The main goal of this study was to examine how stress reactivity and family adversity influence socio-emotional behavior and school readiness skills in children attending kindergarten. Several significant main effects emerged. Consistent with the broad literature on the effects of risk and adversity (e.g., Luthar, 2006
), family adversity was associated with concurrent externalizing symptoms and lower school engagement, and with increases in externalizing symptoms, and decreases in prosocial behaviors and school engagement over the kindergarten year. However, the effect of adversity on fall school engagement and change in externalizing symptoms was significant only for boys. Adversity did not have a main effect on early kindergarten indices of academic competence. This discrepancy with previous research could be due to the fact that the majority of children in this study come from highly educated families, which may support academic success despite adversity exposure.
Low RSA reactivity was associated with higher levels of externalizing symptoms in the fall, corroborating previous findings of low stress reactivity and under-arousal in children with externalizing symptoms (Boyce et al., 2001
; Calkins et al., 2007
; Mezzacappa et al., 1997
; Pine et al., 1996
) and the idea that low RSA reactivity is an index of emotional dysregulation and lability (Beauchaine, 2001
). On the other hand, high cortisol reactivity was associated with high levels of externalizing behaviors and low levels of school engagement and academic competence. These findings are consistent with studies that have linked elevated cortisol in response to classroom challenges to poor self-regulation, impulsivity, peer rejection and more solitary, negative behaviors (Dettling et al., 1999
; Gunnar et al., 1997
) and the notion that hyperresponsivity of the HPA axis signals a risk for maladaptation (Gunnar & Vazquez, 2006
). However, these main effects and the absence of others in these models should be interpreted with caution, as they were conditioned by significant interactions. For example, the effect of RSA reactivity on change in academic competence significantly varied across sex, with high reactivity being promotive for boys but a risk factor for girls. In addition, high cortisol reactivity was related to higher levels of fall externalizing symptoms only in boys. These findings suggest that children’s sex may be important factor to consider when examining the effects of stress reactivity on adaptation.
The study’s most novel and salient findings emerged when adversity and stress reactivity were considered together, as components of interactions between environmental exposures and measures of biological sensitivity. Stress reactivity moderated the negative effect of family adversity across various domains of adaptation. Overall, the findings are consistent with the ‘stress diathesis’ hypothesis that high reactive children show worse adaptive functioning in the context of high adversity. Indeed, such children generally evinced the lowest levels of adaptive functioning of the entire study sample.
However, equally reactive children in settings of low adversity showed the highest levels of adaptation, levels even higher than those of their less reactive counterparts. Specifically, in the context of low family adversity, children who showed high RSA reactivity in response to challenges had the lowest levels of externalizing symptoms and the highest levels of prosocial behaviors and school engagement. Although adaptation showed significant stability from fall to spring, high reactive children showed improvement in academic competence in the context of low adversity and a decline in competence in the context of high adversity, whereas the inverse was true for low reactive children. Similarly, children who showed high cortisol reactivity to the challenge protocol had the highest levels of prosocial behaviors in the context of low adversity. Further, children exhibiting low RSA reactivity in response to challenges were fully or partially buffered against the harmful effects of adversity on externalizing symptoms, prosocial behavior, and school engagement. Likewise, among children who showed low cortisol reactivity, levels of prosocial behaviors did not significantly change across different levels of adversity.
These findings support the biological sensitivity to context theory advanced by Boyce and colleagues (Boyce et al., 2005
; Boyce, 2007
) and the concept of differential susceptibility to environmental influences proposed by Belsky and colleagues (Belsky, 2005
; Belsky et al., 2007
). This study illustrates that high reactivity is not merely a pathogenic, risk-amplifying response to adversity but can also promote adaptive functioning. Corroborating Boyce and colleagues’ theoretical perspective, children exhibiting high levels of biological sensitivity to context, as indexed by high autonomic and adrenocortical reactivity, were more susceptible to environmental influences in the context of both low and high family adversity. Thus, biologically sensitive children showed the highest levels of symptoms in the context of high family adversity but the highest levels of competence in the context of low family adversity. However, a lack of family adversity does not necessarily imply the presence of a nurturing family environment. Thus, future studies will need to further examine the role of heightened biological sensitivity to context across both stressful, health-undermining and supportive, health-enhancing contexts.
Together, these findings provide strong support for reframing stress reactivity as a biological sensitivity to context. First, the findings are relatively robust, especially for RSA reactivity, which emerged as a significant moderator of adversity effects across all four domains of adaptation. The evidence was parallel, though less robust, for cortisol reactivity, as the interaction effect was significant only for one domain. Second, each observed interaction was consistent with the biological sensitivity to context theory; high reactivity was promotive in the context of low adversity but a risk factor in the context of high adversity. Third, the interaction effect was found for both positive and negative indices of adaptation. Past studies have rarely examined how interactive processes apply to multiple domains of functioning and have often focused primarily on measures of psychopathology. Fourth, these highly consistent interaction effects were found in a normative, community sample of typically developing children. If stress reactivity moderated the effects of adversity on development in a community sample of children, who on average do not face extreme disadvantage or show clinical levels of problem behavior, the interactive effects are likely to be larger at the extreme ends of such distributions.
