ABSTRACT The overarching goal of this research is to promote children’s positive socioemotional pathways and to prevent maladaptive pathways. We elucidate why some children embark on positive paths toward prosocial, internalized, rule-abiding conduct, and robust social competence, whereas others enter maladaptive paths toward callousness, disregard for conduct rules and others’ feelings, antisocial behavior, and impoverished competence. We focus on the parent–child early relationship, formed in the first years of life, as an influential source of the divergent paths, and we longitudinally chart its complex, indirect yet powerful, long-term legacy. Drawing from our extensive research, correlational and experimental, in low- and high-risk families, we propose that although early relationship may not have long-term unqualified, direct effects, it nevertheless serves as a powerful moderator of future parent–child unfolding dynamics. Specifically, early relationship can set the stage for an adversarial, negative cascade. In suboptimal, insecure parent–child dyads, the child’s difficult temperament easily triggers the parent’s negative, harsh, power-assertive control, which, in turn, leads to detrimental child outcomes. In contrast, an early optimal relationship sets the stage for positive, cooperative, effective socialization, and defuses risks of negative cascades. We proposed that parents’ and children’s differing internal representations, expectations, and perceptions of each other (Internal Working Models, IWMs) that characterize suboptimal and optimal relationships and come to guide parents’ and children’s behavior and interactions are the key mechanisms that account for the divergent cascades. We are testing this framework in an ongoing study of 200 community mothers, fathers, and children, richly assessed at 8, 16, 38, and 50-54 months. This application proposes to leverage those massive data to follow up the families at ages 5-6, 7-8, and 9-10. Using state-of-science measures of parents’ and children’s social representations, in Aim 1 we examine how their unfolding IWMs of each other are linked to their relationship quality, how children’s Theory of Mind contributes to their IWMs, how children’s IWMs of the parents generalize to their representations of the social world, particularly hostile attributional biases, and how the child’s IWMs of two parents become integrated in development. In Aim 2, we examine the parent’s IWM of the child as moderating paths from child difficulty to parental control, and the child’s IWM of the parent as moderating paths from parental control to child outcomes. In Aim 3, we embed our model in the dynamics of the family system. Our multi-method, multi-level approach encompasses observational, genetic, and reported measures of the parent’s and the child’s relational information processing, representations, temperament, relationships, parental control, and child developmental outcomes. Variable- and person-centered a...