Date of Degree


Document Type


Degree Name





Yu Gao

Committee Members

Deborah J. Walder

Jennifer E. Drake

Elisabeth Brauner

Erika Y. Niwa

Subject Categories

Child Psychology | Experimental Analysis of Behavior


Emotion Regulation, Respiratory Sinus Arrhythmia, Social Adversity, Reactive Aggression, Proactive Aggression


Two types of aggression in children and adolescents have been identified: reactive aggression (RA) and proactive aggression (PA). Despite the accumulating evidence suggesting differential temperamental, behavioral, cognitive, social-environmental, and neurobiological correlates in relation to the two types of aggression, no study has examined emotion regulation in children with RA vs. PA using psychophysiological approaches. In this study a sample of eight to 10 years old children participated in an emotion regulation task in which they were required to either induce or inhibit their emotions. They also reported their aggressive behavior using the Reactive-Proactive Aggression questionnaire (RPQ; Raine et al., 2006). Electrocardiogram and respiration were assessed continuously during a 2-min rest period and the emotion regulation task and were used to derive the baseline respiratory sinus arrhythmia (RSA) and RSA reactivity, respectively. Both RSA measures and aggressive behavior were assessed again during the one-year follow-up visit. The aims of the study were threefold: First, we aimed to examine if emotion dysregulation, as reflected by low baseline RSA and/or reduced RSA reactivity, would be differentially associated with RA and PA. Second, the study examined the interaction effects between baseline RSA and RSA reactivity in relation to the two types of aggression. Finally, we explored the moderating effects of psychosocial risk factors on the RSA-aggression associations. Concurrent correlations between RSA, aggression, and social adversity measures at each time point were first examined, and the predictive relations between these measures from Time 1 to Time 2 were then determined using multiple regressions and ANCOVAs. Results showed that after PA was controlled for, high RA was associated with autonomic hyperarousal, i.e., higher baseline RSA and increased RSA reactivity, in boys at Time 1 only. No such effect was found in girls. By contrast, high PA was associated with hypoarousal, i.e., reduced RSA reactivity, only in the conditions of low social adversity and no moderating effect of gender was found. Together, these findings provide initial evidence that the RA and PA are characterized by different patterns of the parasympathetic –mediated emotional regulation processes and that both neurobiological and psychosocial influences are important in understanding the etiology of aggressive behavior in children and adolescents.