Content uploaded by Annette Karmiloff-Smith
Author content
All content in this area was uploaded by Annette Karmiloff-Smith on Jul 18, 2014
Content may be subject to copyright.
From Constructivism to
Neuroconstructivism:
e Activity-Dependent
Structuring of the
Human Brain
Annette Karmilo-Smith
Introduction
In philosophy, psychology, and linguistics, the pendulum regularly
swings from nativist claims regarding hardwired domain-specificity
to empiricist claims regarding domain-general processes that en-
able learning. Although Piaget was by no means an empiricist, his
constructivist theory does embody domain-general mechanisms of
change (assimilation, accommodation, and equilibration), purported
to apply to all cognitive domains such as number, space, physical
causality, social cognition, and language (Piaget, 1936). Several com-
putational modeling approaches to child development are also
domain-general in nature. For example, production system modeling
(Klahr et al., 1987) has been used to account for children’s problem-
solving across a wide variety of domains (Klahr, 1992, 2000; Klahr and
Dunbar, 1989).
While strongly supporting Piaget’s view that infants are active
participants in their own learning and that cognitive structures are
emergent and not innately specified (Piaget, 1936, 1966), I hereby
propose a different view from the domain-general or domain-specific
approaches—a domain-relevant view of progressive change—which
argues that the brain starts out with a number of basic-level biases,
1
1
After Piaget
each of which is somewhat more relevant to the processing of certain
kinds of input over others and which become domain-specific over
time through neuronal competition and a process of gradual modu-
larization (Elman et al., 1996; Karmiloff-Smith, 1992, 1998). But first,
let us briefly examine one of the most influential approaches to infant
cognition in the literature over recent decades.
Nativist Approaches to Cognitive and Neural Development
As the popularity of Piagetian approaches to infant development
began to wane, the nativist approach arguing for innately specified,
cognitive-level core knowledge became particularly influential. At
least four arguments are used to support nativist claims. First, they
drew on the field of adult neuropsychology in which patients whose
brains had previously developed normally subsequently suffer a
brain trauma and end up with a pattern of relatively dissociated im-
pairments, for example, cases of agrammatism, prosopagnosia, or
agnosia. is, theorists argue, indicates that the brain is composed of
independently functioning, domain-specific modules (Baron-Cohen,
1998; Butterworth, 2005; Duchaine, 2000; Gopnik, 1997; Temple, 1997;
Van der Lely, 2005). e second argument emanated from a version
of evolutionary psychology, which maintains that the human brain
has evolved into the equivalent of a Swiss army knife in which each
innately-specified module in the newborn brain is exquisitely adapted
for a specific, independent function (Barkow et al., 1992; Duchaine
et al., 2001). (e analogy ignores the fact that most users of the Swiss
army knife actually employ for all purposes only a few of the numer-
ous special-purpose tools their knife possesses!) e third argument
was based on the capacities of what is known as the “competent
infant,” that is, claims that young infants possess innately specified
core knowledge or core principles (e.g., Butterworth, 2005; Carey,
2009; Kinzler and Spelke, 2007; Pinker, 1999; Spelke, 2000; Spelke and
Kinzler, 2007, 2009). In nativist accounts, learning was banished from
having any explanatory role (Piatelli-Palmerini, 2001). Finally, children
with genetic disorders presenting with uneven cognitive profiles and
displaying a juxtaposition of scores “in the normal range” in one or
more domains alongside serious deficits in others were argued to
illustrate the dissociation of general intelligence from independently
functioning domains like grammar, number, face processing, and the
like. So why should we consider these arguments to be less compelling
than they initially seem?
2
From Constructivism to Neuroconstructivism
3
First and foremost, they are all static. ey ignore what Piaget
deemed to be essential: the developmental history of the organism.
