Role of Inter-Personal Synchrony in Extracting Social Signatures: Some Case Studies

Abstract

Originally studied by developmental psychologists, synchrony has now captured the interest of researchers in such fields as social signal processing (SSP). Developing metrics that can capture dynamics of social interaction is of particular interest, called here Social Signatures (behavioral characteristics). We describe several case studies for extracting robust and distinctive social signatures : early parent-infant interaction, new framework by exploring hormone changes as ground-truth labeling of quality of interaction and a robot learning approach based on neural networks is able to capture distinctive social signatures between clinical groups. Finally, we discuss future perspectives of learning and characterizing social signatures.