It has long been an assumption that serious and chronic
childhood psychiatric disorders reflect, at least in part,
relatively subtle abnormalities of brain development.
Strong indirect support for this has been provided by
the association of childhood psychiatric disorders with
numerous neurological and neurodevelopmental dis-
orders  and from clinical neuropsychological studies
over many decades, indicating abnormal brain function
in child psychiatric populations . Because of the lim-
itations of clinical investigation to validate subtle brain
abnormalities, child psychiatric research has explored
new methods for studying the brain. Brain imaging is an
innovative technology that to date has best furthered the
goal of understanding normal and psychiatrically abnor-
mal brain structure and function. With the advent of
non-invasive brain magnetic resonance imaging (MRI)
Imaging normal and abnormal brain development:
new perspectives for child psychiatry
Judith L. Rapoport, F. Xavier Castellanos, Nitin Gogate, Kristin Janson,
Shawn Kohler, Phillip Nelson
Objective: The availability of non-invasive brain imaging permits the study of normal and
abnormal brain development in childhood and adolescence. This paper summarizes current
knowledge of brain abnormalities of two conditions, attention deficit hyperactivity disorder
(ADHD) and childhood onset schizophrenia (COS), and illustrates how such findings are
bringing clinical and preclinical perspectives closer together.
Method: A selected review is presented of the pattern and temporal characteristics of
anatomic brain magnetic resonance imaging (MRI) studies in ADHD and COS. These
results are discussed in terms of candidate mechanisms suggested by studies in develop-
Results: There are consistent, diagnostically specific patterns of brain abnormality for
ADHD and COS. Attention deficit hyperactivity disorder is characterized by a slightly smaller
(4%) total brain volume (both white and grey matter), less-consistent abnormalities of the
basal ganglia and a striking (15%) decrease in posterior inferior cerebellar vermal volume.
These changes do not progress with age. In contrast, patients with COS have smaller brain
volume due to a 10% decrease in cortical grey volume. Moreover, in COS there is a pro-
gressive loss of regional grey volume particularly in frontal and temporal regions during ado-
Conclusions: In ADHD, the developmental pattern suggests an early non-progressive
‘lesion’ involving neurotrophic factors controlling overall brain growth and selected
dopamine circuits. In contrast, in COS, which shows progressive grey matter loss, various
candidate processes influencing later synaptic and dendritic pruning are suggested by
human post-mortem and developmental animal studies.
Key words: ADHD, brain development, childhood schizophrenia.
Australian and New Zealand Journal of Psychiatry 2001; 35:272–281
Judith L. Rapoport, Chief (Correspondence); F. Xavier Castellanos,
Medical Officer; Nitin Gogate, Clinical Fellow; Kristin Janson, IRTA
Child Psychiatry Branch, National Institute of Mental Health,
Building 10, Room 3N202, 10 Center Drive MSC 1600, Bethesda,
Maryland 20892-1600, USA. Email: email@example.com
Shawn Kohler, Post-Baccalaureate IRTA Fellow; Phillip Nelson, Head
Section on Neurobiology, Laboratory of Developmental Neurobiology
Branch, National Institute of Child Health and Human Development,
Received 17 January 2001; accepted 1 February 2001.
methodology, imaging data can now be acquired for
paediatric populations. Moreover, these data are con-
verging with new information on the organization and
function of circuits in the developing brain, and on the
molecular mediators of these changes. It is hoped that
imaging studies in child psychiatric populations, will not
only define the brain systems underlying illness, but also
suggest candidate molecules for genetic studies. Based
on the nature, location and temporal pattern of these
abnormalities, and preclinical findings, we will be able
to make more specific and testable hypotheses about the
aetiology of these disorders. The quantitative study of
brain development during childhood and adolescence
with MRI began in the late 1980s (e.g. ). Subsequent
cross-sectional [4–7] and mixed longitudinal/cross-
sectional studies (e.g. ) have confirmed that although
total brain volume changes between ages 5 and 18 are
negligible, there are robust and complex changes in white
and grey matter. White matter volume increases linearly
during this age range, reflecting increasing myelination
[4,9], while grey matter volume increases until early
to mid-adolescence before decreasing during late ado-
lescence , presumably from synaptic pruning and
reduction of neuropil. A special feature of these norma-
tive data is that they were acquired in parallel with
prospective clinical studies, so that brain development
for psychiatrically abnormal populations can be com-
pared. Elegant studies of known chromosomal abnor-
malities, such as Down syndrome and Rett’s syndrome,
all testify to abnormal development in these known
retardation syndromes [10,11]. These produce gross dis-
turbances of central nervous system (CNS) development
for which the cause is known and, in principle, relatively
simple screening could be used to detect and prevent
such disorders. More subtle non-dementing disorders,
however, have proven more difficult. There remains con-
siderable controversy, for example, about even the valid-
ity of the diagnosis of attention deficit hyperactivity
disorder (ADHD). Brain morphometric studies may help
validate this diagnosis, and are summarized in detail
here. Since prospective longitudinal rescan data is now
available, we can address not only how the two patient
groups, ADHD and childhood onset schizophrenia
(COS), differed from controls at their initial evaluation
but also examine their differing developmental course.
Anatomic brain magnetic resonance imaging
studies of attention deficit hyperactivity
disorder and childhood onset schizophrenia
Attention deficit hyperactivity disorder
Table 1 summarizes some representative anatomic
brain studies that have been carried out to date in ADHD.