Indeed, a crucial component of Piaget’s epistemology focused on
the growth of knowledge over ontogenetic time, not a snapshot of
knowledge at one specific point in time like birth. Second, those of
nativist persuasion tend to disregard everything we now know about
the progressive development of the infant brain.
e Dynamics of Infant Brain Development
Just as Piaget stressed the ways in which children build their own
cognitive structures, so neuroscientists are increasingly focusing on
how the child’s activities sculpt the resulting structure of the brain.
e brain is rather like a large, very wrinkled walnut! Yet if its wrinkled
layers and folds were to be flattened out to a smooth surface, it would
cover a full-sized football pitch. Within all those complicated folds are
nearly a billion neurons. In fact, the newborn’s brain contains most
of the neurons that it will use throughout life, although research has
recently shown that even in adulthood some areas of the brain continue
to generate new neurons. By about eight months of age, the infant brain
will have about one thousand trillion connections between neurons,
which is roughly twice the amount of connections found in adult
brains. But this is a temporary difference. With time, the connections
that have proven useful will get increasingly strengthened, whereas
those which haven’t been active often will be “pruned” or weakened.
So very gradually over time, the connections in the brain become
increasingly specialized, fine tuned, and more adultlike.
While some macrostructures of the brain, like the overall six-layer
structure of cortex, may be under general genetic constraints, most of
the microcircuitry of the brain turns out to be the result of complex
multilevel interactions over developmental time. Indeed, fine-tuning of
functional brain organization is a progressive, activity-dependent pro-
cess (Kandel et al., 2000). According to Huttenlocher (2002), plasticity
itself changes over developmental time, with some mechanisms avail-
able throughout the lifetime (increase in synaptic strength, decrease in
local inhibition, dendritic sprouting, formation of new synapses, and
formation of new neurons), whereas others are only available to the
early developing brain (utilization of unspecified labile synapses, com-
petition for synaptic sites, persistence of normally transient connec-
tions, and myelination). Moreover, changes in plasticity turn out to be
region specific, not general across the brain, suggesting (omas, 2003)
After Piaget
that there is no such overarching thing as “the brain’s plasticity.” Many
structures (e.g., dendrites, axons, and synapses) initially undergo
exuberant growth, followed by a period of pruning in which the pro-
cessing of environmental input gradually moulds the way in which the
microstructure of the brain emerges.
In young infants, neural processing tends initially to be diffuse
across several regions in both hemispheres, but over developmental
time with the continuous processing of inputs, brain activity becomes
increasingly restricted to more specific networks in the left (LH) or
right hemispheres (RH) (Durston et al., 2006; Johnson, 2001). And
this gradual process of modularization over developmental time
(Karmiloff-Smith, 1992), as opposed to the notion of built-in modules,
improves processing efficiency. A recent study by Minagawa-Kawai
et al. (2007) examined language-specific phonemic contrasts in
infants from three months to twenty-eight months and found that
the onset of activation in different areas of cortex was age-specific.
Another study suggests that comprehension of single words moves
from bilateral processing between thirteen and seventeen months
to left lateralized processing at twenty months (Mills et al., 1997).
Like vocabulary development, processing of human faces starts out
with bilateral activity, with the brain displaying similar signatures
for other stimuli like cars or monkey faces (de Haan et al., 2002;
Pascalis et al., 2001). But by the end of the first year, the brain becomes
increasingly fine-tuned for processing human faces, with other stimuli
displaying different neural signatures, as well as increasing localization
for human faces to specific networks in the RH (de Haan et al., 2002;
Peelen et al., 2009).
Many questions about the developing brain of course remain to be
answered. For example, what explains individual differences across
different brain regions? Even the brains of monozygotic twins end up
rather different, highlighting the role of gene–environment interac-
tions. Further, we need to know more about how hemispheric differ-
ences influence neural change over developmental time. In adults, the
RH seems to be implicated in more parallel, coarse-grained, integrative
processing, whereas the LH is involved in more serial, fine-grained,
predictive processing. How does this develop in children? Is informa-
tion passage through the corpus callosum always faster from RH to
LH than from LH to RH, or does this alter over developmental time?
Certainly the thickness of the corpus callosum fibers changes develop-
mentally over a long period of time between infancy and adolescence
4
From Constructivism to Neuroconstructivism
5
(Keshavan et al., 2002). Finally, short-range gray matter connectivity
is greater in children, while long-range white matter connectivity
develops considerably more slowly over time (Huttenlocher, 2002). All
of these and other developmental changes in the brain must be taken
into account when, for instance, analyzing neuroimaging data over
time, because the brain continues to undergo quite major changes
even at puberty (Blakemore, 2010; Crone et al., 2008).
