This research is designed to develop and test a process model that explains and predicts the maintenance of aggressive mother- son interactions. The significance of examining the relations between attributions and aggressive behavior is that coercive mother-child interactions place children, especially boys, at risk for exhibiting antisocial behavior in other contexts, even into adolescence and adulthood. The specific aims of this study are (1) to examine the relations between mothers's and children's attributions and the coerciveness of their interactions: (2) to chronicle stability and change in conflictual mother-son interchanges and their social cognitions over one year in order to (3) assess the extent to which attributional processes contribute toward aggressiveness (coerciveness) and aggressive interactions contribute toward mothers' and sons' attributions; and (4) assess whether there are particular setting conditions (e.g., maternal depression, spousal/ex-spousal conflict, economic stress) that increase the tendency of some mother-son dyads to make negative attributions and engage in coercive interactions. Mother-son dyads from married, separated, and divorced families will be studied prospectively. Two assessments (each separated by one year) will be made of mothers' and their sons' (7-9 years of age) attributions and coercive interactions. The subjects will be divided into two cohorts of 100 families each. Cohort assessments are staggered such that replication of findings in each wave can be accomplished within 12 months. In addition, assessments will be made of personal-social and socio-ecological setting conditions that may contribute toward some mother-son dyads being particularly vulnerable to making negative attributions and engaging in coercive interactions. This research will extend our understanding of the relationship between social cognitions, coercive interactions, and ecological circumstances.