Dramatic gains in knowledge have been made in the fields of neuroscience, humandevelopment, and developmental psychopathology during the past quarter of a century. Despitethe advances that have been achieved in each discipline separately, considerably less progress hasoccurred in understanding the relation between neurobiological and behavioral development innormal and atypical populations. Research has increasingly demonstrated that abnormalities thatoccur early in development may result in the emergence of aberrant neural circuitry thateventuates in relatively enduring forms of psychopathology. Knowledge of normalneurobiological development provides a powerful foundation for understanding the contributionsthat neurodevelopmental processes make to the etiology and sequelae of psychopathology acrossthe life course. An integrated perspective wherein an appreciation of the complex neural,psychological, and social-contextual processes that cohere to bring about normal andpathological outcomes is necessary in order to advance understanding of the genesis andepigenesis of mental disorders. Such an approach will require a reduction of the schisms that sooften separate neurobiological and behavioral research.

The basic principles of the development of the central nervous system (CNS) are reviewed,and their implications for both normal and abnormal development of the brain are discussed. Thegoals of this review are (a) to provide a set of concepts to aid in understanding the variety ofcomplex processes that occur during CNS development, (b) to illustrate how these conceptscontribute to our knowledge of the normal anatomy of the adult brain, and (c) to provide a basisfor understanding how modifications of normal developmental processes by traumatic injury, byenvironmental or experiential influences, or by genetic variations may lead to modifications inthe resultant structure and function of the adult CNS.

In this paper I review evidence and viewpoints on developmental plasticity in the cerebralcortex. Although there is some degree of plasticity in the cortex during early postnatal life in thehuman infant, this plasticity is constrained by various factors. Three working hypotheses aboutpostnatal cortical specialization of function are advanced, and some specific predictions aboutthe limits and extent of plasticity are assessed through both empirical evidence from infants andsimulations on simple cortical network models.

Growing epidemiological, genetic, and clinical neurobiological evidence indicates thatabnormalities in brain development play determining roles in the pathobiology of schizophrenia.Neuropathological research has made significant progress in delineating cellular and molecularabnormalities in schizophrenia that have relevance to neurodevelopment. This paper reviews theneurodevelopmental processes of neurogenesis, neuronal migration, differentiation,synaptogenesis, neuron and synaptic pruning, and myelination and the reported neuropathologicalfindings in schizophrenia that may be a consequence of disturbances in these processes. Whilemany neuropathological findings in schizophrenia are controversial or await confirmation,reported abnormalities in neuron density, number and morphology, cytoarchitecture, dendriticarbors and spines, synapse-related proteins, and the well-established absence of gliosis or anyother evidence of neurodegeneration or neural injury all provide support for theneurodevelopmental model of schizophrenia.

Our findings in the Helsinki Influenza Study and the Danish Forty Year Study lead us toconclude that a 2nd-trimester maternal influenza infection may increase risk for adultschizophrenia or adult major affective disorder. More recently we have also reported an increaseof unipolar depression among offspring who were exposed prenatally to a severe earthquake (7.8on the Richter scale) in Tangshan, China. Among the earthquake-exposed males (but not thefemales), we observed a significantly greater depression response for those individuals exposedduring the 2nd trimester of gestation. These findings suggest that maternal influenza infectionand severe maternal stress may operate (in different ways) as teratogens, disrupting thedevelopment of the fetal brain and increasing risk for developing schizophrenia or depression inadulthood.

A number of lines of evidence converge in implicating neurodevelopmental processes in theetiology and epigenesis of schizophrenia. In this study we used a prospective, longitudinal designto examine whether adverse obstetric experiences predict schizophrenia and whether there is adeviant functional–developmental trajectory during the first 7 years of life amongindividuals who manifest schizophrenia as adults. The 9,236 members of the Philadelphia cohortof the National Collaborative Perinatal Project were screened for mental health service utilizationin adulthood, and chart reviews were performed to establish diagnoses according to DSM-IVcriteria. The risk for schizophrenia increased linearly with the number of hypoxia-associatedobstetric complications but was unrelated to maternal infection during pregnancy or fetal growthretardation. Preschizophrenic cases (and their unaffected siblings who were also cohortmembers) manifested cognitive impairment, abnormal involuntary movements and coordinationdeficits, and poor social adjustment during childhood. There was no evidence of intraindividualdecline in any domain, but preschizophrenic cases did show deviance on an increasing number offunctional indicators with age. Together, these findings suggest that both genetic and obstetricfactors participate in creating a neural diathesis to schizophrenia, the phenotypic expressions ofwhich are age dependent, probably reflecting the maturational status of a number ofinterconnected brain systems.