Importance of the Brain’s Resting State
Functional Connectivity
Since self-organizing processes are a necessary part of the explana-
tion of how the brain changes over developmental time, it is critical
to understand the spontaneous neural activity that occurs without
external stimuli. In adult neuroscience, a resting state circuit has been
identified, comprising a large network of brain regions, associated with
task-irrelevant mental processes: precuneus/posterior cingulate cor-
tex, medial prefrontal cortex, and medial, lateral, and inferior parietal
cortex. It turns out that more of the brain’s energy is spent on intrinsic
rather than evoked activity; in fact that brain is never at rest. Studies
point to a high degree of functional connectivity during rest, that is,
interregional temporal synchrony (Raichle et al., 2001), indicating that
this spontaneous neural activity is not merely random activation.
Spontaneous brain activity during sleep, for instance, plays a criti-
cal role in the consolidation of memory, involving redistribution of
memory representations from temporary hippocampal storage to
neocortical long-term storage sites. e dialogue between neocortex
and hippocampus generates sharp-wave ripples and is orchestrated by
the <1 Hz EEG slow oscillation during slow-wave sleep. Unlike adults
whose sleep patterns involve cycles of slow-wave sleep to rapid eye
movement (REM) sleep, young infants fall directly into REM sleep,
with the proportion of slow-wave sleep increasing only very progres-
sively over the first years of life (Hill et al., 2007). How this affects the
resting state neural processes involved in sleep-related consolidation
of learning in children remains to be more fully elucidated.
e importance of developmental changes in resting state brain
activity will, I believe, become very prominent in the next few years.
Years ago (Karmiloff-Smith, 1986, 1992), I put forward a cognitive-
level developmental hypothesis—the representational redescription
(RR) hypothesis—postulating that what is specifically human to
human intelligence is a process by which task-specific representations
After Piaget
stored as procedures in the brain become, via an internally-generated
process of RR, domain-general knowledge to the brain. is internal
self-organizing process, I argued, is generated by behavioral mastery,
not by negative feedback, and allows knowledge relevant to one
domain to become transportable to other domains without the need
to process new external input. In other words, RR was argued to be
an internally-generated process occurring outside the processing of
external stimuli. With the current advances in developmental brain
imaging, it should be possible to assess the hypothesis by detecting
specific networks in cerebral resting state underlying RR.
Gradual Developmental Process of Modularization
Neuroconstructivism, with many epistemological overlaps with
Piagetian constructivism but incorporating knowledge of brain
structure and function, argues that if the adult brain is in any way
modular, it is the product of an emergent developmental process of
modularization, not its starting point (Karmiloff-Smith, 1992, 1998,
2006, 2007; Elman et al., 1996; Johnson et al., 2002; Westermann
et al., 2007). A crucial error is to conflate the specialized brains of
adults, which have developed normally prior to damage in later life,
with those of infants and children, which are still in the process of
developing (Karmiloff-Smith et al., 2002). To date, there is no evidence
to suggest functional specificity of gene expression in the brain, that
is, no evidence that individual genes which are expressed in the brain
target discrete cortical regions. Rather, gene expression seems to be
widespread showing diffuse, large-scale gradients across cortex (Kings-
bury and Finlay, 2001). Moreover, genetic mutations contributing to
developmental disorders in infants are likely to affect widespread
systems within the brain (Karmiloff-Smith, 1998). is does not pre-
clude that the outcome of the dynamic developmental process could
end up with some areas being more impaired than others, but this
would not be a pattern necessarily apparent at the outset but due to
the result of processing demands of certain kinds of inputs to those
areas and to differences in synaptogenesis across various cerebral
regions (Huttenlocher and Dabholkar, 1997). By contrast, the nativist
modular view underestimates the dynamics of the changing patterns
of connectivity within and across different brain areas during develop-
ment. Indeed, the same overt behavior may be subserved by different
underlying neural substrates at different ages during development
(Karmiloff-Smith, 1998).