Finally, the reported interaction effects complement recent work examining individual differences in markers of behavioral and genetic susceptibility to environmental influences. The current findings are consistent with research on behavioral reactivity indicating that children with high levels of negative affectivity are particularly susceptible to both negative and positive experiences (Belsky, 2005
; Belsky et al., 2007
; Klein Velderman et al., 2006
). In addition, the current findings parallel recent research showing that genetic polymorphisms can moderate the effects on adaptive functioning of family adversity, unresolved trauma, or abusive parenting (Bakermans-Kranenburg & Van IJzendoorn, 2007
; Caspi et al., 2002
; Taylor et al., 2006
; van IJzendoorn & Bakermans-Kranenburg; 2006
) and resonate with the observations of Rutter and colleagues on the previously overlooked protective effects of ‘risk-engendering’ genetic variants (Rutter, Moffitt & Caspi, 2006
). Future studies should examine whether these behavioral, physiological, and genetic markers of susceptibility to contextual factors represent the same phenomena expressed at different levels of assessment or whether they represent different types of susceptibility that may have a cumulative effect on development.
For the purposes of future studies, it is important to note that the nature of the reported interaction effects may depend on several dimensions depicted in the conceptual model of . Single studies may be capable of capturing only a portion of the underlying, paradigmatic interaction shown in , as the ‘observational window’ shifts from study to study, depending on sampling characteristics, the outcomes of the interest, the measure of reactivity, and the type of challenge. For example, the observational window may shift to the right in studies of highly disadvantaged child populations or upwards in studies of children with severe behavior problems. Similarly, the slopes of the two component lines may be determined by the researchers’ ability to capture the full range of stress reactivity responses and thus distinguish between high and low reactive children. It is critical, therefore, that developmental researchers examine and compare the interactive effects of stress reactivity and context across a range of outcomes and environmental influences and that the full scope of existing literature be taken into account in the interpretation of such effects.
Conceptual model of the interactive effects between biological sensitivity to context (BSC) and environmental influences.
The relations between stress reactivity and context are also likely to be dynamic and evolving over time and the present study provides only a snapshot into a potentially complex cycle of bi-directional and reverberatory influences. Boyce and colleagues (Boyce, 2006
; Boyce et al., 2005
) have argued, for example, that stress reactivity may not only moderate effects of early experience, but may itself be influenced by
the quality and character of early experience. While our analyses addressed stress reactivity as a moderator of adversity, given the developmental stage of the participants and the nature of the study data, we are mindful that adversity exposure could also play a role in shaping individuals’ stress reactivity. From the perspective of evolutionary biology (Ellis, Jackson, & Boyce, 2006
), children growing up in supportive and nurturing families might be expected to develop high levels of biological sensitivity to context in order to take greater advantage of the positive and stimulating features of their environments. As a result of both high sensitivity and protective environments, these children would show high levels of competence and low rates of mental and health problems. On the other hand, children exposed to high levels of early risk and adversity may also develop high biological sensitivity to context, as a means of sustaining vigilance for environmental threats and hazards. Although such vigilance may be adaptive in the short run, over a longer period of time it could augment children’s vulnerability to the deleterious effects of adversity. It is not surprising then that these children tend to have higher rates of mental and health problems. On the other hand, children who develop low biological sensitivity to context seem to be less affected by both positive and negative environmental influences. These children may consequently demonstrate resilience—adaptation that is better than expected given their adversity exposures—or vulnerability—low levels of adaptation despite growing up in environments abundant with resources and support (Luthar, 2006
; Masten & Obradović, 2006
). Future studies are needed to examine the processes by which early experiences shape children’s biological stress responses.
Strengths, Limitations and Future Directions
This study had several unique strengths as well as notable limitations that highlight important directions for future improvements. First, the study was based on a large, ethnically diverse sample of kindergarten children; however, a majority of children came from families with relatively high socio-economic status, as indicated by income and parents’ education. Second, the study was unique in that it employed a carefully controlled measure of autonomic reactivity that took into account the possible effects of motor activity and engagement demands elicited by challenge tasks. However, only a third of sample had elevated cortisol in response to stress reactivity protocol, which may have contributed to less robust findings for cortisol reactivity. Future studies should attempt to raise the responder rate, as lack of response to the current protocol does not necessarily indicate a low reactivity to more intense, real-life stressors. Third, we included the various sources of family adversity as an index of early experience. However, future researchers should explicitly assess positive environmental influences in children’s lives by including measures of social support and resources. Another strength of the current study was the use of multiple informants. While it may not always be feasible to use multiple informants when assessing family adversity, given the young age of the participants and the nature of the construct, future studies should, where possible, triangulate the child, parent, and teacher reports of adaptation, thus minimizing informant variance. Fourth, the study was unique in its effort to compare how both indices of autonomic and adrenocortical reactivity interact with early family adversity to predict adaptation. In addition to examining how indices of different types of stress reactivity affect children’s behaviors and skills, future research should examine how different stress response systems interact to produce unique stress reactivity profiles, which may be related to different adaptive patterns across adversity exposure. As discussed above, researchers might use person-focused analytic approaches to examine whether the same children express analogous behavioral, physiological, and genetic sensitivity to contextual factors. Finally, future research should examine transactional relations between stress reactivity and adversity across longer periods of development, as adaptation in this study showed high stability over the 6-month period.
Given the pervasive and long-lasting effects that adversity can have on adaptation (Luthar, 2006
; Obradović et al., in press
; Sameroff, 2006
), it is important to elucidate the processes that buffer or exacerbate these effects. This study represents a rare attempt to examine complex interactions between biological and environmental factors. The findings indicate that children’s biological sensitivity to social context played an important role in moderating the effects of early experiences of family adversity on positive and negative indices of adaptive functioning. We believe that studies such as the one reported here can advance future prevention and intervention efforts, to address more holistically the social disparities in children’s competence and psychopathology.