Attentional deficits are well-established characteristics of patients with schizophrenia andtheir at-risk offspring, suggesting a biological connection between attention and schizophrenia.The goal of this study is to clarify the developmental role of attention in the illness. Data hasbeen collected from 87 subjects at high and low risk for schizophrenia who have participated inthe New York High-Risk Project from 1977 to the present. Individuals are considered to be athigh risk if either or both of their parents has schizophrenia. Analyses of attention and globalbehaviors, measured at intervals from about 12 to 26 years of age, indicate (a) attentional deficitscan be reliably detected in high-risk children who will develop future schizophrenia-spectrumdisorders (the prespectrum [PSP] group); (b) these deficits are stable, enduring over time, andappear to reflect a compromised attentional capacity; (c) attention is not affected by the onset ofillness in the PSP group; (d) for all subjects, attention and global behaviors follow independentdevelopmental pathways; and (e) behavioral difficulties, but not attention deficits, appear to behighly sensitive to environmental factors, especially rearing by a mentally ill parent. It isconcluded that in PSP individuals impaired attention probably results from prenataldevelopmental abnormalities (possibly on the cellular level) and is likely to be a marker of abiological vulnerability to schizophrenia. In addition, attentional deficits, as opposed to earlybehavioral difficulties, are concluded to be a useful first step in screening for youngsters in needof early intervention.

Motor dysfunction is associated with schizophrenia, and recent longitudinal studies indicatethat it precedes the onset of clinical symptoms. Of particular interest is the heightened occurrenceof involuntary movements, which are apparent as early as infancy and suggest the presence ofsubcortical brain abnormalities. In this article, we present the results of a study of spontaneousmovements in adolescents with schizotypal personality disorder (SPD). SPD is a syndrome thathas been shown to be genetically linked with schizophrenia and is often observed prior to theearly adult onset of schizophrenia. Systematic coding of videotapes of diagnostic interviewsrevealed that the SPD group showed significantly more involuntary movements of the head,trunk, and upper limbs than did normals and adolescents with other personality disorders. Therewere no diagnostic group differences in the rate of voluntary movements. Salivary cortisol,measured before the interview, was positively correlated with involuntary movements. Takentogether, the findings provide further support for the hypothesized etiologic relation betweenSPD and schizophrenia. Based on a neural diathesis-stress model, potential underlyingmechanisms are discussed.

The central feature of schizophrenia is its onset in adolescence. Although this clinicalobservation is consistent with the view that schizophrenia may be a neurodevelopmentaldisorder, debate has focused on when the proposed brain maturational deviations may begin andwhat might be the nature of such defective development. Conflicting models of this illness (e.g.,the early and late neurodevelopmental models) have been proposed. In this paper, we will firstreview concepts from basic developmental neurobiology pertinent to these issues; we thensummarize aspects of the neurobiology of schizophrenia that have a particular bearing on theadolescent onset of this illness. We propose that the schizophrenic syndrome may result fromearly brain adversity and late maturational processes of brain development interacting withadverse humoral, biochemical, and psychosocial factors during adolescence and early adulthood.The onset of schizophrenia in adolescence may be related to the “plasticity switch”secondary to the peripubertal brain maturational changes, perhaps involving an alteration inglutamate receptor function. This loss of plasticity could result in social and nonsocial cognitivedeficits that are central to the pathophysiology of schizophrenia; the vulnerable person maytherefore utilize prepubertal processing styles that are insufficient to the adaptive and“gistful” abstraction requirements of adult cognition. Schizophrenia onset mightoccur in the context of psychosocial developmental challenges to a delayed social cognitivecapacity among neurodevelopmentally compromised individuals. We review therapeuticimplications as well as testable predictions generated by this model, and discuss researchstrategies that might further our understanding of the brain maturational abnormalities inschizophrenia.

