and a mental process, such as rehearsal in working memory, that
may be used to attain the goal.
The second question was concerned with the relations between these processes.
What is general, what is specific and what is holding them together? In the cognitive
tradition, most scholars would be neutral, probably arguing that any of the four
types of processes may be central, depending on the task at hand or the stage of
processing: for instance, attention at the beginning (when focusing on a problem),
working memory at a next stage (when representing the problem), and reasoning
at the end (when solving the problem). If pressed to choose one of the processes,
perhaps many scholars would select the central executive or working memory as a
whole. Differential psychologists do have a more straightforward answer. There are
several specialized processes which are hierarchically organized under g. What is in
g, however, is still under dispute. For Spearman it was basically inductive and
analogical reasoning. Piaget would agree but he would emphasize some core
processes such as reversibility. Neo-Piagetians would again invoke working memory
and executive control. For modern scholars g is the dynamic interaction between all of
these processes, with some of them probably in a more central role in these inter-
actions, such as working memory. Our studies showed that none of these processes
is a proxy of g. All of the central processes underlying broad domains, such as fluid
intelligence, spatial, or quantitative reasoning, are about equally and very highly
related to g (> 65% of common variance); all of the various fundamental processes
thought to be reductive of g (i.e., attention control, flexibility or shifting, and
working memory) may share a good (~15%-35%) portion of the variance of g but
none of them emerges as the sole predictor. Interaction may be the way out of the
impasse but it is not blind. It is directed by a mental mechanism holding the processes
together and in liaison with the environment.
We suggested that AACog (abstraction, alignment, and cognizance) stands in the
centre of the four-fold model, allowing and constraining the interaction of the four
types of systems. Therefore this is the mechanism underlying g. Minimally, AACog
allows search, interlinking, and reduction of information or representations into
new representations. AACog is the hard core of the language of thought (and
natural language) because it allows its basic syntactic principles (compositionality,
recursion, generativity, and hierarchical integration) to apply. AACog is compositional
because it puts informational or representational patterns together; it is recursive
because it can take the patterns over and over again and embed them into each
other, forming compositional chains; it is generative because it can introduce
Conclusions: towards an overarching theory 283
variations in any of the patterns; and it is hierarchical in that it can embed compositions
into reductions accounting for them.
In conclusion, there is a Spearmanian reality in the mind. General intelligence
does exist and it is a very powerful factor in intellectual functioning in real time,
intellectual development, and individual differences in rate and eventual achieve-
ments. Interestingly the processes involved are even more general than Spearman
himself assumed; his eduction of relations and correlates is an implementation of
AACog that must be constructed in development, like many other important
implementations. As a result, different measures of general mental ability, such as
the Raven test, the WISC test, and many other tests including academic achievement
tests, may reflect well the state of g in an individual but they are just like X-ray
pictures. They show bones or tissue structures but not their underlying components
or functioning.
The very operation of g involves a Kantian reality which escaped Spearman:
cognizance. “The highest principle of Kant’s theoretical philosophy is that all
cognition must ‘be combined in one single self-consciousness’” (Kitcher, 1999). We
showed here that there is a powerful cognizance mechanism intertwined with
search, alignment, and abstraction processes guiding and taking stock of their
functioning throughout developmental cycles. In fact, to a large extent, this
awareness defines the subjective aspect of mental functioning, raising it from simple
computation to representation where information and mental functioning is
subjectively meaningful.
The development of the mind

The third question was about development. What is stable and what is changing in
the mind as time goes by? AACog is always present. However, it is expressed
differently with development. Early in life it is expressed in statistical learning and
basic algebraic rule abstraction. The first generates concepts on the basis of frequency
and recurrence of properties; similarity is the most obvious expression of this
mechanism. The second generates frames of meaning that have an increasingly
obligatory character. As the infant moves through the second year of life, statistical
regularities and rules are aligned into episodic representations giving stability to the
infant’s conception of the world and allowing relatively complex interactions.
Awareness at this level is minute and transient, most probably used in mental
constructions as they take place rather than staying available and recalled later on.
Thus the major priority in the episodic cycle is the shift from the “here and now”
of present action and perception to the mental and the representational. This comes
with language and representational intelligence. Until then the episodic mind is
captive of environmental variation, guided by it as much as it errs because of it.
With entrance to the third year of life, toddlers move into an explicitly
representational world. Thus the major priority in the representational cycle is
attentional control and flexibility. Awareness of perceptual processes and the
perceptual origins of knowledge are the major markers of the formation of general
mental ability in this cycle. Gradually statistical learning ascends into inductions
and rule-based learning ascends into logical assertions, such as conjunctions or
disjunctions. The dimension of time comes in their inferences, allowing pragmatic
deals predating modus ponens reasoning. However, the realistic representational
mind is often trapped into the representational world in the fashion the episodic
mind is trapped in the sensory world, blurring the boundaries between imagination
and reality, enjoying the imaginary world as much as it may be deceived by it.
Inferences are systematically made in this cycle but their validity (for the preschool
child) derives from the dominant beliefs of the moment rather than from any
objective validity system.
The beginnings of a validity system appear in the rule-based mind. By the sixth
year of life the threads connecting representations are compacted into inferential
scaffolds which are evident in both inductive and deductive reasoning. In inductive
reasoning children make the relational shift, looking for relations between relations
(Gentner, 1983; Ratterman & Gentner, 1998); in deductive reasoning they reach
bi-conditional reasoning where modus ponens (p and q; p; q) and modus tollens
(p and q; not q; not p) are integrated into a common scheme (Barrouillet & Lecas,
1999: Taplin, Staudenmayer, & Taddonio, 1974). Concomitantly, children become
increasingly aware of the mental world, flexible in shifting between mental spaces
and able in relating them. Thus inductive inference and awareness of it are the major
markers of g. However, the representation of the world in this cycle often lacks
cohesion and the ability for logical validation may suffer because there are no general
principles to connect conceptual spaces and evaluate inferences.
The principle-based mind, in adolescence, grasps the principles connecting infer-
ential scaffolds of rule-based thought, cognizes mental processes explicitly, and it is
sensitive to its own mental strengths and weaknesses. Thus adolescents fully command
conditional reasoning, which is a major marker of g in this cycle (e.g., p and q; not
p; unknown if q). The principle-based mind adopts a suppositional stance allowing
the viewing of realities from multiple points of view. Illusion and error always loom,
however, because exhaustive consideration of alternatives and painstaking evaluation
of each evades most minds, even the most trained ones.
Thus, with development, AACog operates on different representations, abstract-
ing from, aligning, and cognizing different types of relations. Therefore psychometric
g is never the same in development. It is re-morphed in each developmental cycle
because it is infused by different processes. The processes infusing g tend to
intertwine with it and the processes that are consolidated tend to become freer from
its influence, as explicated in Chapter 11. Thus our theory resolves the differentiation
dispute.
In conclusion, there is also a Piagetian reality in the mind. There are four
developmental cycles with two phases in each that look close to the four Piagetian
stages of cognitive development (sensori-motor, preoperational, concrete, and
formal operational intelligence). These cycles are representationally rather than
logically defined. That is, they are distinguished from each other in reference to the
type of representation dominating in each cycle (i.e., episodic schemes, mentations,
rules, and principles) and by the relations connecting representations (i.e., spatially
and time-based associations, representational mappings, inferential links, truth- or
validity-based inferential constraints). However, although overlapping in time, they
follow a necessary sequence and each next cycle integrates all earlier ones.
However, there is a neo-Piagetian reality as well, which we call a Pascual-Leonian
reality, after Pascual-Leone (1970; Arsalidou & Pascual-Leone, 2016) who initiated
the neo-Piagetian movement. Specifically, development through the cycles is
associated with increasing mastery of complexity in the structures of information
that may be dealt with. We stress, however, that we take relational complexity as a
tool for analysing the representational dimensionality of concepts rather than the
representational capacity of the individual. The findings summarized here strongly
suggested that changes in reasoning are not driven by changes in working memory.
In fact, working memory was a rather weak player in the formation of general
mental ability in each cycle, especially in the early phase of each cycle. Differences
in representational resolution at successive cycles are associated with a different level
of complexity in the concepts that may be grasped because they point to different
underlying dimensions. This in turn is reflected in differences in the executive and
awareness profile of each phase. In line with this approach, Shipstead, Harrison,
and Engle (2016) maintained recently that working memory and fluid intelligence
involve processes that are not causally related but they are organized around top-
down processing goals in problem-solving. The first allows the person to represent
information so that solutions can be envisaged. The second involves the ability to
disengage from rejected solutions and envisage new ones. This explicates why the
role of cognizance and inference strengthens with development. Thus it might well
be that relational complexity may set constraints on what may be represented and
reflected on. However, the direction of causality may go either way, because
relational complexity may also expand because increasing resolution of cognizance
allows more focused scanning of and identification of dimensions in information
structures and more precise alignment and encoding into new concepts.
Why some people are smarter than others

The fourth question was about individual differences—why do individuals differ in
intelligence and learning possibilities?—and the fifth question concerned genome-
brain-mind relations. It is clear that individuals differ in all three important
dimensions influencing intellectual functioning and development: genes, the brain,
and the social environment. It is safe to conclude, based on the evidence discussed
in Chapters 13 and 14, that there are several genes which are explicitly related to
several aspects of brain structure, functioning, and development. These genes
constrain the formation, functioning, and development of various aspects of the
brain, which are important for the representational and processing capacity of
the brain. These include the formation of total brain volume, grey matter complexity
and white matter structure, and various aspects of connectivity, such as axons lengths
or synaptic formation. In turn these aspects of brain structure and functioning are
explicitly related to several aspects of general mental ability, such as abstraction and
handling novelty. Specifically, the brain builds neuronal networks and patterns of
activity to stand for realities, aligns and structurally or functionally connects them
into further networks, and abstracts patterns from them. Structural or functional
interlinking occurs by overlaying patterns of activation onto each other or tuning
them to co-activate in recursive sequences. Synaptic formation and brain rhythms
of various frequencies constitute links and syntactic elements underlying the
formation of these networks (Chapter 14). In these recursive brain activation
sequences, individual patterns are both increasingly refined and individually labelled
but also hierarchically integrated so that they may be individually picked up, if
needed, varied vis-à-vis each other, or co-activated in increasingly flexible new
combinations.
Constraints to variations and co-activations of this kind derive from the resolution
of access to the individual elements of network hierarchy and the refinement of
paths between them. The dominant level of organization (corresponding to episodic,
reality-based representations, rules, and principles) and related cognizance operate
as advanced pointer- and place-holder generators that may direct variations and
co-activations. For instance, activation of a rule-founding network can generate a
large variation of rule-instantiating cases and their connections; activation of a
principle-founding network can generate a large variation of principle-instantiating
rules, which in turn may activate relevant instantiation cases. However, as noted in
Chapter 13, there are no thoughts or mental processes in the genome or in the
brain. Genes are made up of molecular structures prescribing a developmental
programme for a body, including a brain among other organs; brains are made up
of neuronal structures prescribing how the environment may be represented and
how experience from it may be used to deal with recurrent or unexpected patterns
of information generated from interactions with the environment. In their turn
these aspects of the brain are explicitly related to real-world mental outcomes, such
as IQ and educational attainment. There is an environment at all levels though:
genes are expressed in a body which functions in a nurturing (or under- or mal-
nurturing) environment; brains interact with environments setting the experience
and problem context in which they operate; mental processes take place in learning
environments that contribute directly to their formation and development. It is
obvious that there is even a general human-kind cultural ideal for what is intelligent
and mentally necessary for individuals to function in human culture; this is served by
a general master plan implemented by education and both implicitly and explicitly
by other cultural institutions, such as the family. Obviously there may be differences
across cultures or social groups or individuals in a culture in how much they serve
this general ideal. All in all, general intelligence does exist all the way from genes to
culture: we showed that there is a genetic g, a neuronal (brain) g, a psychometric g,
a subjective (self) g, and a cultural g. Causality runs both ways, bottom-up and top-
down. Bottom up, genetic g is expressed in and causally affects brain g; brain g is expressed
and causally affects the psychometric g; in turn, psychometric g is expressed and causally affects
self g. Each of these different levels of g interact directly but also indirectly, through the rest,
with the environment, cultural and natural. As a result, effects also run top-down. Thus
differences in mind-related genes or the gene-related environment, brain structures
and brain-related environment, or the thought-related learning environment do
cause differences in intelligence and learning possibilities.
Specifying the relations between components of g within each of the levels above
(e.g., interactions between genes defining genetic g, or between brain regions
defining brain g, or between mental processes involved in psychometric g, or
between self-representations involved in subjective g) or components across these
levels (e.g., between genes, brain regions, mental processes, and self-representations)
is not an easy task. We showed in Chapter 14 that graph analysis is used to explore
how local, very specific, units are interlinked to form progressively more complex
networks which are themselves interconnected into increasingly inclusive systems
of networks. These networks may be specified at all levels of importance from the
perspective of this theory: genes, brain, cognition, social groups, and culture (see
Figure 14.3 in Chapter 14). We suggested that there are close analogies between
these networks on at least three levels: brain networks, cognitive-psychometric
networks, and self-representation networks. Moreover we suggested that the
changes in the relations between g and different forms of executive processes and
reasoning at successive cycles of cognitive development are related to changes in the
interlinking between nodes both within and across brain networks. Obviously we
have a long way to go before these assumed correspondences are established and
fully specified.
How is the developmental formation of these increasingly complex networks at
the brain and the cognitive level translated into actual differences between
individuals? The answer lies in the very nature of representations feeding into
AACog at the successive cycles of development. Specifically, each next cycle’s
representations are more difficult to visualize by the mind’s eye. For instance,
episodic representations have a compulsory nature of their own. Their properties
are directly readable from the physical stimuli. Relations between them are part of
their physical organization: same colour is physically present to the eyes; patterns in
colour or sound are physically present in their deployment in space or time. Thus
their alignment is directly guided by perceptual search; their abstraction only
requires encoding their physical similarity or pattern similarity; awareness of them
may emerge by the time the individual reiterates what was seen or heard or turns
to comparisons between matching episodes.
Implementing AACog on realistic representations is already more difficult by
several orders of magnitude. Abstraction over them requires holding them mentally
together; this may go astray for several reasons, including external interference or
forgetting; alignment requires mental search guided by an implicit rule used to hook
relevant instances as they appear. Cognizance of representations requires their
availability. Lack of any of them would cause delays in awareness and related meta-
representations that may be generated. For instance, lack of vision in congenitally
blind 4-year-old children causes delays in the emergence of theory of mind (Minter,
Hobson, & Bishop, 1998).
Implementing rule-based AACog is more demanding in many respects. Abstraction

here requires the relational shift that would direct search and alignment of relations
across representations rather than properties in representations. The options for
relating relations are, by definition, many more than relating representations because
pairs of representations may be related in many alternative ways, depending on their
properties selected for processing. Awareness of inferential processes requires
differentiating objects and representations from the mental processes applied on them.
This is often difficult, as indicated by the fact that children often focus on the content
characteristics of tasks rather than underlying processes.
Finally, implementing principle-based AACog is much more difficult than
implementing rule-based AACog for the very reasons stated above. That is,
abstracting over principles multiplies the options of abstracting over rules because
principles bridge several rules. Awareness of inferential processes underlying
principles requires by definition registering multiple processes. For instance,
awareness of mental processes involved in conditional reasoning requires activating
all four schemes discussed extensively in this book.
In conclusion, abstracting over representations and integrating them into higher
levels of executive and reasoning schemes becomes increasingly difficult with
increasing developmental level because options increase exponentially, rendering
errors more likely. In other words, a decreasing likelihood of attaining later develop-
mental levels is related to the very nature of the main factor of developmental
transition in cognitive development. In line with this interpretation, the probability
of attaining each next level of intellectual development at the age modally associated
with this level decreases in the general population. As a reminder, the reader is
referred to Figure 9.2 in Chapter 9 and Figure 10.1 in Chapter 10. It is particularly
notable that attainment of the cycle of principle-based thought is rare in the general
population, limited to the upper 5% of the population at 11–12 years and the upper
25% at 16–17 years. Therefore the sparsity of higher intelligence scores in the
population is associated with developmental deceleration. That is, higher scores of
intelligence require solving problems associated with later developmental phases.
Therefore high scores are constrained by developmental constraints.
One might ask here if there is any particular process that may be singled out as
a good index of individual differences in mental attainment and another process that
may be singled out as a good index of developmental attainment. Some scholars
suggested that processing speed is the individual differences index and executive control is the
developmental index (Anderson, 2017; Coyle, 2017). According to these scholars, at
any time, faster individuals are more privileged, compared to slower individuals, to
build better and wider brain networks and, by implication, cognitive networks
to serve meaning-making and problem-solving. This is due to their efficiency to
operate within almost always present time limits and construct the necessary
concepts or processes possible at the occasion or activate them later and properly
tune them to the specificities of the current occasion. The complexity of executive
control reached at a particular age reflects the upper limit of cognitive routines that
may be assembled under a goal at this age. Anderson (2017) suggested that speed
and executive control are grounded in different aspects of the brain: speed reflects
white matter integrity and relates to the establishment of long-distance connections;
it is primarily expressed via alpha rhythms in the brain and IQ which reflect
individual differences. Executive function reflects grey matter connectivity and it is
primarily expressed via theta rhythms integrating alpha rhythms into hierarchies of
activation, thereby building executive ensembles.
We would endorse this approach under a specific condition. Our research
summarized in Chapter 10 (see Figure 10.1) suggests that changes in speed predict
changes in reasoning at the beginning of cycles, when new representational units
are formed; changes in working memory predict changes at the end of cycles,
when the new representational units are interconnected into broader systems. Thus
there is a developmental and an individual differences dimension in both constructs.
Specifically, as an index of individual differences, speed may always reflect the ease
and efficiency in effecting new mental constructions by sculpting the necessary
new networks. In age windows associated with major cognitive transitions, speed
becomes, additionally, a developmental signifier. Working memory, as a major
envelope of executive control, may reflect the ease of interweaving and reshaping
networks into new ones, accelerating developmental progression. In age windows
associated with major cognitive expansions, working memory (and thus executive
control) becomes, additionally, an index of individual differences in that individ-
uals with better executive control can more efficiently build the cycle-specific
networks.
This very reason also explains the Flynn effect already discussed. In a sense the
Flynn effect is, at the level of general population, the inverse of the phenomenon
above at the individual level. Flynn ascribed the phenomenon to the expansion of
education and the increasing symbolic demands of technologically advanced
cultures. These changes drive individuals to use and refine relational thought related
to fluid intelligence. According to the present theory, the Flynn effect would be
associated with both direct training of relational thought but also reflection and
cognizance that would be required to metarepresent and organize knowledge
and problem-solving that was associated with social and educational changes in the
twentieth century.
The theory would also predict an inversion of the changes, with decreases in
secular IQ in the industrial nations. In line with this prediction, Flynn and Shayer
(submitted) found that the process underlying the Flynn effect came to an end in
Scandinavian and British children in about 1995 with a subsequent negative trend
of a projected 30-years drop of 6.85 IQ points. They also found that this drop is
associated with processes acquired at the transition between Piagetian concrete and
formal thought. That is, in our time, sound rule-based thought associated with
world exploration is compromised by emerging cultural practices that are not very
favourable for them. What social change might account for this year-by-year extra
decrease since 1995? Flynn and Shayer assume that this may be related to children’s
use of their spare time: hours of TV watching, computer gaming, and smartphones
have already replaced much old-fashioned “play”. Perhaps these new habits lessen
cognizance and metarepresentational needs for the individual as a search for location,
retrieval, and even evaluation and use of knowledge is delivered by smart machines.
Probably phase sensitive explicit instruction and training of these mechanisms along
the lines discussed in Chapters 17 and 18 may compensate for these new cultural
practices.
Instruction and learning

The last question was about learning. Can intelligence be enhanced by specific
interventions? Noticeably the same factor which accounts for decreasing likelihood
of advancing to higher developmental levels also explicates the effects of instruction
and learning. That is, development is boosted and accelerated when the AACog
processes are specifically targeted by instruction. The research presented in Chapter
16 showed that reasoning processes may be enhanced if training focuses specifically
on relational processes (abstraction and alignment) and cognizance. Thus this core
mechanism can be translated into all sorts of intellectual ensembles if domesticated
in special domains by specifically adding domain-specific qualifiers. We chose
deductive reasoning as an example to demonstrate that a type of inference that
appears general emerges as a specialization translating the central core above into a
representational ensemble of a specific set of relations in the environment that may
be expressed into a suitable syntax in language. In fact, the other learning studies
summarized in Chapter 16, which focused on mathematics, showed that learning
in every domain is a matter of translating and naturalizing the central core into the
routines and language of each domain. These studies also showed that enhancement
generalizes to other processes, such as working memory and attention control which
derive from these basic AACog processes.
However, these studies suggested that learning gains are developmentally specific
and, often, domain-specific. Affecting an earlier cycle would not necessarily transfer
to the next cycle or to another domain: the next cycle operates on a different level
of representation which needs special practice to be established; another domain may
require different skills to deal with the domain-specific representations and relations
involved. A cycle-specific learning programme may also change a process at the level
targeted, but gains do not fully consolidate unless they are embedded in the supportive
frame of operating at a next higher-level developmental cycle. Therefore, transfer to
processes specific to the next cycle would not be attained unless learning comes
repetitively in accordance with the needs of each cycle, until gains are locked into
the system as habitual ways of dealing with problems. Often this would occur if a
cycle’s achievements are locked into the operation of the next cycle. Practically
speaking, these findings imply that learning programmes must recycle along the cycles
of development themselves. That is, they must be tailored to successive developmental
cycles through the end, each time boosting the processes that relate to the emergence
and consolidation of each cycle, as explicated in Chapters 17 and 18.
If stated in terms of the graph modelling of networks outlined in Chapter 14,
efficient learning must satisfy two conditions. First, to obtain sustainability of learning
a particular cognitive process, both component nodes and hub nodes must be
specifically targeted in the module concerned. Training component nodes (e.g.,
training how to execute different arithmetic operations on different forms of fractions)
would ensure that the hub node (understanding the general concept of fraction)
would have enough connected components to stabilize; training hub nodes would
ensure that the component nodes would receive enough higher-level feedback to
endure. Second, to render a specific learning experience developmental, hub nodes
of the trained network must be connected to hub nodes (and probably also local
component nodes) of other related networks (e.g., the principle underlying the
concept of number as a variable). This would transfer the stabilization and endurance
gains noted above from the local to a higher level and thus stronger network.
Thus it seems that both developmental progression and learning depend on a
Gödelian reality. That is, each current phase acquires its full potential in the next
phase and each cycle comes to a closure only when moving into the first phase of
the next cycle. Thus, although there are four Piagetian-like cycles, there are only
three Gödelian cycles, if not extended into a post-principle cycle. The episodic cycle
comes to a closure only in the first phase of realistic representation because episodes
can be autonomously and independently examined only when representation allows
them to be revisited beyond “here and now”. Realistic representational thought
comes to a closure only in the first phase of rule-based thought. It is only then that
representations may be organized and intentionally used for purposes other than the
events or episodes that gave birth to them. Rule-based thought comes to a closure
only in the first phase of principle-based thought because it is only then that a
personal theory of rules can intentionally be used to evaluate incoming or self-
generated representations, concepts, decisions, and actions. It might, of course, be
the case that principle-based thought comes to a closure in an epistemological
environment where principles are arranged according to related general theories.
Personality and mind

