Annotating and Generating Posture from Discourse Structure in Embodied Conversational Agents

This paper addresses the problem of designing embodied conversational agents that exhibit appropriate posture shifts during dialogues with human users. Previous research has noted the importance of hand gestures, eye gaze and head nods in conversations between embodied agents and humans. However, this research has neglected the role of other body movements, in particular postural shifts. We present an analysis of human monologues and dialogues that suggests that postural shifts can be predicted as a function of discourse state in monologues, and discourse state and conversation state in dialogues. On the basis of these findings, we have implemented an embodied conversational agent that uses a dialogue manager called Collagen in such a way as to generate postural shifts.