Capturing the Social Spectrum in ASD Rats*

Social interaction is a core component of animal behavior. Describing it in rigorous and quantitative ways requires unsupervised tracking of spontaneous social behaviors. We achieve this by extending a recently developed technique for 3D kinematic tracking of single animals (DANNCE) to freely interacting animals (social-DANNCE). Using this method, we acquired a rich dataset of interactions across pairings from autism spectrum disorder (ASD) knockout rats and their wild-type counterparts. Using a multi-scale dynamical embedding approach to parse animal movement in both solitary and social contexts, we describe differential expression of both individual and social behavior. Our framework for quantifying and comparing interactions in paired animals revealed a spectrum of changes in social behavior that depend both on the strain of the animal and the identity of its social partner. This approach promises to greatly facilitate quantitative studies of social behaviors and their neurobiological underpinnings.