How is intelligence related to personality? The will and the self-discipline of the
individual to set demanding mental and life goals and self-organize to pursue and
attain them are highly important. We showed that all aspects of personality which
relate to these forces of mental attainment are highly instrumental, conscientiousness
in particular. Additionally there is a strong link between mental functioning and
personality: self-concept. Self-concept carries information about how individuals
represent themselves across mental and personality dimensions and this is involved
in choices of activities, eventually expressed in actual performance in life-important
domains. Cognitive self-concept itself may be minimally predictive of actual
performance before adolescence. However, aspects of self-concept related to self-
esteem and executive processes in personality are highly related from rather early in
primary school.
In modern terms, the model summarized here shares a fundamental postulate
with many models that ascribe change to self-reflection which generates increasingly
higher levels of awareness. The present theory models the role of this factor in
development from infancy to adulthood, accommodating changes in reasoning and
other aspects of mental processing, such as working memory and intelligence,
and specifying how they may relate to personality. Kantian dynamics operate under
strong Freudian constraints. Cognizance is not always accurate and it is often only
weakly related to actual performance. This reflects the fact that mental processing
does not always reach awareness; when it does it is not properly recognized, it is
transient, and often twisted. Thus Freud’s unconscious always looms large to
swallow experiences having the potential to surface to consciousness. When seen
from a Freudian perspective, g is the cognitive ego which is shaped in every cycle
by cognitive experiences developing individuals become aware of.
The theory advanced in this book owes much to earlier theories. To honour
the lasting contributions of some of the great thinkers of the field, we named the
realities discovered by them and which have stood up well to the test of time.
Hopefully our theory constructively integrated these realities and expanded our
understanding of the growing mind further than the earlier theories allowed,
uncovering realities that passed unnoticed or better illuminating those already
known. The ultimate aim was to expand our understanding of the growing human
mind and support it to fully deploy and develop its potential.
REFERENCES
Abu-Akel, A., & Shamay-Tsoory, S. (2011). Neuroanatomical and neurochemical bases of

theory of mind. Neuropsychologia, 49, 2971–2984.
Ackerman, P. L., Beier, M. E., & Boyle, M. O. (2005). Working memory and intelligence:
The same or different constructs? Psychological Bulletin, 131, 30–60.
Alexander, P., Dumas, D., Grossnickle, E. M., List, A., & Firetto, C. M. (2015). Measuring
relational reasoning. The Journal of Experimental Education. doi:10.1080/00220973.2014.
963216.
Allport, G. W. (1937). Personality: A psychological interpretation. Oxford: Holt.
Anderson, J. R., & Fincham, J. M. (2014). Extending problem-solving procedures through
reflection. Cognitive Psychology, 74, 1–34.
Anderson, M. (2015). After phrenology: Neural reuse and the interactive brain. New York:
Bradford.
Anderson, M. (2017). Binet’s error: Developmental change and individual differences in
intelligence are related to different mechanisms. Journal of Intelligence, 5, 24. doi:10.3390/
jintelligence5020024.
Anderson, M., Kinnison, J., & Pescoa, L. (2014). Describing functional diversity of brain
regions and brain networks. Neuroimage, 73, 50–58.
Andreou, M. (2009). The development of emotional intelligence and the influence of cognitive
development on emotional intelligence. Doctoral Dissertation. University of Cyprus.
Andrews, G., Halford, G. S., Bunch, K. M., Bowden, D., & Jones, T. (2003). Theory of
mind and relational complexity. Child Development, 74, 1478–1499.
Antinori, A., Carter, O. L., & Smillie, L. D. (2017). Seeing it both ways: Openness to
experience and binocular rivalry suppression. Journal of Research in Personality, 68, 15–22.
Arffa, S. (2007). The relationship of intelligence to executive function and non-executive
function measures in a sample of average, above average, and gifted youth. Archives of
Clinical Neuropsychology, 22, 969–978.
Arsalidou, M., & Pascual-Leone, J. (2016). Constructivist developmental theory is needed in
developmental neuroscience. npj Science of Learning, 1, 16016; doi:10.1038/npjscilearn.
2016.16.
Asbury, K., Wachs, T. D., & Plomin, R. (2005). Environmental moderators of genetic
influence on verbal and nonverbal abilities in early childhood. Intelligence, 33, 643–661.
294 References
Asendorph, Y., & van Aken, M. (2003). Validity of Big Five Personality judgments in
childhood: A 9-year longitudinal study. European Journal of Personality, 17, 1–17.
Au, J., Buschkuehl, M., Duncan, G. J., Jaeggi, S. M. (2015a). There is no convincing evidence
that working memory training is NOT effective: A reply to Melby-Lervåg and Hulme
(2015). Psychonomic Bulletin & Review. pmid:26518308.
Au, J., Sheehan, E., Tsai, N., Duncan, G. J., Buschkuehl, M., Jaeggi, S. M. (2015b). Improving
fluid intelligence with training on working memory: A meta-analysis. Psychonomic Bulletin
& Review, 22, 366–377.
Baars B. J. (1989). A cognitive theory of consciousness. Cambridge: Cambridge University
Press.
Baars, B. J. (1997). In the Theater of Consciousness. New York: Oxford University Press.
Baddeley, A. D. (1990). Human memory: Theory and practice. Hillsdale: Erlbaum.
Baddeley, A. D. (2000). The episodic buffer: A new component of working memory? Trends
in Cognitive Sciences, 4, 417–423.
Baddeley, A. D. (2012). Working memory: Theories, models, and controversies. Annual
Review of Psychology, 63, 1–29.
Baddeley, A. D., & Hitch, G. H. (2000). Development of working memory: Should the
Pascual-Leone and the Baddeley and Hitch models be merged? Journal of Experimental and
Child Psychology, 77, 128–137.
Baillargeon, R. (1995). Physical reasoning in infancy. In M. S. Gazzaniga (editor-in-chief),
The cognitive neurosciences (pp. 181–204). Cambridge: The MIT Press.
Baillargeon, R., Scott, R. M., & Bian, L. (2016). Psychological reasoning in infancy. Annual
Review of Psychology, 67, 159–186.
Baldwin, J. M. (1968). Mental development in the child and the race methods and processes. New
York: Augustus M. Kelley (Original work published 1894).
Baltes, P. B. (1991). The many faces of human aging: Toward a psychological culture of old
age. Psychological Medicine, 21, 837–854.
Bardin, J. (2012). Neurodevelopment: Unlocking the brain, Nature, 487(7405), 24–26.
Barrett, H. C., & Kurzban, R. (2006). Modularity in cognition: Framing the debate.
Psychological Review, 113, 628–647.
Barrouillet, P., & Lecas, J.-F. (1999). Mental models in conditional reasoning and working
memory. Thinking & Reasoning, 5, 289–302.
Barrouillet, P., Gavens, N., Vergauwe, E., Gaillard, V., & Camos, V. (2009). Working
memory span development: A time-based resource-sharing model account. Developmental
Psychology, 45, 477–490.
Barrouillet, P., Grosset N., Lecas J.-F. (2000). Conditional reasoning by mental models:
Chronometric and developmental evidence. Cognition, 75, 237–266.
Becker, A. H., & Ward, T. B. (1991). Children’s use of shape in extending novel labels to
animate objects: Identity versus postural change. Cognitive Development, 6(1), 3–16.
Beilin, H., & Lust, B. (1975). A study of the development of logical and linguistics connectives:
Linguistics data. In Harry Beilin (Ed.), Studies in the cognitive basis of language development
(pp. 76–120). New York: Academic Press.
Benoit, L., Lehalle, H., Molina, M., Tijus, C., & Jouen, F. (2013). Young children’s mapping
between arrays, number words, and digits. Cognition, 129, 95–101. doi:10.1016/j.
cognition.2013.06.005.
Benyamin, B., Pourcan, B. St., (2014). Childhood intelligence is heritable, highly polygenic
and associated with FNBP1L. Molecular Psychiatry, 19, 253–258.
Bjorklund, D. F., & Harnishfeger, K. K. (1995). The evolution of inhibition mechanisms
and their role in human cognition and behavior. In F. N. Dempster & C. J. Brainerd
(Eds.), Interference and inhibition in cognition (pp. 141–173). New York: Academic Press.
References 295
Black, P., & Wiliam, D. (2009). Developing the theory of formative assessment. Educational
Assessment, Evaluation, and Accountability, 21, 5–13.
Blair, C. (2006). How similar are fluid cognition and general intelligence? A developmental
neuroscience perspective on fluid cognition as an aspect of human cognitive ability.
Behavioral and Brain Sciences, 29, 109–160.
Bliss, J., & Ogborn, J. (1994). Force and motion from the beginning. Learning and Instruction,
4, 7–25.
Bloom, P., & German, T. P. (2000). Two reasons to abandon the false belief task as a test of
theory of mind. Cognition, 77, B25–B31.
Boly, M., Seth, A. K., Wilke, M., Ingmundson, P., Baars, B., Laureys, S., Edelman, D. B.,
& Naotsugu, T. (2013). Consciousness in humans and non-human animals: Recent
advances and future directions, Frontiers in Psychology. doi.org/10.3389/fpsyg.2013.00625.
Boot, W. R., Blakely, D., P., & Simons, D. J. (2011). Do action video games improve
perception and cognition? Frontiers in Psychology. doi:10.3389/fpsyg.2011.00226.
Bouchard, T. J. (2004). Genetic influence on human psychological traits. Directions in
Psychological Science, 13, 148–151.
Bouchard, T. J., & McGue, M. (2003). Genetic and environmental influences on human
psychological differences. Developmental Neurobiology, 54, 4–45.
Boyd, R., & Silk, J. B. (2014). How humans evolved. New York: Norton.
Brainder, C. J. (1972). Neo-Piagetian training experiments revisited: Is there any support for
the cognitive-developmental stage hypothesis? Cognition, 2, 349–370.
Braine, M. D. S. (1990). The “natural logic” approach to reasoning. In W. F. Overton (Ed.),
Reasoning, necessity, and logic: Developmental perspectives (pp. 133–157). Hillsdale: Erlbaum.
Braine, M. D. S., & Rumain, B. (1983). Logical reasoning. In J. H. Flavell & E. M. Markman
(Eds.), Handbook of child psychology (Vol. III). Cognitive development. New York: Wiley.
Brainerd, C. J. (1973). Neo-Piagetian training experiments revisited: Is there any support for
the cognitive-developmental stage hypothesis? Cognition, 2, 349–370.
Brainerd, C. J. (1978). The stage question in cognitive-developmental theory. The Behavioral
and Brain Sciences, 2, 173–2013.
Braithwaite, D. W., Pyke, A. A., & Siegler, R. S. (2017). A Computational model of fraction
arithmetic. Psychological Review, 124, 603-625. doi.org/10.1037/rev0000072.
Bressler, S. L., & Menon, V. (2010). Large-scale brain networks in cognition: Emerging
methods and principles. Trends in Cognitive Sciences, 14, 277–290.
Broadmann, K. (1909). Vergleichende lokalisationslehre der grosshirnrindle. Leipzig: Barth.
Brod, G., Bunge, S. A., & Shing, Y. L. (2017). Does one year of schooling improve children’s
cognitive control and alter associated brain activation? Psychological Science. doi:10.1177/
0956797617699838.
Brody, N. (2008). Does education influence intelligence? In P. C. Kyllonen, R. D. Roberts,
& L. Stankov (Eds.), Extending intelligence: Enhancement and new constructs (pp. 85–91). New
York: Lawrence Erlbaum Associates.
Brooks, R., & Meltzoff, A. N. (2015). Connecting the dots from infancy to childhood: A
longitudinal study connecting gaze following, language, and explicit theory of mind.
Journal of Experimental Child Psychology, 130, 67–78.
Brown, J. D. (1998). The self. New York: McGraw-Hill.
Bryant, P. E., & Trabasso, T. (1971). Transitive inference and memory in young children.
Nature, 232, 456–458.
Brydges, C. R., Reid, C. L., Fox, A. M., & Anderson, M. (2012). A unitary executive
function predicts intelligence in children. Intelligence, 40, 458–469.
Burkart, J. M., Schubinger, M. N., & van Schaik, C. P. (2017). The evolution of general
intelligence. Behavioral and Brain Sciences, 40, 1–67. doi:10.1017/S0140525X16000959,
e195.
296 References
Buschkuehl, M., & Jaeggi, S. M. (2010). Improving intelligence: A literature review. Swiss
Medical Weekly, 140(19–20), 266−272.
Buss, A. H., & Plomin, R. (1984). Temperament: Early developing personality traits. London:
Wiley.
Butler, H. A., Pentoney, C., & Bong, M. P. (2017). Predicting real-world outcomes: Critical
thinking ability is a better predictor of life decisions than intelligence. Thinking Skills and
Creativity. http://dx.doi.org/10.1016/j.tsc.2017.06.005
Butterworth, G. (1998a). Origins of joint visual attention in infancy: Commentary on
Carpenter et al. Monographs of the Society for Research in Child Development, 63(4),
144–166.
Butterworth, G. (1998b). Perceptual and motor development. In A. Demetriou, W. Doise,
& C. F. M. van Lieashout (Eds.), Life-span developmental psychology (pp. 101–136).
Chichester: Wiley & Sons.
Buzsaki, G., & Brendon, W. O. (2012). Brain rhythms and neural syntax: Implications for
efficient coding of cognitive content and neuropsychiatric disease. Dialogues in Clinical
Neuroscience, 14, 345–367.
Cahan, E. D. (1984). The genetic psychologies of James Mark Baldwin and Jean Piaget.
Developmental Psychology, 20, 128–135.
Campbell, F. A., & Burchinal, M. (2008). Early childhood interventions: The Abecedarian
Project. In P. C. Kyllonen, R. D. Roberts, & L. Stankov (Eds.), Extending intelligence:
Enhancement and new constructs (pp. 61–84). New York: Lawrence Erlbaum Associates/
Taylor & Francis Group.
Cao, M., Wang, J.-H., Dai, Z.-J., Cao, X.-Y., Jiang, L.-L., Fan, F.-M., Song, X.-W., Xia,
M.-R., Shu, N., Dong, Q., Milham, M. P., Castellanos, F. X., Zuo, X.-N., & He, Y.
(2014). Topological organization of the human brain functional connectome across the
lifespan. Developmental Cognitive Neuroscience, 7, 76–93.
Caravita, S., & Hallden, O. (1994). Re-framing the problem of conceptual change. Learning
and Instruction, 4, 89–111.
Carey, S. (1985). Conceptual change in childhood. Cambridge: MIT Press.
Carey, S. (2009). The origin of concepts. New York: Oxford University Press.
Carlozzi, N. E., Tulsky, D. S., Kail, R. V., & Beaumont, J. L. (2013). VI. NIH Toolbox
Cognition Battery (CB): Measuring processing speed. Monographs of the Society for Research
in Child Development, 78, 88–102.
Caro, T. M., & Hauser, M. D. (1992). Is there teaching in nonhuman animals? Quarterly
Review of Biology, 67, 151–174.
Carpendale, J. I., & Chandler, M. J. (1996). On the distinction between false belief
understanding and subscribing to an interpretive theory of mind. Child Development, 67,
1686–1706.
Carraher, T. N., Carraher, D., & Schliemann, A. D. (1985). Mathematics in the streets and
in schools. British Journal of Developmental Psychology, 3, 21–29.
Carroll, J. B. (1993). Human cognitive abilities: A survey of factor-analytic studies. New York:
Cambridge University Press.
Carroll, J. B. (1997). Psychometrics, intelligence, and public perception. Intelligence, 24,
25–52.
Carruthers, P. (2002). The cognitive functions of language & author’s response: Modularity,
language, and the flexibility of thought. Behavioral and Brain Sciences, 25, 657–719.
Carruthers, P. (2006). The architecture of the mind: Massive modularity and the flexibility of thought.
Oxford: Oxford University Press.
Carruthers, P. (2009). How we know our own minds: The relationship between mindreading
and metacognition. Behavioral and Brain Sciences, 32, 121–182.
References 297
Case, R. (1985). Intellectual development. Birth to adulthood. New York: Academic Press.
Case, R. (1992). The mind’s staircase: Exploring the conceptual underpinnings of children’s thought
and knowledge. Hillsdale: Erlbaum.
Case, R., & Okamoto, Y. (1996). The role of central conceptual structures in the development
of children’s thought. Monographs of the Society for Research in Child Development, 61, (1–2,
Serial No. 246).
Case, R., Demetriou, A., Platsidou, M., & Kazi, S. (2001). Integrating concepts and tests of
intelligence from the differential and the developmental traditions, Intelligence, 29,
307–336.
Case, R., Okamoto, Y., Griffin, S., McKeough, A., Bleiker, C., Henderson, B., &
Stephenson, K. M. (1996). The role of central conceptual structures in the development
of children’s thought. Monographs of the Society for Research in Child Development, 61, (1–2,
Serial No. 246).
Casey, B. J. Tottenham, N., Liston, C., and Durston, S. (2005). Imaging the developing
brain: What have we learned about cognitive development? Trends in Cognitive Sciences,
9, 104–110.
Cattell, R. B. (1957). Personality and motivation structure and measurement. Oxford, UK: World
Book Co.
Cattell, R. B. (1963). Theory of fluid and crystallized intelligence: A critical experiment.
Journal of Educational Psychology, 54, 1–22.
Cattell, R. B. (1965). The Scientific Analysis of Personality. New York: Penguin Group.
Cattell, R. B., & Horn, J. L. (1978). A check on the theory of fluid and crystallized intelligence
with description of new subtest designs. Journal of Educational Measurement, 15, 139–164.
Ceci, S. J. (1991). How much does schooling influence general intelligence and its cognitive
components? Developmental Psychology, 27, 703–722.
Chalvo, D. R. (2014). Critical brain dynamics at large scale. In E. Niebur, D. Plenz, &
H. G. Schuster (Eds.), Criticality in neural systems (pp. 1–24). London: Wiley.
Chamorro-Premuzic, T., & Furnham, A. (2006). Intellectual competence and the intelligent
personality: A third way in differential psychology. Review of General Psychology, 10,
251–267.
Chandler, M., Fritz, A. S., & Hala, S. (1989). Small scale deceit: Deception as a marker of
2-, 3-, and 4-year-olds early theories of mind. Child Development, 60, 1263–1277.
Chandler, M. J., Hallett, D., & Sokol, B. W. (2002). Competing claims about competing
knowledge claims. In B. K. Hofer & P. R. Pintrich (Eds.), Personal epistemology: The
psychology of beliefs about knowledge and knowing (pp. 145–168). Mahwah, NJ: Lawrence
Erlbaum Associates.
Chein, J. M., & Morrison, A. B. (2010). Expanding the mind’s workspace: Training and
transfer effects with a complex working memory span task. Psychonomic Bulletin & Review,
17, 193–199.
Cheng, P. W., & Holyoak, K. J. (1985). Pragmatic reasoning schemas. Cognitive Psychology,
17, 391–416.
Cheung, A. K., Harden, K. P., & Tucker-Drob, E. M. (2015). From specialist to generalist:
Developmental transformations in the genetic structure of early child abilities.
Developmental Psychobiology, 57, 566–553.
Chevalier, N., & Blaye, A. (2016). Metacognitive monitoring of executive control
engagement during childhood. Child Development, 87, 1264–1276. doi:10.1111/
cdev.12537.
Chomsky, N. (1986). Knowledge of language: Its nature, origin and use. New York: Praeger.
Christoforides, M., Spanoudis, G., & Demetriou, A. (2016). Coping with logical fallacies:
A developmental training program for learning to reason. Child Development, 87,
1856–1876.
298 References
Chuderska, A., & Chuderski, A. (2009). Executive control in analogical reasoning: Beyond
interference resolution. Proceedings of the Annual Meeting of the Cognitive Science Society, 31.
Chuderski, A. (2013). When are fluid intelligence and working memory isomorphic and
when are they not? Intelligence, 41, 244–262.
Chugani, H. T., Phelps, M. E., & Mazziotta, J. C. (1987). Positron emission tomography
study of human brain functional development. Annals of Neurology, 22, 487–497.
Cimpian, A., Hammond, M. D., Mazza, G., & Corry, G. (2017). Young children’s self-
concepts include representations of abstract traits and the global self. Child Development,
88, 1786-1798.
Cohn, L. D., & Westenberg, P. M. (2004). Intelligence and maturity: Meta-analytic evidence
or the incremental and discriminant validity of Loevinger’s measure of ego development.
Journal of Personality and Social Psychology, 86, 760–772.
Colby, J. B., O’Hare, E. D., Bramen, J. E., & Sowell, E. R. (2013). Structural brain
development. In J. Rubenstein & R. Pasko (Eds.), Neural circuit development and function in
the brain: Comprehensive developmental neuroscience (pp. 207–230). Burlington: Academic
Press.
Cole, M., Gay, J., Glick, J., & Sharp, D. W. (1971). The cultural context of learning and thinking.
New York: Basic Books.
Comalli, P. E. Jr., Wapner, S., & Werner, H. (1962). Interference effects of Stroop color-
word test in childhood, adulthood, and aging. The Journal of Genetic Psychology, 100,
47–53.
Commons, M. L., & Rodriguez, J. A. (1990). “Equal access” without “establishing” religion:
The necessity for assessing social perspective-taking skills and institutional atmosphere.
Developmental Review, 10, 323–340.
Cosmides, L., & Tooby, J. (1994). Origins of domain specificity: The evolution of functional
organization. In L. Hirschfeld & S. Gelman (Eds.), Mapping the mind: Domain specificity in
cognition and culture (pp. 85–116). Cambridge: Cambridge University Press.
Costa, P. T., Jr., & McCrae, R. R. (1997). Longitudinal stability of adult personality. In
R. Hogan, J. Johnson, & S. Briggs (Eds.), Handbook of personality psychology (pp. 269–290).
San Diego: Academic Press.
Couchman, J. J., Beran, M. J., Coutinho, M. V. C., Boomer, J., & Smith, D. J. (2012).
Evidence for animals metaminds. In M. J. Beran, J. L. Brandl, J. Perner, & J. Proust (Eds.),
Foundations of metacognition (pp. 61–97). Oxford: Oxford University Press.
Cowan, N. (2010). The magical mystery four: How is working memory capacity limited,
and why? Current Directions in Psychological Science, 19(11), 51–57.
Cowan, N. (2016). Working memory maturation: Can we get at the essence of cognitive
growth? Perspectives on Psychological Science, 11(2), 239–64. doi:10.1177/1745691615621279.
Cowey, C. M. (1996). Hippocampal sclerosis on working memory. Memory, 4, 19–30.
Coyle, T. R. (2017). A differential-developmental model (DDM): Mental speed, attention
lapses, and general intelligence (g). Journal of Intelligence, 5, 25. doi:10.3390/
Cramer, P. (2006). Ego functions and ego development: Defense mechanisms and intelligence
as predictors of ego level. Journal of Personality, 67, 735–760.
Cramer, P. (2007). Longitudinal study of defense mechanisms: Late childhood to late
adolescence. Journal of Personality, 75, 1–23.
Cramer, P. (2008). Identification and the development of competence: A 44-year longitudinal
study from late adolescence to late middle age. Psychology and Aging, 23, 410–421.
Cramer, P. (2015a). Defense mechanisms: 40 years of empirical research. Journal of Personality
Assessment, 97, 114–122.
Cramer, P. (2015b). IQ and defense mechanisms assessed with the TAT. Rorschachiana, 36,
40–57.
References 299
Crick, F. C., & Koch, C. (2005). What is the function of the claustrum? Philosophical
Transactions of the Royal Society: Brain and Biological Sciences, 30, 1271–1279.
Csapó, B. (1992). Improving operational abilities in children. In A. Demetriou, M. Shayer,
& A. Efklides (Eds.), Neo-Piagetian theories of cognitive development. Implications and applications
for education (pp. 144–159). London: Routledge.
Currie, J. & Duncan, T. (1995). Does Head Start make a difference? American Economic
Review, 85(3), 341–364.
Daniels, P. T., & Bright, W. (1996). The world’s writing systems. Oxford: Oxford University
Press.
Dasen, P. R. (1977). Piagetian psychology: Cross-cultural contributions. New York: Gardner
Press.
Dasen, P. R. (1994). Culture and cognitive development from a Piagetian perspective. In
W. J. Lonner & R. Malpass (Eds.), Psychology and culture (pp. 141–150). Boston: Allyn and
Bacon.
de Ribaupierre, A., & Bailleux, C. (1994). Developmental change in a spatial task of
attentional capacity: An essay toward an integration of two working memory models.
International Journal of Behavioral Development, 17, 5–35.
de Ribaupierre, A., & Bailleux, C. (1995). Development of attentional capacity in childhood:
A longitudinal study. In F. E. Weinert & W. Schneider (Eds.), Memory performance and
competencies: Issues in growth and development (pp. 45–70). Hillsdale: Erlbaum.
de Ribaupierre, A., & Pascual-Leone, J. (1984). Pour une intégration des methods en
psychologie: Approaches expérimentale, psycho-génétique et différentielle. L´Année
Psychologique, 84, 227–250.
Deak, G. O., & Wiseheart, M. (2015). Cognitive flexibility in young children: General or
task-specific capacity? Journal of Experimental Child Psychology, 138, 31–53.
Deary, I. J., Egan, V., Gibson, G. J., Austin, E. J., Brand, C. R., & Kellaghan, T. (1996).
Intelligence and the differentiation hypothesis. Intelligence, 23, 105–132.
Deary, I. J., Strand, S., Smith, P., & Fernandes, C. (2007). Intelligence and educational
achievement. Intelligence, 35, 13–21. doi:10.1016/j.intell.2006.02.001.
Dehaene, S. (2011). The number sense (2nd ed.). New York: Oxford University Press.
Dehaene, S. (2014). Consciousness and the brain: Deciphering how the brain codes our thoughts. New
York: Penguin.
Dehaene, S., & Nacchache. L. (2001). Towards a cognitive neuroscience of consciousness:
Basic evidence and a workspace framework. Cognition, 79, 1–37.
DeLoache, J. S. (2000). Dual representation and young children’s use of scale models. Child
Development, 71, 329–338.
DeLoache, J. S., Miller, K. F., & Pierroutsakos, S. L. (1998). Reasoning and problem solving.
In W. Damon (Ed.), Handbook of child psychology: Volume 2: Cognition, perception, and
language, (pp. 801–850). Hoboken: John Wiley & Sons Inc.
Demetriou, A. (1990). Structural and developmental relations between formal and postformal
capacities: Towards a comprehensive theory of adolescent and adult cognitive develop-
ment. In M. L. Commons, C. Armon, L. Kohlberg, F. A. Richards, T. A. Grotzer, &
J. D. Sinnott (Eds.), Adult development: Vol. 2., Models and methods in the study of adolescent
and adult thought (pp. 147–173). New York: Praeger.
Demetriou, A. (1998). Cognitive development. In A. Demetriou, W. Doise, & K. F. M. van
Lieshout (Eds.), Life-span developmental psychology (pp. 179–269). London: Wiley.
Demetriou, A., (2000). Organization and development of self-understanding and self-
regulation: Toward a general theory. In M. Boekaerts, P. R. Pintrich, & M. Zeidner
(Eds.), Handbook of self-regulation (pp. 209–251). Academic Press.
300 References
Demetriou A., & Bakracevic K. (2009). Reasoning and self-awareness from adolescence to
middle age: Organization and development as a function of education. Learning and
Individual Differences, 19, 181–194.
Demetriou, A., & Efklides, A. (1981). The structure of formal operations: The ideal of the
whole and the reality of the parts. In J. A. Meacham & N. R. Santilli (Eds.), Social
development in youth: Structure and content (pp. 20–46). Basel: Karger.
Demetriou, A., & Efklides, A. (1985). Structure and sequence of formal and postformal
thought: General patterns and individual differences. Child Development, 56, 1062–1091.
Demetriou, A., & Efklides, A. (1988). Experiential structuralism and neo-Piagetian theories:
Toward an integrated model. In A. Demetriou (Ed.), The neo-Piagetian theories of cognitive
development: Toward an integration (pp. 173–222). Amsterdam: North-Holland.
Demetriou A, & Efklides, A. (1989). The person’s conception of the structures of develop-
ing intellect: Early adolescence to middle age. Genetic, Social, and General Psychology
Monographs, 115, 371–423.
Demetriou, A., & Kazi, S. (2001). Unity and modularity in the mind and the self: Studies on the
relationships between self-awareness, personality, and intellectual development from childhood to
adolescence. London: Routledge.
Demetriou, A., & Kazi, S. (2006). Self-awareness in g (with processing efficiency and
reasoning). Intelligence, 34, 297–317.
Demetriou, A., Kyriakides, L., & Avraamidou, C. (2003). The missing link in the relations
between intelligence and personality. Journal of Research in Personality, 37, 547–581.
Demetriou, A., & Kyriakides, L. (2006). A Rasch-measurement model analysis of cognitive
developmental sequences: Validating a comprehensive theory of cognitive development.
British Journal of Educational Psychology, 76, 209–242.
Demetriou, A., & Raftopoulos, A. (1999). Modeling the developing mind: From structure
to change. Developmental Review, 19, 319–368.
Demetriou, A., & Spanoudis, G. (2017). Mind and intelligence: Integrating developmental,
psychometric, and cognitive theories of human mind. In M. Rosén (Ed.), Challenges in
educational measurement—contents and methods, (pp. 39–60). New York: Springer.
Demetriou, A., Christou, C., Spanoudis, G., Pittalis, M., & Mousoulides, N. (in preparation).
Learning to reason in mathematics.
Demetriou, A., Christou, C., Spanoudis, G., & Platsidou, M. (2002). The development of
mental processing: Efficiency, working memory, and thinking. Monographs of the Society
of Research in Child Development, 67, Serial Number 268.
Demetriou, A., Efklides, A., & Gustafsson, J. (1992). Training, cognitive change, and
individual differences, in A. Demetriou, M. Shayer, & A. Efklides (Eds.), Neo-piagetian
theories of cognitive development (pp. 122–144), New York: Routledge.
Demetriou, A., Efklides, A., & Platsidou, M. (1993). The architecture and dynamics of
developing mind: Experiential structuralism as a frame for unifying cognitive developmental
theories. Monographs of the Society for Research in Child Development, 58, Serial Number 234.
Demetriou, A., Kui, Z. X., Spanoudis, G., Christou, C., Kyriakides, L., & Platsidou, M.
(2005). The architecture, dynamics, and development of mental processing: Greek,
Chinese, or universal? Intelligence, 33, 109–141.
Demetriou, A., Mouyi, A., & Spanoudis, G. (2008). Modeling the structure and development
of g. Intelligence, 5, 437–454.
Demetriou, A., Mouyi, A., & Spanoudis, G. (2010). The development of mental processing.
In W. F. Overton (Ed.), Biology, cognition and methods across the lifespan. Vol. 1: Handbook
of life-span development (pp. 306–343), Editor-in-chief: R. M. Lerner. Hoboken: Wiley.
Demetriou, A., Pachaury, A., Metallidou, Y., & Kazi. S. (1996). Universal and specificities
in the structure and development of quantitative-relational thought: A cross-cultural study
in Greece and India. International Journal of Behavioral Development, 19, 255–290.
References 301
Demetriou, A., Spanoudis, G., & Mouyi, A. (2011). Educating the developing mind:
Towards an overarching paradigm. Educational Psychology Review, 23(4), 601–663.
Demetriou, A., Spanoudis, G., & Shayer, M. (2014). Inference, reconceptualization, insight,
and efficiency along intellectual growth: A general theory. Enfance, issue 3, 365–396.
Demetriou, A., Spanoudis, G., Kazi, S., Mouyi, A., Žebec, M. S., Kazali, E., Golino, H.,
Bakracevic, K., & Shayer, M. (2017). Developmental differentiation and binding of
mental processes with g through the life-span. Journal of Intelligence. 5, 23. doi:10.3390/
Demetriou, A., Spanoudis, G., Kazi, S., Žebec, M., & Andreou, M. (submitted). Intelligence,
personality, and emotions: Deciphering their interactions in development from 7 to 22
years of age.
Demetriou, A., Spanoudis, G., Shayer, M., Mouyi, A., Kazi, S., & Platsidou, M. (2013).
Cycles in speed-working memory-G relations: Towards a developmental-differential
theory of the mind. Intelligence, 41, 34–50. doi:10.1016/j.intell.2012.10.010.
Demetriou, A., Spanoudis, G., Shayer, M., van der Ven, S., Brydges, C. R., Kroesbergen,
E., Podjarny, G., & Swanson, H. L. (2014). Relations between speed, working memory,
and intelligence from preschool to adulthood: Structural equation modeling of 15 studies.
Intelligence, 46, 107–121.
Dempster, F. N. (1991). Inhibitory processes: A neglected dimension of intelligence.
Dempster, F. N. (1992). The rise and fall of the inhibitory mechanism: Toward a unified
theory of cognitive development and aging. Developmental Review, 12, 45–75.
Dempster, F. N. (1993). Resistance to interference: Developmental changes in a basic
processing mechanism. In M. L. Howe & R. Pasnak (Eds.), Emerging themes in cognitive
development, Vol.1: Foundations (pp. 3–27). New York: Springer.
Destan, N., & Roebers, C. M. (2015). What are the metacognitive costs of young children’s
overconfidence? Metacognition and Learning, 10, 347–374.
Detterman, D. K. (1987). Theoretical notions of general intelligence and mental retardation.
American Journal of Mental Deficiency, 92, 2–11.
Diamond, A. (2013). Executive functions. Annual Review of Psychology, 64, 135–168.
Donaldson, M. (1978). Children’s minds. Glasgow: Fontana/Collins.
Dumas, D., & Alexander, P. A. (2016). Calibration of the test of relational reasoning.
Psychological Assessment, 28, 1303–1318.
Dumontheil, I. (2014). Development of abstract thinking during childhood and adolescence:
The role of rostrolateral prefrontal cortex. Developmental Cognitive Neuroscience, 10, 57–76.
Duncan, J., Chylinski, D., Mitchell, D. J., & Bhandari, A. (2017). Complexity and
compositionality in fluid intelligence. Proceedings of the National Academy of Sciences. https://
doi.org/10.1073/pnas.1621147114.
Edelman, G. M., & Tononi, G. A. (2000). Universe of consciousness. New York: Basic
Books.
Efklides, A. (2008). Metacognition: Defining its facets and levels of functioning in relation
to self-regulation and co-regulation. European Psychologist, 13, 277–287.
Efklides, A., Demetriou, A., & Gustafsson, J.-E. (1992). Training, cognitive change, and
individual differences. In A. Demetriou, M. Shayer, & A. Efklides (Eds.), Modern theories
of cognitive development go to school (pp. 122–143). London: Routledge.
Ektstrom, R. B., French, J. W., & Harman, H. H. (1976). Manual for Kit of factor-referenced
cognitive tests. Princeton, NJ: Educational Service.
Emerson, H. F. (1980). Children’s judgments of correct and reversed sentences with “if.”
Journal of Child Language, 7, 137–155.
Ennis, R. H. (1962). A concept of critical thinking. Harvard Educational Review, 32, 81–111.
302 References
Ennis, R. H. (1996). Critical thinking. Upper Saddle River: Prentice-Hall.