6
From Constructivism to Neuroconstructivism
7
In studies of typically developing infants and of those with
developmental disorders, researchers have shown how different
cortical pathways become increasingly specialized and localized as a
result of being recruited for specific tasks over developmental time
(Elman et al., 1996; Johnson, 2001). Various areas of the brain start
out by competing to process different inputs (Karmiloff-Smith, 1998),
because cortical regions initially respond to a wide variety of different
stimuli and task situations. In other words, the infant brain displays
more widespread activity than the older child or adult brain when
processing specific kinds of inputs. With time, however, the develop-
ing brain starts to show increasing specialization and localization of
function as certain areas win out in the competitive processing. How
does neuroconstructivism explain this?
It is important to stress that the neuroconstructivist approach does
not imply that the neonate brain is a blank slate with no structure, as
empiricists would claim. Nor does it entertain the possibility that just
any part of the brain can process any and all inputs. On the contrary,
neurconstructivism maintains that the neonate cortex has some
regional differentiation in terms of types of neuron, density of neurons,
firing thresholds, and so on. ese differences are not domain-specific
aimed at the sole processing of proprietary inputs, nor do they amount
to domain-general constraints. Rather, they are ‘domain-relevant,’
that is, different parts of the brain have small structural differences,
which turn out to be more appropriate or relevant to certain kinds of
processing over others. But initially, brain activity is widespread for
processing all types of input, and competition between regions gradu-
ally settles which domain-relevant circuits become domain-specific
over time. Emergent specialization of function (e.g., for faces) can be
viewed as the fine-tuning of initially domain-relevant but coarsely
coded systems (e.g., for visual patterns), but this is for visual patterns
in general, not for faces in particular. e face specialization emerges
from the interaction between the environment (huge numbers of
face stimuli over time) and the initial visual processing constraints,
not from an innately-specified, dedicated face-processing module, as
some would argue (e.g., Duchaine and Nakayama, 2006).
So starting out with tiny differences across brain regions in terms of
the patterns of connectivity, the balance of neurotransmitters, synaptic
density, neuronal type or orientation, and the like, some areas of the
brain are somewhat more suited (i.e., more relevant in terms of their
computational properties) than others to the processing of certain
After Piaget
kinds of input, and over time they ultimately win out. In other words,
the computational properties of a particular brain circuit may be more
relevant to certain types of processing (e.g., holistic vs. componential
processing) than others, although they are initially not specific to
that type of processing only. It is only after developmental time and
repeated processing that such a circuit becomes domain specific as
ontogenesis proceeds (Karmiloff-Smith, 1992, 1998). ere is thus a
gradual process of recruitment of particular pathways and structures
for specific functions (Elman et al., 1996) such that brain pathways
that were previously partially activated in a wide range of task contexts
increasingly confine their activation to a narrower range of inputs and
situations (Johnson et al., 2002).
e neuroconstructivist position is supported by neuroimaging
research showing that the functional specialization of brain regions
is highly context sensitive and depends on interactions with other
brain regions through feedback processes and top-down modulation
(Friston and Price, 2001). is process becomes most evident in brain
organization in people who lack one sensory modality. For example,
in individuals who have been blind from an early age, visual cortex
is recruited for the tactile modality instead, that is, braille reading
(Sadato et al., 1996). Moreover, using transcranial magnetic stimulation
(TMS) to block processing in this area affects tactile identification of
braille letters in the blind but not in seeing people, who instead display
impaired visual processing when stimulated in this area (Cohen et al.,
1997). It therefore appears that the functional development of cortical
regions is strongly constrained by available sensory inputs and that the
final organization of the cortex is an outcome of interactive processes
such as competition for space.
