Abstract

he idea that delusions often reflect an abnormal attitude towards the self has a long history. Consistent with this idea, the most common delusional themes-2013; persecution and grandiosity – seem to reflect individuals' preoccupations about their position in the social universe. Psycho-analysts have interpreted persecutory delusions as the consequence of defensive strategies that serve the function of protecting the individual from negative perceptions of the self, and the well-replicated finding of an abnormal stye of explaining negative events in paranoid patients seems consistent with this idea. However, attempts to test the hypothesis of a paranoid defence by measuring self-esteem in deluded patients has led to inconsistent and even contradictory results.T

These findings may reflect false assumptions about the unity and stability of selfrepresentations. The self consists of a cluster of fluidly related constructs, for example the self as it actually is, the ideal self and the self as it ought to be. Furthermore, evaluations of the self can change dramatically from day to day, and some people consistently make less stable self-evaluations than others. These observations point to the need for a dynamic account of the relationship between self-representations, other cognitive structures and environmental influences.

In this chapter a model is presented of the dynamic interactions between causal explanations (attributions) and self-representations. It will be shown that this kind of model can accommodate the apparently inconsistent findings obtained when self-esteem has been measured in paranoid patients, and leads to novel predictions that can be experimentally tested.

In the last two decades, research in developmental psychopathology and neuroscience has made considerable progress. Although we have learned a great deal about the time course of neural development and of psychopathology, some basic questions concerning the interaction between neurobiological and behavioral development still remain unanswered (Cicchetti, 1993). For example, we still do not know whether the emergence of a specific psychopathology requires activation of maladaptive processes during a time-critical window of vulnerability or whether simple alterations in the order or timing of neurobiological events during normal development are sufficient to induce psychopathology.

The differentiation of species, organ systems, and neural systems all involve sequencing the activation and suppression of a suite of genes and their products (Eisenberg, 1995). Changes in the endogenous timing mechanisms that regulate this programmed gene expression could affect the physical organization of the developing nervous system so that the organism's sensitivity to environmental stimuli is permanently enhanced or reduced. Altered sensitivity to the environment could then lead the organism to a developmental pathway that is atypical for the species. Alternatively, a change in timing may not be reflected in the physical organization of the developing nervous system, but advancing or delaying the deployment of a neural system could make a different system vulnerable to modification. In this view, the change in timing affects the developmental context of systems that are already in place, ultimately altering their function.

Even if scientists uncover a single, powerful determinant of behavior, whether it be a gene linked to a disorder, a toxic biological event that goes awry during fetal development, or an aspect of parental caregiving that interferes with the development of healthy relationships, it is now clear that not all individuals who share such risks will emerge with the same degree of mental health or disorder nor stay on the same, predicted developmental trajectory. In particular, little attention has been paid to the role of neurodevelopmental processes in enhancing or decreasing the opportunity for healthy development in individuals with environmental or genetic risks and vice versa. If we accept that neither environmental qualities nor genes or other biological mechanisms directly determine behavior or developmental course, then we are challenged to consider the potential explanatory power of the complexities of their possible interplay.

It is the purpose of this chapter to accept the challenge of such an integrative perspective for the understanding of risk for the development of psychopathology or other adverse outcomes in children with depressed mothers. Specifically, this chapter will examine current knowledge on outcomes and mechanisms of risk to children with depressed mothers from this perspective and propose future directions for such integrative research. An effort will be made to explore the influences on outcomes not only as a static event but also on alternative developmental pathways.

It needs to be stated at the outset that ideas about complex determinants of behavior are not new.

Research has identified disturbances in the nervous system as underlying schizophrenia since Kraepelin (1919) first proposed the term dementia praecox. Ample evidence now supports this earlier speculation. More recent evidence has established a neurodevelopmental basis for schizophrenia and other major mental disorders (Akbarian et al., 1993; Mednick, Cannon, Barr, & Lyon, 1991; Mednick & Hollister, 1995; Machón, Mednick, & Huttunen, 1997). In this chapter we will examine some of this evidence establishing a neurodevelopmental basis for major mental disorders, including schizophrenia, as well as propose two related mechanisms involving maternal stress and immune response processes that may mediate the maternal infection and subsequent, adult mental disorder outcome.

