Social Context in Developmental Psychopathology: Recommendations for Future Research from the MacArthur Network on Psychopathology and Development

Accumulating evidence suggests that social contexts in early life have important and complex effects on childhood psychopathology. Spurred by the lack of an explicit operational definition that could guide the study of such effects, we define a social context operationally as “a set of interpersonal conditions, relevant to a particular behavior or disorder and external to, but shaped and interpreted by, the individual child.” Building on this definition, we offer a series of recommendations for future research, based on five theoretically derived propositions: (a) Contexts are nested and multidimensional; (b) contexts broaden, differentiate, and deepen with age, becoming more specific in their effects; (c) contexts and children are mutually determining; (d) a context’s meaning to the child determines its effects on the child and arises from the context’s ability to provide for fundamental needs; and (e) contexts should be selected for assessment in light of specific questions or outcomes. As reflected in an increasingly rich legacy of literature on child development and psychopathology, social contexts appear to influence emerging mental disorders through dynamic, bidirectional interactions with individual children. Future research will benefit from examining not only statistical interactions between child-and context-specific factors, but also the actual transactions between children and contexts and the transduction of contextual influences into pathways of biological mediation. Because adverse contexts exert powerful effects on the mental health of children, it is important for the field to generate new, more theoretically grounded research addressing the contextual determinants of psychological well-being and disorder.

Since many of the most disabling psychiatric provide conceptual and methodologic tools for integrating the psychosocial and biological disorders begin in childhood, an understanding of psychopathology in children and ado-influences on early development are also critical to constructing models of psychiatric dis-lescents is an essential requirement for the creation of new and specific interventions for ease onset.In an effort to advance a broader and more general understanding of how envi-neuropsychiatric syndromes.Approaches that ronments exert effects on mental disorder and how "context" might best be conceptualized been.Nearly 13 million U.S. children live in million) have mental disorders meeting DSM criteria (U.S.Department of Health and Hu-poverty, with young people less than 18 years of age constituting the single most impover-man Services, 1990b).Affective, conduct, and anxiety disorders, which together comprise ished demographic group in the nation (Bane & Ellwood, 1989; U.S. Department of the largest proportion of childhood psychiatric disturbances, often produce adaptive difficul-Health and Human Services, 1990a).African-American and Hispanic children, moreover, ties in the teenage years and, particularly in girls, lead to chronic mental health disorders are two to three times more likely to be impoverished than White children (U.S. House in adult life (Institute of Medicine, 1989; U.S. Department of Health and Human Services, of Representatives Select Committee on Children, 1989).Over one third of children expe-1990b).Further, data from the 1988 National Health Interview Survey reveal that 4% of rience a parental divorce (Solomon & Wallerstein, 1994), and approximately 25% live in U.S. children have developmental delays, 6.5% learning disabilities, and 13.4% emo-single parent homes (Richmond, 1990).More than one million cases of child abuse or neg-tional or behavioral disorders, leading Zill and Schoenborn (1990) to conclude that "psycho-lect are confirmed per year, and between 3 and 10 million children are exposed annually logical disorders rank among the most prevalent health conditions of modern childhood."to domestic violence (Children's Defense Fund, 1995).The social environments in The declining economic, material, and social environments in which contemporary which many children now live offer minimal support for healthy emotional development children are raised have almost certainly played an etiologic role in the escalation of and are often openly and multiply aversive.
As if these longstanding threats to the childhood mental disorders (Jensen, Richters, Ussery, Bloedau, & Davis, 1991).As a result, well-being of children were not sufficiently disquieting, more recent social changes have investigators in the field of developmental psychopathology have focused attention on produced issues of equal or even greater concern.An estimated 1.2 million families are on the ways in which social environmental forces interact with children's biology to shape ac-waiting lists for public housing, and one in four persons reported as homeless is a child tively the emergence of psychopathology in early development (Bornstein, 1995; Harris, younger than 18 years of age (Children's Defense Fund, 1995).The annual number of run-1995;Hiday, 1995).While there is new recognition of biological-genetic factors in the away youths is approximately one million (U.S. House of Representatives Select Com-etiology of mental disorder, there is also evidence that constitutional elements seldom act mittee on Children, 1989), and nearly 2.1 million children between the ages of 5 and 13 in isolation, and that interactions between biology and psychosocial factors often have have no adult to care for them after school (U.S. House of Representatives Select Com-greater explanatory power in the prediction of psychopathology (Sameroff, 1983).Jensen et mittee on Children, 1989).A rise in parental substance abuse has affected the biological or al. (1993) thus summarized research needs emanating from the 1990 National Plan for emotional development of as many as one fifth to one third of inner city newborns Research on Child and Adolescent Mental Disorders by highlighting the importance of (Frank et al., 1988), and the escalation of violence in urban neighborhoods has become an relationships and social environments as determinants of child and adolescent psycho-urgent public health concern (Koop & Lundberg, 1992).
pathology.Perhaps related to these troubling forces in the environments of the nation's children, Purpose childhood mental disorders have become increasingly prevalent and compelling prob-The publication of this Special Issue of Development and Psychopathology provides a lems.Recent studies suggest that between 17 and 22% of children (approximately 11-14 timely opportunity to review, synthesize, and elaborate upon several dominant themes psychopathology, may prompt some readers to view our work as preaching to the con-emerging over the recent past in research on context and psychopathology.The past two verted.While we recognize that some will find some of the article's content to be old decades, in particular, have witnessed important shifts in the ways researchers conceptual-news, we nonetheless believe that there is considerable value in an occasional stock-tak-ize, assess, and study the settings in which children grow and psychopathology develops.ing, if chiefly for the benefit of those readers whose familiarity with the field is either lim-Owing to the empirical work of such scholars as Belsky (1988), Cicchetti (1993), Garmezy ited or uneven.Indeed, only cursory examination of research published in many current (Garmezy, Masten & Tellegen, 1984), Patterson (1982), Rutter (1990), Sameroff (1983), and credible scholarly journals reveals that some of the principles assumed to be common and Sroufe (1989), among others, and drawing on the theoretical perspectives of those knowledge are, in fact, not universally adhered to by the majority of practicing investi-such as Baltes (1997), Bronfenbrenner (1979), Lewin (1951), andMinuchin (1974), the study gators.Thus, one of our hopes is that this publication will help turn common knowledge of psychopathology and development has been transformed from one focused on the into common practice.As to the apology, the brevity of this article, especially in relation to "afflicted" individual, into a field that examines transactions between developing individ-the size of the field, necessitates that our discussion of specific research be illustrative, not uals and the complex, multilayered environments in which they live.