Epstein, H. T. (1986). Stages in human development. Developmental Brain Research, 30,
114–119.
Erdle, S., Irwing, P., Rushton, J. P., & Park, J. (2010). The general factor of personality and
its relation to self-esteem in 628,640 internet respondents. Personality and Individual
Differences, 48, 343–346.
Evans, J. (2003). In two minds: dual-process accounts of reasoning. Trends in Cognitive
Sciences, 7(10), 454–459. doi:10.1016/j.tics.2003.08.012.
Eysenck, H. (1997). Dimensions of personality. London: Routledge.
Fabricius, W. V., & Schwanenflugel, P. J. (1994). The older child’s theory of mind. In
A. Demetriou & A. Efklides (Eds.), Intelligence, mind, and reasoning: Structure and development
(pp. 111–132). Amsterdam: North-Holland.
Facon, B. (2006). Does age moderate the effect of IQ on the differentiation of cognitive
abilities during childhood? Intelligence, 34, 375–386.
Falmagne, R. J. (1990). Situations, statements, and logical relations. Paper presented at the 20th
annual symposium of the Jean Piaget Society, Philadelphia, June.
Fischer, K. W. (1980). A theory of cognitive development: The control and contribution of
hierarchies of skills. Psychological Review, 87, 477–531.
Fischer, K. W., & Bidell, T. R. (1998). Dynamic development of psychological structures in
action and thought. In R. Lerner (Ed.), W. Damon (Series Ed.), Handbook of child
psychology (5th ed.): Vol. 1: Theoretical models of human development. New York: Wiley.
Flage, D. E. (1990). David Hume’s theory of mind. London: Routledge.
Flavell, J. H. (1963). The developmental psychology of Jean Piaget. New York: D. Van Nostrand.
Flavell, J. H. (1979). Metacognition and cognitive monitoring: A new area of cognitive
developmental inquiry. American Psychologist, 34, 906–911.
Flavell, J. H., Green, F. L., & Flavell, E. R. (1986). Development of knowledge about the
appearance reality distinction. Monographs of the Society for Research in Child Development,
51, (Serial No. 212).
Flavell, J. H., Green, F. L., & Flavell, E. R. (1995). Young children’s knowledge about
thinking. Monographs of the Society for Research in Child Development, 60(1) (Serial No. 243).
Flavell, J. H., Green, F. L., Wahl, K. E., & Flavell, E. R. (1986). The effects of question
clarification and memory aids on young children’s performance on appearance-reality
tasks. Cognitive Development, 2, 127–144.
Fleming, K. A., Heintzelman, A., J., & Bartholow, B. D. (2015). Specifying associations
between conscientiousness and executive functioning: Mental set shifting, not-prepotent
response inhibition or working memory updating. Journal of Personality. doi:10.1111/
jopy.12163.
Flynn, J. R. (2009). What is intelligence: Beyond the Flynn effect. Cambridge: Cambridge
University Press.
Flynn, J. R., & Shayer, M. (submitted). IQ decline and Piaget: Does the rot start at the top?
Fodor, J. A. (1975). The language of thought. Hassocks: Harvester Press.
Fodor, J. A. (2008). LOT 2: The language of thought revisited. Cambridge: MIT Press.
Fonlupt, P. (2003). Perception and judgement of physical causality involve different brain
structures. Cognitive Brain Research, 17, 248–254.
Fortey, R. (1997). Life: An unauthorised biography. New York: Harper & Collins.
Franson, P., Aden, U., Blenow, M., & Lagercrantz, H. (2011). The functional architecture
of the infant brain as revealed by resting-state fMRI. Cerebral Cortex January, 21,
145–154.
Freud, A. (1966). The ego and the mechanisms of defence. London: Karnac Books.
Freud, S. (1927). The ego and the id. London: Hogarth Press and Institute of Psycho-Analysis.
References 303
Friedman, H. R., & Goldman-Rakic, P. S. (1988). Activation of the hippocampus and

dentate gyrus by working memory: A 2-deoxyglucose study of behaving rhesus monkeys.
The Journal of Neuroscience, 8, 4693–4706.
Frith, U, & Frith, C. D. (2003). Development and neurophysiology of mentalizing.
Philosophical Transactions of the Royal Society of London. Series B, Biological Sciences, 358,
459–473.
Frith, U., & Happe, F. (1999). Theory of mind and self-consciousness: What is it like to be
autistic? Mind & Language, 14, 1–22.
Fry, A. F., & Hale, S. (1996). Processing speed, working memory, and fluid intelligence:
Evidence for a developmental cascade. Psychological Science, 7, 237–241.
Fuster, J. M. (2002). Frontal lobe and cognitive development. Journal of Neurocytology, 31,
373–385.
Galaburda, A. M. (2002). The neuroanatomy of categories. In A. M. Galaburda, S. M.
Kosslyn, & Y. Christen (Eds.), The languages of the brain (pp. 23–42). Cambridge: Harvard
University Press.
Gallup, G. G. (1982). Self-awareness and the emergence of mind in primates. American Journal
of Primatology, 2, 237–248.
Galotti, K. M., Komatsu, L. K., & Voelz, S. (1997). Children’s differential performance on
deductive and inductive syllogisms. Developmental Psychology, 33, 70–78.
Gardner, H. (1983). Frames of mind: The theory of multiple intelligences. New York: Basic Books.
Garon, N. (2016). Review of hot executive functions in preschoolers. Journal of Self-Regulation
and Regulation, 2, 57–79. doi:10.11588/josar.2016.2.34354.
Garon, N., Bryson, S. E., & Smith, I. M. (2008). Executive function in preschoolers: A
review using an integrative framework. Psychological Bulletin, 134, 31–60.
Gauffroy, C., & Barrouillet, P. (2009). Heuristic and analytic processes in mental models for
conditionals: An integrative developmental theory. Developmental Review, 29, 249–282.
doi:10.1016/j.dr.2009.09.002.
Gauthier, I., Tarr, M. J., Moylan, J., Skudlarski, P., Gore, J. C., & Anderson, A. W. (2000).
The fusiform “face area” is part of a network that processes faces at the individual
level. Journal of Cognitive Neuroscience, 12(3), 495–504.
Gelman, R. (1990). First principles organize attention to and learning about relevant data:
Number and the animate inanimate distinction as examples. Cognitive Science, 14, 79–106.
Gelman, R., & Gallistel, R. (1978). The child’s understanding of number. Cambridge: Harvard
University Press.
Gelman, S. A. (1988). The development of induction within natural kind and artifact
categories. Cognitive Psychology, 20, 65–95.
Gelman, S. A. (2003). The essential child: Origins of essentialism in everyday thought. Oxford:
Oxford University Press.
Gelman, S. A. (2005). The essential child: Origins of essentialism in everyday thought. Oxford:
Gelman, S. A., & Coley, J. D. (1990). The importance of knowing a dodo is a bird: Categories
and inferences in 2-year-old children. Developmental Psychology, 26, 796–804.
Gentner, D. (1983). Structure-mapping: A theoretical framework for analogy. Cognitive
Science, 7, 155–170.
Gentner, D. (2005). The development of relational category knowledge. In L. Gershkoff-
Stowe & D. H. Rakison (Eds.), Building object categories in developmental time (pp. 245–275).
Hillsdale: Erlbaum.
German, T. P., & Hehman, J. A. (2006). Representational and executive selection resources
in “theory of mind”: Evidence from compromised belief-desire reasoning in old age.
Cognition, 101, 129–152.
304 References
Gignac, G. E. (2014). Dynamic mutualism versus g factor theory: An empirical test.