FETAL NEURAL DEVELOPMENT

The development of fetal neural structures is a delicate process. Neurons creating the human neocortex proliferate by the fifth month of gestation. Rapid migration and differentiation occur in the central nervous system during the second trimester (CNS; Nowakowski, 1991). The cortical subplate, essential for the formation of the cerebral cortex, also develops during the second trimester. Disruption during a critical period of proliferation may dramatically disorganize neural development leading to observable physical anomalies. Jones and Akbarian (1995) conclude, however, that the existing data suggest as “unlikely that the proliferative phase of forebrain ontogenesis is compromised in any major way in schizophrenia” (p. 29).

Kovelman and Scheibel (1984) suggest that disrupted neural migration may result in the excess of ectopic neurons found in the hippocampi of schizophrenia patients.

It is often said that the only constant in life is change – in fact, responding to change might be considered the most fundamental aspect of life. As early as the 1930s, Selye used the term stress to refer to any condition (perceived or real) that threatens homeostasis (Selye, 1952, 1978), or in other words, signals the need for adaptive change. Responses to stress signals are believed to have developed as an alarm system for animals caught in a potentially dangerous situation, and as an activation system for animals faced with a competitive situation. These physiological responses to stress are coordinated by a rapidly responding sympathetic nervous system (SNS), and by a slower but longer-acting second stage involving the hypothalamus, the pituitary gland, and the outer cortex of the adrenal glands (collectively termed the hypothalamic–pituitary–adrenal axis, or HPA; Lopez, Akil, & Watson, 1999). Restorative biological processes (e.g., digestion, immune function, tissue building) are temporarily suppressed so that energy resources can be mobilized for activity (Sapolsky, Romero, & Munck, 2000). In the extreme, the stress response becomes a stereotypical set of behaviors and physiological reactions initially described by Cannon (1932) as the “fight or flight” response. Generally the physiological reaction returns to baseline levels soon after the offset of stress cues. If a stressful situation is prolonged or inescapable, the initial high state of physiological activation cannot be indefinitely sustained – eventually, exhaustion sets in.

Cicchetti and Lynch (1995) asserted that child maltreatment may represent the greatest failure of the caregiving environment to provide many of the expectable experiences that are necessary to facilitate normal developmental processes. Maltreating parents also may be viewed as an aberration of the supportive, nurturant, sensitive, and protective adults that are expected by children in the evolutionary context of species-typical development (Belsky, 1984; Cicchetti & Lynch, 1995; Howes, Cicchetti, Toth, & Rogosch, 2000; Rogosch, Cicchetti, Shields, & Toth, 1995).

In contrast to what is anticipated in response to an average expectable environment, the ecological, social, biological, and psychological conditions that are associated with maltreatment set in motion a probabilistic path of epigenesis for maltreated children characterized by an increased likelihood of failure and disruption in the successful resolution of major stage-salient tasks of development, resulting in grave implications for functioning across the lifespan (Cicchetti, 1989; Cicchetti & Lynch, 1993; Egeland, 1997; Malinosky-Rummell & Hansen, 1993). These repeated developmental disruptions create a profile of relatively enduring vulnerability factors that increase the probability of the emergence of maladaptation and psychopathology as negative transactions between the child and the environment continue (Cicchetti & Lynch, 1993; Cicchetti & Rizley, 1981).

The notion that an average expectable environment is required for species-typical development suggests that competent outcomes in maltreated children should be highly improbable due to wide-ranging disturbances in the maltreatment ecology (Cicchetti & Lynch, 1993).