exhaustive.The particular studies chosen to illustrate the general points we raise were se-Our purpose in this paper is not to review the immense literature on contextual influ-lected because they are good examples of the principles we advocate, not because they are ences in psychopathology, a literature that has burgeoned over the last 20 years.Indeed, such the only examples.We apologize, in advance, to the many researchers whose exemplary a review would be beyond the reasonable scope of a single volume, much less a single work our discussion appears to ignore.article.Rather, our intent here is to integrate and draw readers' attention to a collection of Themes and Definition central assertions about the role of context in the development of psychopathology.We While interest in and appreciation for the influences of early psychosocial environments have tried to approach that objective in a manner that acknowledges the ideas undergirding have grown in recent years, a significant proportion of research on childhood psychopath-this emerging focus (e.g., Cicchetti & Richters, 1997;Goldsmith, Gottesman, & Lemery, ology has proceeded without a unifying framework for conceptualizing and operationalizing 1997; Kagan, 1997;Rutter et al., 1997;Sroufe, 1997), that provides ready reference the relations among environmental factors, child development, and psychopathology.More for researchers unfamiliar with these intellectual developments, and that marks a starting specifically, the absence of a conceptual and operational definition of the term context can point for a new generation of research on contexts and psychopathology.Our hope is that lead to misunderstanding when less developmentally guided investigators, in an effort to this integrative summary will both codify a set of guiding propositions and stimulate fur-be comprehensive, assess as many known contextual risk factors for child psychopathol-ther empirical and theoretical work grounded within them.ogy as possible.In this sense, context is often used as a form of jargon for anything environ-Both an apologia and an apology are in order before turning to the business at hand.mental, as if invoking the term suggests compliance with current scientific and conceptual With respect to the former, our attempt to summarize familiar, yet fundamental, tenets canons.The inexact use of such terms and the accumulation of findings without a guiding that have emerged over a span of years, and that undergird contemporary study of child conceptual framework can become, however, a cacophony of uninterpretable observations.behavior or disorder and external to, but shaped and interpreted by, the individual We therefore offer an approach to conceptualizing and operationalizing social context and child.So defined, social contextual factors help us to understand for whom, or under present a series of recommendations for future research, in an effort to lend increased coher-what conditions, a given outcome will hold.
Contexts inherently imply contingencies, in ence to the flourishing literature on psychopathology in development.
that they describe the circumstances or settings in which individual predispositions may Implicit within our proposals are several key themes that have guided our thinking in emerge and become visible.Additionally, this definition of context emphasizes the trans-this work.First, we believe that, while an exhaustive consideration of environmental influ-actional nature of individual-environment interaction; put most simply, contexts affect ences is beyond the scope of any study, a set of general principles can be articulated that individuals and individuals affect contexts (Sameroff, 1983).Further, our definition of can guide investigators' selection of critical social-psychological factors and contextual social context implies a universal placement of persons, objects, and events into a frame constructs within a given area of research.These principles, presented as five proposi-of reference from which interpretation and meaning are derived (Cole, 1995).With these tions, are discussed in detail below.Second, while we emphasize the importance of under-defining conditions specified, we offer five strategic propositions, together comprising standing children's social contexts in our proposal, we do not in any way suggest that the our recommendations for the treatment of social context in future research within develop-effects of social-psychological factors on children's development are fully separable or in-mental psychopathology.dependent from a child's unfolding biologic substrate.Such simplistic assumptions are contradicted by the burgeoning literature on Proposition 1: Contexts are nested social environments and brain development and multidimensional (Cicchetti & Tucker, 1994;Eisenberg, 1995;Nelson & Bloom, 1997), and by recent evi-Although the literature on psychopathology in childhood has frequently investigated social dence for interactions between social environments and genetically regulated endogenous environmental influences on the development of disorder, many studies have been relatively factors (e.g., Boyce et al., 1995).
Finally, we believe that antecedent social narrow in scope.While there are important exceptions to this empirical shortcoming (e.g., environmental experiences significantly influence and are incorporated into later develop- Belsky & Rovine, 1988;Cairns & Cairns, 1994; Conger, Ge, Elder, Lorenz, & Simons, mental structures and functions (Erikson, 1963).Heritable and early experiential influ-1994; Patterson, 1982), and while many writers acknowledge the need for more complex ences, interpreted by the child and transcribed into individual biology in the form of memory studies of the interplay among settings, even a cursory review of the research published in traces and neural processes, are brought forward in time into dynamic interactions with many leading clinical psychology and psychiatry journals indicates that context often con-new experience, as the child interprets, acts upon, and is affected by these new experi-sists of a single setting, the child's family, and within that setting, only one dimension is typ-ences.Such interactions result in the elaboration of increasingly distinct and differentiated ically examined.Interestingly, this description is less characteristic of research on normative developmental outcomes.
For the purposes of the present paper, we development, which has been faster than research on psychopathology to embrace the offer the following operational definition of social context for future studies of child psy-ecological approach articulated by Bronfenbrenner (1986b), Sameroff (1983), and others.chopathology: A social context is a set of interpersonal conditions, relevant to a particular Jessor (1993) has called attention to this "tra-ditional preoccuption with intrafamilial inter-externalizing problems might examine the convergent and divergent impacts of disci-actions" in the study of psychopathology and suggests the need for new and broader scien-plinary laxity (Patterson, 1986), parent criminality (Farrington & West, 1991), or harsh tific attention to extrafamilial transactions.Further, even when multiple settings are con-parenting (Conger et al., 1994), and peer/ neighborhood influences.While investiga-sidered, examination of contextual effects in many studies of childhood psychopathology tions of single aspects of the environment may be informative, they are likely to be only may resemble a simple "laundry list" of environmental risk factors for various disorders, partially so.As Bronfenbrenner (1979) has pointed out, any given context has multiple rather than an integrative understanding of the processes through which complex environ-dimensions, and these dimensions are likely to interact with one another in etiologically ments and developing children interact over time.Thus, studies that simply rely upon mul-important ways.
