We examined emotion regulation (ER) processes in three high-risk samples, with two primary aims: (1) to examine the effects of concentric levels of influence on ER, from within the individual (i.e., genotype, temperament), to aspects of the proximal environment (i.e., parenting behaviors), and to broader contextual factors (e.g., income, family composition); and (2) to determine whether various levels of influence affect developmental trajectories of ER, in addition to initial levels. We advanced these aims in samples of children with and without developmental delay, low-income children enrolled in Early Head Start, and children adopted from foster care.
In Study 1, we used latent growth curve modeling (LGCM) to model yearly change in emotion dysregulation from child age 3 to 6 years and to test predictors of initial levels of and change in dysregulation. Dysregulation was found to decrease overall across early childhood, and serotonin transporter genotype X parenting interactions predicted individual differences in growth curves. Dysregulation trajectories of children with the SL/LL genotype were minimally related to positive and negative parenting behavior, whereas dysregulation decreased more precipitously among children with the SS genotype when exposed to low negative parenting or high positive parenting.
In Study 2, latent class growth analysis (LCGA) was used to identify trajectory groups of ER across toddlerhood and examine predictors of those trajectory groups from child temperament, parenting behaviors, and environmental risk. LCGA supported a three-class model, with a Stable-High ER group, a Low-to-High group, and a High-to-Low group. The Stable-High group was characterized by high positive parenting, the Low-to-High group by high child negative emotionality, and the High-to-Low group by high negative parenting and high environmental risk. Lastly, membership in these trajectory groups was found to be predictive of resilient functioning in the 5th grade.
In Study 3, we examined factors that predict ER among children adopted from foster care, at the time of the adoptive placement and then 5 years post-placement. Among child temperament, pre/perinatal risk, and pre-placement environmental risk, reactive temperament emerged as the sole predictor of ER at the time of adoptive placement. When looking at ER after 5 years of adoptive placement, aspects of the adoptive family environment predicted changes in ER, though the nature of the changes depended on the temperament of the child.
Results from the three studies were discussed in terms of theoretical and practical implications.