Using automated interactional synchrony to predict complex behaviors


A team of students led by researchers in the Social Science Research Institute and Departments of Mathematics and Statistics will curate a unique video data set for studying how different aspects of social interactions relate to social and psychiatric variables, like trust, empathy, and scores on clinical social competence and autism scales. Students will prepare the data set by performing machine vision and audio analysis, signal processing techniques, and novel data cleaning techniques, and analyze the data set using interpretable machine learning.  This work will help generate a one-of-a-kind data set to be used by researchers across the world and will provide proof-of-concept demonstrations for how operationalizing social interactions mathematically can help us understand the human brain and diagnose and treat psychiatric disease.

Project Leads: Jana Saich-Borg & Hau-Tieng Wu