Given measurement of any two hypothe-tiple regression approaches to examining the presumed linear, additive effects of multiple sized contextual factors, three general models of multidimensional influence may be identi-environmental factors on child outcomes may not sufficiently advance our understanding of fied: additive models, moderated models, and mediated models (Baron & Kenny, 1986).how unique and specific combinations of social-psychological factors shape the outcomes Consider, as a specific example, the effects of parental emotional rejection and lax disci-of individual children.Alternative research approaches would carefully assesses the con-pline, both of which have been independently documented as risk factors for psychological ditions under which these relations do and do not apply, while simultaneously incorporating disorder in children (Becker, 1962;Blumenthal & Kupfer, 1988; Miller, Cowan, Cowan, the child's, family's, and/or community's attribution of meaning to these social-psycho- Hetherington, & Clingempeel, 1993;Patterson & Stouthamer-Loeber, 1984; Petersen et logical factors.
In our view, several specific changes in al., 1993).On the one hand, additive models would assume that both parental rejection and perspective would advance research on the role of social context in psychopathological lax discipline increase childhood psychopathology, with each factor elevating the risk for development.First, investigators could take into account more systematically the multidi-disorder to the same degree in isolation as they do when present in combination.On the mensionality of contextual influence.For example, within any given environment in other hand, one might find that the two factors function interactively, so that the effect of lax which children develop, an investigator may identify one or more particular environmental discipline on psychopathology is diminished under conditions of parental acceptance but dimensions as a focus for study in relation to one or more outcome variables.Once these amplified under conditions of parental rejection (a moderated model).Yet a third possi-are identified, the investigator can then examine the relations among two or more social bility is that the effect of one variable is mediated entirely through a process involving the environmental dimensions and the conditions under which the purported relations among other.For example, parental rejection could elevate the risk for psychopathology, because contextual factors apply to specific outcomes.Thus, a study of familial influences on the de-rejection leads to lax discipline, which in turn promotes psychological dysfunction.The spe-velopment of childhood anxiety might examine the contingent and interactive roles of cific examples are less important than the general point: Contexts have multiple dimen-marital conflict (e.g., Cummings, Ballard, El-Sheikh, & Lake, 1991), parental intrusive-sions that add to, moderate, and mediate one another in influencing children's behavioral ness (e.g., Stark, Humphrey, Cook, & Lewis, 1990), or maternal depression (e.g., Rutter, and emotional development. Second, research may benefit from investi-1990).Similarly, a study of the role of socialpsychological factors in the development of gators differentiating more clearly between structural versus functional aspects of con-a rope or cord, woven together by different strands of interacting contextual influence texts.Structural aspects refer to the organization and composition of the hierarchical occurring over time.
As an example of such contextual interac-elements that define a context, what Bronfenbrenner (1979) has called the setting's "social tions, consider the relations among familial factors, peer factors, and disorders of conduct.address," whereas functional aspects refer to transactional processes that take place be-In one hypothetical model, we might predict that both lax discipline in the home and asso-tween those elements.Conceptual problems typically arise when a researcher has a theory ciation with deviant peers contribute independently to children's behavior problems.Chil-about the impact of functional variables on psychopathology (e.g., the impact of parental dren exposed to both risk factors would therefore be more likely to develop behavior monitoring on childhood conduct disorder), but is in actuality studying structural, not problems than children exposed to only one.
An alternative account might be developed, functional, variables.Thus, marital status (e.g., whether a child's parents are married, however, in which externalizing problems are associated with lax parental discipline only divorced, or remarried) may be used as a proxy for parenting practices (e.g., how strict when the child has contact with antisocial peers.In either case, a research design that the parent is with the child), or a mother's employment status may be used as a substitute focused on only a single setting might reveal important information about its contribution for a measure of mother-child interaction.A reasonable hypothesis is that the impact of the to conduct problems, but would account for only part of the story.structural characteristics of a setting on the psychological development of the child is me-Studies that consider only one environmental setting (e.g., home, school, or neighbor-diated through one or more functional characteristics of the setting.If marital status does hood) are limited, then, in their capacity to illuminate processes occurring over time, affect child psychopathology, for example, it likely does so through its impact on aspects through which multiple settings may be linked to psychopathology.This point is illus-of family functioning (the parent-child relationship, the marital relationship, etc.) (e.g., trated in a series of studies by Patterson and his colleagues on the joint influence of par -Fauber, Forehand, McCombs, & Wierson, 1990) or the child's attributions about the ents and peers on adolescent misbehavior (e.g., Dishion, Patterson, Stoolmiller, & Skin-meaning of particular marital status categories in comparison with the larger peer milieu. ner, 1991).According to their findings, ineffective discipline in the home leads children Third, as much as possible, research on contextual effects may be enhanced by an ex-to affiliate with antisocial peers, which increases the risk of engaging in misbehavior.amination of interactions among contexts and the ways in which multiple settings influence In this instance, the link between experiences in one context (the family) and psychopathol-each other in affecting child development.In Bronfenbrenners (1979) model of the ecology ogy is mediated through experiences in a second setting (the peer group). of human development, studies of the interrelations among contexts and their joint impacts Finally, studies will be enriched by recognizing the embeddedness of contexts.Proxi-on development are referred to as studies of the "meso-system."Addressing the role of mal settings (e.g., the family) are nested within broader settings (e.g., communities), cultural-historical contexts in child development, Cole (1992) invoked two images of which moderate the influences of proximal processes.The proximal social contexts con-such contextual interrelations.In one, successively broader social contexts form concentric sidered in the preceding example, the family and peer group, are embedded in a broader circles around an individual child, each circle serving as a larger context for the more proxi-context that contains them, shapes them, links them to each other, and moderates their inde-mal ones.In a second analogy, the dynamic exchanges among contexts are envisioned as pendent and joint influences on a child's de-velopment.This broader context is composed • the relative merits of choices between structural and functional measures within a given of the neighborhood, community, culture, and historical epoch in which the child lives.Re-social context, markably little is known about how these • the possible additive, mediating, and moderbroader contexts affect the development or ating relations among social-psychological expression of psychopathology.