Recent studies have focused on the behavioral and neurobiological sequella of exposure toearly adverse events. We hypothesize that early adverse experiences result in an increasedsensitivity to the effects of stress later in life and render an individual vulnerable to stress-relatedpsychiatric disorders. This vulnerability may be mediated by persistent changes incorticotropin-releasing-factor (CRF)-containing neurons, thehypothalamic–pituitary–adrenal axis, and the sympathetic nervous system. Wetherefore present an overview of the CRF system and its role as a mediator in the development ofthe stress response, major depression, and posttraumatic stress disorder. The literature pertainingto behavioral and neurobiological alterations associated with exposure to early adverse life eventsin rodents, nonhuman primates, and humans is reviewed. We focus on animal models thatprecipitate depressive and anxiety symptoms while producing neuroendocrine alterations thatmimic those seen in adults with those disorders. The literature integrating neurobiological andbehavioral consequences of early life stress is also reviewed, focusing primarily on infants bornto mothers with depression and on infants who were abused or neglected.

Autismis a developmental disorder with variable severity, occurring at all levels ofcognitive ability and having a number of slightly different clinical presentations. It is associatedwith neuropsychological deficits that occur in other conditions also, but its patternmay be specific to autism. Genetic and environmental early insults to brain development areetiological determinants of the disorder. Brain circuitries important for social, communicative,and integrational purposes have been suggested to be dysfunctional in autism. There could be atleast two different pathways to autism, one connected with primary temporofrontal dysfunction(and late prenatal–early postnatal origins) and another linked to primary brain-stemdysfunction (and early prenatal origins). Further study of neurodevelopmental andneuropsychological processes in autism will help elucidate not only the pathological mechanismsinvolved in the specific syndromes but also the underpinnings of normal brain development.

In previous studies, infants of depressed mothers have been found to exhibit reduced leftfrontal brain electrical activity (EEG). The left frontal region has been hypothesized to mediatesocial approach behaviors and positive affective expression. These findings raise importantquestions about the cause and nature of atypical EEG patterns in infants of depressed mothers.The present study begins to address some of these questions by examining whether or notvariations in patterns of frontal brain activity in infants of depressed and nondepressed mothersare related to variations in infant behavior as observed in naturalistic situations. If such relationsexist, are they specific to certain behaviors hypothesized to be mediated by the frontal region(i.e., positive approach behaviors)? Frontal and parietal brain electrical activity was recordedfrom 14- to 15-month old infants of depressed versus nondepressed mothers during a baselinecondition and during conditions designed to elicit interest and positive affect. Infant behaviorwas observed in naturalistic play conditions, with and without mother, on a separate day fromEEG testing. Mothers provided information on infant temperament. Infants of depressed mothersshowed less affection and touching of their mothers. For infants of depressed mothers only,reduced left frontal brain activity was found to be related to lower levels of affection towardmother, but not to infant temperament. Furthermore, increased generalized frontal activation wasfound to be related to higher levels of negative affect, hostility, and tantrums and aggression.Relations between infant brain activity and behavior were not found for parietal EEG activity.These results suggest that infant frontal electrical brain activity is related to variations in infantbehavior, especially those involved in positive affiliative behavior and the expression andregulation of negative affect. The nature and cause of atypical patterns of brain activity andquestion of whether such atypical patterns of frontal brain activity predispose infants to affectivedisorders in later life are discussed.

Recent research on the disorders of attention and activity has indicated inherited variants ofgenes controlling aspects of neurotransmission, abnormalities of structure and function in regionsof frontal lobes and basal ganglia, failures to suppress inappropriate responses, and a cascade offailures in various kinds of cognitive performance and organization of behavior. This reviewintegrates the neurodevelopmental findings with findings from developmental psychopathology.It outlines several developmental tracks by which constitutional factors interact with thepsychological environment. In one set of tracks, altered brain states lead to cognitive alteration.An understimulating environment is evoked by (and may be genetically associated with) aninattentive and cognitively impulsive style during early childhood. In another track, impulsiveand inattentive behavior shows direct continuity through childhood into late adolescence. In yetanother track, impulsiveness evokes (and may be genetically associated with) critical expressedemotion from parents and inefficient coping strategies, which in turn contribute to thedevelopment of antisocial conduct. This formulation emphasizes the need for several types ofresearch: the mapping of biological findings onto different components of disorder, thecombination of genetically informative designs with direct measurement of relevant aspects ofthe environment, and the use of longitudinal studies to examine predictive and mediating factorsseparately for different aspects of outcome.

This paper reviews what is known about developmental dyslexia at three levels of analysis:cognitive, neurological, and genetic. It also considers the difficult problem of establishing causallinks between these levels of analysis, and argues that solving the gene–behavior problemis paradoxically easier than solving the brain–behavior problem.