Communicative criteria for processing time/space-varying information

This paper describes our on-going project for investigating the communicative criteria for human-robot face-to-face communication like that with animated or robotic talking-heads. We assume that people recover linguistic and affective meaning not from static or nearly static configurations such as still pictures or short sound segments, but in the temporal and spatial patterns of the visible and/or audible information associated with the face and the surrounding environment. Based on this assumption, we discuss what spatio-temporal patterns convey communicative information and introduce some of our preliminary attempts using animated and robotic talking-heads.