As seen in Table 1, several independent studies have
found a smaller total brain volume. This represents a
global reduction of grey and white equally (not shown
here) [12–16]. There are also subtle and not entirely con-
sistent abnormalities of various basal ganglia structures
(Table 2) [12,13,15,17] and, most striking, a consistent
and significant reduction of the volume of the posterior
inferior cerebellar vermis (Table 2) [16,18,19]. These
findings support other biological models of ADHD
implicating frontal–basal ganglia and dopaminergic
These abnormalities appear to be a fixed, rather than
an ongoing, process. Longitudinal changes during child-
hood and adolescence did not differ between our 73
ADHD subjects, and 75 healthy matched controls studied
prospectively with 2 and 4 year follow-up rescan .
These anatomic abnormalities are not due to stimulant
drug effects since the 17 medication-naïve patients
showed the same brain pattern. Thus, in contrast to COS
(described below), the smaller total brain and cerebellar
vermis in ADHD, seems due to an earlier process (at
least before age 4, the earliest age at which these scans
were obtained). Moreover, since the trajectories of the
total and regional brain development does not differ
between ADHD patients and controls, severe inattention
or impulsivity per se is not likely to cause the late pro-
gressive abnormalities seen for the schizophrenic group.
J.L. RAPOPORT, F.X. CASTELLANOS, N. GOGATE, K. JANSON, S. KOHLER, P. NELSON273
Table 1.Anatomic brain magnetic resonance imaging studies in ADHD
% SmallerEffect size
Aylward et al. 1996 
Filipek et al. 1997 
Bullmore et al. 1999 
Castellanos et al. 1996, 2000 [15,21]
Representative brain volume
Right hemisphere volume
Grey and white matter voxels
Grey and white matter voxels
ADHD, attention deficit hyperactivity disorder.
NORMAL AND ABNORMAL BRAIN DEVELOPMENT274
Based on these findings, candidate processes for the
abnormalities in ADHD focus on prenatal or early post-
Childhood onset schizophrenia
Schizophrenia is a heterogeneous illness both with
respect to clinical phenomenology and aetiology. Age of
onset of illness has provided an avenue to understanding
of disease across all of medicine, with earlier onset cases
often having more striking or homogeneous risk factors
and/or differing pathophysiologies. For this reason, a
study of COS, defined as onset of psychotic symptom by
the 13th birthday, has been ongoing since 1990 at the
National Institute of Mental Heath. These severely ill
patients are less contaminated with substance abuse and
other factors seen in later-onset populations. Moreover,
the childhood onset illness appears continuous with
respect to clinical and neurobiological measures .
A comparison of 46 COS and 84 healthy volunteers
extends our previous studies [23,24] showing that COS
had decreased total brain volume and increased lateral
ventricular volume as seen in adult onset schizophrenia.
Unlike the subtle global decrease in grey and white
matter seen in ADHD, the decreased brain volume here
is due exclusively to the robust 10% decrease in cortical
Table 2.Anatomic brain magnetic resonance imaging studies in ADHD: basal ganglia and cerebellum findings
Aylward et al.
ADHD probands Contrast subjectsFindingsComments
10 boys11 normal controls;
16 boys with ADHD + TS
Globus pallidus volume
smaller in ADHD
(significant on left)
Normal symmetry in
prefrontal brain, caudate,
and globus pallidus
significantly decreased in
ADHD; cerebellum volume
Caudates smaller (only
significant left) in ADHD;
right anterior superior white
matter also significantly
diminished; posterior white
matter volumes decreased
only in stimulant non-
Right caudate larger in
ADHD; larger caudate
nucleus areas associated
with poorer performance
on tests of attention and
higher ratings on Conner
Teachers Rating Scale
Caudate and putamen
also smaller in ADHD,
although not significantly
Data support hypothesis
cortical circuitry mediates
ADHD, particularly on
Castellanos et al.
55 health controls 55 healthy controls
Filipek et al.
15 males (same
subjects as Semrud-
Clikeman et al.
15 normal controls First study to quantify
grey and white matter
suggest using medication
response to subtype
Mataro et al.
11 adolescents with
19 healthy control
subjects (three girls in
Single slice axial MRI-
not volumetric measure;
only study to find larger
size in ADHD
Berquin et al.
46 right-handed boys
Castellanos 1996) 
47 right-handed boysPosterior inferior cerebellar
vermis volume and area
covariance for total
covariance for total
convariance for total
cerebral volume; most
robust and replicated
finding in ADHD
Mostofsky et al.
12 males 23 malesPosterior inferior cerebellar
vermis area significantly
Castellanos et al.
50 girls 50 girlsPosterior inferior cerebellar
vermis volume significantly
smaller (–12%, same as
ADHD, attention deficit hyperactivity disorder; MRI, magnetic resonance imaging; TS, Tourette’s Syndrome.
grey matter, as the total white matter volume does not
differ significantly between the COS and healthy groups.
These findings are summarized in Table 3.
In addition, Table 4 and Figure 1 show prospective
longitudinal brain MRI rescan measures for these COS
cases. There is increasing ventricular volume, and decreas-
ing cortical grey and medial temporal lobe structures
across 2, 4 and (for a smaller number of cases), 6 years
after their initial scan. Here, too, regional grey–white
segmentation showed the progressive loss to be for grey
matter only [25–28]. Figure 1 shows this progressive
loss during adolescence for regional cortical grey matter.