factors at different hierarchical levels, and One welcome trend in the exploration of the larger contexts of children's lives and • the embeddedness of each context in a their implications for psychological dysfuncbroader social milieu and the influence of tion is reflected in the growth of research on distal contexts on more proximal settings neighborhood and community factors in child and processes.and adolescent development (e.g., Brooks-Gunn, Duncan, & Aber, 1995; Sampson & While no study can or should address every Laub, 1994).Rooted in sociological studies of plausible context and contextual dimension of urban neighborhoods during the 1930s and influence, it will be important in future work 1940s (e.g., Faris & Dunham, 1939), this reto consider the complexity and interrelatedsearch is examining how structural characterness of multiple, hierarchically ordered conistics of communities (e.g., unemployment, textual systems.Among the many issues in racial mix, and crime rates) affect children's need of further research are (a) the ways in internalizing and externalizing problems which cultural variation in concepts of normal through effects on the behavior of families child development affect the definition and and peer groups.Some community effects are identification of various childhood disorders direct; for example, repeated exposure to vio-(e.g., Weisz et al., 1993); (b) the role of lence is likely to elevate children's levels of broader economic factors (such as unemploydepression, anxiety, and fear (Fitzpatrick, ment) in the development of childhood dis-1993).Other community effects are mediated turbance (e.g., Brody et al., 1994;Conger et through influences on proximal environments;al., 1994;McLoyd, 1990a; Sampson & Laub, thus, crime in a community may heighten par-1994; Steinberg, Catalano, & Dooley, 1981); ents' protectiveness to the point of restricting and (c) historical factors that influence the the development of autonomy and indepenprevalence and expression of childhood psydence (Elder & Ardelt, 1992;McCarthy, chopathology (e.g., Lewinsohn, Rohde, Seel-Lord, Eccles, Kalil, & Furstenberg, 1992). ey, & Fischer, 1993).Finally, some community effects operate by moderating processes that occur within more immediate settings.For example, the relation Proposition 2: Contexts broaden, between certain parenting practices, such as differentiate, and deepen with age, parental authoritativeness, and children's becoming more specific in their effects mental health may vary as a function of features of the neighborhood, such as the propor-Social contexts change over time in response to a child's growth, development, or disorder, tion of other authoritative families (Darling & Steinberg, 1993).
to changes in other setting participants (e.g., parents, peers, or teachers), and to alterations in the broader environment (community, Recommendations.Research on the role of economy, etc.).Study of psychopathology in social contexts in childhood psychopathology development is thus enriched by adopting a may benefit from consideration of the followtransactional approach (e.g., Sameroff & ing issues: Chandler, 1975) that examines changes in context over time and the effects of such changes on the course of development.We • the multidimensionality of social-psychological influences within and across hierar-know, for example, that the onset or progression of psychological disorder in one family chical levels of children's environments, member is likely to influence patterns of rela-of development necessarily or uniformly wane in influence.For example, although tionships within the family system, which will create further effects on the disordered indi-peer groups become increasingly important during the transition to early adolescence, the vidual's mental health (e.g., Patterson & Stouthamer-Loeber, 1984;Rutter, 1990).Simi-family's influence remains as strong in adolescence as it was in early and middle child-larly, transformations in family relations in response to normative developmental change hood.Nor is it the case that relevant contexts broaden over the entire life cycle.After late (e.g., the onset of puberty) often have implications for the mental health of family members adolescence, the relevant context for the development of psychopathology probably con- (Silverberg & Steinberg, 1990).Four particular aspects of contextual change that are tracts, as individuals narrow their choices and limit the settings in which they spend their worthy of more systematic investigation in research on psychopathology may be highlighted.time.Research on contextual influences on psychopathology, with its focus on the family First, the contexts of childhood broaden, differentiate, and deepen with age, as children system, has not paid sufficient attention to the expanding and evolving social worlds of move from infancy through adolescence.Contexts broaden with development when they the child and their influences on development (Rubin, Hymel, Mills, & Rose-Krasnor, grow distally, into wider arenas of the child's external world.Differentiation occurs when a 1991).Especially needed are more studies of children as they make transitions into increas-set of existing relationships disaggregates into subsets of functionally distinct social con-ingly complex and differentiated settings, such secondary school or part-time employ-texts.At high school entry, for example, a young person's early peer group typically re-ment.
Second, the contexts of normal and patho-distributes itself into new and often unpredictable clusters of friendships and alliances.A logical development shift in response to psychological and physical changes in the child deepening of contextual influences refers to change, over time, in the functions of social over time.Transformations in family and peer relations and in classroom interactions take contexts, from those addressing basic, primal needs, such as protection and nutrition, to place, at least in part, in response to the child's development and maturation (Brown, those filling more complex, elaborated needs, such as emotional regulation and interper-1990;Eccles et al., 1993;Steinberg, 1990).
Other changes in interpersonal relations may sonal support.While contextual influences in early infancy, for example, are dominated by occur specifically in response to the early emergence of psychopathology (e.g., Patter-biological imperatives in the dyadic, caregiver-infant relationship, the broader family son, 1986).In addition to examining static features of contexts that contribute to the de-system takes on greater salience later in the 1st year of life as an influence on develop-velopment of childhood disorders, future research is needed to investigate how patterns mental outcomes (Belsky & Rovine, 1988).As the child matures, the relevant contexts ex-of psychopathology emerge, are consolidated, or remit as settings change in response to the pand to include childcare and school settings (in early childhood), peer groups and neigh-child's behavior.Far more research has focused on contextual responses (in the family, borhoods (in elementary school), work and extracurricular settings (in adolescence), as peer group, or classroom) to children's externalizing problems (e.g., Cairns & Cairns, well as broader influences derived from the surrounding culture and society.The roles and 1994; Patterson, 1982) than on contextual responses to internalizing problems.Not only effects of such contexts change with development as well, becoming richer and more dif-should researchers continue to ask how family relationships affect child anxiety and depres-ferentiated in their interactions with the child.
The broadening and deepening of develop-sion (e.g., Rutter, 1990); studies are also needed on how internalizing behavior on the mentally relevant contexts over time does not imply that salient settings during early periods part of the child affects family relationships.
Third, children's contexts also change fol-• the emergence of behavior problems and other early signs of mental disorder in the lowing changes in a setting's other participants.A few studies of such contextual child, changes and their implications for psycho-• the presence or emergence of psychopathopathology have been conducted (e.g., research logic conditions in other members of the on the developmental effects of parental dechild's social contexts, pressive episodes) (Dawson, Hessl, & Frey, 1994;Hammen, Burge, Burney, & Adrian, • dynamic changes in the breadth or roles of 1990a; Rutter, 1990) and parental alcoholism particular social contexts, and (Chassin, Rogosch, & Barrera, 1991), but • secular changes in the characteristics of the much more systematic work is needed.Espebroader social ecology.cially important will be research that examines these changes as dynamic processes (e.g., studies of the impact of changes in parental Proposition 3: Contexts and children are depressive symptomatology on children's mutually determining well-being, as opposed to studies comparing children of depressed versus nondepressed Individual developmental trajectories are responsive to children's experiences within the parents).Very little is currently known, for example, about the impact of parental psycho-multiple contexts of culture, community, and family.Even at the cellular level, however, logical development on child well-being, about the relation between childhood psycho-complex biological systems have been observed to change states in accord with the pathology and changes in peer group relationships, or about changes in teachers' behavior broader environments in which such systems operate (Fentress, 1991).At a much higher and children's concomitant responses.