Ginsburg, H., & Opper, S. (1988). Piaget’s theory of intellectual development (3rd ed.). Englewood
Cliffs: Prentice-Hall.
Gladwin, T. (1970). East is a big bird. Cambridge: Harvard University Press.
Glasser, M. F., Coalson, T. S., Robinson, E. C., Hacker, C. D., Harwell, J., Yacoub, E.,
Ugurbil, K., Anderson, J., Beckmann, C., Jenkinson, M., Smith, S. M., & Van Essen, D.
C. (2016). A multi-modal parcellation of human cerebral cortex. Nature, 536(7615),
171–178.
Goel, V. (2007). Anatomy of deductive reasoning. Trends in Cognitive Science, 11, 435–441.
http://dx.doi.org/10.1016/j.tics.2007.09.003.
Goel, V., & Dolan, R. J. (2000). Anatomical segregation of component processes in an
inductive inference task. Journal of Cognitive Neuroscience, 12, 1–10.
Goel, V., & Dolan, R. J. (2001). Functional neuroanatomy of three-term relational reasoning.
Neuropsychologia, 39, 901–909.
Goel, V., Buchel, C., Frith, C., & Dolan, R. J. (2000). Dissociation of mechanisms underlying
syllogistic reasoning. NeuroImage, 12, 504–514.
Goswami, U. (1992). Analogical reasoning in children. Hillsdale: Erlbaum.
Goswami, U. & Brown A. L. (1989) Melting chocolate and melting snowmen: Analogical
reasoning and causal relations. Cognition, 35, 69–95.
Gottfredson, L. S. (2016). A g theorist on Why Kovacs and Conway’s Process Overlap
Theory amplifies, not opposes, g theory. Psychological Inquiry, 27, 210–217.
Gottlieb, G. (2007). Probabilistic epigenesis. Developmental Science, 10(1), 1–11. doi:10.1111/
j.1467-7687.2007.00556.x.
Graziano, W. G., Jensen-Campbell, L. A., & Finch, J. F. (1997). The self as a mediator between
personality and adjustment. Journal of Personality and Social Psychology, 73, 392–404.
Grimm, K. J., Ram, N., & Hamagani, F. (2011). Nonlinear growth curves in developmental
research. Child Development, 82, 1357–1371.
Gruber, O., & Goschke, T. (2004). Executive control emerging from dynamic interactions
between brain systems mediating language, working memory and attentional processes.
Acta Psychologica, 115, 105–121.
Gruber, O., & von Cramon, Y. D. (2003). The functional neuroanatomy of human working
memory revisited. Evidence from 3-T fMRI studies using classical domain-specific
interference tasks. NeuroImage, 19, 797–809.
Gustafsson, J.-E. (1984). A unifying model for the structure of intellectual abilities. Intelligence,
8, 179−203.
Gustafsson, J.-E. (2008). Schooling and intelligence: Effects of track of study on level and
profile of cognitive abilities. In P. C. Kyllonen, R. D. Roberts, & L. Stankov (Eds.),
Extending intelligence: Enhancement and new constructs (pp. 37–59). New York: Lawrence
Erlbaum Associates.
Gustafsson, J.-E., & Undheim, J. O. (1996). Individual differences in cognitive functions. In
D. C. Berliner & R. C. Calfee (Eds.), Handbook of educational psychology (pp. 186–242).
New York: Macmillan.
Halford, G. S., & Boulton-Lewis, G. M. (1992). Value and limitations of analogs in teaching
mathematics. In A. Demetriou, A. Efklides, & M. Shayer (Eds.), Neo-Piagetian theories of
cognitive development: Implications and applications for education (pp. 183–209). London:
Routledge.
Halford, G. S., Cowan, N., & Andrews, G. (2007). Separating cognitive capacity from
knowledge: A new hypothesis. Trends in Cognitive Science, 11, 236–242.
Halford, G. S., Maybery, M. T., O’Hare, A. W., & Grant, P. (1994). The development of
memory and processing capacity. Child Development, 65, 1330–1348.
References 305
Halford, G. S., Wilson, W. H., Andrews, G., & Phillips, S. (2014). Categorizing cognition:
Toward conceptual coherence in the foundations of psychology. Cambridge: MIT Press.
Hall, P. A., & Fong, G. T. (2013). Conscientiousness versus executive function as predictors
of health behaviors and health trajectories. Annals of Behavioral Medicine, 45, 398–399.
Hall, P. A., Fong, G. T., & Epp, L. J. (2013). Cognitive and personality factors in the
prediction of health behaviors: An examination of total, direct and indirect effects. Journal
of Behavioral Medicine. Advance online publication. doi:10.1007/s10865-013-9535-4.
Halpern, D. F. (2006). The nature and nurture of critical thinking. In R. J. Sternberg, H. L.
Roediger III, & D. F. Halpern (Eds.), Critical thinking in psychology (pp. 1–14). Cambridge:
Hansell, N. K., Halford, G. S., Andrews, G., Shum, D. H. K., Harris, S. E., Davies, G.,
Franic, S., Christoforou, A., Zietsch, B., Painter, J., Medland, S. E., Ehli, E. A., Davies,
G. E., Steen, V. M., Lundervold, A. J., Reinvang, I., Montgomery, G. W., Espeseth, T.,
Hulshoff H. E., Starr, J. M., Martin, N. M., Le Hellard, S., Boomsma, D. I., Deary, I. J.,
& Wright, M. J. (2015). Genetic basis of a cognitive complexity metric. Plos One.
doi:10.1371/journal.pone.0123886.
Hanson, J. L., Hair, N., Shen, D. G., Shi, F., Gilmore, J. H., Wolfe, B. L., & Pollak, S. D.
(2013). Family poverty affects the rate of human infant brain growth. Plos One, 8, 12.
e80954, doi:10.1371/journal.pone.0080954.
Harnishfeger, K. K. (1995). The development of cognitive inhibition: Theories, definitions,
and research evidence. In F. N. Dempster & C. J. Brainerd (Eds.), Interference and inhibition
in cognition (pp. 175–204). New York: Academic Press.
Harris, I. M., Egan, G. F., Sonkkila, C., Tochon-Danguy, H. J., Paxinos, G., & Watson, D. G.
(2000). Selective right parietal lobe activation during mental rotation. Brain, 123, 65–73.
Harris, P. L. (2007). Trust. Developmental Science, 10, 135–138.
Harris, P. L., & Núñez, M. (1996). Understanding of permission rules by preschool children.
Child Development, 67, 1572–1591.
Harris, P. L., Brown, E., Marriott, C., Whittall, S., & Harmer, S. (1991). Monsters, ghosts
and witches: Testing the limits of the fantasy–reality distinction in young children. British
Journal of Developmental Psychology, 9, 105–123.
Harter, S. (2012). The construction of the self: Developmental and sociocultural foundations (2nd ed.).
New York: Guilford Press.
Hartman, P. (2006). Spearman’s Law of Diminishing Returns: A look at age differentiation.
Journal of Individual Differences, 27, 199–207.
Hattie, J. (1992). Measuring the effects of schooling. Australian Journal of Education, 36(1),
5–13.
Haxby, J. V., Ungerleider, L. G., Horwitz, B., Maisog, J. M., Rapoport, S. I., & Grady, C. L.
(1996). Face encoding and recognition in the human brain. Proceedings of the National
Academy of Sciences, 93, 922–927.
Heavey, C. L., & Hurlburt, R. T. (2008). The phenomena of inner experience. Consciousness
and Cognition. doi:10.1016/j.concog.2007.12.006.
Hermer, L., & Spelke, E. (1996). Modularity and development: The case of spatial
reorientation. Cognition, 61, 195–232.
Herrnstein, R. J., & Murray, C. (1994). The bell curve: Intelligence and class structure in American
life. New York: The Free Press.
Heyman, G. D. (2008). Children’s critical thinking when learning from others. Current
Directions in Psychological Science, 17, 344–347.
Hill, P. L., & Jackson, J. J. (2016). The invest-and-accrue model of conscientiousness. Review
of General Psychology, 20, 141–154.
306 References
Hill, W. D., Davirs, G., McIntosh, A. M., Gale, C. R., & Deary, I. J. (2017). A combined
analysis of genetically correlated traits identifies 107 loci associated with intelligence.
bioRxiv preprint, 7 July. doi: http://dx.doi.org/10.1101/160291.
Hoff, G. E. A-J., van den Heuvel, M. P., Benders, M. J. N. L., Kersbergen, K. J., & De Vries,
L. S. (2013). On development of functional brain connectivity in the young brain.
Frontiers in Human Neuroscience. doi:10.3389/fnhum.2013.00650.
Holland, J., Holyoak, K., Nisbett, R., & Thagard, P. (1989). Induction: Processes of inference,
learning, and discovery. Cambridge: MIT Press/Bradford Books.
Hudspeth, W. J., & Pribram, K. H. (1992). Psychophysiological indices of cerebral maturation.
International Journal of Psychophysiology, 12, 19–29.
Hunt, E. (2011). Human intelligence. Cambridge: Cambridge University Press.
Hupp, J. H., & Sloutsky, V. M. (2011). Learning to learn: From within-modality to cross-
modality transfer during infancy. Journal of Experimental Child Psychology, 110, 408–421.
Hurlburt, R. (1993). Sampling inner experience with disturbed affect. New York: Plenum Press.
Hwang, K., Hallquist, M. N., & Luna, B. (2012). The development of hub architecture in
the human functional brain network. Cerebral Cortex. doi:10.1093/cercor/bhs227.
Hy, L. X., & Loevinger, J. (1996). Measuring ego development. London: Psychology Press.
Ide, J. S., Shenoy, P., Yu, A. J., & Li, C. S. (2013). Bayesian prediction and evaluation in
the anterior cingulate cortex. The Journal of Neuroscience: The Official Journal of the Society
for Neuroscience, 33, 2039–2047.
Inagaki, K., & Hatano, G. (2002). Young children’s naive thinking about the biological world. New
York: Psychology Press.
Inhelder, B., & Piaget, J. (1958). The growth of logical thinking from childhood to adolescence. New
York: Basic Books. (Original work published 1955).
Inhelder, B., & Piaget, J. (1969). The early growth of logic in the child: Classification and seriation.
New York: Norton. (Original work published 1959).
Jaeggi, S. M., Buschkuehl, M., Jonides, J., & Perrig, W. J. (2008). Improving fluid intelligence
with training on working memory. Proceedings of the National Academy of Sciences of the
United States of America, 105(19), 6829–6833.
James, W. (1890). Principles of psychology. New York: Henry Holt & Co.
Jensen, A. R. (1998). The g factor: The science of mental ability. Westport: Praeger.
Jensen, A. R. (2006). Clocking the mind: Mental chronometry and individual differences. Amsterdam:
Elsevier.
Jensen, O., & Lisman, J. E. (2005). Hippocampal sequence-encoding driven by cortical
multi-item working memory buffer. Trends in Neurosciences, 28, 67–72.
Johansson, B. S., & Sjolin, B. (1975). Preschool children’s understanding of the coordinators
“and” and “or”. Journal of Experimental Child Psychology, 19, 233–240.
Johnson, J., Pascual-Leone, J., & Agostino, A. (2001). Solving multiplication word problems: The
role of mental attention. Presented at the meeting of the Society for Research in Child
Development, Minneapolis.
Johnson, J., Im-Bolter, N., & Pascual-Leone, J. (2003). Development of mental attention in
gifted and mainstream children: The role of mental capacity, inhibition, and speed of
processing. Child development, 74, 1594–1614.
Johnson, M. H. (2011). Interactive specialization: A domain-general framework for human
functional brain development? Developmental Cognitive Neuroscience, 1, 7–21.
Johnson-Laird, P. N. (1983). Mental models: Towards a cognitive science of language, inference, and
consciousness. Cambridge: Harvard University Press.
Johnson-Laird, P. N. (2006). How we reason. Oxford: Oxford University Press.
Johnson-Laird, P. N. (2012). Inference with mental models. In K. J. Holyoak & R. G.
Morrison (Eds.), The Oxford handbook of thinking and reasoning (pp. 134–145). New York:
References 307
Johnson-Laird, P. N., & Khemlani, S. S. (2014). Toward a unified theory of reasoning. In

B. H. Ross (Ed.), The psychology of learning and motivation. Elsevier Inc: Academic Press,
1–42.
Johnson-Laird, P. N., and Wason, P. C. (1970). A theoretical analysis of insight into a
reasoning task. Cognitive Psychology, 1, 134–138.
Jonides, J., Lacey, S. C., & Nee, D. E. (2005). Processes of working memory in mind and
brain. Current Directions of Psychological Science, 14, 2–5.
Jung, R. E., & Haier, R. J. (2007). The Parieto-Frontal Integration Theory (P-FIT) of
intelligence: Converging neuroimaging evidence. Behavioral and Brain Sciences, 30,
135–187.
Kahneman, D. (2011). Thinking: Fast and slow. New York: Farrar, Straus and Giroux.
Kail, R. (1991). Developmental functions for speed of processing during childhood and
adolescence. Psychological Bulletin, 109, 490–501.
Kail, R. (2000). Speed of information processing: Developmental change and links to
intelligence. Journal of School Psychology, 38, 51–61.
Kail, R., (2007). Longitudinal evidence that increases in processing speed and working
memory enhance children’s reasoning. Psychological Science, 4, 312–313.
Kail, R. V., & Ferrer, E. (2007). Processing speed in childhood and adolescence: Longitudinal
models for examining developmental change. Child Development, 78, 1760–1770.
Kail, R., & Salthouse, T. A. (1994). Processing speed as a mental capacity. Acta Psychologica,
86, 199–225.
Kail, R. V., Lervåg, A., & Hulme, C., (2015). Longitudinal evidence linking processing speed
to the development of reasoning. Developmental Science, 1–8. doi:10.1111/desc.12352.
Kaldy, Z., & Leslie, A. M. (2005). A memory span of one? Object identification in
6.5-month-old infants. Cognition, 97, 153–177.
Kane, M. J., Bleckley, M. K., Conway, A. R. A., & Engle, R. W. (2001). A controlled-
attention view of working-memory capacity. Journal of Experimental Psychology: General,
130, 169–183.
Kant, I. (1902). Kants gesammelte schriften, Akademie Ausgabe, 29 vols. Ed. Königlichen
Preussichen Akademie der Wissenschaften. Berlin and Leipzig: Walter de Grunter and
presecessors.
Kanwisher, N., McDermott, J., & Chun, M. M. (1997). The fusiform face area: A module
in human extrastriate cortex specialized for face perception. The Journal of Neuroscience, 17,
4302–4311.
Kargopoulos, P., & Demetriou, A. (1998). Logical and psychological partitioning of mind:
Depicting the same picture? New Ideas in Psychology, 16, 61–88 (with commentaries by J.
Pascual-Leone, M. Bickhard, P. Engel, and L. Smith).
Kaufman, S. B., DeYoung, C. G., Reis, D. L., & Gray, J. R. (2011). General intelligence
predicts reasoning ability even for evolutionarily familiar content. Intelligence, 39, 311–322.
Kazali, E. (2016). Development of inductive and deductive reasoning from 4 to 10 years: Interactions
with executive control and cognizance. Doctoral dissertation. Panteion University of Social
Sciences, Athens, Greece.
Kazi, S., Demetriou, A., Spanoudis, G., Zhang, X. K., & Wang, Y. (2012). Mind–culture
interactions: How writing molds mental fluidity. Intelligence, 40, 622–637.
Keil, F. C. (1989). Concepts, kinds, and cognitive development. Cambridge: MIT Press.
Keith, T. Z., Fine, J. G., Taub, G. E., Reynolds, M. R., & Kranzler, J. H. (2006). Higher
order, multisample, confirmatory factor analysis of the Wechsler Intelligence Scale for
Children, (4th ed.): What does it measure? School Psychology Review, 35, 108–127.
Kemps, E., De Rammelaere, S., & Desmet, T. (2000). The development of working
memory: Exploring the complementarity of two models. Journals of Experimental Child
308 References
Kiefer, M., & Pulvermüller, F. (2012). Conceptual representations in mind and brain:
Theoretical developments, current evidence and future directions. Cortex, 48, 805–825.
Kievit, R. A., Davis, S. W., Griffiths, J., Correia, M. M., Cam-Can, & Henson, R. N.
(2016). A watershed model of individual differences in fluid intelligence. Neuropsychologia,
91, 186–198.
King, P. M., & Kitchener, K. S. (2002). The reflective judgment model: Twenty years of
research on epistemic cognition. In B. K. Hofer & P. R. Pintrich (Eds.), Personal
epistemology: The psychology of beliefs about knowledge and knowing (pp. 37–61). Mahwah, NJ:
Lawrence Erlbaum Associates.
Kitcher, P. (1999). Kant on self-consciousness. Philosophical Review, 108(3), 345–386.
Kitcher, P. (2011). Kant’s thinker. Oxford: Oxford University Press.
Klauer, K. J., & Phye, G. (1994). Cognitive training for children: A developmental program of
inductive reasoning and problem solving. Seattle: Hogrefe & Huber.
Klauer, K., & Phye, G. D. (2008). Inductive reasoning: A training approach. Review of
Educational Research, 78(1), 85–123.
Klingberg, T. (1998). Concurrent performance of two working memory tasks: Potential
mechanisms of interference. Cerebral Cortex, 8, 593–601.
Koch, C. (2012). Consciousness: Confessions of a romantic reductionist. Cambridge: MIT Press.
Koenig, M. A., & Harris, P. L. (2005). Preschoolers mistrust ignorant and inaccurate speakers.
Child Development, 76, 1261–1277.
Kohlberg, L., Levine, C., & Nucci, L. (1983). Moral stages: A current formulation and a
response to critics. Contributions to human development, Vol. 10. Basil: Karger.
Kosslyn, S. M. (1980). Image and mind. Cambridge: Harvard University Press.
Kovacs, K., & Conway, A. R. A. (2016). Process overlap theory: A unified account of the
general factor of intelligence. Psychological Inquiry, 27, 151–177.
Kovas, Y., Haworth, C. M. A., Harlaar, N., Petrill, S. A., Dale, P. S., & Plomin, R. (2007).
Overlap and specificity of genetic and environmental influences on mathematics and
reading disability in 10-year-old twins. Journal of Child Psychology and Psychiatry, 48,
914–922.
Krumm, S., Ziegler, M., & Buehner, M. (2008). Reasoning and working memory as
predictors of school grades. Learning and Individual Differences, 18, 248–257.
Kuhn, D. (1999). A developmental model of critical thinking. Educational Researcher, 28,
16–25.
Kuhn, D. (2005). Education for thinking. Cambridge: Harvard University Press.
Kyllonen, P., & Christal, R. E. (1990). Reasoning ability is (little more than) working
memory capacity? Intelligence, 14, 389–433.
Kyllonen, P., & Kell, H. (2017). What is fluid intelligence? Can it be improved? In
M. Rosén, K. Yang Hansen, & U. Wolff (Eds.), Cognitive abilities and educational outcomes.
methodology of educational measurement and assessment (pp. 15–37). Cham: Springer.
Kyriakides, L., & Luyten, H. (2009). The contribution of schooling to the cognitive
development of secondary education students in Cyprus: An application of regression-
discontinuity with multiple cut-off points. School Effectiveness and School Improvement,
20(2), 167–186.
Lackner, C., Sabbagh, M. A., Hallinan, E., Liu, X., & Holden, J. J. A. (2012). Dopamine
receptor D4 gene variation predicts preschoolers’ developing theory of mind. Developmental
Science, 15, 272–280.
Lake, B. M., Salakhutdinov, R., & Tenenbaum, J. B. (2015). Human-level concept learning
through probabilistic program induction. Science, 350, 1332–1338.
Lamb, M. E., Chuang, S. S., Wessles, H., Broberg, A. G., & Hwang, C. P. (2002). Emergence
and construct validation of the Big Five factors in early childhood: A longitudinal analysis
of their ontogeny in Sweden. Child Development, 73, 1517–1524.
References 309
Lamme, V. A. F. (2006). Towards a true neural stance on consciousness. Trends in Cognitive

