Interparental violence, maternal emotional unavailability and children's cortisol functioning in family contexts

Abstract

Our goal in the present study was to examine the specificity of pathways among interparental violence, maternal emotional unavailability, and children's cortisol reactivity to emotional stressors within interparental and parent-child relationships. The study also tested whether detrimental family contexts were associated, on average, with hypocortisolism or hypercortisolism responses to stressful family interactions in young children. Participants included 201 toddlers and their mothers who were from impoverished backgrounds and who experienced disproportionate levels of family violence. Assessments of interparental violence were derived from maternal surveys and interviews, whereas maternal emotional unavailability was assessed through maternal reports and observer ratings of caregiving. Salivary cortisol levels were sampled at 3 time points before and after laboratory paradigms designed to elicit children's reactivity to stressful interparental and parent-child contexts. Results indicated that interparental violence and the mother's emotional unavailability were differentially associated with children's adrenocorticol stress reactivity. Furthermore, these family risk contexts predicted lower cortisol change in response to distress. The results are interpreted in the context of risky family and emotional security theory conceptualizations that underscore how family contexts differentially impact children's physiological regulatory capacities.

Figure 1

Timeline for task procedures and cortisol samples across study visits.

Figure 2

Unconditional latent growth curve model of children’s trajectories of cortisol reactivity to the interaparental disagreement task and the strange situation paradigm. SPAT = interparental disagreement task, SS = Strange Situation, Time = Time of day, μ = sample average, σ = sample variance. † p < .10, ^*p < .05.

Figure 3

A structural equation model testing associations among interparental violence, maternal emotional unavailability, family SES, and children’s cortisol reactivity within stressor paradigms. Parameter estimates for the structural paths are standardized path coefficients and robust standard errors are presented in (). All possible pathways between the three predictor variables and the four cortisol variables were estimated, however for ease of presentation only significant pathways are presented in the figure. In addition, the model presented in Figure 1 was estimated within this process model, however to reduce the complexity of the model within the figure, only the latent variables for cortisol constructs are included. Pathways constrained (time loadings on growth curve variables) and freely estimated within Figure 1 were modeled the same way in the process model presented in Figure 2. ^† p < .10, ^* p < .05.