These observations have led us to view adolescence as
a time-limited window in which progressive brain
changes in schizophrenia may be observed.
Clinically these changes parallel a decline in full-scale
IQ  and lack of normal maturation of neurological
status  for these patients. These progressive changes
are not likely to be due to medication. Data from a
contrast group of 17 children with transient psychotic
symptoms treated with the same typical and atypical
antipsychotic drugs, but without clinical progression to
schizophrenia (MDI or ‘multidimensionally impaired’
group)  show no progressive ventricular volume
J.L. RAPOPORT, F.X. CASTELLANOS, N. GOGATE, K. JANSON, S. KOHLER, P. NELSON275
Table 3.Brain MRI volumes (mL) for childhood onset schizophrenics (n = 46) and healthy controls (n = 82)
(n = 46)
Mean (± SD)
(n = 82)
Mean (± SD)
Region of the brain
Total cerebral volume
Total white 389.3 (51.8)383.8 (50.5)
Regional grey volumes
Superior temporal gyrus
*All probabilities are ANCOVAs except for ANOVAs as indicated by *. COS, childhood onset schizophrenia; MRI, magnetic
increase or grey matter volume loss relative to healthy
These findings are consistent with the neuropathology
of schizophrenia [33,34]. The loss of cortical volume is
consistent with models of progressive widespread, subtle
disruption and decreasing connectivity of multiple cor-
tical regions hypothesized for schizophrenia [35,36].
Regulation of brain development: clinical
A combination of genetic and environmental factors
controls the development of the CNS. The process is
highly plastic, involving a series of sequential and par-
allel events, and a disruption of one step can greatly
influence later processes .
Because of high malleability in the human brain, it
would seem possible that the stress and abnormal thought
and behaviour experienced by psychotic patients might
be responsible for the progressive loss of grey matter and
cytoarchitectonic deficiencies found in these individuals.
For example, could schizophrenic behaviour be a cause
rather than an effect of the decreased and decreasing
frontal grey brain volume? Could severe inattention
and impulsivity produce small grey matter volume?
Although the influence of plasticity is most pronounced
during the critical periods of postnatal development and
declines during adolescence, a significant level of adapt-
ability continues to exist even into adulthood. The fact
that developed brains can be induced into schizophrenic
states  is at least consistent with the idea that changes
in synaptic efficacy could play a critical role in the onset
of schizophrenia [39,40].
Important preclinical demonstrations of brain plastic-
ity have suggested two distinct forms. A classic study
by Hubel, Wiesel and LeVay exemplifies the first .
Specifically, monkeys with one eye completely covered
from birth were found to develop a greater number of
afferents from the lateral geniculate nucleus to the visual
cortex of the open eye than to the sutured eye. In other
words, synaptic pathways failing to attain the expected
level of stimulation by the environment lost efficiency,
while those that were properly activated increased in
connectivity and gained efficiency . This process has
been called experience-expectant maturation. A second
form of plasticity can be demonstrated by studies with
rats which have suggested that different levels of general
sensory stimulation have distinct influences on brain
development. Animals raised in ‘enriched’environments
have been found to have significantly greater cerebral
cortical weight and thickness, dendritic length and spine
density, total dendritic material per neuron, acetyl-
cholinestrase activity, RNA/DNA ratio, RNA diversity
NORMAL AND ABNORMAL BRAIN DEVELOPMENT276
Table 4.Progressive brain changes during adolescence in COS
Findings for COS vs controls
Progressive reduction in superior temporal gyrus, lateral
hippocampus at 2-year follow up (n = 10)
Increased lateral ventricular volume at 2-year follow up
(COS n = 16; normal controls n = 24)
Progressive decrease in TCV and hippocampus; increase
in ventricular volume (COS n = 42; normal controls n = 74)
Frontal and temporal grey volume loss for COS ages 13–18
Loss = 5% more than for healthy controls
(COS n = 15; normal controls n = 43)
Only two time points 
Clear progression across ages 14–16 
No progression after age 19; longitudinal and cross-sectional
(208 scans) 
Differences most significant for frontal and temporal grey
Longitudinal (Dx X time X region, p = 0.004) 
COS, childhood onset schizophrenia; TCV, total cerebral volume.
adolescence for childhood onset schizophrenia.
*p = 0.6, **p = 0.2, ***p = 0.001. Based on 172 scans
from 98 healthy controls and 98 scans from 48 patients
with childhood onset schizophrenia. (Based on
Rapoport et al.  and Giedd et al. 
Regional cortical grey matter loss during
and brain-specific protein concentrations than animals
raised in ‘impoverished’environments . These results
suggest that environmental stimuli that are not predeter-
mined and vary between individuals in a species con-
tribute to brain structure: a second form of plasticity
described as experience-dependent maturation. Applied
to the schizophrenic patient, then, could experience-
dependent plasticity cause the cytoarchitechtonic decline
and grey matter deficiencies?
Probably not. In general, experiments involving plas-
ticity apply drastic conditions. In the aforementioned
examples, animals were exposed to highly unusual if
not devastating visual and sensory environments. Other
studies involve equally extreme measures. For instance,
plasticity is often studied through the induction of lesions.
