Psychology

Automating the analysis of facial emotion expression dynamics: A computational framework and application in psychotic disorders

N. T. Hall, M. N. Hallquist, et al.

We introduce a machine-learning and network-modeling method to quantify the dynamics of brief facial emotion expressions using video-recorded clinical interviews. Applied to 96 people with psychotic disorders and 116 never-psychotic adults, the approach reveals distinct expression trajectories—schizophrenia toward uncommon emotions, other psychoses toward sadness—and offers broad applications including telemedicine. This research was conducted by Nathan T. Hall, Michael N. Hallquist, Elizabeth A. Martin, Wenxuan Lian, Katherine G. Jonas, and Roman Kotov.