With Piaget, Taking Development Seriously
Paradoxically, numerous studies of infants and children (typical
or atypical) are not developmental at all because they take a static
approach. e truly developmental, constructivist and neuroconstruc-
tivist perspectives embrace a developmental way of thinking, irrespec-
tive of the age of the population studied; even studies of infants can be
nondevelopmental and studies of adults can focus on developmental
change. In my view, to understand developmental outcomes, it is vital
to identify full developmental trajectories, to assess how progressive
change occurs from infancy onwards, and how parts of the develop-
ing system may interact with other parts differently at different times
8
From Constructivism to Neuroconstructivism
9
across ontogenesis. A process that is vital, say, at time 2 may no longer
play a role at time 5. Yet its delay at time 2 may have been crucial to a
healthy developmental trajectory and outcome. Indeed, developmental
timing is amongst the most important of factors that need to be taken
into account when endeavoring to understand human development,
particularly in the atypical case. Even when scores are “in the normal
range,” this doesn’t necessarily imply a normal developmental trajec-
tory, without examining the cognitive and neural levels underlying
the behaviour. Yet researchers of a nativist persuasion will often label
the domain in which scores fall in the normal range as “intact” or
“preserved” alongside other domains considered to be impaired.
In my view, the very notion of “intactness/preservation” has a static
connotation and implies genetic determinism, as if states in the brain
were hardwired, unchanging, and unaffected by developmental or
environmental factors. e neuroconstructivist view, by contrast,
considers the brain as a self-structuring, dynamically changing organ-
ism over developmental time as a function of multiple interactions at
multiple levels, including gene expression (e.g., Casey, 2002; Johnson,
2001). Research on birds and mammals eloquently illustrates this
point. Extensive evidence from studies of the neural and epigenetic
consequences of song listening and song production in passerine
birds (Bolhuis et al., 2000) shows how gene expression changes
over developmental time and may be significantly more important
during learning than during final production. Rather than something
fixed and predetermined, gene expression in the birds turned out to
be a function of how many elements the bird copied from its tutor.
A second example comes from early mammalian development and
also underlines the potential role of the environment in shaping
long-term patterns of gene expression (Kaffman and Meaney, 2007).
ese authors studied brain development in rodent pups and traced
how differences in maternal grooming behavior influence patterns
of gene expression in their pups, which have lifelong effects. e
researchers showed that rather than thinking of gene expression as
preprogrammed, differences in the amount of postnatal pup grooming
and stroking change the amount of gene expression of genes involved
in the body’s responses to stress, and that these changes last the pups’
lifetime. ese kinds of dynamic environment–gene relations are likely
to be a pervasive feature of mammalian brain development, includ-
ing that of humans. In general, epigenesis is not deterministic under
tight genetic control. Rather, as Gottlieb stressed (Gottlieb, 2007),
After Piaget
epigenesis is probabilistic and only under very broad genetic control,
a point Piaget would certainly have endorsed.
A Computational Connectionist Model of Emergent
Brain Specialization
Many scientists consider the different functions of the ventral
and dorsal streams in the brain to be built in. e possibility that
their differing functions might emerge is rarely considered. Yet an
eloquent, in principle illustration of emergent specialization comes
from a computational model of these two streams (O’Reilly and
McClelland, 1992). e network had an input layer, two hidden layers
(A and B), and an output layer. A small difference in activation levels
(equivalent to a difference in neuronal firing thresholds) between A
and B sufficed, after both competing to process identical inputs, to
result in one stream ending up processing where objects were and
the other stream processing the features of objects. In other words,
the ‘where’ and ‘what’ pathways were not built into the network, with
one only processing spatial information about objects and the other
only processing featural information about objects. On the contrary,
early on, both streams started off by processing all inputs. However,
an initial, tiny difference in the speed of activation levels was suf-
ficient, over time, to progressively give rise to gradual specialization
of the where/what pathways. Without this firing threshold difference,
both streams would have continued to process all inputs in a domain-
general way. So a tiny bias in the initial state can suffice to give rise
to a major domain-specific difference in the end state after a lengthy
period of processing. In this sense, it is possible that domain-specific
outcomes may not even be possible without the gradual process of
development over time.