Neuronal input to a target zone is severed, and anatomi-
cal, biochemical and electrophysiolocial data are col-
lected to determine whether re-innervation has occurred
with other healthy afferent areas . ‘Postlesional plas-
ticity’ has been found in many regions including the
cerebral cortex  and cerebellum .
In considering the relationship between plasticity and
the onset of schizophrenia, therefore, it is important to
recognize that even the abnormality in environmental
input to the brain of a schizophrenic patient is consider-
ably less severe than the conditions involved with most
of the experiments exploring plasticity. The possibility
that abnormal behaviour per se could induce abnormal
connectivity exists, and experimental studies of plastic-
ity using functional imaging may demonstrate such
effects. For structural findings reviewed here, however, it
seems more probable that schizophrenic behaviours
result from these abnormalities rather than the reverse.
Processes that may underlie the brain abnormalities
in attention deficit hyperactivity disorder
While total brain volume is slightly but significantly
decreased in ADHD, the total and regional growth curves
for this group run parallel to normal brain curves .
Thus, the events leading to the anatomic differences in
ADHD probably occur early in neurodevelopment. The
period of critical neurodevelopmental organizational
events  is influenced by a complex interplay of
various genetic and epigenetic factors such as neurotrans-
mitters, neurotrophins and growth factors, and cytokines,
along with hormonal influences [48,49]. Even a subtle
injury during this vulnerable process of neurodevelop-
ment and organization in utero (second or third trimester)
can affect the brain development and size globally, thus
explaining the changes seen in ADHD . Various phe-
nomenological observations in ADHD may help us
understand the mechanism of this global size reduction.
Incidence of pregnancy-related complications, and pre-
maturity are slightly higher in ADHD . However, the
strongly genetic nature of ADHD, and lack of specificity
of findings for what mediates the effects, for example
prematurity, leaves us without a specific hypothesis of
what might mediate this subtle global reduction in
brain volume. The striking volume reduction of the pos-
terior inferior lobule of cerebellar vermis in ADHD are
also important in this regard (see Table 1). This region
of cerebellar vermis is highly dopaminergic  and
appears, like most brain volumetric measures to be
highly heritable, [Giedd J, Castellanos X: unpublished
data]. The posterior inferior vermis, thus, may be an
important part of cerebello-striato-frontal circuitry and
hence in the aetiopathogenesis of ADHD. Recently, it
was observed that brain-derived neurotrophic factor
(BDNF) and neurotrophin-3 (NT-3) mRNAs (important
brain growth modulators) are colocalized to specific
ventral mesencephalic dopaminergic neurons . Thus
any defect in dopaminergic circuitry could alter these
growth modulators and ultimately the brain size and
development. Allelic variations of several dopaminergic
genes (DA transporter, DA receptor 4) have been asso-
ciated with ADHD , however, these alleles have
been examined in ADHD populations and show no sig-
nificant association with any of these brain volumetric
measures . Conversely, subtle abnormalities in
BDNF, NT-3 and others may cause localized abnormal-
ity of dopamine circuitry. This awaits exploration. Finally,
differences in the volume of caudate nucleus could also
point us toward aetiopathogenic mechanisms.
The caudate nucleus volume is decreased in boys with
ADHD [56,57]. This is interesting as healthy girls have a
larger caudate nucleus, probably owing to higher concen-
tration of oestrogen receptors in the region. In the CNS,
oestrogen plays a role in regulation of gene transcription,
can act as an antioxidant for toxic substances, and can
also act as a neuroprotectant . In theory, this might
explain the smaller caudate volume in ADHD boys and
lower incidence of ADHD in girls. However, this would
not account for the fact that brain MRI findings for
ADHD boys and girls are very similar (see Table 1).
Candidate processes for childhood onset
The smaller brain sizes in COS and ADHD indicate
that some compartment or compartments in the brain are
being reduced. Two major theories have been put forth
that name different compartments as the source of the
loss . The first proposes that the cause is either a
developmental failure to produce the proper number of
neurons or it is neurodegenerative disease that results in
J.L. RAPOPORT, F.X. CASTELLANOS, N. GOGATE, K. JANSON, S. KOHLER, P. NELSON277
an overall neuronal loss in the brain. The second theory
proposes that the differences in brain volume are due to
reduction of or reduced formation of the neuropil, which
consists of the axonal and dendritic arbours that make up
a large fraction of the grey matter in the cortex [34,60].
This reduction of the neuropil could result from reduc-
tion in the numbers of synaptic connections made between
neurons. In humans, detailed post-mortem electron micro-
scopy work has shown that there is normally a postnatal
increase in synaptic connections followed by a decrease
in synaptic connections that extends as far as mid-
adolescence in some parts of the brain .
As noted above, the decreased volume in ADHD may
be related to the first factor (a reduced number of neurons
generated during early neurodevelopment) while COS
is due to the second factor (an excessive degree of syn-
apse reduction of an initially fairly normal number of
synapses and neurons.