Finally, a context may broaden, differenti-level of organization, the cultural context of childhood exerts important influences on chil-ate, or deepen in reaction to events within the broader ecology.As Elder and others have el-drens acquisition and use of language, their adoption of gender-specific roles, their attri-oquently pointed out (e.g., Elder, van Nguyen, & Caspi, 1985;McLoyd, 1990b) families bution of meaning to life experience, and their incremental construction of a coherent world-often change in structure and function in response to changing economic conditions.view (Mead, 1961;Minturn & Lambert, 1964;Super & Harkness, 1980).Children growing Similarly, the behavior of peer groups may change as a consequence of shifts in commu-up in economically blighted urban communities, for example, have futures that are clearly nity crime rates or adult supervision (e.g., Sampson & Laub, 1994).Schools may re-jeopardized by the poverty of early life experiences (Adler et al., 1994; Parker, Greer, & spond to changes in educational policies or practices.Although such changes may place Zuckerman, 1988).Abusive family settings create lifelong emotional and behavioral diffi-severe adaptive demands on children and adults, research on how alterations in the culties (Andrews, Valentine, & Valentine, 1995;Fox & Gilbert, 1994), while supportive broader environment affect the development of psychopathology is surprisingly sparse.
parents can positively influence the structure and affective coloring of a child's convictions about the world (Boyce, 1985).At multiple Recommendations.Future studies could be enriched by considering and measuring the levels of social organization, the contexts and settings in which children are reared wield ways in which social contexts and their effects on child psychopathology change over powerful influences on the direction, course, and trajectory of developmental change.time in response to dynamic, developmental factors, such as Generally less well recognized, however, are the equally important, reciprocal ways in which children influence, interpret, choose • the child's age and changing developmental status, and construct their own social contexts.In so doing, children play an active role in shaping ior, may have more frequent exposures to chaotic homes or unruly peers.Plausible accounts their own development (Lerner & Busch-Rossnagel, 1981;Sameroff, 1983).A growing for such gene-context correlations include the possibilities that children with the DRD4 body of infancy research, for example, suggests that early behavior serves to regulate in-marker gene might elicit less effective parental supervision or may select more antisocial terpersonal interactions in order to achieve a communicative and nurturant reciprocity with peers (Plomin & Rutter, in press).
As described by Hinde (1992) and Samer-the parent (Gianino & Tronick, 1988).In the course of interactions with their mothers, in-off (1983), there is continuous interplay between features of the child and aspects of the fants play an active role in achieving such reciprocity through differential responses to social context in which the child is raised.
This bidirectional transaction of context-mothers, fathers, and strangers (Dixon et al., 1981), through contingencies in vocalization upon-child and child-upon-context results in a continuous, mutually influential exchange be-and smiling (Mayer & Tronick, 1985), and through specific behaviors capable of modify-tween the developing individual and his or her social and physical environments.Child psy-ing parental responses (Beebe, Jaffe, Feldstein, Mays, & Alson, 1985).As noted by Gi-chopathology research would benefit greatly, we believe, from efforts to move beyond the anino and Tronick (1988), "This affords the infant significant communicative power, par-positions of individual constructivism and social determinism, to embrace a vision of dy-ticularly with a sensitive and responsive partner, enabling him to initiate, modify, and namic interplay between an active, transforming child and the child's powerful but pliable maintain the exchange." Beyond infancy, children continue to mod-social contexts.Such a vision has been more often apparent within research on normative ify and select their environments.Beginning with Bell's early paper (Bell, 1968) question-aspects of child and adolescent development than in studies of psychopathology.ing the unidirectionality of effects in studies of child socialization, a variety of investiga-Less apparent or resolved on the basis of existing evidence are two aspects of these dy-tors have produced strong evidence of children's and adolescents' influences on the namic, transactional effects: first, the modalities of influence in both directions (i.e., the character of their social environments (e.g., Lytton, 1980).As developed further by Scarr means by which contextual experience and child characteristics are mutually constructed), and McCartney (1983), child effects on social contexts are produced through at least two and second, the relative strength or importance of child → environment effects versus identifiable mechanisms.First, evocative child-context effects result from the differen-environment → child effects.As examples of the first of these dilemmas, the mental health tial responsivity of social environments to children of different temperaments and per-effects of socioeconomic status and parental depression are well recognized, yet the actual sonalities.Thus, friendly, happy babies are more likely than irritable, difficult babies to means by which poverty or maternal depression affect psychopathological processes in evoke positive and stimulating responses from their families and other social settings.Sec-children remain largely obscure (Adler, Boyce, Chesney, Folkman, & Syme, 1993; Offord & ond, active child-context effects result from the process of niche-picking: the growing ca-Fleming, 1995).Consistent associations have been demonstrated between child psychopath-pacity of the child for choosing or selectively engaging those environments or aspects of en-ology and both family economic disadvantage (Offord, 1990) and parental affective disor-vironments most pleasing, affirming, or comforting, given the temperamental or personal-ders (Leckman, Weissman, & Merikangas, 1985; Merikangas, Prusoff, & Weissman, ity characteristics of the individual child.For example, children with the 7-repeat allele of 1988), but hypotheses regarding the pathways through which these associations arise have the dopamine receptor gene DRD4, a fixed marker for ADHD and novelty seeking behav-only recently been tested.Candidate media-tors include the poor quality of mother-child als.The genetic data themselves thus suggest a role for experiential influence.Further, ge-interactions (Dawson, 1994;Hammen, Burge, & Stansbury, 1990a;Harnish, Dodge, & Va-netic studies indicate that the most important of the contextual effects are those that are dif-lente, 1995), a lack of maternal warmth (Dodge, Pettit, & Bates, 1994), or marital ferent for different siblings in the same family (i.e., the nonshared environmental effects).discord and family dysfunction (Lipman, Offord, & Boyle, 1994).Parenting style and so-Environmental influences on psychopathologic disorders generally operate in the direc-cial referencing are two other mechanisms through which context may influence aspects tion of making children from the same family different, as opposed to similar.Finally, chil-of social and emotional development (Darling & Steinberg, 1993;Hinde, 1992).The dren are, without question, much more than the products of their assembled genomes, and young of both human and nonhuman primates utilize the responses of trusted others (e.g., a child's interpretation and assignment of meaning to contextual experiences may alter parents) to signal appropriate reactions to emotionally evocative events.While some of profoundly the dynamic interchange between child and social context.Siblings who have the mechanisms through which social contexts exert their effects have been explored, different perceptions of their family environments (setting aside whether the psychologi-much less is understood about the ways in which individual differences among children cally relevant environments are objectively different) have different developmental out-produce developmentally salient effects on social environments.