Implications of a Neuroconstructivist Approach
for Atypical Development
Rather than invoking “intact” and “impaired” modules, assessing
atypical development in terms of cascading developmental effects
of tiny perturbations should result in a better understanding of
genetic disorders in children. However, perhaps the notion of impaired
versus intact brain systems in uneven cognitive profiles might be
useful for clinical practice, even if theoretically it totally underplays
the role of development. If a patient has scores in the normal range
in a specific domain, surely there is no need to consider remediation
10
From Constructivism to Neuroconstructivism
11
in that domain? e nativist would probably agree and focus solely
on the domains of deficit. However, the neuroconstructivist would
not rule out intervention also in a proficient domain. For instance,
take a patient who presents with a serious deficit in, say, number,
yet scores in the normal range for all other domains. It would be
tempting in such a case to tailor remediation solely to the domain
of number. But that misses the very point of the neuroconstructivist
framework. First, the scientist would need to trace back to infancy
the origins of the number deficit which might not be in the number
domain directly; it could be a deficit in the visual system in scanning
arrays of objects. But a scanning deficit might affect other domains,
although to a lesser degree, meaning that these other domains could
look normal in subsequent development but may camouflage subtle
deficits. Once one explores multiple, low-level interacting processes
that underpin early development, this leads to a more dynamic view
of remediation.
Neuroconstructivism does not rule out domain-specificity; it
argues that it cannot be taken for granted and must always be ques-
tioned. Unlike the nativist perspective, neuroconstructivism—like
Piaget’s constructivism—offers a truly developmental approach that
focuses on change and emergent outcomes. And, every aspect of
development turns out to be dynamic and interactive. Genes do not
act in isolation in a predetermined way. Even the FOXP2 gene, about
which there was much excitement regarding its role in human lan-
guage, must be thought of in terms of the downstream gene targets
to which FOXP2 binds. e profiles of those downstream genes sug-
gest roles in a wide range of general, not domain-specific, functions
including morphogenesis, neurite growth, axon guidance, synaptic
plasticity, and neurotransmission (Teramitsu and White, 2007).
is is a very different level from theorizing at the level of cognitive
modules and making claims about “a gene for language” and points
to the multilevel complexities of understanding human development
in any domain.
It is clear that development—whether typical or atypical, whether
human or nonhuman—is fundamentally characterized by plasticity
for learning, with the infant brain dynamically structuring itself over
the course of ontogeny. e infant brain is not a collection of static,
built-in modules handed down by evolution. Rather, the infant brain
is the emergent property of dynamic multidirectional interactions
between biological, physical, and social constraints.
After Piaget
References
Barkow, J. H., Cosmides, L. and Tooby, J. (Eds.) (1992). e Adapted Mind:
Evolutionary Psychology and the Generation of Culture. New York, NY:
Oxford University Press.
Baron-Cohen, S. (1998). Does the study of autism justify minimalist innate
modularity? Learning and Individual Differences, 10, 179–91.
Blakemore, S. J. (2010). e developing social brain: Implications for education.
Neuron, 65, 744–47.
Bolhuis, J. J., Zijlstra, G. G. O., den Boer-Visser, A. M. and van der Zee, E. A.
(2000). Localized neuronal activation in the zebra finch brain is related to the
strength of song learning. PNAS, 97, 5, 2282–85.
Butterworth, B. (2005). Developmental dyscalculia. In: J. I. D. Campbell
(Ed.) Handbook of Mathematical Cognition. London: Psychology Press,
pp. 455–67.
Carey, S. (2009). e Origin of Concepts. Oxford: Oxford University Press.
Casey, B. J. (2002). Windows into the human brain. Science, 296, 1409–10.
Cohen, L. G., Celnik, P., Pascual-Leone, A., Corwell, B., Falz, L., Dambrosia, J.,
Honda, M., Sadato, N., Gerloff, C., Catalá, M. D., and Hallett, M. (1997).
Functional relevance of cross-modal plasticity in blind humans. Nature, 389,
180–83.
Crone, E. A. M., Bullens, L., Plas, E. A. A., van der Kijkuit, E. J., and Zelazo, P. D.
(2008). Developmental changes and individual differences in risk and perspec-
tive taking in adolescence. Developmental Psychopathology, 20, 1213–29.
de Haan, M., Humphreys, K., and Johnson, M. H. (2002). Developing a brain
specialized for face perception: A converging methods approach. Develop-
mental Psychobiology, 40, 200–12.
Duchaine, B. (2000). Developmental prosopagnosia with normal configural
processing. NeuroReport, 11, 1, 79–83.
Duchaine, B., Cosmides, L., and Tooby, J. (2001). Evolutionary psychology and
the brain. Current Opinion in Neurobiology, 11, 225–30.
