SUMMARY STATEMENT Aggression is an innate social behavior seen across vertebrate species. Although an essential means to compete for resources and defend oneself, aggression is also a costly behavior with potentially deadly consequences. Hence, it is important for animals to direct aggression toward the correct targets to maximize benefits and minimize risk associated with fighting. In humans, misdirected aggression is particularly destructive as it often targets the most important people in one’s life—spouse, children, caregivers, or even oneself. Such misdirected aggression can strain relationship and cause tremendous physical and mental stress. For example, individuals with certain psychiatric disorders, such as bipolar disorder and post-traumatic stress disorder, are more likely to express misdirected aggression, jeopardizing their own and others' lives. The goal of the current project is to elucidate the neural mechanisms that allow an animal to generate properly directed aggression. Doing so involves multiple behavioral processes. In mice, for example, individuals in a social group first learn the social status of each member through repeated agonistic interactions. Then, in subsequent encounters, they recognize the social status of an opponent and use this information to determine whether to initiate attack. Here, we will investigate the neural mechanisms supporting each of these behavioral processes that ultimately allow mice to selectively initiate aggression. Given the well-known role of dopamine in reinforcement learning, Aim 1 will address whether dopamine from the ventral tegmental area (VTA) plays an important role in fighting-outcome-based social learning and aggression expression. We will use a systematic approach to identify the key site(s) of dopamine action in modulating aggression. Aim 2 will examine the neural representation of social status information, with particular focus on the posterior part of the medial preoptic nucleus (pMPOA), a key upstream region of multiple nodes in the aggression circuit. In Aim 3, we will examine the functional role of pMPOA projections in directing aggression via pathway-specific optogenetic manipulation. In summary, this study will reveal how social status information is represented in the brain and used to direct aggression toward the right social targets. It further promises to provide new understanding of dopamine's role in aggression and associated social learning. If successful, our proposed investigations will provide new insight into the fine neural control of aggression, and may yield new therapeutic targets for suppressing the exaggerated and mistargeted aggression accompanying some psychiatric conditions.