comes as a consequence (Daniels, Dunn, Furstenberg, & Plomin, 1985).Although some debate has centered upon the relative magnitudes of contextual versus child effects, it is increasingly clear that de-Recommendations.Future research on the joint and interactive influences of child-spe-velopment and psychopathology are always the products of both organism and context.cific and contextual factors would benefit from considering Past views almost undoubtedly erred in ascribing too much potency to environmental effects.As noted by Plomin and Rende • measurement of both social contextual effects on the child and child effects on social (1991), for example, monozygotic twins reared together are as much alike on IQ as the context, same person measured twice, and twins reared • modalities of such effects and the means in different families are only slightly less simthrough which children and contexts are ilar.Such observations led Scarr (1992) to mutually determining, and conclude that "ordinary differences between families have little effect on children's devel-• differentiation of genetically driven child effects and the individual child's values, opment, unless the family is outside of a normal developmental range."Baumrind (1993), dispositions, and propensities that lie, in a sense, between biology and context.on the other hand, has argued that, even within a normative range of parenting practices, skills such as persuasive communica-Proposition 4: A context's meaning to the tion, contingent reinforcement, and monitoring child determines its effects on the child and produce powerful effects on developmental arises from the context's ability to provide endpoints.
for fundamental needs Rejecting the predominance of either genetic or contextual influences, Reiss et al.The developmental impact of social context is often, though not exclusively, mediated (1991) point out that, while there is irrefutable evidence of hereditary effects on the develop-through a child's subjective reality.Optimal assessments of contextual influences must ment of psychopathology, no studies have indicated that genetic factors account entirely therefore take into account the child's construction and interpretation of contextual ex-for differences between ill and well individu-perience.Lewin (1951) and later Bronfen-ences within a given social context, however, is neither simple nor straightforward, and brenner (1986a; Bronfenbrenner & Crouter, 1983) and Jessor (1981) argued, for example, must often reach beyond the use of self-report measures.Although self-reports remain im-that contexts should be described as perceived and experienced by the child.While there are portant and often singular windows into aspects of children's personal experiences, they objective aspects of children's environments that affect children directly (e.g., the impact can be contaminated by a variety of factors, including pressures for social desirability, dif-of poverty and malnutrition on intellectual development), many environmental factors rele-ficulties in accurate recall, unconscious processes that lead to exaggeration, minimization vant to psychopathology are psychosocial in nature and open to individual interpretation.or denial, and problems in verbal articulation.
Researchers must therefore strive to find Jessor (1981) suggested that three different contexts-the demographic, social structural, clever and original ways of assessing children's perceptions, drawing on methodologi-and perceived-all differ in their proximity to the immediate, direct experience of the child.cal techniques used in clinical interviews, projective testing, experimental paradigms, and It is the perceived or interpretive social context, however, that lies closest to the child's psychophysiological assessment.
Several recent studies illustrate innovative actual experience of the world and is, of these three, most likely to effect psychopathological approaches to augmenting the insights garnered from simple self-reports.In a study de-development.To paraphrase the Thomases' classic dictum, if children define the condi-signed to eliminate problems with children's verbal articulation of psychological con-tions under which they are reared as real, those conditions will be real in their conse-structs, Eder (1990) obtained information on preschoolers' self-conceptions.She hypothe-quences (Thomas & Thomas, 1928).
Among the factors that influence the mean-sized that standard methods used to elicit information from young children about them-ing a child assigns to a particular facet of social context are the child's developmental sta-selves (i.e., interviews and self-report scales) were underestimating the sophistication of tus, past experiences, and reference groups.Parental unemployment, for example, will children's views.Rather than asking children to answer questions about their self-concep-likely have different effects on a younger versus older child, on a child who has previously tions directly, Eder presented a series of displays involving two puppets who differed on experienced parental unemployment versus one who is experiencing it for the first time, specific personality dimensions.Based on children's answers, she was able to show that and on a child who lives in a community with a high unemployment rate versus one in a their self-conceptions were highly differentiated, internally coherent, and consistent over community in which virtually all adults are working. time.
A number of other researchers have fol-The importance of "meaning-making" in the development of psychopathology is well-lowed Eder's lead and adapted her "puppet methodology" to the study of children's per-illustrated in the work of Brown and colleagues (e.g., Brown & Harris, 1989), who ceptions of the family environment and the links between their perceptions and psycho-have argued that the effect of life stress on psychological well-being operates through the logical functioning.Such investigations have provided evidence that even preschool age meaning of the event to the individual experiencing it.From this perspective, the predic-children can provide reliable reports on different aspects of their family environments, and tion of psychopathology from contextual factors hinges on the researcher's ability to that such reports may be more highly correlated than are parents' reports with the objec-evaluate individual interpretation and experience.Prediction is less likely to be enhanced tive assessments of trained observers (e.g., Ablow, Measelle, Cowan, & Cowan, 1998; by an objective assessment of context, independent of the individual's perception.Sessa, Avenevoli, & Steinberg, 1998).Clever approaches to eliciting respondents' Assessing a child's true subjective experi-perceptions of their environment may be able through the process of development.An infant's needs for basic life sustaining provis-to address some of the well-known problems inherent in self-report research.In one study ions, such as food and physical protection, give way in the preschool years to a young constructed to overcome problems of social desirability in self-reports of parenting prac-child's needs for adult constraints on behavior and access to fantasy and play.Similarly, tices, Kagan (Kagan et al., 1986) used an experimental paradigm to examine social class needs in middle childhood for the support and guidance of trustworthy adults are transposed differences in parental socialization techniques.Instead of asking parents to report di-steadily and visibly into the adolescent's requirements for strong peer relationships, a rectly on their behavior, Kagan presented audiotapes of experts giving child-rearing ad-sense of personal identity, and latitude for individuation from the family system.Social vice and varied the content of the advice along several key dimensions of child-rearing contexts thus generate meanings across all developmental periods, but the implications of (e.g., discipline and acceptance).He then tested parents' recall of the taped messages.such meanings change considerably with entry into new stages of motor, cognitive, emo-As hypothesized, significant socioeconomic differences were found in the patterns of re-tional, and social development.