Duchaine, B. and Nakayama, K. (2006). Developmental prosopagnosia: a window
to content-specific face processing. Current Opinion in Neurobiology, 16,
2, 166–73.
Durston, S., Davidson, M. C., Tottenham, N., Galvan, A., Spicer, J., Fosella,
J. A., and Casey, B. J. (2006). A shift from diffuse to focal cortical activity with
development. Developmental Science, 9, 1, 1–8.
Elman, J. L., Bates, E., Johnson, M. H., Karmiloff-Smith, A., Parisi, D., and
Plunkett, K. (1996). Rethinking Innateness: A Connectionist Perspective on
Development. Cambridge, MA: MIT Press.
Friston, K. J. and Price, C. J. (2001). Dynamic representations and generative
models of brain function. Brain Research Bulletin, 54, 275–85.
Gopnik, M. (1997). Language deficits and genetic factors. Trends in Cognitive
Science, 1, 1, 5–9.
Gottlieb, G. (2007). Probabilistic epigenesis. Developmental Science, 10, 1–11.
Hill, C. M., Hogan, A. M., and Karmiloff-Smith, A. (2007). To sleep, perchance
to enrich learning? Archives of Disease in Childhood, 92, 7, 637–42.
Huttenlocher, P. R. (2002). Neural Plasticity: e Effects of the Environment on
the Development of the Cerebral Cortex. Cambridge, MA: Harvard University
Press.
12
From Constructivism to Neuroconstructivism
13
Huttenlocher, P. R. and Dabholkar, A.S. (1997) Regional differences in syn-
aptogenesis in human cerebral cortex. Journal of Comparative Neurology,
387, 167–78.
Johnson, M. H. (2001). Functional brain development in humans. Nature Reviews
Neuroscience, 2, 475–83.
Johnson, M. J., Halit, H., Grice, S. J., and Karmiloff-Smith, A. (2002). Neuroimaging
and developmental disorders: a perspective from multiple levels of analysis.
Developmental Psychopathology, 14, 521–36.
Kaffman, A. and Meaney, M. J. (2007). Neurodevelopmental sequelae of
postnatal maternal care in rodents: clinical and research implications of
molecular insights. Journal of Child Psychology and Psychiatry, 48, 3–4,
224–44.
Kandel, E. R., Jessell, T. M., and Sanes, J. R. (2000). Sensory experience and
the fine tuning of synaptic connections. In: E. R. Kanel, J. H. Schwarz, and
T. M. Jessell (Eds.), Principles of Neural Science (4th edition). New York, NY:
Elsevier.
Karmiloff-Smith, A. (1986). From metaprocesses to conscious access: Evidence
from children’s metalinguistic and repair data. Cognition, 23, 2, 95–147.
Karmiloff-Smith, A. (1992). Beyond Modularity: A Developmental Approach to
Cognitive Science. Cambridge, MA: MIT Press.
Karmiloff-Smith, A. (1998). Development itself is the key to understanding
developmental disorders. Trends in Cognitive Science, 2, 10, 389–98.
Karmiloff-Smith, A. (2006). e tortuous route from genes to behaviour: A
neuroconstructivist approach. Journal of Cognitive and Affective Behavioural
Neuroscience, 6, 1, 9–17.
Karmiloff-Smith, A. (2007). Atypical epigenesis. Developmental Science,
10, 1, 84–88.
Karmiloff-Smith, A., Scerif, G., and omas, M.S.C. (2002). Different approaches
to relating genotype to phenotype in developmental disorders. Developmental
Psychobiology, 40, 311–22.
Keshavan, M. S., Diwadkar, V. A., DeBellis, M., Dick, E., Kotwal, R., Rosenberg, D. R.,
Sweeney, J. A., Minshew, N., and Pettegrew, J. W. (2002). Development of the
corpus callosum in childhood, adolescence and early adulthood. Life Sciences,
70, 16, 1909–22.
Kingsbury, M. A. and Finlay, B. L. (2001). e cortex in multidimensional space:
Where do cortical areas come from? Developmental Science, 4, 2, 125–42.
Kinzler, K.D. and Spelke, E.S. (2007). Core systems in human cognition. Progress
in Brain Research, 164, 257–64.