Post-mortem studies to determine if there is a reduced
cell number in the brains of people with schizophrenia
have not produced consistent results in many parts of the
brain [34,60]. Conversely, a number of studies using
stereological methods have found increased densities of
neurons in the prefrontal cortex and temporal lobe
[62–64]. In addition, these neurons have smaller cell
bodies, and it has been shown that the cell bodies are
proportional to the level of dendritic and axonal arbour-
ization . Other evidence shows decreased loss of
expression of synaptic markers in schizophrenia  and
most recently, decreased expression of several functional
genes important in presynaptic function and develop-
ment . Thus, while there is not direct evidence for
synaptic loss in COS there is a good deal of evidence
that supports the reduced neuropil hypothesis. Detailed,
quantitative electron microscopy studies are needed to
directly address this crucial question as to the cellular
locus of the brain abnormality in COS. The two theories
are not mutually exclusive, but the evidence is strong in
favour of a reduced neuropil playing a prominent role in
schizophrenia (see [34,60] for details).
In normal development it is known that excess
numbers of synaptic connections are formed initially and
as development progresses the extra connections are
eliminated or pruned. Considerable information is avail-
able concerning some of the mechanisms regulating the
generation of neurons, development of the synapses
between the neurons and the process by which some
synapses are lost during development. We will briefly
review some of this information, focusing primarily on
the process of synapse elimination, since the available
evidence suggest that this may play a key role in COS.
Synapse elimination is particularly well characterized
in two systems, namely in the climbing fibre (CF)
system which innervates Purkinje cells (PC) in the cere-
bellum, and in motor neurons which innervate skeletal
muscle fibres, although synapse elimination has been
demonstrated in many systems . The possible mech-
anisms behind pathologic elimination of the synapses
are too numerous and complex to thoroughly review
here. However, a general idea of the possibilities can be
gleaned from the well-studied model of elimination
at the neuromuscular junction (NMJ). In this system, at
birth in the rodent, all muscle fibres are innervated by
at least two axons, and by 3 weeks postnatally all but one
axon has been eliminated from all of the muscle cells .
This process requires activation of the system because it
fails to take place in paralysed preparations. Some of the
steps in this coupling between activation and synapse
elimination are becoming clarified, and seem to involve
a number of protein kinases, enzymes which regulate
protein function by altering their state of phosphoryla-
tion. In particular, neurotransmitter receptors are known
to be targets for kinase action, and their physiology and
stability in the membrane of nerve and muscle are
greatly affected by the addition of phosphate groups.
One major link in the chain leading to synapse elimina-
tion, therefore, may be activation of appropriate kinases,
phosphorylation of neurotransmitter receptors and sub-
sequent, selective destabilization of the synapses involv-
ing those receptors . Loss of synaptic acetylcholine
receptors has been shown to be an early step in the loss
of synapses at the NMJ . Selective loss of synapses
may be due to differential activation of localized kinases,
which have different effects on receptor stability.
Evidence for the involvement of a kinase (protein kinase
C or PKC) in synapse elimination in the central nervous
system has been obtained in experiments on the CF/PC
system mentioned above. On average PCs in the cere-
bellum are innervated by 3.5 CFs at birth but as normal
development proceeds, the number of CFs forming a syn-
aptic connection with each PC is reduced to one .
Genetic manipulations have given some clues as to the
mechanisms involved in synapse elimination. Mutation
of one isoform of PKC which inactivates the kinase has
been shown to block synapse elimination in the mouse
cerebellum . In the mouse cerebellum the expression
of the PKC-gamma isoform goes up in PCs during the
period during which multiple synapses from CFs are
reduced to single innervations. In mutant mice in which
PKC-gamma is inactivated there is marked reduction in
the elimination of the initial multiple innervations of PCs
by CFs. Thus the activity of this particular molecule,
PKC, is essential for normal elimination of redundant
synapses. It might be that excessive activity of the enzyme
would produce the abnormally high degree of synapse
elimination postulated to be related to COS.
NORMAL AND ABNORMAL BRAIN DEVELOPMENT 278
Trophic factors have been shown to have powerful
effects on neuronal survival and synaptic structure and
function during development [50,72–74]. Competition
for a limited supply of trophic material has been postu-
lated to account for at least a portion of the synapse loss
that occurs during development. Blockade of the trophin
BDNF has been shown to prevent the normally occurring
pruning that is essential for development of the normal
architecture of the visual cortex  and an inadequate
supply could result in inadequate development or sur-
vival of cortical synapses.
Cortical neurons are generated during a sharply delim-
ited time window relatively early in development and the
total number of neurons can be drastically affected by
various manipulations during this period. For instance, it
has been shown that a neuropeptide, vasoactive intestinal
peptide (VIP) controls the duration of the mitotic cycle
in the neuroblasts in the ventricular germinal zone and
that this affects the total number of neurons that get born
during development. Antagonists of VIP given during the
critical neuron-generating period (and only during this
period) result in markedly microcephalic animals .
Some such interference with the process of neuron gen-
eration could be involved in the early deficit in brain size
seen in ADHD.
Following on the initial speculation of Feinberg ,
McGlashan and Hoffman  and others have used
computer modelling of neural networks to test the
plausibility of the ‘over-pruning’ hypothesis of schizo-
phrenia. Their experiments characterized the behaviour
of over-pruned neural networks and found strong paral-
lels between schizophrenia and the networks’behaviour.
In summary, as we hope this report illustrates, clinical
brain imaging studies are bringing child psychiatry and
the developmental neurosciences ever closer. We present
here some possible candidates, but expect that much
greater specificity and converging information will
emerge from genetic linkage and association studies that
will be carried out over the next decade. Some of these
studies are ongoing with these populations at the NIMH.
