Project Summary Social interactions are critical to the physical and emotional health of a wide variety of species. Perturbations in social functioning, a hallmark symptom of many psychiatric and neurodevelopmental disorders such as autism and schizophrenia, can profoundly impair an individual’s ability to sustain healthy social relations. While a growing body of literature has elucidated neural circuits for dyadic social interactions (interactions between two individuals), our understanding of higher order interactions at the level of larger groups is remarkably weak. Humans and other species organize themselves into social groups, in which the behavior of each individual both contributes to and benefits from the cohesiveness and well-being of the collective. Social groups serve as a context for sharing of resources, buffering of stress, regulation of homeostatic needs, and a reservoir of cognitive capacity to solve problems and respond to environmental challenges. In order to address this gap in the literature, I am using a novel behavioral approach to study how groups of mice self-organize into huddles in response to a cold temperature thermal challenge. Prior studies examining social groups have been limited by lack of technology to track the pose and identity of each mouse over the duration of a session. To address these challenges, I am using novel multi-animal pose estimation tools developed through computer vision to quantitatively identify huddling configurations in groups of four mice. Furthermore, I am combining this automated behavioral tracking with circuit dissection tools to understand which circuits in the brain coordinate huddling in response to thermal challenge. Published work from our lab and others suggests that medial prefrontal cortex (mPFC) is a critical node involved in regulating group level behaviors and inter-brain dynamics across species. However, the contribution of mPFC and its descending projections to group huddling has never been explored. Furthermore, although whole-brain knockout studies have found that the social hormone oxytocin is necessary for huddling in mice, the precise neural circuits that are engaged by oxytocin to promote huddling never been examined. The experiments described in this proposal will fill a critical gap in our understanding of neural mechanisms for collective behavior by performing detailed quantitative behavioral analysis and neural circuit dissection to physiologically observe, computationally model, and functionally manipulate individual descending projections of the dmPFC during huddling. Using an ethologically relevant group behavior, in vivo freely moving calcium imaging, chemogenetic manipulations, and anatomical mapping, the proposed study will test the hypothesis that oxytocin engages discrete, anatomically-defined pathways descending from the dmPFC to promote group huddling. These results will set the foundation for a more incisive analysis of how dmPFC circuits shape social ...