Science, 10, 494–501.
Landau, B., Smith, L. B., & Jones, S. S. (1988). The importance of shape in early lexical
learning. Cognitive Development, 3(3), 299–321.
Lehman, D., Lembert, R. O., & Nisbett, R. E. (1988). The effects of graduate training on
reasoning: Formal discipline and thinking about everyday-life events. American Psychologist,
43, 431–442.
Leslie, A. M., Friedman, O., & German, T. P. (2004). Core mechanisms in “theory of
mind.” Trends in Cognitive Sciences, 8, 528–533.
Levesque, H. J. (1989). A knowledge-level account of abduction. Toronto: Dept. of Computer
Science, University of Toronto.
Liddle, B., & Nettle, D. (2006). Higher-order theory of mind and social competence in
school-age children. Journal of Cultural and Evolutionary Psychology, 4, 231–246.
Lisman, J. (2005). The theta/gamma discrete phase code occuring during the hippocampal
phase precession may be a more general brain coding scheme. Hippocampus, 15,
913–922.
Livingstone, M., & Hubel, D. (1988). Segregation of form, color, movement, and depth:
Anatomy, physiology, and perception. Science, 240, 740–749.
Loevinger, J. (1976). Ego development: Conceptions and theories. San Francisco: Jossey-Bass.
Lourenço, O. (2016). Developmental stages, Piagetian stages in particular: A critical review.
New Ideas in Psychology, 40, 123–137.
Luria, A. (1968). The mind of a mnemornist: A little book for a vast memory. Cambridge: Harvard
University Press.
Luria, A. (1976). Cognitive development: Its cultural and social foundations. Cambridge: Harvard
University Press.
Lynn, R. (2008). The global bell curve: Race, IQ, and inequality worldwide. Washington:
Washington Summit Publishers.
Lyons, K. E., & Zelazo, P. D. (2011). Monitoring, metacognition, and executive function:
Elucidating the role of self-reflection in the development of self-regulation. In J. Benson
(Ed.), Advances in child development and behavior, 40, (pp. 379–412). Burlington: Academic
Press.
Lysaker, P., Dimaggio, G., & Brüne, M. (2014). Social cognition and metacognition in
schizophrenia: Psychopathology and treatment approaches. New York: Academic Press.
Machery, E. (2016). The amodal brain and the offloading hypothesis. Psychological Bulletin &
Review, 23, 1090–1095. doi:10.3758/s13423-015-0878-4.
Mackey, A. P., Hill, S. S., Stone, S. I., & Bunge, S. A. (2010). Differential effects of reasoning
and speed training in children. Developmental Science, 14, 582–590.
Mackey, A. P., Miller Singley, A. T., & Bunge, S. A. (2013). Intensive reasoning training
alters patterns of brain connectivity at rest. The Journal of Neuroscience, 33, 4796–4803.
Mackey, A. P., Park, A. T., Robinson, S. T., & Gabrieli, J. D. E. (2017). A pilot study of
classroom-based cognitive skill instruction: Effects on cognition and academic
performance. Mind, Brain, and Education, 11, 85–95.
Mackintosh, N. J. (1998). IQ and human intelligence. Oxford: Oxford University Press.
MacLeod, C. M. (1991). Half a century of research on the Stroop Effects: An integrative
review. Psychological Bulletin, 109, 163–203.
MacLeod, C. M., & Dunbar, K. (1988). Training and Stroop-like interference: Evidence for
a continuum of automaticity. Journal of Experimental Psychology: Learning, Memory, and
Cognition, 14, 126–135.
Macnamara, J. (1986). A border dispute: The place of logic in psychology. Cambridge: MIT
Press.
310 References
Mahy, C. E. V., Moses, L. J., & Pfeifer, J. H. (2014). How and where: Theory-of-mind in
the brain. Developmental Cognitive Neuroscience, 9, 68–81.
Makris, N. (1995). Personal theory of mind and its relation with cognitive abilities. Doctoral
dissertation. Aristotle University of Thessaloniki, Greece.
Makris, N., Tahmatzidis, D., Demetriou, A., & Spanoudis, G. (2017). Mapping the evolving
core of intelligence: Relations between executive control, reasoning, language, and
awareness. Intelligence, 62, 12–30.
Mandler, J. M. (1992). How to build a baby: II. Conceptual primitives. Psychological Review,
99, 587–604.
Marcus, G. F., Fernandes, K. J., & Johnson, S. P. (2007). Infant rule learning facilitated by
speech. Psychological Science, 18, 387–391.
Marcus, G. F., Vijayan, S., Bandi Rao, S., & Vishton, P. M. (1999). Rule learning in
7-month-old infants. Science, 283, 77–80.
Markovits, H. (2014). On the road toward formal reasoning: Reasoning with factual causal
and contrary-to-fact causal premises during early adolescence. Journal of Experimental Child
Psychology, 128, 37–51. doi:10.1016/j.-jecp.2014.07.001.
Markovits, H., & Vachon, R. (1990). Conditional reasoning, representation, and level of
abstraction. Developmental Psychology, 26, 942–951. doi:10.1037/0012-1649.26.6.942.
Markovits, H., Thomson, V. A., & Brisson, J. (2015). Metacognition and abstract reasoning.
Memory and Cognition, 43, 681–693.
Markus, H. R., & Wurf, E. (1987). The dynamic self-concept: A social psychological
perspective. Annual Review of Psychology, 38, 299–337.
Massey, C. M., & Gelman, R. (1988). Preschooler’s ability to decide whether a photographed
unfamiliar object can move itself. Developmental Psychology, 24, 307–317.
Matzel, L. D., & Kolata, S. (2010). Selective attention, working memory, and animal
intelligence. Neuroscience & Biobehavioral Reviews, 34, 23–30.
Mayer, J. D. & Salovey, P. (1993). The intelligence of emotional intelligence. Intelligence, 17,
433–442.
McArdle, J. J., & Nesselroade, J. R. (2003). Growth curve analysis in contemporary
psychological research. In J. Schinka & W. Velicer (Eds.), Comprehensive handbook of
psychology, Vol. 1 (pp. 447–480). New York: Pergamon Press.
McBride, C. A. (2015). Is Chinese special? Four aspects of Chinese literacy acquisition that
might distinguish learning Chinese from learning alphabetic orthographies. Educational
Psychology Review. doi:10.1007/s10648-015-9318-2.
McBride-Chang, C., Zhou, Y., Cho, J.-R., Aram, D., Levin, I., & Tolchinsky, L. (2011).
Visual spatial skills: A consequence of learning to read? Journal of Experimental Child
McCrae, R. R., & Costa, P. T., Jr. (1999). A Five-Factor theory of personality. In
L. A. Pervin & O. P. John (Eds.), Handbook of personality: Theory and research (2nd ed.,
pp. 139–153). New York: Guilford.
McCrae, R. R., Costa, P. T., Ostendorf, F., Angleitner, A., Hrebickova, H., Avia, M. D.,
Sanz, J., & Sanchez-Bernardos, M. L. (2000). Nature over nurture: Temperament,
personality, and life span development. Journal of Personality and Social Psychology, 78,
173–186.
McGrew, K. S. (2009). CHC theory and the human cognitive abilities project: Standing on
the shoulders of the giants of psychometric intelligence research. Intelligence, 37, 1–10.
McIntyre, M., & Graziano, W. G. (2016). Seeing people, seeing things: Individual differences
in selective attention. Personality and Social Psychology Bulletin, 1–14. doi:10.1177/
0146167216653937.
McLaughlin, G. H. (1963). Psycho-logic: A possible alternative to Piaget’s formulation.
British Journal of Educational Psychology, 33, 61–67.
References 311
Mealor, A. D., & Dienes, Z. (2013). The speed of metacognition: Taking time to get to
know one’s structural knowledge. Consciousness and Cognition, 22, 123–136.
Meir, M. E., Smeekers, B. A., Silvia, P. J., Kwapil, T. R., & Kane, M. J. (2017). Working
memory capacity and the antisaccade task: A microanaltyic-macroanalytic investigation
of individual differences in goal activation and maintenance. Journal of Experimental
Psychology: Learning, Memory, and Cognition. doi:10.1037/xlm0000431.
Melby-Lervåg, M., Redick, T. S., & Hulme, C. (2016). Working memory training does not
improve performance on measures of intelligence or other measures of “far transfer”:
Evidence from a meta-analytic review. Perspectives on Psychological Science, 11, 512–534.
Menon, V., & Uddin, L. Q. (2010). Saliency, switching, attention and control: A network
model of insula function. Brain Structure and Function, 214, 655–667.
Miller, G. A. (1956). The magical number seven, plus or minus two: Some limits on our
capacity for processing information. Psychological Review, 63, 81–97. http://dx.doi.
org/10.1037/h0043158.
Miller, S. A., Custer, W. L., & Nassau, G. (2000). Children’s understanding of the necessity
of logically necessary truths. Cognitive Development, 15, 383–403.
Minter, M., Hobson, P. R., & Bishop, M. (1998). Congenital visual impairment and “theory
of mind”. British Journal of Developmental Psychology, 16, 183–196.
Mithen, S. (1996). The prehistory of the mind: The cognitive origins of art, religion and science.
London: Thames and Hudson.
Miyake, A., & Friedman, N. P. (2012). The nature and organization of individual differences
in executive functions: Four general conclusions. Current Directions in Psychological Science,
21(1), 8–14.
Miyake, A., Friedman, N. P., Emerson, M. J., Witzki, A. H., Howerter, A., & Wager, T. D.
(2000). The unity and diversity of executive functions and their contributions to complex
“Frontal Lobe” tasks: A latent variable analysis. Cognitive Psychology, 41, 49–100.
Moreno-Juan, V., Filipchuk, A., Antón-Bolaños, N., Mezzera, C., Gezelius, H., Andrés, B.,
Rodríguez-Malmierca, L., Susín, R., Schaad, O., Iwasato, T., Schüle, R., Rutlin, M.,
Nelson, S., Ducret, S., Valdeolmillos, M., Rijli, F. M., & López-Bendito, G. (2017) Prenatal
thalamic waves regulate cortical area size prior to sensory processing. Nature Communications,
8, 14172. doi:10.1038/ncomms14172.
Morra, S. (2000). A new model of verbal short-term memory. Journal of Experimental Child
Moshman, D. (1990). The development of metalogical understanding. In W. F. Overton
(Ed.), Reasoning, necessity, and logic: Developmental perspectives (pp. 205–225). Hillsdale:
Erlbaum.
Moshman, D. (1994). Reasoning, metareasoning and the promotion of rationality. In
A. Demetriou & A. Efklides (Eds.), Mind, intelligence, and reasoning: Structure and development
(pp. 135–150). Amsterdam: Elsevier.
Moshman, D. (2011). Adolescent rationality and development: Cognition, morality, and identity (3rd
ed.). New York: Psychology Press.
Moshman, D., & Tarricone, P. (2016). Logical and causal reasoning. In A. Greene, W. A.
Sandoval, & I. Braten (Eds.), Handbook of epistemic cognition (pp. 54–67). London:
Routledge.
Motes, M. A., Gamino, J. F., Chapman, S. B., Rao, N. K., Maguire, M. J., Brier, M. R.,
Kraut, M. A., Hart, & J. Jr. (2014). Inhibitory control gains from higher-order cognitive
strategy training. Brain and Cognition, 84, 44–62.
Müller, U., Overton, W. F., & Reene, K. (2001). Development of conditional reasoning: A
longitudinal study. Journal of Cognition and Development, 2, 27–49. doi:10.1207/
S15327647JCD0201_2.
312 References
Murray, C. (2009). Real education: Four simple truths for bringing America’s schools back to reality.
New York: Three Rivers.
Na, J., Grossmann, I., Varnum, M. E. W., Kitayama, S., Gonzalez, R. & Nisbett, R. E. (2010).
Cultural differences are not always reducible to individual differences. PNAS, 107, 6192–
6197. doi/10.1073/pnas.1001911107.
Naglieri, J. A. (1997). Naglieri nonverbal ability test. San Antonio: The Psychological
Corporation.
Naglieri, J. A., & Kaufman, J. C. (2001). Understanding intelligence, giftedness and creativity
using the PASS theory. Roeber Review, 23, 151–156.
Naglieri, J. A., & Ronning, M. E. (2000). Comparison of White, African-American,
Hispanic, and Asian children on the Naglieri nonverbal ability test. Psychological Assessment,
12, 328–334.
Neill, W. T., Valdes, L. A., & Terry, K. M. (1995). Selective attention and the inhibitory
control of cognition. In F. N. Dempster & C. J. Brainer (Eds.), Interference and inhibition
in cognition (pp. 207–226). San Diego: Academic Press.
Neisser, U. (Ed.) (1998). The rising curve: Long-term gains in IQ and related measures. Washington:
American Psychological Association.
Neisser, U., Boodoo, G., Bouchard, T. J., Jr., Boykin, A. W., Brody, N., Ceci, S. J., &
Urbina, S. (1996). Intelligence: Knowns and unknowns. American Psychologist, 51, 77–101.
doi:10.1037/0003-066X.51.2.77.
Nieder, A., & Dehaene, S. (2009). Representation of number in the brain. Annual Review of
Neuroscience, 32, 185–208.
Nisbett, R. E. (2003). The geography of thought: Why we think the way we do. New York: Free
Press.
Nisbett, R. E., Aronson, J., Blair, C., Dickens, W., Flynn, J., Halpern, D. F., & Turkheimer,
E. (2012). Intelligence: New findings and theoretical developments. American Psychologist,
67, 130–159. doi:10.1037/a0026699.
Nisbett, R. E., Fong, G. T., Lehman, D. R., & Cheng, P. W. (1987). Teaching reasoning.
Science, 238, 625–631.
O’Boyle, M. W., Cunnington, R., Silk, T. J., Vaughan, D., Jackson, G., Syngeniotis, A., &
Egan, G. F. (2005). Mathematically gifted male adolescents activate a unique brain
network during mental rotation. Cognitive Brain Research, 25, 583–587.
Oberauer, K. (2006). Reasoning with conditionals: A test of formal models of four theories.
Cognitive Psychology, 53, 238–283.
Oberauer, K., Farrell, S., Jarrold, C., & Lewandowsky, S. (2016). What limits working
memory capacity? Psychological Bulletin, 142, 758–799.
O'Brien, D., & Overton, W. F. (1982). Conditional reasoning and the competence-
performance issue: A developmental analysis of a training task. Journal of Experimental Child
Olson, D. R., & Astington, J. W. (2013). Preschool children conflate pragmatic agreement
and semantic truth. First Language, 33, 617–627.
O’Neill, D. K. (1996). Two-year-old children’s sensitivity to parents’ knowledge state when
making requests. Child Development, 67, 659–677.
Onishi, K. H., & Baillargeon, R. (2005). Do 15-month-old infants understand false beliefs?
Science, 308(5719), 255–258.
Osherson, D., & Markman, E. (1975). Language and the ability to evaluate contradictions
and tautologies. Cognition, 3, 213–226.
Osherson, D., Perani, D., Cappa, S., Schnur, T., Grassi, F., & Fazio, F. (1998). Distinct brain
loci in deductive versus probabilistic reasoning. Neuropsychologia, 36, 369–376.
Osman, M. (2004). An evaluation of dual-process theories of reasoning. Psychonomic Bulletin
& Review, 11(6), 988–1010. doi:10.3758/bf03196730.
References 313
Overton, W. F. (1990). Competence and procedures: Constraints on the development of

logical reasoning. In W. F. Overton (Ed.), Reasoning, necessity, and logic (pp. 1–30).
Hillsdale: Erlbaum.
Overton, W., Byrnes, J. P., & O'Brien, D. P. (1985). Developmental and individual
differences in conditional reasoning: The role of contradiction training and cognitive
style. Developmental Psychology, 21, 692–701.
Panaoura, A., Gagatsis, A., & Demetriou, A. (2009). An intervention to the metacognitive
performance: Self-regulation in mathematics and mathematical modelling. Acta Didactica
Universitatis Comenianae, 9, 63–79.
Pang, W., Esping, A., & Plucker, J. A. (2017). Confucian conceptions of human intelligence.
Review of General Psychology, 21, 161–169.
Papageorgiou, E., Christou, C., Spanoudis, G., & Demetriou, A. (2016). Augmenting
intelligence: Developmental limits to learning-based cognitive change. Intelligence, 56,
16–27.
Pascual-Leone, J. (1970). A mathematical model for the transition rule in Piaget’s
developmental stages. Acta Psychologica, 32, 301–345.
Pascual-Leone, J. (1988). Organismic processes for neo-Piagetian theories: A dialectical causal
account of cognitive development. In A. Demetriou (Ed.), The neo-Piagetian theories of
cognitive development: Toward an integration (pp. 25–64). Amsterdam: North-Holland.
Pascual-Leone, J., & Baillargeon, R. (1994). Developmental measurement of mental
attention. International Journal of Behavioral Development, 17, 161–200.
Pascual-Leone, J., & Goodman, D. R. (1979). Intelligence and experience: A neo-Piagetian
approach. Instructional Science, 8, 301–367.
Pascual-Leone, J., & Morra, S. (1991). Horizontality of water level: A neo-Piagetian
developmental review. Advances in Child Development and Behavior, 23, 231–275.
Patterson, K., Nestor, P. J., & Rogers, T. T. (2007). Where do you know what you know?
The representation of semantic knowledge in the human brain. Nature Reviews Neuroscience,
8, 976–987.
Paulus, M., Proust, J., & Sodian, B. (2013). Examining implicit metacognition in 3.5-year-
old children: An eye-tracking and pupillometric study. Frontiers in Psychology: Cognition
4:145. doi:10.3389/fpsyg.2013.00145.
Paulus, M., Tsalas, N., Proust, J., & Sodian, B. (2014). Metacognitive monitoring of oneself
and others: Developmental changes in childhood and adolescence. Journal of Experimental
Child Psychology, 122, 153–165.
Perner, J. (1991). Understanding the representational mind. Cambridge: MIT Press.
Perner, J., & Dienes, Z. (2003). Developmental aspects of consciousness: How much
theory of mind do you need to be consciously aware? Consciousness and Cognition, 12,
63–82.
Perone, P., Molitor, S., Buss, A. T., Spencer, J. P., & Samuelson, L. K. (2015). Enhancing
the executive functions of 3-year-olds in the dimensional change card sort task. Child
Development, 86, 812–827.
Petersen S., & Posner, M. I. (2012). The attention system of the human brain: 20 years after.
Annual Review of Neuroscience, 35, 73–89.
Piaget, J. (1952). The origins of intelligence in children. New York: International Universities
Press. (Original work published 1936).
Piaget, J. (1968). Psychology of Intelligence. Totowa: Littlefield, Adams and Company.
Piaget, J. (1970). Piaget’s theory. In P. H. Mussen (Ed.), Carmichael’s handbook of child
development (pp. 703–732). New York: Wiley.
Piaget, J. (1976). The grasp of consciousness. Cambridge: Harvard University Press.
Piaget, J. (1977). To understand is to invent: The future of education. New York: Penguin.
314 References
Piaget, J. (2001). Studies in reflecting abstraction. London: Psychology Press.

