(PS3-62) Knowledge structures moderate the relationships between testosterone, cortisol, and aggressive behavior

Abstract Text: The dual hormone hypothesis (DHH) posits that cortisol (C) moderates the influence of testosterone (T) on aggression. Specifically, high T is associated with increased aggression when C levels are low, whereas high C inhibits this relationship. However, conflicting findings in previous studies suggest that additional moderating variables may be involved. This study examined whether specific knowledge structures (KSs)—justification of violence, grandiosity, and mistrust—moderate the relationship between T, C, and two types of aggression: proactive and reactive. The study included 419 adolescents (51.1% girls; 48.9% boys) aged 13–18 (M = 14.95; SD = 0.72). KSs were assessed at a single time point, while aggressive behavior was measured at two time points, three months apart. Additionally, saliva samples were collected to assess hormonal levels. KSs were evaluated using validated measures: the Justification of Violence subscale of the Irrational Beliefs Scale for Adolescents (Cardeñoso & Calvete, 2004) and relevant subscales from the Young Schema Questionnaire-3 (Young, 2006). Proactive and reactive aggression were measured with the Reactive–Proactive Aggression Questionnaire (Raine et al., 2006). Saliva samples were collected in classrooms and analyzed for T (pg/ml) and C (nmol/L) using the Salimetrics® Testosterone Enzyme Immunoassay Kit and the Cortisol II ECLIA method. The study was approved by the institutional ethics committee, and active informed consent was obtained from both parents and adolescents. To test the hypothesized path model, data were analyzed using Mplus 8.9 with a full information maximum likelihood estimator, running separate models for each KS. Results indicated that all three KSs predicted both proactive and reactive aggression and, moreover, moderated the hormone–aggression relationship. Specifically, justification of violence moderated the associations between C and proactive aggression, T and both forms of aggression, and also showed a significant three-way interaction among T, C, and justification of violence in predicting proactive aggression. Grandiosity moderated the association between T and C with proactive aggression but not with reactive aggression. Mistrust moderated only the relationship between C and proactive aggression. In all cases, KSs amplified the relationship between T and aggression. However, for C, the link between KSs and aggression was stronger when C levels were low. The significant three-way interaction indicated that justification of violence predicted proactive aggression only when T levels were high and C levels were low, aligning with the DHH model. These findings highlight the need for an integrative biopsychosocial approach to understanding adolescent aggression. By considering hormonal and psychological factors as well as their complex interactions, this study contributes to a more comprehensive understanding of the mechanisms underlying aggressive behavior in adolescence. The results offer valuable insights for developing targeted interventions aimed at reducing aggression in this age group.