To some degree, the fundamental needs for call, with individuals more likely to remember advice consistent with the predominant values healthy development will also vary from culture to culture and within cultures over time. of their social class.
Implicit in our interest in children's inter-On the other hand, certain core developmental needs are ubiquitous and may be required pretations of social contexts is the idea that aspects of context critical to the development from a child's social context in some form, almost regardless of the individual culture and of psychopathology are those most closely related to children's most fundamental needs.society.Examples of such needs might include those for predictability and consistency Research on contextual factors in the development of psychological disorder might produc- (Ainsworth, Blehar, Waters, & Wall, 1978;Katz, 1971;Martin, 1975), warmth and sup-tively begin with an examination of what these needs are and the features of the envi-port (Boyce, 1985;Cohen & Wills, 1985;Parry & Shapiro, 1986;Siegelman, 1966), en-ronment most relevant to their satisfaction or frustration.In our view, the processes by couragement of mastery (Bandura, 1995;Nolen-Hoeksema, Girgus, & Seligman, 1986), which social contexts foster or undermine healthy development are those in which the physical safety (Garbarino & Kostelny, 1993), and a sense of identity (Butler, 1975; Panak & child appraises the degree to which basic psychological needs are being met.While in in-Garber, 1992;Zucker, 1990).A social context could thus be usefully evaluated in terms of fants and toddlers this process is largely nonverbal and preconscious, older children may the degree to which it meets or fails to meet critical developmental needs that play an im-be fully conscious of such appraisals and may be capable of verbalizing both their contents portant, collective role in the emergence or prevention of psychopathology.and conclusions.Even in an older child, however, meaning may be not simply cognitive, Recommendations.Several methodological imbut may involve working models derived plications follow from this view of the imporfrom preconscious processes (see, for examtance of subjective meanings assigned to exple, Bowlby, 1988).Finally, it is important to periences within contexts.Among design note that positive meanings do not always features that may add substantially to the inarise from adequate provision for needs; a sights derived from study data are the folchild may cling to an abusing parent, for exlowing: ample, despite active and malevolent disregard for the child's safety.
• collection of subjective as well as objective As Erikson (1963) and others in his tradiinformation about children's social contexts, tion have noted, the child's fundamental needs reliably shift as he or she moves • innovative methods, beyond self-report, for assessing children's actual or likely subjec-work of Cowan and Cowan (Cowan, Cowan, Schulz, & Heming, 1994) has documented the tive appraisals of their environments, and advent of powerful new contextual influences • new methodologies that attend to the basic as children move from preschool to primary psychological and physical needs addressed school settings.A complete developmental by childhood social contexts.
analysis of contextual influences would change its focus as it charts the effects of context on successive developmental phases (Bronfen-Proposition 5: Contexts should be selected brenner, 1986b).

for assessment in light of specific questions
There is also evidence that within a particor outcomes ular level of context the most important dimensions differ depending on the outcome In all studies of context and childhood psychopathology, it is important that the social under investigation.For example, Furstenberg and Hughes (1995) have shown that parental environment be defined and measured according to the specific questions or outcomes un-warmth is a critical aspect of family context for predicting high school graduation, but that der investigation.There is evidence, for example, that peer groups are more important parental involvement and role modeling are more important aspects of family context in than parents and families in influencing the adolescent behaviors of drug use and delin-promoting stable work after the completion of schooling.A combination of these and other quency, while parents are more important than peers in influencing long-term values family context effects are important in understanding complex developmental outcomes, and future orientations (Kandel & Lesser, 1972;Youniss, 1980).Although recent re-as such outcomes often require success in a number of domains with different contextual search has shown that there are often powerful interactions between peer and parent ef-determinants.Successful transition to adulthood, for example, is associated with a multi-fects (e.g., Oswald & Suss, 1994), the fact that one domain is usually dominant means dimensional profile of outcomes that includes such factors as avoiding teen childbearing, that a researcher constrained to study a single context type might best select peer groups in completing school, establishing an intimate relationship with a partner, and developing a study of delinquency, and families in a study of educational and occupational aspira-stable membership in the labor force.Each of these outcomes is affected by somewhat dif-tions.For all of the reasons enumerated above, however, studying more than a single ferent contextual influences, which should be studied in disaggregated form to understand contextual level and employing multidimensional views of those chosen would substan-the developmental processes involved in the larger outcome.tively advance research on social contexts.
A variation on this theme is that develop-A more complex within-context problem for the researcher involves choosing the rele-ment itself is accompanied by shifts in the relative importance of children's various social vant level of aggregation to define a particular context for the outcomes of interest.Affilia-contexts.For example, the relative importance of peer groups versus parents would increase tion with a peer group that values success in school, for example, is known to be an impor-in a longitudinal study, over the course of adolescence (Bowerman & Kinch, 1959).Shifts tant determinant of individual commitment and motivation at school (Ianni, 1983).As a in relevant contexts can also be seen in the phases through which problem behaviors result, the friendship network might be considered the most relevant level of aggregation progress.For example, while parental attitudes toward substance use are important de-for defining the school/peer context in a study of investment in school work.However, the terminants of adolescent drug use initiation, peer influences are dominant in affecting the general level of commitment and motivation among all classmates-not only the few who course of continuing use (Kandel & Andrews, 1987).In another example, the longitudinal might share a child's high commitment to ed-ucation, but others in the class who also deter-Sealand, 1993).Crane's (1991) contagion theory about the effects of neighborhood poverty mine the speed and intensity with which the teacher can proceed-is likely to be more rel-and Wilson's (1987) theory of persistent poverty in inner-city neighborhoods both suggest evant in a study of learning.In a similar way, the appropriate level of aggregation used to that a critical mass of poverty (e.g., the percent of community residents living in poverty) define a neighborhood can differ according to the outcome of interest.If exposure to hazards is more important than exposure to middle class neighbors in producing school dropout is the focus, then the block might be the most relevant unit of analysis.If access to resources and teenage childbearing.Research based on collective socialization theories, in compari-is the focus, then the area defining a police precinct, school district, or health department son, would be more interested in evaluating options for positive role models, such as the catchment area may be a more suitable unit of analysis.Finally, if a researcher is inter-number of middle class neighbors, since the proportion of families with higher incomes is ested in peer influences in the neighborhood, then the area where youth congregate would a better predictor of adolescent outcomes than the proportion living in poverty (Brooks-be most appropriate, even if this area is geographically removed from the neighborhood Gunn et al., 1993). in which the youth live.Because several of these influences may be of interest in a single Recommendations.Taken together, these obstudy, accurate assessment of even a single servations indicate that constructs utilized as level of context will, at times, require several measures of contextual influence are best chodifferent definitions tailored to the concepts sen to reflect the outcome or question of relethe investigator seeks to operationalize.