Piaget, J., & Inhelder, B. (1974). The child’s construction of quantities. London: Routledge &
Kegan Paul. (Original work published 1941).
Pillow, B. L. (2008). Development of children’s understanding of cognitive activities. The
Journal of Genetic Psychology, 169, 297–321.
Plomin, R., & Spinath, F. M. (2002). Genetics and general cognitive ability (g). Trends in
Cognitive Sciences, 6, 169–176.
Pollak, S. D., Nelson, C. A., Schlaak, M. F., Roeber, B. J., Wewerka, S. S., Wiik, K. L.,
Frenn, K. A., Loman, M. L., & Gunnar, M. R. (2010). Neurodevelopmental effects of
early deprivation in postinstitutionalized children. Child Development, 81, 224–236.
Posner, M. I., & Raichle, M. (1997). Images of mind (2nd ed.). New York: Scientific American
Library.
Posner, M. I., & Rothbart, M. K. (2006). Educating the human brain. New York: American
Psychological Association.
Povinelli, D. J. (2001). The self: Elevated in consciousness and extended in time. In
C. Moore & K. Lemmon (Eds.), The self in time: Developmental perspectives (pp. 75–95).
Mahaw: Lawrence Erlbaum Associates.
Protzko, J. (2015). The environment in raising early intelligence: A meta-analysis of the
fadeout effect. Intelligence, 53, 202–210.
Protzko, J. (2016). Effects of cognitive training on the structure of intelligence. Psychological
Bulletin & Review. doi:10.3758/s13423-016-1196-1.
Raganath, C., & D’Esposito, M. (2005). Directing the mind’s eye. Prefrontal, inferior, and
medial temporal mechanisms for visual working memory. Current Opinion in Neurobiology,
15, 175–182.
Rakoczy, H., Fizke, E., Bergfeld, D., & Schwarz, I. (2015). Explicit theory of mind is even
more unified than previously assumed: Belief ascription and understanding aspectuality
emerge together in development. Child Development, 86(2), 486–502. doi:10.1111/
cdev.12311.
Ratterman, M. J., Gentner, D. (1998). More evidence for a relational shift in the development
of analogy: Children’s performance on a causal-mapping task. Cognitive Development, 13,
453–478.
Raven, J. (2000). The Raven’s Progressive Matrices: Change and Stability over Culture and
Time. Cognitive Psychology, 41, 1–48.
Repovš, G., & Baddeley, A. (2006). The multi-component model of working memory:
Explorations in experimental cognitive psychology. Neuroscience, 139, 5–21.
Reverberi, C., Pischedda, D., Burigo, M., & Cherubini, P. (2012). Deduction without
awareness. Acta Psychologica, 139(1), 244–253.
Reynolds, M. R. (2013). Interpreting the g loadings of intelligence test composite scores in
light of Spearman’s Law of Diminishing Returns. School Psychology Quarterly, 28, 63–76.
Ricco, R. (2010). The development of deductive reasoning across the life span. In
W. F. Overton & R. M. Lerner (Eds.), The handbook of life-span development: Cognition,
biology and methods (Vol. 1, pp. 391–430). Hoboken: Wiley.
Rinaldi, L., & Karmiloff-Smith, A. (2017). Intelligence as a developing function: A
neuroconstructivist approach. Journal of Intelligence, 5, 18. doi:10.3390/jintelligence5020018.
Rindermann, H., & Neubauer, A. C. (2004). Processing speed, intelligence, creativity, and
school performance: Testing of causal hypotheses using structural equation models.
Intelligence, 32(6), 573−589.
Rips, L. J. (1994). The psychology of proof. Deductive reasoning in human thinking. Cambridge:
MIT Press, Bradford.
Rips, L. J. (2001). Two kinds of reasoning. Psychological science, 12, 129–134.
References 315
Rivera, S., & Sloutsky, V. M. (2016). Salience versus prior knowledge—how do children
learn rules? Cognitive Science Society.
Roberts, B. W., Walton, K. E., & Viechtbauer, W. (2006). Patterns of mean-level change
in personality traits across the life course: A meta-analysis of longitudinal studies.
Psychological Bulletin, 132, 1–25.
Rocadin, C., Pascual-Leone, J., Rich, J. B., & Dennis, M. (2007). Developmental rela-
tions between working memory and inhibitory control. Journal of the International
Neuropsychological Society, 13, 59–67.
Rochat, P. (1998). Self-perception and action in infancy. Experimental Brain Research, 123,
102–109.
Roebers, C. M., Cimeli, P., Röthlisberger, M., & Neuenschwander, R. (2012). Executive
functioning, metacognition, and self-perceived competence in elementary school
children: An explorative study on their interrelations and their role for school achievement.
Metacognition Learning. doi:10.1007/s11409-012-9089-9.
Rohde, T. E., & Thomson, L. A. (2007). Predicting academic achievement with cognitive
ability. Intelligence, 35, 83–92.
Ronald, A. (2011). Is the child “father of the Man”? Evaluating the stability of genetic
influences across development. Developmental Science, 14, 1471–1478.
Rothbart, M. A. (2011). Becoming who we are: Temperament and personality in development. New
York: The Guilford Press.
Rothbart, M. K., & Bates J. E. (1998). Temperament. In W. Damon & N. Eisenberg (Eds.),
Handbook of child psychology: Vol. 3, Social, emotional, personality development. New York:
Wiley & Sons.
Rothbart, M. K., & Posner, M. I. (2015). The developing brain in a multitasking world.
Developmental Review, 35, 42–63.
Rothbart, M. K., Ahadi, S. A., & Evans, D. E. (2000). Temperament and personality: Origins
and outcomes. Journal of Personality and Social Psychology, 78, 122–135.
Rubinov, M., Sporns, O., van Leeuwen, C., & Breakspear, M. (2009). Symbiotic relationship
between brain structure and dynamics. BMC Neuroscience, 10, 55. doi:10.1186/1471-
2202-10-55.
Rueda, M., Posner, M. I., & Rothbart, M. K. (2005). The development of executive
attention: Contributions to the emergence of self-regulation. Developmental Neuropsychology,
28, 573–594. doi:10.1207/s15326942dn2802_2.
Rumbaugh, D. M., & Washburn, D. (2003). Intelligence of apes and other rational beings. New
Haven: Yale University Press.
Rushton, J. P., & Irwing, P. (2009). A general factor of personality in 16 sets of the Big Five,
the Guilford–Zimmerman Temperament Survey, the California Psychological Inventory,
and the Temperament and Character Inventory. Personality and Individual Differences, 47,
558–564.
Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old
infants. Science, 274, 1926–1928.
Sala, G., & Gobert, F. (2017). Working memory training in typically developing children:
A meta-analysis of the available evidence. Developmental Psychology, 53, 671–685.
Salomon, G., & Perkins, D. N. (1989). Rocky roads to transfer: Rethinking mechanisms of
a neglected phenomenon. Educational Psychologist, 24, 113–142.
Salthouse, T. A. (1996). The processing-speed theory of adult age differences in cognition.
Psychological Review, 103, 403–428.
Salthouse, T. A. (2000). Aging and measures of processing speed. Biological Psychology, 54,
35–54.
Sampson, G., Gil, D., & Trudghill, P. (Eds.). (2009). Language complexity as an evolving variable.
Oxford: Oxford University Press.
316 References
Sauseng, P., Griesmayer, B., Freunberger, R., & Klimesch, W. (2010). Control mechanisms
in working memory: A possible function of EEG theta oscillations. Neuroscience and
Biobehavioral Reviews, 34, 1015–1022.
Saxe, R., & Carey, S. (2006). The perception of causality in infancy. Acta Psychologica,
123(1–2), 144–165.
Schaafsma, S. M., Pfaff, D. W., Spunt, R. P., & Adolphs, R. (2015). Deconstructing and
reconstructing theory of mind. Trends in Cognitive Sciences, 19, 65–72.
Schaie, K. W. (1996). Adulthood and old age. In T. Husen & T. N. Postlewaithe (Eds.),
International encyclopedia of education (pp. 163–168). Oxford: Pergamon Press.
Schaie, K. W., Willis, S. L., Jay, G., & Chipuer, H. (1989). Structural invariance of cognitive
abilities across the adult life span: A cross-sectional study. Developmental Psychology, 25,
652–662.
Schmitt, D. P., Allik, J., McCrae, R. R., Benet-Martinez, V. (2007). The geographic
distribution of Big Five personality traits: Patterns and profiles of human self-description
across 56 nations. Journal of Cross-Cultural Psychology, 38, 173–212.
Schneider, S., & Katz, M. (2011). Rethinking the language of thought. WIREs Cognitive
Science. doi:10.1002/wcs.1155.
Schneider, W., & Lockl, K. (2007). Knowledge about the mind: Links between theory of
mind and later metamemory. Child Development, 78, 148–167.
Schubert, A.-L., Hagemann, D., & Frischkorn, G. T. (2017, July 13). Is general intelligence
little more than the speed of higher-order processing? Journal of Experimental Psychology:
General. http://dx.doi.org/10.1037/xge0000325.
Schweizer, K., & Koch, W. (2002). A revision of Cattell’s investment theory: Cognitive
properties influencing learning. Learning and Individual Differences, 13, 57–82.
Scott, R. B., Dienes, Z., Barrett, A. B., Bor, D., & Seth, A. K. (2014). Blind insight:
Metacognitive discrimination despite chance task performance. Psychological Science,
25(12), 2199–2208. doi:10.1177/0956797614553944.
Sergent, C., & Naccache, L. (2012). Imaging neural signatures of consciousness: ‘What’,
‘When’, ‘Where’ and ‘How’ does it work? Archives Italiennes de Biologie, 150, 91–106.
Shah, P., & Miyake, A. (1996). The separability of working memory resources for spatial
thinking and language processing: An individual differences approach. Journal of
Experimental Psychology: General, 125(1), 4–27.
Shayer, M. (2003). Not just Piaget; not just Vygotsky, and certainly not Vygotsky as
alternative to Piaget. Learning and Instruction, 13, 465–485.
Shayer, M., & Adey, P. S. (1993). An exploration of the long-term far-transfer effects
following an extended intervention program in the high school science curriculum. In
L. Smith (Ed.), Critical readings on Piaget (pp 66–95). London: Routledge.
Shayer, M., & Adey, P. S, (Eds.). (2002). Learning intelligence: Cognitive acceleration across the
curriculum from 5 to 15 years. Milton Keynes: Open University Press.
Shayer, M., & Adhami, M. (2007). Fostering cognitive development through the context of
mathematics: Results of the CAME project. Educational Studies in Mathematics, 64,
265–291.
Shayer, M., Demetriou, A., & Pervez, M. (1988). The structure and scaling of concrete
operational thought: Three studies in four countries. Genetic, Social and General Psychology
Monographs, 114, 307–376.
Shayer, M., Ginsburg, D., & Coe, R. (2007). Thirty years on – a large anti-Flynn effect? The
Piagetian test Volume & Heaviness norms 1975–2003. British Journal of Educational
Sheppard, L. D., & Vernon, P. A. (2008). Intelligence and speed of information-processing:
A review of 50 years of research. Personality and Individual Differences, 44, 535–551.
References 317
Shipstead, Z., Harrison, T. L., & Engle, R. W. (2016). Working memory capacity and fluid
intelligence: Maintenance and disengagement. Perspectives on Psychological Science, 11,
771–799.
Shipstead, Z., Redick, T. S., & Engle, R. W. (2012). Is working memory training effective?
Psychological Bulletin, 138, 628–654. doi:10.1037/a0027473.
Shultz, T. R. (1982). Rules of causal attribution. Monographs of the Society for Research in Child
Development, 47(1), 1–51.
Siegal, M., & Varley, R. (2002). Neuronal systems involved in “theory of mind”. Nature
Reviews, 3, 463–471.
Siegel, H. (1989). The rationality of science, critical thinking, and science education, Synthese,
80, 9–41.
Siegler, R. S. (2016). Continuity and change in the field of cognitive development and in
the perspectives of one cognitive developmentalist. Child Development Perspectives, 10(2),
128–133. doi:10.1111/cdep.12173.
Simion, F., Macchi Cassia, V., Turati, C., & Valenza, E. (2001). The origins of face
perception: Specific vs. nonspecific mechanisms. Infant and Child Development, 10, 59–65.
Simoneau, M., & Markovits, H. (2003). Reasoning with premises that are not empirically
true: Evidence for the role of inhibition and retrieval. Developmental Psychology, 39,
964–975.
Skoritch, D. P., Gash, T. B., Stalker, K. L., & Zheng, L. (2017). Exploring the cognitive
foundations of the shared attention mechanism: Evidence for a relationship between self-
categorization and shared attention across the autism spectrum. Journal of Autism and
Developmental Disorders, 47, 1341–1353. doi:10.1007/s10803-017-3049-9
Sniekers, S., Stringer, S., Watanabe, K., Jansen, P. R., Coleman, J. R. I., Krapohl, E., Taskesen,
E., Hammerschlag, A. R., Okbay, A., Zabaneh, D., Amin, N., Breen, G., Cesarini, D.,
Chabris, C. F., Iacono, W. G., Ikram, M. A., Johannesson, M., Koellinger, P., Lee, J. J.,
Magnusson, P. K. E., McGue, M., Miller, M., Ollier, W. E. R., Payton, A., Pendleton,
N., Plomin, R., Rietveld, C. A., Tiemeier, H., van Duijn, C. M., & Posthuma, D. (2017).
Genome-wide association meta-analysis of 78,308 individuals identifies new loci and
genes influencing human intelligence. Nature: Genetics. doi:10.1038/ng.3869.
Somsen, R. J. M., van’t Klooster, B. J., van der Molen, M. W., van Leeuwen, H. M. P., &
Licht, R. (1997). Growth spurts in brain maturation during middle childhood as indexed
by EEG power spectra. Biological Psychology, 44, 187–209.
Spanoudis, G., Demetriou, A., Kazi, S., Giorgala, K., & Zenonos, V. (2015). Embedding
cognizance in intellectual development. Journal of Experimental Child Psychology, 132,
32–50.
Spearman, C. (1904). “General intelligence,” Objectively determined and measured. The
American Journal of Psychology, 15, 201–292.
Spearman, C. (1927). The abilities of man: Their nature and measurement. London: Macmillan.
Sporns, O. (2012). Discovering the human connectome. Cambridge: MIT Press.
Sporns, O., & Betzel, R. F. (2016). Modular brain networks. Annual Review of Psychology,
67, 613–640.
Squire, L., R., (1992). Memory and the hippocampus: A synthesis from rats, monkeys, and
humans. Psychological Review, 92, 195–231.
Stanovich, K., & West, R. (2000). Individual differences in reasoning: Implications for the
rationality debate? Behavioral and Brain Sciences, 23(5), 645–665.
Staudenmayer, H., & Bourne, L. E. (1977). Learning to interpret conditional sentences:
A developmental study. Developmental Psychology, 13, 616–623.
Sternberg, R. J. (2006). The nature of creativity. Creativity Research Journal, 18, 87–98.
Sternberg, R. J. (2011). Cognitive psychology. New York: Cengage Learning.
318 References
Sternberg, R. J., & Downing, C. J. (1982). The development of higher-order reasoning in

adolescence. Child Development, 53, 209–221.
Stiles, J., & Jernigan, T. L. (2010). The basics of brain development. Neuropsychological Review,
20, 327–348.
Stone, V. E., & Gerrans, P. (2006). Does the normal brain have a theory of mind? Trends in
Cognitive Sciences, 10(1), 3–4.
Storrs, C. (2017). How poverty affects the brain. Nature, 547, 150–152.
Strauss, S. (1972). Learning theories of Gagne and Piaget: Implications for curriculum
development. Teachers College Record, 74(1), 81–102.
Stroop, J. R. (1935). Studies of interference in serial verbal reactions. Journal of Experimental
Taplin, J. E., Staudenmayer, H., & Taddonio J. L. (1974). Developmental changes in
conditional reasoning: Linguistic or logical? Journal of Experimental Child Psychology, 17,
360–373.
Tau, Z. G., & Peterson, B. S. (2010). Normal development of brain circuits. Neuropharmacology,
35(1), 147–168.
Tenenbaum, J. B., Kemp, C., Griffiths, T. L., & Goodman, N. D. (2011). How to grow a
mind: Statistics, structure, and abstraction. Science, 331, 1279–1285. doi:10.1126/science.
1192788.
Thatcher, R. W. (1992). Cyclic cortical reorganization during early childhood. Brain and
Cognition, 20, 24–50.
Thatcher, R. W. (1994). Cyclic cortical reorganization: Origins of human cognitive
development. In G. Dawson & K. Fischer (Eds.), Human behavior and the developing brain
(pp. 232–266). New York: Guilford.
Thompson, P. M., Gledd, J. N., Woods, R. P., MacDonald, D., Evans, A. C., & Toga, A.
W. (2000). Growth patterns in the developing brain detected by using continuum tensor
maps. Nature, 404, 190–193.
Thornton, A., & Raihani, N. J. (2008). The evolution of teaching. Animal Behaviour, 75,
1823–1836.
Thurstone, L. L. (1935). The vectors of mind: Multiple-factor analysis for the isolation of primary
traits. Chicago: University of Chicago Press.
Thurstone, L. L. (1938). Primary mental abilities. Chicago: University of Chicago Press.
Thut, G., & Miniussi, C. (2009). New insights into rhythmic brain activity from TMS-EEG
studies, Trends in Cognitive Sciences, 13(4), 182–189.
Tideman, E. & Gustafsson, J. E. (2004). Age-related differentiation of cognitive abilities in
ages 3–7. Personality and Individual Differences, 36, 1965–1974.
Toplak, M. E., & Stanovich, K. E. (2002). The domain specificity and generality of disjunctive
reasoning: Searching for a generalizable critical thinking skill. Journal of Educational
Tran, R., & Pashler, H. (2017). Learning to exploit a hidden predictor in skill acquisition:
Tight linkage to conscious awareness. Plos One 12(6): e0179386. https://doi.org/10.1371/
journal.pone.0179386.
Tsalas, N., Sodian, B., & Paulus, M. (2017). Correlates of metacognitive control in 10-year-
old children and adults. Metacognition and Learning. doi:10.1007/s11409-016-9168-4.
Tucker-Drob, E. M. (2009). Differentiation of cognitive abilities across the lifespan.
Developmental Psychology, 45, 1097–1118.
Vallotton, C. (2008). Infants take self-regulation into their own hands. Zero to Three, 29–34.
van der Linden, D., te Nijenhuis, I., & Bakker, A. B. (2010). The general factor of personality:
A meta-analysis of Big Five intercorrelations and a criterion-related validity study. Journal
of Research in Personality, 44, 315–327.
References 319
van der Maas, H. L. J., Dolan, C. V., Grasman, R. P. P. P., Wicherts, J. M., Huizenga, H. M.,
& Raijmakers, M. E. J. (2006). A dynamical model of general intelligence: The positive
manifold of intelligence by mutualism. Psychological Review, 113, 842–861.
van der Maas, H. L. J., Kan, K., Marsman, M., & Stevenson, C. E. (2017). Network models
for cognitive development and intelligence. Journal of Intelligence, 5, 16. doi:10.3390/
van Geert, P. (1998). A dynamic systems model of basic developmental mechanisms: Piaget,
Vygotsky and beyond. Psychological Review, 105, 634–677.
van Geert, P. (2000). The dynamics of general developmental mechanisms: From Piaget
and Vygotsky to dynamic systems models. Current Directions in Psychological Science, 9,
64–68.
Vendetti, M. S., and Bunge, S. A. (2014). Evolutionary and developmental changes in the
lateral frontoparietal network: A little goes a long way for higher-level cognition. Neuron,
84, 906–917.
Vendetti, C., Kamawar, D., Podjarny, G., & Astle, A. (2015). Measuring preschoolers’
inhibitory control using the black/white Stroop. Infant and Child Development. doi:10.1002/
icd.1902.
Vergauwe, E., Hartstra, E., Barrouillet, P., & Bras, M. (2015). Domain-general involvement
of the posterior frontolateral cortex in time-based resource-sharing in working memory:
An fMRI study. NeuroImage. doi.org/10.1016/j.neuroimage.2015.04.059.
Vértes, P. E., & Bullmore, E. T. (2015). Annual research review: Growth connectomics—the
organization and reorganization of brain networks during normal and abnormal
development. Journal of Child Psychology and Psychiatry, 56(3), 299–320.
Vincent, J. L., Snyder, A., Fox, M. D., Shannon, B. J. Andrews, J. R., Raichie, M. E., &
Buckner, R. L. (2006). Coherent spontaneous activity identifies a hippocampal-parietal
memory network. Journal of Neurophysiology, 96, 3517–3531.
Visser, B., Ashton, M. C., & Vernon, P. (2006). Beyond g: Putting multiple intelligences
theory to the test, Intelligence, 34, 487–502.
von Bastian, C. C., & Druey, M. D. (2017). Shifting between mental sets: An individual
differences approach to commonalities and differences of task switching components.
Journal of Experimental Psychology: General. http://dx.doi.org/10.1037/xge0000333.
Vosniadou, S. (1994). Capturing and modeling the process of conceptual change. Learning
and Instruction, 4, 45–69.
Vosniadou, S. (2013). International handbook of research on conceptual change. London: Routledge.
Vygotsky, L. (1986). Thought and language. Cambridge: MIT Press.
Wagner, S., Winner, E., Cicchetti, D., & Gardner, H. (1981). “Metaphorical” mapping in
human infants. Child Development, 52, 728–731.
Wason, P. C., & Evans, J. St. B. T. (1975). Dual processes in reasoning? Cognition, 3,
141–154.
Waterhouse, L. (2006). Multiple intelligences, the Mozart effect, and emotional intelligence:
A critical review. Educational Psychologist, 41, 207–225.
Watson, G. B., & Glaser, E. M. (1980). WGCTA Watson-Glaser Critical Thinking Appraisal
manual: Forms A and B. San Antonio: The Psychological Corporation.
Wechsler, D. (1950). The range of human abilities. Baltimore: William & Wilkins.
Wellman, H. M. (1990). The child’s theory of mind. Cambridge: MIT Press.
Wellman, H. M. (2014). Making minds: How theory of mind develops. Oxford: Oxford
University Press.
Wellman, H. M., Cross, D., & Watson, J. (2001). Meta-analysis of theory-of-mind
development: The truth about false belief. Child Development, 72, 655–684.
Wellman, H. M., Fang, F., & Peterson, C. C. (2011). Sequential progressions in a theory-
of-mind scale: Longitudinal perspectives. Child Development, 82, 780–792.
320 References
Wendelken, C., Ferrer, E., Whitaker, K. J., & Bunge, S. A. (2015). Fronto-Parietal network
reconfiguration supports the development of reasoning ability. Cerebral Cortex, 1–13.
doi:10.1093/cercor/bhv050.
West, R. F., Toplak, M. E., & Stanovich, K. E. (2008). Heuristics and biases as measures of
critical thinking: Associations with cognitive ability and thinking dispositions. Journal of
Educational Psychology, 100, 930–941.
Whorf, B. (1956). Language, thought, and reality. London: Wiley.
Wildenger, L. K., Hofer, B. K., & Burr, J. E. (2010). Epistemological development in very
young knowers. In Lisa D. Bendixen & Florian C. Feucht (Eds.), Personal epistemology
in the classroom: Theory, research, and implications for practice (Ch. 8). Cambridge:
Wilhem, O., & Oberauer, K. (2006). Why are reasoning ability and working memory
capacity related to mental speed? An investigation of stimulus-response compatibility in
choice reaction time tasks. European Journal of Cognitive Psychology, 18, 18–50.
Wimmer, R. D., Schmitt, L. I., Davidson, T. J., Nakajima, M., Deisseroth, K., & Halassa,
M. M. (2015). Thalamic control of sensory selection in divided attention. Nature, 526,
705–709. doi:10.1038/nature15398.
Winship, C., & Korenman, S. (1997). Does staying in school make you smarter? The effect
of education on IQ in The Bell Curve. In B. Devlin, S. E. Fienberg, D. P. Resnick, &
K. Roeder (Eds.), Intelligence, genes, and success. Scientists respond to The Bell Curve
(pp. 215–234). New York: Springer-Verlag.
Woodley of Menie, M. A., Younuskunja, S., Balan, B., & Piffer, D. (2017). Holocene
selection for variants associated with cognitive ability: Comparing ancient and modern
genomes. bioRxiv preprint, 21 February. doi:http://dx.doi.org/10.1101/109678.
Yoon, T., Okada, J., Jung, M. W., & Kim, J. J. (2008). Prefrontal cortex and hippocampus
subserve different components of working memory in rats. Learning and Memory, 15,
97–105.
Yuan, L., Uttal, D., & Gentner, D. (2017). Analogical processes in children’s understanding
of spatial representations. Developmental Psychology, 53, 1098–1114.
Zabaneh, D., Krapoli, E., Gaspar, H. A., Curtis, C., Lee, S. H., Patel, H., Newhouse, S., Wu,
H. M., Simpson, M. A., Putallaz, M. A., Lubinski, D., Plomin, R., & Breen, G. (2017).
A genome-wide association study for extremely high intelligence. Molecular Psychiatry, 1–7.
doi:10.1038/mp.2017.121.
Žebec, M. S., Demetriou, A., & Kotrla-Topić, M. (2015). Changing expressions of general
intelligence in development: A 2-wave longitudinal study from 7 to 18 years of age.
Zelazo, P. D. (2004). The development of conscious control in childhood. Trends in Cognitive
Sciences, 8, 12–17.
Zelazo, P. D. (2015). Executive function: Reflection, iterative reprocessing, complexity, and
the developing brain. Developmental Review, 38, 55–68.
Zelazo, P. D., Anderson, J. E., Richler, J., Wallner-Allen, K., Beaumont, J. L., & Weintraub,
S. (2013). II. NIH toolbox cognition batter (CB): Measuring executive function and
attention. Monographs of the Society for Research in Child Development, 78, 16–33.
Zheng, X., & Rajapakse, J. C. (2006). Learning functional structure from fMRI images.
NeuroImage, 31, 1601–1613.
Zuo, X. N., He, Y., Betzel, R. F., Colcombe, S., Sporns, O., & Milham, M. P. (2017).
Human connectomics across the life span. Trends in Cognitive Sciences, 21(1), 32–45,
doi:10.1016/j.tics.2016.10.005.
INDEX
Numbers in bold italics refer to boxes; numbers in italics refer to figures; numbers in bold
refer to tables.
a priori awareness 125 agency 252