The major mental illnesses, most notably schizophrenia and other psychotic disorders, typically have their onset in young adulthood, and often lead to a lifetime of chronic disability. The possibility of preventing these illnesses has received increasing attention in the past few years (McGorry & Edwards, 1998; Wyatt, Apud, & Potkin 1996). This trend has been fueled by evidence that the longer the duration of the initial untreated episodes of psychosis, the worse the long-term prognosis (Wyatt, 1995). Also, the availability of atypical antipsychotic medications that have fewer immediate side effects has contributed to interest in psychosis prevention.

The first step in the prevention process is the identification of vulnerable individuals. It is well established that the clinical onset of schizophrenia is preceded by behavioral dysfunction. In some cases, preschizophrenic individuals manifest consistent dysfunction that is apparent within the first few years of life, extends throughout childhood, and becomes more pronounced in adolescence (Larsen, McGlashan, Johannessen, & Vibe-Hansen, 1996; Walker, Baum, & Diforio, 1998). Others show relatively normal childhood development, then a precipitous decline that begins in adolescence. Based on the best available evidence, about 70 percent of adult-onset patients manifested behavioral dysfunction in adolescence (Larsen et al., 1996; Neumann, Grimes, Walker, & Baum, 1995; Yung & McGorry, 1996). Thus, many view adolescence/early adulthood as the most plausible developmental stage for initiating prevention.

In parallel with the increasing emphasis on prevention, there has been a resurgence of interest in the neurodevelopmental changes that accompany pubertal maturation.

In this chapter, we will discuss the long-term consequences of being born very early or very small. This issue is relevant to those interested in developmental psychopathology as this area of research emphasizes the understanding of the mechanisms underlying the development of pathology in high-risk individuals, as well as the pathways to competent adaptation despite exposure to conditions of adversity.

Studies in the United States and others have shown that rates of low birth weight increase with decreasing socioeconomic status (e.g., Paneth, 1995). At the individual level, women belonging to lower socioeconomic strata are at significantly higher risk of preterm delivery, even after controlling for other known risk factors such as weight, weight gain, alcohol and tobacco consumption, ethnicity, parity, and source of prenatal care. At the geographic level, for any defined area, the more socioeconomically disadvantaged the population, the higher the incidence of preterm delivery and low birth weight. Socio-demographic factors, such as maternal education and family income, may be related to a less favorable neuropsychological outcome in low birth weight and preterm children, as well as in normal birth weight children (Sameroff, Seifer, Barocas, Zax, & Greenspan, 1987). In addition, a considerable proportion of preterm children are treated by intensive care and sustain some brain injuries and subsequent neurological and neuropsychological impairment. However, even in the absence of major neurological deficits, preterm and low birth weight children seem to be at risk of developing a variety of cognitive and behavioral problems in childhood.

This volume represents a burgeoning perspective on the origins of psychopathology, one that focuses on the development of the human central nervous system (CNS). The contemporary neurodevelopmental perspective assumes that mental disorders result from etiologic factors that alter the normal course of brain development. Defined here in its broadest sense, neurodevelopment is a process that begins at conception and extends throughout the lifespan. We now know that it is a complex process and that its course can be altered by a host of factors, ranging from inherited genetic liabilities to psychosocial stressors. This knowledge has challenged clinical researchers to devise novel methodologies aimed at identifying links in this chain of events that can lead to psychopathology.

Neurodevelopmental perspectives on psychopathology have become increasingly dominant as a consequence of major advances in both basic animal research and clinical investigations of human populations. Basic research efforts have succeeded in elucidating amazing facets of brain development that extend from the molecular to the behavioral levels of analysis. For example, using animal models, basic scientists have documented the long-term effects of prenatal and postnatal events on brain structure and function and have shown how these effects vary as a function of hereditary factors. They also have demonstrated that there are significant behavioral sequela of aberrant neurodevelopment. At the same time, clinical research has yielded extensive evidence that prenatal and early childhood factors are associated with subsequent risk for psychopathology.

Over the past decade our research efforts have focused on the identification of prenatal risk factors for schizophrenia. In this chapter, we summarize our study designs and methods, describe our findings, discuss the implications of these findings for the field, and present our plans for future investigations. In order to provide a conceptual framework for these findings, we shall first discuss the neurodevelopmental hypothesis of schizophrenia, and briefly review previous studies by other investigators in this research domain.