1. Rutter M, Graham P,Yule W. A neuropsychiatric study in
Childhood. Clinics in Developmental Medicine Nos 35, 36.
London: S.I.M.P./Heinemann, 1970.
2. Goodman R. Brain disorders. In: Rutter M, Taylor E, Hersov L,
eds. Child and adolescent psychiatry: modern approaches. 3rd
edn. London: Blackwell, 1994:172–190.
3. Jernigan TL, Tallal P. Late childhood changes in brain
morphology observable with MRI. Developmental Medicine and
Child Neurology 1990; 32:379–385.
4. Reiss AL, Abrams MT, Singer HS, Ross JL, Denckla MB. Brain
development, gender and IQ in children. A volumetric imaging
study. Brain 1996; 1119:1763–1774.
5. Giedd J, Snell J, Lange N et al. Quantitative magnetic resonance
imaging of human brain development ages 4–18. Cerebral
Cortex 1996; 6:551–560.
6. Sowell ER, Thompson PM, Holmes CJ, Jernigan T, Toga AW.
In vivo evidence for post-adolescent brain maturation in frontal
and striatal regions. Nature Neuroscience 1999; 2:859–861.
7. Sowell ER, Thompson PM, Holmes CJ, Batth R, Jernigan TL,
Toga AW. Localizing age-related changes in brain structure
between childhood and adolescence using statistical parametric
mapping. Neuroimage 1999; 9:587–597.
8. Giedd JN, Blumenthal J, Jeffries NO et al. Cerebral cortical gray
matter changes during childhood and adolescence: a longitudinal
MRI study. Nature Neuroscience 1999; 2:861–863.
9. Paus T, Zijdenbos A, Worsley K et al. Structural maturation of
neural pathways in children and adolescents: in vivo study.
Science 1999; 283:1908–1911.
10. Schapiro M, Luxenberg J, Kay J, Haxby J, Friedland R,
Rapoport SI. Serial quantitative estimates of brain morphmetrics
in Downs syndrome at different ages. Neurology 1989;
11. Reiss A, Eliez S, Schmitt JE, Patwardhan A, Haberecht M.
Brain imaging in neurogenic conditions: realizing the potential
of behavioural neurogenetics research. Mental Retardation and
Developmental Disabilities Research Reviews 2000; 6:186–197.
12. Aylward EH, Reiss AL, Reader MJ, Singer HS, Brown JE,
Denckla MB. Basal ganglia volumes in children with
attention-deficit hyperactivity disorder. Journal of Child
Neurology 1996; 11:112–115.
13. Filipek PA, Semrud-Clikeman M, Steingard RJ, Renshaw PF,
Kennedy DN, Biederman J. Volumetric MRI analysis comparing
attention-deficit hyperactivity disorder with normal controls.
Neurology 1997; 48:589–601.
14. Bullmore ET, Suckling J, Overmeyer S, Rabe-Hesketh S,
Taylor E, Brammer MJ. Global, voxel, and cluster tests, by
theory and permutation, for a difference between two groups of
structural MR images of the brain. IEEE Transactions of
Medical Imaging 1999; 18:32–42.
15. Castellanos FX, Giedd JN, Marsh WL et al. Quantitative brain
magnetic resonance imaging in attention-deficit hyperactivity
disorder. Archives of General Psychiatry 1996; 53:607–616.
16. Castellanos FX, Giedd JN, Berquin PC et al. Quantitative brain
magnetic resonance imaging in girls with attention-deficit/
hyperactivity disorder. Archives of General Psychiatry, 2001;
17. Mataró M, García-Sánchez C, Junqué C, Estévez-González A,
Pujol J. Magnetic resonance imaging measurement of the caudate
nucleus in adolescents with attention-deficit hyperactivity
disorder and its relationship with neuropsychological and
behavioral measures. Archives of Neurology 1997; 54:963–968.
18. Berquin PC, Giedd JN, Jacobsen LK et al. Cerebellum in
attention-deficit hyperactivity disorder: a morphometric MRI
study. Neurology 1998; 50:1087–1093.
19. Mostofsky SH, Reiss AL, Lockhart P, Denckla MB. Evaluation
of cerebellar size in attention-deficit hyperactivity disorder.
Journal of Child Neurology 1998; 13:434–439.
20. Castellanos FX, Swanson J. Biological underpinnings of ADHD.
In: Sandberg S, ed. Hyperactivity and attention disorders of
children. Cambridge University Press, Cambridge UK (in press).
21. Castellanos FX, Giedd J, Walter J et al. Longitudinal MRI in
attention-deficit/hyperactivity disorder (ADHD): effects of prior
stimulant treatment in cerebellum and total brain [Abstract].
Society for Neuroscience Abstracts 2000; 26:1327.
22. Nicolson R, Rapoport J. Childhood onset schizophrenia: rare but
worth studying. Biological Psychiatry 1999; 46:1418–1428.
23. Frazier JA, Giedd JN, Hamburger SD et al. Brain anatomic
magnetic resonance imaging in childhood-onset schizophrenia.
Archives of General Psychiatry 1996; 53:617–624.