AACog mechanism 118–23, 134–5, aggressiveness 172
188–90, 200, 211–13, 223, 282–4, agreeableness 164–6, 168, 170, 172,
287–90 227–8
Abecedarian study 230 alerting system 198
ability differentiation index 151–2 algebraic reasoning 203
ability emotional intelligence 170 alignment flexibility 250
abstraction 288 alignment of reasoning development 131,
AC see affirming the consequent 136–7
academic performance 226–7; see also Allport, G. W. 162
school performance aloofness 168
accessibility of education 248–9 alpha (α) factor 168, 227, 289
accommodation 33 alpha oscillations 201
accretion of knowledge 277 alternative worlds 127
Ackerman, P. L. 28 American Psychological Association
adaptive functions of intelligence 32–3 179
Adey, P. S. 234 analogical reasoning 41–3, 57–9, 67, 139,
adolescent critical thinking 277–80 259; development of 57–9
adolescent education 266–8; educating Anderson, J. R. 19
awareness 268; educating reasoning Anderson, M. 127, 206, 211, 288–9
268; educating representational capacity Andrews, G. 72
267–8; educating SCS 267 animal g 183–4
adopting a critical stance 274–5, 277 animate–inanimate distinction 85
adoption 184–5 Antinori, A. 167
adult personality factors 164–5 APA see American Psychological
adverse life conditions 209–210 Association
affiliativeness 164 Aristotle 1
affirming the consequent 13–14, 21, 38, arousal 164
64, 66, 109, 112, 237–8 Asendorph, Y. 171
age differentiation index 151–2 aspects of mental functioning 175–6
322 Index
assessment for learning 268–70; cognitive brain architecture 24, 193–201, 211;
development diagnostic tool 269–70; communication between brain
overcoming mental difficulties 270 networks 200–201; locating cognitive
assimilation 33 functions in brain 193–200
association 11–15 brain bases of mental processes 20–21
atomic symbols see language of thought brain connectivity 206–8
attainment of logical schemes 112 brain development 201–210; changes in
attention mechanisms 8, 46, 54, 97 mind-related brain networks 202–9;
attention–inhibition mechanism 77 under adverse life conditions 209–210
attunement 148 brain development cycles 207–9
autism 82 brain functioning underlying intelligence
autobiographical memory 194 199–200
automated tendencies 175–6 brain morphometry 193
autonomy 82, 173, 272 brain networks 193–201; communication
awareness 17–19, 68–81, 198–9, 254, between 200–201
260–62, 268; brain networks 198–9; brain rhythms 200–201
and consciousness 17–19; educating Braine, M. D. S. 59, 63
254, 260–62, 268; see also knowledge Brendon, W. O. 201, 207
about mind bridging experimental/psychometric
awareness-raising 239 traditions 26–30
axon guidance 182 bridging Piaget and Vygotsky 234–7
Broberg, A. G. 172
Baars, B. J. 199 Brod, G. 222
Baar’s theatre metaphor 18 Brooks, R. 74
Baddeley’s model of WM 9–10, 9, 10, 19, Brown, A. L. 58
47–8, 50, 52, 95 Brown, E. 85
Baillargeon, R. 57, 72 Bryant, P. E. 40
Baldwin, James 31 Buehner, M. 221
Barrouillet, P. 66–7 bundle of perceptions 176
Bayesian probabilistic covariation 196–7 Bunge, S. A. 197, 208–9, 222
belief systems 101 Butterworth, G. 86
The Bell Curve 178 Buzsaki, G. 201, 207
Benoit, L. 126
Benyamin, B. 183 capability to inference 67
beta (β) factor 168 Carey, S. 252
beta oscillations 201 Carraher, T. N. 185–6
bifocal coordination 50 Carroll, J. B. 30, 41, 132, 150, 178–9, 225
Big Five factors of adult personality 79, Carroll’s three-strata model 24–7, 25, 27,
164–72, 165, 180, 187–8, 227; 41, 95
characteristics of 165 Carruthers, P. 16
binding mental processes 147–61; see also Carter, O. L. 167
differentiation of mental processes cascade model 54–5, 55, 117
Binet, Alfred 31, 131 CASE experimental schools 237
biological adaptation 83 Case, Robbie 48–51, 58–9, 87–8, 104,
biological causality 87 106–7
biological make-up 163–4 catching fallacies 277–8
biological worlds 85–6 categorical reasoning 12–13, 262, 267
blind insight 19 categorical specialized capacity system
bodily-kinaesthetic intelligence 23 252–3, 255, 262
boosting intelligence 227–8 categorical thought 101–2, 104–5, 107,
bottom-up causal effects 230–31, 262
286–7 categories of reason 107
bottom-up mediation 144–6 Cattell, R. B. 23–4, 164
Bouchard, T. J. 180, 182 Cattell-Horn-Carroll model of intelligence
brain activation 167–8 24, 25, 26, 113, 168, 230
Index 323
causal distinctions 86–8; central conceptual Cohn, L. D. 173

structures 87–8 Cole, M. 185
causal role of mind 70–74 Coley, J. D. 58
causal specialized capacity system 110–111, colour perception 100, 103
253, 256, 263 combinatorial thought 38, 109, 158
causal thought 102–5, 267 Commons, M. L. 63
causal–experimental reasoning 156–7 communication between brain networks
cause–effect relations 279 200–201
Ceci, S. J. 222 complexity of taught concepts 249–50
central conceptual structures 87–8 composition of growing mind 281–3
central executive 10 compositionality 16, 21, 42, 73, 282–3
Chamorro-Premuzic, T. 175–6 concepts of personality 162–6
Chandler, M. 72 conceptual fluency programme 127, 207
change mechanisms in Piaget’s theory 39 conceptual similarity 57–8
changes in factor structure with growth conceptual stability 38–9
148–50 concrete thought 35–9, 52, 127; from
changes in mind-related brain networks preoperational thought to 35–7; to
202–9; cycles in brain development formal thought 37–9
207–9 conditional reasoning 66–7, 222, 237–43,
changing intelligence 243–6 242, 284; improving 237–43
characteristics of Big Five personality confidence 125
factors 165 confirmatory factor analysis model 98
cheating 103 conflict handling system 198
childhood temperament 163–4 Confucianism 185, 189
choice 21 congenital blindness 287
choices of choices 281–2 conscientiousness 164–5, 167–8, 171–4,
Chomsky, Noam 16–17 227–8
Christal, R. E. 27 consciousness 17–19, 198–9; brain
Chuang, S. S. 172 networks 198–9
Chuderska, A. 66 conservation 232–3
Chuderski, A. 28, 66 constructed reasoning 63–6
Cicchetti, D. 58 constructivism 271
CNV see contingent negative variation contingency 63
cognitive acceleration 232–7, 235; contingent negative variation 200
bridging Piaget with Vygotsky 234–7; control of attentional focus 160–61
testing for transfer 233–4 control of impulse 172–3
cognitive cueing 75 controlled attention 8
cognitive development diagnostic tool controlled conceptual change 250
269–70 convergence of SCS dimensions 106–7
cognitive dimensions of SCSs 106–7 Conway, A. R. A. 28–9
cognitive functions in the brain 193– core domains 82–91, 100–101; causal
200209; awareness/consciousness distinctions 86–8; conclusions 90–91;
198–9; domain-specific brain networks explaining development of 88–90;
193–4; inferential capacity 196–8; physical/biological/psychological
parieto-frontal integration theory worlds 85–6
199–200; representational capacity core operators in SCS 251–6; educating
194–6 251–3
cognitive performance 223–6 Corsi task 139
cognitive self-concept 228 “coupling” 20
cognizance 116–22, 127, 137–45, 173–4, Cowan, N. 50, 72
237–46; in 4–7 year-olds 138–43; Cramer, P. 172, 175
manipulating 237–46; mediating role of Crick, F. C. 199
137–45 critical thinking education 275–80; in
cognizance resolution 214 adolescence 277–80; in pre-school
Cohen’s d 222 275–6; in primary school 276–7
324 Index
critical thought 273–80; see also educating 59–63; inductive/analogical reasoning

critical thought 57–9; origins of reasoning 63–6;
Critique of Pure Reason 106–7 reasoning and process efficiency 66–7
crystallized intelligence 23–4, 28, 168 developmental changes in personality
cultural aspects of the mind 177–90; see 171–2
also psychological aspects of the mind developmental cycles 131–2, 147–61;
culture 184–8 binding of mental processes 147–61;
curiosity 164 relation to IQ 131–2; see also
current status of Piaget’s theory 40–41 differentiation of mental processes
cycles in brain development 207–9 developmental g markers 154–60
cycles of development 121–33; see also developmental pattern mapping 150–54
phases of development developmental relations between school
cycles in development of g 122–30; and cognitive performance 223–6
episodic thought 123–4; principle-based developmental theory of instruction
thought 129–30; realistic 217–92; conclusions 281–92; educating
representational thought 124–6; critical thought 273–80; enhancing
rule-based thought 126–9 intelligence in laboratory 229–46;
school and intellectual development
DA see denying the antecedent 219–28; towards theory of instruction
Dasen, Pierre 187 247–72
Deary, I. J. 220 dichotomous understanding 70, 173
debating origins of mental processes 178–9 Dienes, Zoltan 118
deception 72–3 differential tradition 22–30; conclusions
decontextualized language 230 30; research bridging experimental/
decoupling of factors 87 psychometric tradition 26–60
deductive inference 125–6 differentiation of mental processes 147–61;
deductive reasoning 11–12, 59–66, 108, changes of factor structure with growth
139–43, 156–7, 186, 197, 224, 259–60, 148–50; conclusions 160–61; mapping
265, 268; development of 59–63; changes in structural relations 150–54;
explicit content–implicit inference mapping integration–differentiation
60–62; explicit inference–implicit logic patterns 154–60
62; explicit logic–implicit metalogic difficulty estimations 138
62–3; explicit metalogic 63 Dimensional Change Card Sorting task
deductive syllogism 65 256–7, 269
defence mechanisms 174–5 dimensionality 52
defining critical thought 274–5 dimensions in g 113–19
Dehaene, S. 209 disjunctive syllogism 66
DeLoache, J. S. 89–90 distance from truth 277
Demetriou, Andreas 98, 103 distraction 46–7
Democritus 96 DNA 180
denial 174 domain-specific brain networks 81, 193–4
denying the antecedent 13, 21, 38, 64, domain-specificity see core domains
109, 112, 237–8 domains of thought 98–107
Descartes, René 1, 176 dopamine 183, 199
describing human mind 210–215 Downing, C. J. 59
Detterman’s theory of mental retardation dual encoding 89
148 dual representation 89–90
developing mind, school performance, Duncan, J. 43
learning 220–21 dynamic interaction 282
development of brain under adversity
209–210 early childhood cognizance 143–5
development of core theories 88–90 educating critical thought 273–80;
development of mind 283–5 conclusions 280; defining critical
development of reasoning 57–67, 61, 131; thinking 274–5; educating critical
conclusions 67; deductive reasoning thinking 275–80
Index 325
educating infants 251–4; educating conclusions 20–21; language of thought

awareness 254; educating core operators 15–17; mechanisms of attention/
in SCS 252–3; educating reasoning inhibition 8; mechanisms of integration
253–4; educating representational 11–15; mechanisms of representation/
capacity 253 processing 8–10; research in 232
educating the SCS 262–4, 267; in explaining development of core theories
adolescents 267; in primary school 88–90
children 262–4 explicit content–implicit inference
educational timing 250 60–62
eductive ability 24, 41, 52 explicit inference–implicit logic 62
effortful control 164 explicit logic–implicit metalogic 62–3
ego 163, 172–4 explicit metalogic 63
ego-defences see defence mechanisms extraversion 164–6, 169, 171–2
elaborated coordination 50 Eysenck, H. 162–4, 166–7, 171
elements of meaning 15–16; see also
language of thought Fabricius, W. V. 75
Emerson, H. F. 60 facial expression 167, 194
emotional information 170 facility to learn 226–7
emotional intelligence 165–6, 169–71 factor structure with growth 148–50
emotionality 167 factorial structures across cycles 149
emotions in the mind 162–76; see also factors of processing 55–6
personality Falmagne, R. J. 64
empirical mapping of dimensions in g false belief tasks 70–74, 71
113–19 Fernandes, C. 220
empirical mapping of four-fold Ferrer, E. 208
architecture 97–8 Fibonacci numbers sequence 245
empirical substantiation of SCSs 103–6 figurative intelligence 43
encoding relations 244 Fincham, J. M. 19
Engle, R. W. 285 “fixing” activities 78
enhancing intelligence in laboratory Flavell, J. H. 74–5, 86, 89
229–46; accelerating cognitive fluid intelligence 23–8, 42–3, 54, 66, 116,
development 232–7; conclusions 246; 120, 168, 185, 199–200, 222–3, 226,
enhancing intelligence 230–32; 230, 285
manipulating cognizance 237–46 Flynn effect 185, 223, 227, 289–90
enhancing mathematical thought 243–6 Flynn, J. R. 289
environment 179–84, 186–7, 206, 209 fMRI data 194
epigenesis framework 192 Fodor, J. A. 16, 42
episodic buffer 10 formal thought 37–9, 52, 59–63
episodic representations 123–4, 196, fostering intelligence 186
206–7, 284–5 four-fold architecture mapping 96, 97–8,
episodic thought 123–4 133, 211, 227, 231, 248, 251, 270,
Epstein, H. T. 207–8 273–4, 281–2
equation for logistic growth curve 152 four–seven year olds’ cognizance
evaluation of neo-Piagetian theories 134–43
53–5 framing debate about origins of mental
Evans, A. C. 207 processes 178–9
evolutionary endowment 272 Freud, Anna 163
executive control 30, 48–51, 53, 79, Freud, Sigmund 163, 172, 292
256–7; educating 256–7; structures Frischkorn, G. T. 198
48–51 Fritz, A. S. 72
existence of experience 106–7 frontal lobe grey matter volumes 210
existence–non-existence 107 functional classification criteria 185
experience 106–7, 126, 167, 172, 192 functional specializations 84
experimental cognitive tradition 7–21, functioning of mind 74–6
232; awareness/consciousness 17–19; Furnham, A. 175–6
326 Index
g see general intelligence Hagemann, D. 198

Gabrieli, J. D. E. 246 Haier, R. J. 199–200
Galotti, K. M. 65 Hala, S. 72
gamma oscillations 212 Halford, Graeme 51–3, 72, 183
Gardner, H. 23–4, 58, 88, 106 Hansell, N. K. 183
Gc see crystallized intelligence Hanson, J. L. 209
GCE see General Certificate of Education hard capacity theory 52
Gelman, R. 85 hard wiring 99
Gelman, Susan 58 Harmer, S. 85
General Certificate of Education 220 Harris, P. 60, 85
general factor of personality 168–71, 169, Harrison, T. L. 285
172 Hatano, G. 85
general intelligence 22–30, 41–2, 66, Head Start Program 230
89–90, 98, 105–6, 113–130, 154–60, heritability of g 180, 189
183–4, 230–31, 281–92; animal g Herrnstein, R. J. 178–9
183–4; central processes of 105–6; hidden relations 156
cycles in development of g 122–30; hierarchical organization 16, 21, 42, 73,
empirical mapping of dimensions in g 282–3
113–19; heritability of 180; markers of higher-order reasoning 197–8
developmental g 154–60; proxy for higher-order theory of mind tasks 73
113, 115–16; psychometric g 28, 113, Hill, W. D. 182
121, 135, 148, 177–81, 181, 192, 232, hippocampus 194–6, 201, 203, 211
284–7; re-morphing 177, 284 history of inquiry into human minds 1–4;
general model of changes 122 what is involved in intelligence 3–4
general principles of education 248–68; Horn, J. L. 23–4
educating adolescents 266–8; educating how developmental cycles relate to IQ
infants 251–4; educating pre-schoolers 131–2
254–62; educating primary school how personality, emotions, cognition
children 262–6 relate 166–71
generativity 16, 21, 282–3 how schooling influences cognition 221–3
generic rules organizing representations how to enhance intelligence 230–32
123, 129 Hudspeth, W. J. 208
genes 180–83 Hume, David 1, 176
genetic aspects of the mind 177–90; see also Hunt, E. 30
psychological aspects of the mind Hwang, C. P. 172
genetically determined constraints 271 hypothetico-deductive stance 129
genome-wide association studies 182
Gf see fluid intelligence ICNs see intrinsic connectivity networks
GFP see general factor of personality iconic components 14
Gladwin, T. 185 id 163
Gledd, J. N. 207 idealized development 155
global self-worth 79 idiot savants 17
glucose metabolism 207 implicit inference 60–62
Go/No-Go tasks 232, 269 implicit logic 62
goal monitoring 66 implicit metalogic 62–3
Gödelian dimensions 128–9, 291 implied relations 156
Godfredson, L. S. 22 improving conditional reasoning 237–43
Goswami, U. 58 Inagaki, K. 85
graduate training 222 incipient recognition 128
Graziano, W. G. 167–8 individual differences 30, 288
The g factor 162 inductive reasoning 11, 22–3, 57–9, 108,
guided reflection 233 142–3, 155, 186, 224, 259, 265, 268–9;
Gustafsson, J.-E. 222 development of 57–9
GWAS see genome-wide association inductive syllogism 65
studies Industrial Revolution 185
Index 327
infant education 251–4 IQ see intelligence quotient

infant temperament 153–4 iteration of alternatives 63
inference 11–17, 62, 67, 78, 107–8,
118–20, 127–9, 143–6, 155, 242–3; James, William 75–6, 174, 176
language of thought 15–17 Jensen, A. R. 22, 27, 30, 162, 221
inference-based reasoning 107–113 Johnson-Laird, P. N. 238
inferential awareness 67, 144–5, 155, 242 joint iteration 63
inferential identity 242 Jonides, J. 195
inferential relevance 130 Jouen, F. 126
inferential relevance mastery programme Jung, R. E. 199–200
209
inferential scaffolding 284 Kail, R. 44–5, 45
influence of culture 189–90 Kant, Immanuel 1, 17, 32, 76, 82, 106–7,
influence of intelligence on schooling 172, 176, 283, 292
220–21 Keil, F. C. 58, 87
informative convergence 149 Kid of Factor-Referenced Test 233–4
inhibition 8, 44–55, 46, 171–2, 223–4; Kievit, R. A. 180
mechanisms of 8; role of 44–55 knowing self 76–7
innate reasoning 63–6 “knowing self” 76
“I–self” 76–7, 174, 176 knowledge about mind 68–81; conclusions
insight 125 80–81; theory of mind 69–80; what
institutionalization 209–210 children know 68–9
instruction 247–72, 290–91; towards knowledge validation 268, 279
theory of 247–72 “known self” 76
integrated model of brain functioning Koch, C. 199
199–200 Koch, W. 221
integrating experience 172 Komatsu, L. K. 65
integration mechanisms 11–15 Korenman, S. 222
integration–differentiation pattern mapping Kovacs, K. 28–9
153, 154–60 Kovas, Y. 183
integrative inferential processes brain Krumm, S. 221
networks 196–8 Kyllonen, P. 27
intellectual development 180–88; culture Kyriakides, L. 103, 222
and 184–8
intellectual realism error 86 laboratory-enhanced intelligence 229–46;
intelligence 180–88, 220–21, 226–7; see also enhancing intelligence in
culture and 184–8; enhancement of laboratory
229–46; influence on schooling 220–21; Lacey, S. C. 195
personality and academic performance lack of theory of mind 275, 287; see also
226–7 theory of mind
intelligence quotient 26–7, 104, 131–2, Lamb, M. E. 172
131–2, 175, 178–90, 219–29, 286; and language of thought 15–17, 42–3, 120;
environmental impact 184–8; how elements of meaning 15–16; modularity
development cycles relate to 131–2; 16–17
increase in 219–22, 227–9; mental lapses of attention 280
scores with different IQ 132; reasoning late childhood cognizance 143–5
development 131 learning 220–21, 290–91; vs. instruction
intelligence–personality nexus 162 290–91; see also assessment for learning
intentional lies 276 learning to learn 268–70
interpersonal intelligence 23 Lehalle, H. 126
intervention 235–41, 240–41, 245 Lehman, D. 222
intrinsic connectivity networks 194–5 Lembert, R. O. 222
introspection 78 Leslie, A. M. 77
intuition 52, 72, 124 levels of consciousness model 79
inversion 37–8 limitations of control 278–9
328 Index
limited time mechanism 45, 51 mechanisms of attention 8