Klahr, D. (1992). Information processing approaches to cognitive development.
In: M. H. Bornstein and M. E. Lamb (Eds.), Developmental Psychology: An
Advanced Textbook (3rd edition). Hillsdale, NJ: Erlbaum. pp. 273–335.
Klahr, D. (2000). Exploring Science: e Cognition and Development of Discovery
Processes. Cambridge, MA: MIT Press.
Klahr, D. and Dunbar, K. (1989). Developmental differences in scientific dis-
covery processes. In: D. Klahr and K. Kotovsky (Eds.), Complex Information
Processing: e Impact of Herbert A. Simon. Hillsdale, NJ: Lawrence Erlbaum
Associates. pp. 109–43.
Klahr, D., Langley, P., and Neches, R. (Eds.) (1987) Production System Models of
Learning and Development. Cambridge, MA: MIT Press.
After Piaget
Minagawa-Kawai, Y., Mori, K., Naoi, N., and Kojima, S. (2007). Neural attunement
processes in infants during the acquisition of a language-specific phonemic
contrast. Journal of Neuroscience, 27, 315–21.
Mills, D. L., Coffy-Corins, S., and Neville, H. (1997). Language comprehension
and cerebral specialization from 13 to 20 months. Developmental Psychology,
13, 397–445.
O’Reilly, R. C. and McClelland, J. L. (1992). e self-organization of spatially
invariant representations. Technical Report PDP.CNS.92.5, Carnegie Mellon
University.
Pascalis, O., Demont, E., de Haan, M., and Campbell, R. (2001). Recognition of
faces of different species: A developmental study between 5 and 8 years of age.
Infant and Child Development, 10, 39–45.
Peelen, M. V., Glaser, B., Vuilleumier, P., and Eliez, S. (2009). Differential develop-
ment of selectivity for faces and bodies in the fusiform gyrus. Developmental
Science, 12, 12, 16–25.
Piaget, J. (1936). La naissance de l’intelligence chez l’enfant. Delachaux & Niestlé
(e Origins of Intelligence in Children. New York: International University
Press, 1952).
Piaget, J. (1966). Biologie et connaissance. Paris: Gallimard (Biology and Knowl-
edge: An Essay on the Relations between Organic Regulations and Cognitive
Processes. Chicago: e University of Chicago Press).
Piatelli-Palmerini, M. (2001). Speaking of learning. Nature, 411, 887–88.
Pinker, S. (1999). Words and Rules. London: Weidenfeld & Nicolson.
Raichle, M., MacLeod, A. M., Snyder, A. Z., Powers, W. J., Gusnard, D. A.,
and Shulman, G. L. (2001). A default mode of brain function. PNAS, 98, 2,
676–82.
Sadato, N., Pascual-Leone, A., Grafman, J., Ibañez, V., Deiber, M.-P., Dold, G., and
Hallett, M. (1996). Activation of the primary visual cortex by Braille reading
in blind subjects. Nature, 380, 526–28.
Spelke, E. S. (2000). Core knowledge. American Psychologist, 55, 1233–43.
Spelke, E. S. and Kinzler, K. D. (2007). Core knowledge. Developmental Science,
10, 89–96.
Spelke, E. S. and Kinzler, K. D. (2009). Innateness, learning and rationality. Child
Development Perspectives, 3, 96–98.
Temple, C. M. (1997). Cognitive neuropsychology and its application to children.
Journal of Child Psychology and Psychiatry, 38, 1, 27–52.
Teramitsu, I. and White S. A. (2007). Motor learning: e FoxP2 puzzle piece.
Current Biology, 18, R335–R337.
omas, M. S. C. (2003). Book reviews: Limits on plasticity. Journal of Cognition
and Development, 4, 1, 95–121.
Van der Lely, H. K. J. (2005). Domain-specific cognitive systems: Insight from
grammatical specific language impairment. Trends in Cognitive Sciences,
9, 2, 53–59.
Westermann, G., Mareschal, D., Johnson, M. H., Sirois, S., Spratling, M.,
and Thomas, M. (2007). Neuroconstructivism. Developmental Science,
10, 75–83.
14