J.L. RAPOPORT, F.X. CASTELLANOS, N. GOGATE, K. JANSON, S. KOHLER, P. NELSON279
24. Kumra S, Giedd JN, Vaituzis AC et al. Childhood-onset
psychotic disorders: quantitative magnetic resonance imaging of
structural brain abnormalities. American Journal of Psychiatry
25. Jacobsen LK, Giedd JN, Castellanos FX et al. Progressive
reduction of temporal lobe structures in childhood onset
schizophrenia. American Journal of Psychiatry 1998;
26. Rapoport JL, Giedd J, Kumra S et al. Childhood-onset
schizophrenia: progressive ventricular change during
adolescence. Archives of General Psychiatry 1997;
27. Giedd JN, Jeffries NO, Blumenthal J et al. Childhood onset
schizophrenia. Progressive brain changes during adolescence.
Journal of Biological Psychiatry 1999; 46:892–898.
28. Rapoport JL, Giedd JN, Blumenthal J et al. Progressive cortical
change during adolescence in childhood-onset schizophrenia: a
longitudinal magnetic resonance imaging study. Archives of
General Psychiatry 1999; 56:649–654.
29. Bedwell J, Keller B, Smith A, Hamburger S, Kumra S,
Rapoport JL. Childhood onset schizophrenia: post-psychotic
decline in IQ. American Journal of Psychiatry 1999;
30. Karp BI, Garvey M, Jacobsen LK et al. Abnormal neurological
maturation in early-onset schizophrenia. American Journal of
Psychiatry 2001; 158:118–122.
31. Nicolson R, Lenane M, Brookner F et al. Children and
adolescents with psychotic disorder not otherwise specified: a
two to eight year follow-up. Comprehensive Psychiatry (in
32. Nicolson R, Giedd JN, Blumenthal J et al. Lack of progressive
cortical gray matter with non-schizophrenic psychotic disorders.
Biological Psychiatry (in press).
33. Glantz L, Lewis D. Decreased dendritic spine density on
prefrontal cortical pyramidal neurons in schizophrenia. Archives
of General Psychiatry 2000; 57:65–73.
34. Selemon L, Goldman-Rakic P. The reduced neuropil hypothesis:
a circuit based model of schizophrenia. Biological Psychiatry
35. McGlashan TH, Hoffman RE. Schizophrenia as a disorder of
developmentally reduced synaptic connectivity. Archives of
General Psychiatry 2000; 57:637–648.
36. Woods BT. Is schizophrenia a progressive neurodevelopmental
disorder? Toward a unitary pathogenetic mechanism. American
Journal of Psychiatry 1998; 155:161–170.
37. Keshavan MS, Hogarty GE. Brain maturational processes and
delayed onset in schizophrenia. Development and
Psychopathology 1999; 11:525–543.
38. Allen RW,Young SL. Phencyclidine-induced psychosis.
American Journal of Psychiatry 1978; 135:1081–1084.
39. Friston KJ. Schizophrenia and the disconnection hypothesis.
Acta Psychiatrica Scandinavica 1999; 99:68–79.
40. Benes FM, Davidson J, Bird ED. Quantitative cytoarchitectural
studies of the cerebral cortex of schizophrenics. Archives of
General Psychiatry 1986; 43:31–35.
41. Hubel DH, Wiesel TN, LeVay S. Plasticity of ocular dominance
columns in monkey striate cortex. Philosophical Transactions of
the Royal Society of London, Series B. Biological Sciences 1977;
42. Singer W. Neuronal mechanisms in experience dependent
modification of visual cortex function. Progress in Brain
Research 1979; 51:457–477.
43. Greenough W, Black J. Induction of brain structure by
experience: substrate for cognitive development. In: Gunnar MR,
Nelson CA, eds. Minnesota symposia on child psychology 24:
developmental behavioral neuroscience. Hillandale, NJ:
Lawrence Erlbaum, 1991:155–200.
44. Haracz JL. Neural plasticity in schizophrenia. Schizophrenia
Bulletin 1985; 11:191–240.
45. Rutledge LT. Effects of cortical denervation and stimulation on
axon, dendrites, and synapses. In: Cotman CW, ed. Neural
plasticity. New York: Raven, 1978:273–289.
46. Chen S, Hillman DE. Plasticity of the parallel fiber-Purkinje cell
synapse by spine takeover and new synapse formation in the
adult rat. Brain Research 1982; 240:205–220.
47. Volpe JJ. Overview: normal and abnormal human brain
development. Mental Retardation and Developmental
Disabilities Research Reviews 2000; 6:1–5.
48. Evrard P, Marret S, Gressens P. Environmental and genetic
determinants of neural migration and postmigratory survival.
Acta Paediatrica 1997; 422 (Suppl.):20–26.
49. Riddle DR, McAllister AK, Lo DC, Katz LC. Neurotrophins in
cortical development. Cold Spring Harbor Symposium on
Quantitative Biology 1996; 61:85–93.
50. Johnston MV. Neurotransmitters and vulnerability of the
developing brain. Brain and Development 1995; 17:301–306.
51. Milberger S, Biederman J, Faraone SV, Guite J, Tsuang MT.
Pregnancy, delivery and infancy complications and attention
deficit hyperactivity disorder: issues of gene–environment
interaction. Biological Psychiatry 1997; 41:65–75.
52. Melchitzky DS, Lewis DA. Tyrosine hydroxylase- and dopamine
transporter-immunoreactive axons in the primate cerebellum.
Evidence for a lobular- and laminar-specific dopamine
innervation. Neuropsychopharmacology 2000; 22:466–472.