vance within a given study design.Optimal The complexity associated with the fact selection of context measures will be most efthat different aspects of context are relevant fectively served by investigators specifying for different outcomes can be challenging, but and taking into account theoretical considerations can often be helpful.For example, four classes of theory re-• the most important level of the contextual garding the contextual effects of neighbordomain, hoods are (a) contagion theories, which • the salient dimensions within that domain, suggest that peer influences can lead to the spread of behaviors by simple exposure; (b) • subjects' developmental stage, and collective socialization theories, which argue • theoretical assumptions underlying the search that neighborhood role models and behavior for contextual effects.monitoring are more important than behavioral exposure; (c) competition theories, which emphasize the scarcity of resources Social context in light of developmental within the neighborhood; and (d) relative depneurobiology rivation theories, which suggest that neighborhood influences operate largely by creating Finally, there is growing recognition that early psychopathology is often the interactive standards, against which individuals evaluate their relative position (Jencks & Mayer, product of adverse social contexts and learned or genetically determined organismic predis-1990).Each category of theory hypothesizes different associations between neighborhood positions to disorder (Dawson et al., 1994;Jensen et al., 1991;Kagan, 1994).In light of characteristics and child outcomes.Contagion and collective socialization theories, for ex-new awareness for the individual, biological aspects of childhood mental disorders, it is ample, suggest that exposure to middle class neighbors will benefit lower class children, imperative that research on the social contexts of psychopathology be grounded in and alert while most versions of competition and relative deprivation theories suggest just the op-to the known effects of environment on neurobiologic processes and events.While better posite (Brooks-Gunn, Duncan, Klebanov, & Brain development depends critically on con-what value might this definitional and operational perspective on social context and its re-textual input throughout the childhood years, and unless this input is of sufficient quality lation to childhood psychopathology offer at this time in the evolution of a developmental and appropriate in timing, much of normative development can go awry.The developmental view of psychopathology?We believe that such a perspective could be useful at present effects of maternal depression, for example, may be due to the paucity or ill-timing of for a variety of reasons.First, following the seminal work of Cassel (1976) on the health caretaking processes critical for the normative sequence of emotional and social develop-effects of social support, a large body of both observational and experimental evidence has ment (Dawson et al., 1994).
now accumulated documenting the protective, salutary influences of supportive social inter-Recommendations.Observations such as these indicate the importance of viewing social con-actions on a broad variety of health outcomes.
While researchers have studied childhood so-texts as having not only psychological influences, but neurobiological effects as well.cial contexts intensively for many years, few efforts have yet been made to examine com-Emerging neuroscience suggests that social contexts in early life can exert critical influ-monalities in approach, to search for areas of conceptual dissonance, or to identify themes ences on the course of developmental change in neural structures and processes and thereby in the findings from a variety of fields and laboratories.Second, much research to date augment or diminish risk for mental disorder.We have thus argued that researchers address-has focused principally on either aspects of context that promote individual resilience or ing contextual effects on psychopathological development should be knowledgeable of and on those that enhance vulnerability.Less attention has been paid to issues such as the attuned to the crucial and interactive influences of neurobiological processes.We would independence of risk and protective factors in social contexts or the degree to which they also affirm, however, the obverse: neuroscientists studying biological and genetic risk fac-are simply conceptual mirror images of one another.Third, behavioral genetics and twin tors for psychopathology should attend, with new purpose, to the contextual influences that studies have emerged as powerful methodologic approaches to ascertaining the propor-moderate or amplify biological risk.
tions of outcome variance assignable to genetic versus environmental forces.Behavioral Conclusions genetics may provide an unprecedented view into the effects of contextual variation by Our objective in formulating these recommendations for research on social context in child-eliminating the influences of biological inheritance.Finally, many investigators have be-hood psychopathology has been to assemble, within a single statement, the various princi-come convinced that the most interesting and powerful accounts for developmental differ-ples we believe would strengthen studies of contextual influence on child mental health.ences will be found in interactions between intrapersonal, biologic contexts and the char-The five presented propositions are our attempt to delineate these principles and the lit-acter of the individual's multiple social contexts.The processes that govern development eratures that support their empirical validity.It is also our hope that a presentation of prin-are far more complex, however, than those represented by the statistically interactive in-ciples will benefit the field by stimulating further discourse on the character of contextual fluences of child and environmental factors.
What is now needed are studies that examine effects, by highlighting the areas where new or innovative research may be needed, and by not simply the mathematical interactions among organismic and contextual variables, guiding the selection of constructs and measures appropriate to such research.
but the visible and experiential transactions between child and context.What is generally The question might logically be asked, lacking in most current approaches to person-hood, as the most effective and rational means of reducing the burden of the more prevalent environment interaction is a coherent and testable theory regarding how and under what childhood mental disorders.Interventions in all disciplines of medicine require empirical conditions developmentally meaningful interactions take place.tests of their efficacies, and the design of preventive strategies involving social contexts In addition, a more elaborated perspective on childhood social context is needed because will inevitably benefit from increased attention to the character of such contexts and their environments are, at least for the foreseeable future, more accessible and amenable to mod-influences on emotional and behavioral development.It is our strong hope that new knowl-ification than are the genomes that determine, at least in part, the individual differences of edge of children's social contexts will lead ultimately to effective preventive interventions salience to psychopathologic disorders.While genetic interventions may one day become a and to a fuller and richer understanding of the developmental headwaters of childhood psy-plausible approach to the most severe forms of psychopathology, preventive, contextual chopathology.interventions will be regarded, in all likeli-