LOC see levels of consciousness model mediating role of cognizance 137–45,
locating cognitive functions in the brain 141; models of 141; tasks examining
193–200 138–9
Loevinger, H. 172–4 mediation of cognizance 134–46, 243; see
logic schemes 12–13 also recycling
logical dimensions of SCSs 106–7 Meltzoff, A. N. 74
logical fallacies 238–9, 241–3 “Me–self” 76–7, 174, 176
logical insight 65, 67 mental age 30, 131–2, 132, 147–50; trends
logical necessity 128, 239, 266 in IQ scores 132
logical substantiation 274 mental difficulties 270
logistic growth 152–4, 152, 153 mental economy 95
long-term memory 158, 199, 266 mental functions 195
LoT see language of thought mental mapping 101, 107
Luria, A. 185 mental models theory 238–9
Luyten, H. 222 mental power 47–8, 48
Lyons, K. E. 79 mental process training 229–46; see also
enhancing intelligence in laboratory
MA see mental age mental processes 194
MacDonald, D. 207 mental retardation 148
McGue, M. 182, 190 mental rotation 44, 99–101, 104, 108–111,
McIntyre, M. 167 114–16, 157, 193–4, 231–2, 263
Mackey, A. P. 209, 246 mental self-concept 228
Macnamara, J. 63 mentations 284–5
malnutritution 209–210 metacognition 19, 68–9, 79–80, 96
manipulating cognizance 237–46; metalogic 62–3; explicit 63; implicit 62–3
enhancing mathematical thought 243–6; metareasoning 60
improving conditional reasoning metarepresentation 77, 89–90, 119, 129,
237–43 260–62; educating 260–62
mapping changes in structural relations Miller, G. A. 10, 47, 50
150–54; logistic growth 152–4; Miller Singley, A. T. 209
Tucker–Drob differentiation model Miller’s magic number 7 47, 50
150–52 mind-related brain network changes
mapping integration–differentiation 202–9
patterns 154–60; cycle of principle- mindfulness 77–80
based representations 156–60; cycle of mind–brain development 191–215; see also
realistic representations 154–5; cycle of mapping mind–brain development
rule-based representations 155–6 “mind’s eye” 119, 127
mapping mind–brain development Miniusi, C. 201
191–215; architecture 193–201; misconception 87
conclusions 210–215; development modality 107
201–210 model of brain functioning 199–200
Marcus, G. F. 124 model of cognitive accomplishment 132
markers of developmental g 154–60; in model of intelligence 24–6
cycle of principle-based thought model-based reasoning 107–113
156–60; in cycle of realistic models of personality 162–6
representations 154–5; in cycle of modularity 15–17, 82–3
rule-based representations 155–6 modus ponens 12–14, 21, 38, 64–6,
Markman, E. 62 128–30, 237–8; validity of 128–30
Marriott, C. 85 modus tollens 12–14, 21, 38, 64, 108–9,
Massey, C. M. 85 108–9, 112, 237–8
mathematical proportionality 156–7, 203 Molina, M. 126
mathematical thought enhancement 107, monitoring mechanisms 103
243–6 Moshman, D. 60, 62
meaning making 18, 85, 211–12, 280 Mp see mental power
Index 329
MP see modus ponens overarching theory of growing mind

MT see modus tollens 93–216; cycles/phases of development
multiple-order theory of mind tasks 73 121–33; differentiation of mental
Murray, C. 178–9 processes through development cycles
musical intelligence 23 147–61; genetic/psychological/cultural
aspects of mind 177–90; mapping
Naglieri, J. A. 25–6 mind–brain development 191–215;
Naglieri’s PASS theory 25–6 organization of human mind 95–120;
nature of intelligence 32–3 personality/emotions in the mind
nature of knowledge 247–8 162–76; recycling and mediation of
necessity–contingency 107 cognizance 134–46
Nee, D. E. 195 overcoming mental difficulties 270
negative emotionality 164, 170
negative life outcomes 275 P-FIT see parieto-frontal integration
Neisser report 185 theory
neo-Piagetian theories 44–56, 88–90, parieto-frontal integration theory 199–200,
219–20; conclusions 55–6; evaluation of 211
53–5; role of speed of processing/ Park, A. T. 246
inhibition 44–55 parsimonious operation of domains 90–91
Neubauer, A. C. 221 Pascual-Leone, Juan 47–8, 50, 285
neural connectivity 182, 198 Pascual-Leonian reality 285
neural letters of thought 200–201, 211 Pashler, H. 19
neuroticism 164–8, 171–2, 275 pattern alignment 126
“new” knowledge 279–80 Paulus, M. 74, 124
Nisbett, R. E. 179, 185, 188, 222 Pavlovian classical conditioning 11
number 252 perception 16, 33–4, 57, 100–102, 143–6,
number word mapping 126, 129 154, 201
Nunez, M. 60 perceptual control 114
perceptual similarity 57–8
objects 252 personal preferences system 163
“old” knowledge 248 personality 162–76, 226–8, 291–2;
Onishi, K. H. 72 conclusions 175–6; intelligence and
ontogenetic time-scale 2 academic performance 226–7;
ontological status of mind 69–70 intelligence relations 171–2; and mind
openness to experience 164–5, 171 291–2; organization of 162–6; relations
operational consolidation 50 between personality/emotions/
operative development 33–9 cognition 166–71; self, ego, ego-
organization of human mind 74–6, defences 172–5
95–120; conclusions 119–20; empirical personality–intelligence relations 171–2
mapping of dimensions in g 113–19; perspective 34
empirical mapping of four-fold Petersen, S. 198
architecture 97–8; reasoning 107–113; phase-sensitive learning 246
specialized domains of thought 98–107 phases of development 121–33;
organization of personality 162–6; Big conclusions 132; cycles in development
Five factors of adult personality 164–5; of g 122–30; how developmental cycles
emotional intelligence 165–6; relate to IQ 131–2
temperament in infancy/childhood phonological loop 10
163–4 phylogenetic time-scale 1–2
orienting sensitivity 164, 167 physical causality 86–7
orienting system 198 physical resemblance 101–2
origins of mental processes 178–9 physical worlds 85–6
origins of reasoning 63–6 Piaget, Jean 15–16, 31–43, 63–4, 82,
Osherson, D. 62 88–90, 107, 118–19, 219–20, 234–7;
overarching principles integrating bridging with Vygotsky 234–7;
representations 123, 129 conclusions 41–3; current status of
330 Index
Piaget’s theory 40–41; mechanisms of Proust, J. 74, 124

change 39; nature/adaptive functions of proxy for g 113, 115–16
intelligence 32–3; Piaget’s theory 31–2; psychoanalytic theory 163, 172–3
stages of operative development 33–9 psychodynamic forces 163
“Piagetian reasoning” 107, 225, 284 psychological aspects of the mind 177–90;
Piaget’s theory of cognitive development animal g 183–4; conclusions 188–90;
31–2, 39–41, 69, 102, 187, 232–4, 250; culture, intelligence, intellectual
change mechanisms in 39; current status development 184–8; genes, intelligence,
of 40–41 intellectual development 180–83;
planned learning 249 origins of mental processes 178–9
Plato 1 psychological causality 87
plexins 182 psychological worlds 85–6
polis 179 psychometric dimensions of SCSs 106–7
positive life outcomes 227 psychometric g 28, 113, 121, 135, 148,
positive manifold 22 178–81, 181, 192, 232, 284–7
Posner, M. I. 198 psychometric requirements 119–20
possibility–impossibility 107 psychometric standards 103, 119
postnatal brain development 202 psychometric tradition 26–30, 219–20,
Pourcain, B. St. 183 232, 271–2; research in 232
pragmatic reasoning sequences 137, 139
pre-schooler critical thinking 275–6 quality of reason 107
pre-schooler education 254–62; core quantitative specialized capacity system
operators to representations 254–6; 158, 253, 255, 262–3
educating awareness 260–62; educating quantitative thought 102, 104–5, 110–111,
executive control 256–7; educating 139, 233, 245, 262–3, 267
reasoning 257–60 quantity of reason 107
pre-sorting episodic representation 123–4
precocious sensitivity 86–7 Raven’s Progressive Matrices 24–6, 25,
prediction of factors 87 101–2, 119, 126–32, 144, 155, 188,
prenatal thalamic spontaneous calcium 203, 220–21, 224, 232, 269, 283
waves 206 re-chunking 265
preoperational thought 35–7, 41 re-morphing g 177, 284
preservation of truth 15 reactivity 163
Pribram, K. H. 208 readiness to learn 226–7
primary ability 23 reading disability 183
primary reasoning 59–63 realistic classification criteria 185
primary school critical thinking 276–7 realistic mental representations 123, 206–7
primary school education 262–6; educating realistic representational thought 124–6,
reasoning 265; educating 154–5; markers of developmental g
representational capacity 264–5; 154–5
educating SCS 262–4; educating reality constraint 137
self-awareness 265–6 reasoning 11–15, 54, 57–67, 107–113,
principle-based thought 129–30, 136, 139, 150, 253–60, 265–8; development
156–60, 277–80; markers of of 57–67; educating 253–4, 257–60,
developmental g 156–60 265, 268; inference-, rule-, model-
principled reasoning 158 based 107–113; mechanisms of
process of cognizance in early–late integration 11–15; origins of 63–6; and
childhood 143–5 processing efficiency 66–7
processing efficiency 66–7, 133–5, 138–9, recall 264–5
150, 221–2, 221, 224, 246 reciprocity 37–8
processing of emotional information 170 recursion of factors 87
processing mechanisms 8–10 recursivity 16, 21, 42, 73, 77, 282–3, 286
proportional reasoning 231 recycling 48–51, 134–46, 136, 209;
propositional thinking 38 conclusions 145–6; and executive
Protzko, J. 230–31 control structures 48–51; specifying
Index 331
mediating role of cognizance 137–45; Salthouse, T. A. 45, 51

speed, working memory, reasoning SAT see Scholastic Assessment Test
135–7 saturation of developmental needs 249
reductionism 27 schizophrenia 69
reference factors of g 117 Schmitt, D. P. 188
refinement of understanding 80–81 Scholastic Assessment Test 220–22
refining cognitive skills 270–72 school and intellectual development
reflection 119 219–28; conclusions 227–8; developing
reflective abstraction 39, 119 mind, school performance, learning
rehearsal 195–6 220–21; how schooling influences
relation of reason 107 cognition 221–3; personality,
relational complexity 51–3 intelligence, academic performance
relational integration 196 226–7; school performance vs.
relational thought 261–2 cognitive performance 223–6
relations between personality, emotions, school performance 220–26; and cognitive
cognition 166–71 performance 223–6; see also academic
representation mechanisms 8–10 performance
representational alignment 118, 260–62, schooling 220–21; influence of intelligence
261, 283–4 on 220–21
representational capacity brain networks Schubert, A.-L. 198
167, 194–6, 253, 256–7, 264–8; Schweizer, K. 221
educating 253, 256–7, 264–5, 267–8 SCS see specialized capacity systems
representational efficiency 55–6, 134–5 SCS–IQ relation 104, 106
representational ensembles 254–6
second-order theory of mind tasks 73
representational intelligence 35–9; concrete
secondary reasoning 59–63
to formal thought 37–9; preoperational
segmented modelling 153–5, 158
to concrete thought 35–7
self 172–5; defence mechanisms 174–5;
representational resolution 257
representational role of mind 70–74, ego 172–4
254–5 self-assessed intelligence 169–70
reproductive ability 24–5 self-awareness 68–9, 77–81, 97, 114, 171,
research manipulating cognizance 237–46 173, 265–6; educating 265–6
resistance to illusion 232 self-categorization 79–80
response inhibition 66 self-concept 77, 79, 176, 228, 291–2
retention 20, 46, 196 self-consciousness 78, 283
Reverberi, C. 66 self-directed searching 270
reversibility 37, 42, 127, 282 self-discipline 170, 227
Rindemann, H. 221 self-esteem 168, 172, 291–2
Robinson, S. T. 246 self-evaluation 114–15, 130, 157–8
Rodriguez, J. A. 63 self-identity 173
Rohde, T. E. 220 self-inhibition 168
role of speed of processing 44–55 self-initiated movement 83, 85
role of working memory 47–53; executive self-knowledge 17, 23, 76–7, 189–90
control structures 48–51; mental power self-management 168
47–8; relational complexity 51–3 self-monitoring 119, 246
Ronald, A. 183 self-reflection 78–9
rostrolateral prefrontal cortex 212 self-regulation 53, 81, 119, 163, 166, 171,
“royal road” 171, 175, 246, 262 173–4, 188, 265–6
rule-based reasoning 107–113, 155–6, self-representation 68–9, 76, 168, 170,
206–8; markers of developmental g 172, 176, 227
155–6 self-restraint 164
rule-based thought 126–9, 143, 156 self-satisfaction 163
self-systems 168
salience detection 197, 206, 252 self-worth 169
Sally task 70–71, 73, 144; see also false sensorimotor intelligence 34–5, 49, 52, 232
belief tasks serotonin 199
332 Index
SES see socio-economic status speed of processing 8, 44–55, 135–7, 198;

Shayer, M. 234, 289 evaluation of neo-Piagetian theories
Sheppard, L. D. 27 53–5; role of working memory 47–53
Shing, Y. L. 222 spontaneous developmental time 242
Shipstead, Z. 285 SSS see specialized structural systems
short-term storage space 49–51, 50, 52–3, stage theory 39
95–6, 195–6 stages of operative development 33–9;
Shultz, T. R. 86 representational intelligence I 35–7;
similarity estimations 138, 157 representational intelligence II 37–9;
simultaneity mechanism 45 sensorimotor intelligence 34–5
single-nucleotide polymorphisms 182–3 Stanovich, K. 278
Skinnerian operant conditioning 11 Staudenmayer, H. 67
slave systems 10 Sternberg, R. J. 59
SLODRage see Spearman’s Law of Strand, S. 220
Diminishing Returns for Age stream of consciousness 75
small-world networks 203–4, 205 Stroop phenomenon 8, 27, 46, 114, 232
Smillie, L. D. 167 structural relations within phases 135–7
Smith, P. 220 structure mapping 51–3
Sniekers, S. 182 STSS see short-term storage space
SNPs see single-nucleotide polymorphisms subitization 16
social scaffolding 15 subjective experience 192
social scripts 168 subjective logicality 67
social specialized capacity system 253 subjectivity 279
social thought 103–4 substance of g 113–19, 122–30; cycles in
socialization 163 development of g 122–30; empirical
socially acceptable tendencies 163 mapping of dimensions 113–19
socio-economic status 210; see also adverse substantiation of SCSs 103–6
life conditions subvocal rehearsal loop 10
Sodian, B. 74, 124 superego 163
spatial specialized capacity system 158, surgency 164
252–3, 255–6; education of 267 survival training 219
spatial thought 101, 104–5, 107, 111, 139, syllogistic reasoning 107
156–7, 186, 245, 263, 267 synaptic density 203
Spearman, C. 282–3 synaptic pruning 206
Spearman’s Law of Diminishing Returns synaptogenesis 206
for Age 148 synchronization 20, 46, 213–14
Spearman’s two-factor theory 22–4, 41, synthesis of logic 36
52, 65, 119, 148 systematic training of mental processes 246
“special abilities” 82–3
specialized capacity systems 95–121, Taddonio, J. L. 67
149–50, 158–60, 166–8, 188, 225–6, Taplin, J. E. 66–7
231–7, 248–66; association with tasks addressing reasoning 108–9
cognitive functions 159; cognitive style tasks examining mediating role of
and ability 160; educating core cognizance 138–9
operators in 252–3 tasks for learning to reason programmes
specialized domains of thought 95, 258–9
98–107, 99–100; convergence of SCS temperament 163–4, 167; infancy and
dimensions 106–7; domains 101–3; childhood 163–4
empirical substantiation of SCSs tendencies about role of cognizance 145–6
103–6 testing for transfer in SCSs 233–4
specialized storage systems 10 Thatcher, R. W. 207–8
specialized structural systems 98–9 theory of mind 44, 68–80, 199–200, 275,
specific genes, specific aspects 188–90 287; know yourself 76–7; lack of 275;
specifying mediating role of cognizance ontological status of mind 69–70;
137–45; from 4–7 years 138–43; from organization/functioning of mind 74–6;
early to late childhood 143–5 representational/causal role of mind
Index 333
70–74; self-awareness/mindfulness unary relations 52

77–80 unconscious tendencies 163, 292
theory-of-mind tasks 71 understanding mind’s organization
“theory–theory” interpretation of 74–6
developing mind 84 understanding ontological status of mind
thinking about thinking 39–40 69–70
Thompson, P. M. 207 unifocal coordination 50
Thomson, L. A. 220
Thurstone, L. L. 23–4 van Aken, M. 171
Thut, G. 201 van der Maas, H. L. J. 28, 29, 56
Tijus, C. 126 Vendetti, M. S. 197
Toga, A. W. 207 verbal thought 245
ToM see theory of mind Vernon, P. A. 27
top-down causal effects 230–31, 245, visual perception 100, 201
286–7 visuo-spatial sketchpad 10
top-down mediation 144–6 Voelz, S. 65
total processing space 49–50, 52 Vygotsky, Lev 15–16, 234–7, 269–70
towards overarching theory of growing
mind 281–92; composition of growing Wagner, S. 58
mind 281–3; development of the mind Wall Street Journal 178
283–5; instruction/learning 290–91; Wason’s card selection task 13–14, 13, 63,
personality/mind 291–2; why some 107–8
people are smarter 285–90 Wellman, H. M. 73
towards theory of instruction 247–72; Wendelken, C. 208, 213
assessment for learning 268–70; Weschler, D. 162
conclusions 270–72; general principles Wessles, N. 172
of education 248–68 Westenberg, P. M. 173
TPS see total processing space what children know about mind 68–9
Trabasso, T. 40 what is involved in intelligence 3–4
traditions of research on human mind Whitaker, K. J. 208
5–92; awareness about the mind 68–81; Whittall, S. 85
core domains 82–91; development of Whorfian hypothesis 15
reasoning 57–67; differential tradition why some people are smarter 285–90
22–30; experimental cognitive tradition Wimmer, R. D. 198–9
7–21; neo-Piagetian theories 44–56; Winner, E. 58
Piaget’s theory 31–43 Winship, C. 222
trainability 232 WISC-III 104, 105, 119, 131–2, 225,
training intelligence 229–32; research 232, 283
232 WM see working memory
trait emotional intelligence 170 WMC see working memory capacity
Tran, R. 19 Woods, R. P. 207
transcription 213 working memory 8–10, 9, 28–9, 45–55,
transfer in/across SCSs 233–4 95–7, 116, 135–43, 150, 154, 180–86;
transfer to g 243–6 relations between speed, reasoning and
transfer of training effects 236, 246 135–7, 136; role of 47–53
transivity 41, 52 working memory capacity 27–8, 51,
transparent relationship 125 219–20
trial-and-error actions 102
Tucker–Drob differentiation model Zabaneh, D. 182
150–52, 151, 154, 157–8 Zelazo, P. D. 79
turn-taking 260 Ziegler, M. 221
types of thought domain 101–3; zone of proximal development 234–5,
categorical thought 101–2; causal 235, 269–70
thought 102–3; quantitative thought ZPD see zone of proximal
102; social thought 103; spatial thought development
101

Growing Minds A Developmental Theory

Uploaded by

Copyright:

Available Formats

Growing Minds A Developmental Theory

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Growing Minds A Developmental Theory

Uploaded by

Copyright:

Available Formats

GROWING MINDS

Interest in the human mind is a centuries-old fascination, dating back to Plato,

• What are the cognitive processes underlying intelligence?

George Spanoudis is Associate Professor in Psychology at the University

Andreas Demetriou and

ISBN: 978-1-138-68982-4 (hbk)

Powerful contexts allowing our research to go uninterrupted

1 The experimental cognitive tradition 7

2 The differential tradition 22

5 The development of reasoning 57

6 Awareness and knowledge about the mind 68

8 The organization of the human mind 95

9 Cycles and phases of development 121

10 Recycling and mediation of cognizance 134

11 Differentiation and binding of mental processes through

12 Personality and emotions in the mind 162

13 Genetic, psychological, and cultural aspects of the mind 177

14 Mapping mind-brain development 191

15 School and intellectual development 219

16 Enhancing intelligence in the laboratory 229

17 Towards a theory of instruction 247

18 Educating critical thought 273

19 Conclusions: towards an overarching theory of the

6.2 A model generating inferences about mental states without

14.1 Brain regions associated with mental processes presented in

7. Can intelligence be enhanced by specific interventions? That is, is it possible to

What is involved in intelligence?

the developmental tradition focused on cognitive development (Case, 1985; Piaget,

Mechanisms of attention and inhibition

Mechanisms of representation and processing

Mechanisms of integration: association, inference,

Deductive reasoning is a more constrained kind of reasoning. First, it only

FIGURE 1.3 Wason’s selection task examining conditional reasoning

Language of Thought: inference and modularity

Language of Thought (LoT)

handled at different developmental phases. Language and other representational

Awareness and consciousness

making-meaning and problem-solving needs. The input to consciousness is in-

consciousness is always in operation, receiving signals from cognitive functioning

information, to integration and meaning-making. At this level, truth and validity

The differential, or psychometric, tradition focused on individual differences in

FIGURE 2.1 The three stratum Cattell-Horn-Carroll model of intelligence

FIGURE 2.2 Examples of matrices of increasing complexity similar to those included in

problem-solving goals), attention (stimulus selection and inhibition), simultaneous

Research bridging the experimental and the psychometric

Science is reductionist. It seeks to reduce complex phenomena to simpler

because of failing to meet its executive attention demands” (p.163). It is noted,

Developmental psychology, by definition, adopts a developmental perspective to deal

The nature and the adaptive functions of intelligence

The stages of operative development

Representational intelligence I: from preoperational to concrete

Representational intelligence II: from concrete to formal thought

The mechanisms of change in Piaget’s theory

The current status of Piaget’s theory

Hundreds of such studies demonstrated that when extra-logical factors such as

The similarities above (emphasis on reasoning, assumptions about a general LoT,

The role of speed of processing and inhibition

underlying mechanism, which may be correlated to age-related change in the rate

irrelevant information, remove task-irrelevant information from the field or space

1 The experimental cognitive tradition 7

2 The differential tradition 22

5 The development of reasoning 57

6 Awareness and knowledge about the mind 68

8 The organization of the human mind 95

9 Cycles and phases of development 121

10 Recycling and mediation of cognizance 134

12 Personality and emotions in the mind 162

13 Genetic, psychological, and cultural aspects of the mind 177

14 Mapping mind-brain development 191

15 School and intellectual development 219

16 Enhancing intelligence in the laboratory 229

17 Towards a theory of instruction 247

18 Educating critical thought 273

TPS = OS + STSS (1.2)