53. Seroogy KB, Lundgren KH, Tran TM, Guthrie KM, Isackson PJ,
Gall CM. Dopaminergic neurons in rat ventral midbrain express
brain-derived neurotrophic factor and neurotrophin-3 mRNAs.
Journal of Comparative Neurology 1994; 342:321–334.
54. Castellanos FX. Toward a pathophysiology of attention-deficit/
hyperactivity disorder. Clinical Pediatrics 1997; 36:381–393.
55. Castellanos FX, Lau E, Tayebi N et al. Lack of an
association between dopamine-4 receptor polymorphism and
attention-deficit/hyperactivity disorder: genetic and brain
morphometric analyses. Molecular Biology 1998; 3:431–434.
56. Castellanos FX, Giedd JN, Eckburg P et al. Quantitative
morphology of the caudate nucleus in attention deficit
hyperactivity disorder. American Journal of Psychiatry 1994;
57. Hynd GW, Hern KL, Novey ES et al. Attention
deficit-hyperactivity disorder and asymmetry of the caudate
nucleus. Journal of Child Neurology 1993; 8:339–347.
58. Sawada H, Shimohama S. Neuroprotective effects of estradiol in
mesencephalic dopaminergic neurons. Neuroscience and
Biobehavioral Reviews 2000; 24:143–147.
59. Arnold S. Neurodevelopmental abnormalities in schizophrenia:
insights from neuropathology. Development and
Psychopathology 1999; 11:439–456.
60. Harrison P. The neuropathology of schizophrenia: a critical
review of the data and their interpretation. Brain 1999;
61. Huttenlocher PR, Dabholkar AS. Regional differences in
synaptogenesis in human cerebral cortex. Journal of
Comparative Neurology 1997; 387:167–178.
62. Selemon LD, Rajkowska G, Goldman-Rakic PS. Abnormally
high neuronal density in the schizophrenic cortex: a
morphometric analysis of the prefrontal area 9 and occipital area
17. Archives of General Psychiatry 1995; 52:805–818.
63. Selemon LD, Rajkowska G, Goldman-Rakic PS. Elevated neuronal
density in prefrontal area 46 in brains from schizophrenic
patients: application of a three-dimensional stereologic method.
Journal of Comparative Neurology 1998; 392:402–412.
64. Pakkenberg R. Total nerve cell number in neocortex in chronic
schizophrenics and controls estimated using optical dissectors.
Biological Psychiatry 1993; 34:768–772.
NORMAL AND ABNORMAL BRAIN DEVELOPMENT 280
J.L. RAPOPORT, F.X. CASTELLANOS, N. GOGATE, K. JANSON, S. KOHLER, P. NELSON281
65. Glantz LA, Lewis DA. Reduction of synaptophysin
immunoreactivity in the prefrontal cortex or schizophrenic
subjects: regional and diagnostic specificity. Archives of General
Psychiatry 1997; 54:660–669.
66. Mirnics K, Middleton F, Murquez A, Lewis DA, Levitt P.
Molecular characterization of schizophrenia viewed by
microarray analysis of gene expression in prefrontal cortex.
Neuron 2000; 28:53–67.
67. Lohof AM, Delhaye-Bouchaud N, Mariani J. Synapse
elimination in the central nervous system: functional
significance and cellular mechanisms. Reviews in the
Neurosciences 1996; 7:85–101.
68. Thompson WJ. Activity and synapse elimination at the
neuromuscular junction. Cellular and Molecular Neurobiology
69. Lanuza MA, Li M, Jia M et al. Protein kinase C-mediated
changes in synaptic efficacy at the neuromuscular junction in
vitro: the role of postsynaptic acetylcholinereceptors. Journal of
Neuroscience Research 2000; 61:616–625.
70. Balice-Gordon RJ, Lichtman JW. Long-term synapse loss
induced by focal blockade of postsynaptic receptors. Nature
71. Kano M, Hashimato K, Chen C et al. Impaired synapse
elimination during cerebellar development in PKCγ mutant mice.
Cell 1995; 83:1223–1231.
72. Hughes RA, Sendtner M, Thoenen H. Members of several
gene families influence survival of rat motor neurons in vitro
and in vivo. Journal of Neuroscience Research 1993;
73. Thoenen H. Neurotrophins and neuronal plasticity. Science
74. McAllister AK, Katz L, Lo D. Neurotrophins and synaptic
plasticity. Annual Review of Neuroscience 1999; 22:295–318.
75. Cabelli RJ, Shelton DL, Segal DL, Shatz C. Blockage of
endogenous ligand of trI: B inhibits formation of endogenous
ocular dominance columns. Neuron 1997; 19:63–76.
76. Feinberg I. Schizophrenia: caused by a fault in programmed
synaptic elimination during adolescence? Journal of Psychiatric
Research 1982/1983; 17:319–334.
77. Semrud-Clikeman M, Steingard RJ, Filipek P et al. Using
MRI to examine brain-behaviour relationships in males with
attention deficit hyperactivity. Journal of the American
Academy of Child and Adolescent Psychiatry 2000;
78. Mataro M, Garcia-Sanchez C, Junque C, Estevez-Gonzalez A,
Pujol J. Magnetic resonance imaging measurement of the
caudate nucleus in adolescents with attention-deficit
hyperactivity disorder and its relationship with
neuropsychological and behavioural measures. Archives of
Neurobiology 1997: 54:963–968.