NEURODEVELOPMENTAL HYPOTHESIS OF SCHIZOPHRENIA

The neurodevelopmental model of schizophrenia posits that adverse in utero events influence critical processes in the genesis of brain structures, which predispose to the emergence of schizophrenia in adulthood (Brown et al., 1999; Susser, 1999). The evidence that led to this hypothesis derives from many diverse areas of investigation; however, there appear to be three pivotal supportive findings. First, patients destined to develop schizophrenia have a tendency for diminished neurocognitive (David et al., 1997; Jones et al., 1994), neuromotor (Walker et al., 1994), and behavioral (Done et al., 1994) function. Second, patients with schizophrenia, as compared with healthy controls, have an increased frequency and severity of minor physical anomalies, particularly of the craniofacial area, which are suggestive of an in utero developmental disruption (Green et al., 1989; Waddington, 1993). Third, neuroimaging studies indicate that several of the brain abnormalities in schizophrenia, such as ventriculomegaly and diminished hippocampal volume, occur among patients in their first episode of psychosis (Bogerts et al., 1990; DeGreef et al., 1992; Nopoulos et al., 1995).

The past three decades have witnessed exponential growth in our knowledge of the interactions among the central and peripheral nervous systems and the immune system (Ader, 1981, 2000; Ader, Felten, & Cohen, 1991). In particular, the signals and routes via which psychological and physical stressors lead to endocrine and immune responses have been studied extensively. A detailed picture of an intriguing puzzle has now begun to emerge. Theorists, applied researchers, and professionals are extrapolating these basic findings in order to consider how individual differences in psychological traits and states might be associated with immunity, illness susceptibility, and negative health outcomes (see Cohen & Herbert, 1996; Herbert & Cohen, 1993b; Kemeny & Gruenewald, 1999; Kemeny & Laudenslager, 1999; Kiecolt-Glaser & Glaser, 1995). Quite surprisingly, how such processes affect children's immunity has received scant empirical attention (e.g., Adamson-Macedo, 2000; Boyce et al., 1995; Coe, 1996, 1999). Although the field should be concerned with the ultimate impact of these phenomena for children's health, and additional research with that particular focus seems warranted, the focus of this chapter is on the implications of another leading edge of psychoneuroimmunologic research (e.g., Maier & Watkins, 1998a, b; Maier, Watkins, & Fleshner, 1994). Specifically, accumulated findings have sparked a scientific revolution regarding the direction of effects among the brain, behavior, and immunity (e.g., Blalock, 1994a; Dantzer, 2001).

Not much is known about neurodevelopmental factors involved in the pathophysiology of child and adolescent depression. Neuroendocrine and sleep EEG research paradigms have been used in the majority of published studies examining the neurobiological correlates of early-onset depression. Although these procedures have their merits, the “window to the brain” afforded by these methods is extremely limited. Emerging neuroimaging technologies will provide a unique opportunity to investigate the brain mechanisms underlying child and adolescent depression. To date, however, the application of these techniques in the study of early-onset depression is in its infancy.

The existence of major depressive disorder (MDD) in children and adolescents was controversial until relatively recently, and the diagnosis of MDD was not included in any child psychiatric text prior to the late 1970s (Puig-Antich & Gittleman, 1982). Research over the past two decades, however, has clearly demonstrated that children are capable of experiencing episodes of depression which meet standard DSM-IV (Diagnostic and Statistical Manual of Mental Disorders, 4th ed.) criteria for MDD (Birmaher et al., 1996c; Ryan et al., 1987). In addition, MDD in children and adolescents is common, recurrent, and associated with significant morbidity and mortality (Birmaher et al., 1996c). Epidemiological studies estimate that the prevalence of depression is 2 percent in children (Kashani et al., 1983) and 5–8 percent in adolescents (Lewinsohn, Clarke, Seeley & Rohde, 1994).