Understanding Human Social Kinematics Using Virtual Agents

AbstractA pressing issue in both psychology and agent-modeling communities is the inability to account for the wide variance in human variability and individual differences. Added to this is the further complexity of changing goals and social meaning in a dynamic, sequential interaction. While prior work on artificial agent design has prominently addressed physical cues and nonverbal behavior, there is a lack of emphasis on (1) examining cues in combination, and (2) assessing judgments of social situational meaning. In the current work, we present an ontology of physical behavior (Social Kinematics) that accounts for the combinatorial effects of multiple cues, as well as the changing social meaning associated with these different combinations of cues. Here, we assess individuals’ social situational judgments of multiple combinations of ambiguously-defined virtual agent animations. Ultimately, this paper provides a potentially useful framework that has relevance for researchers in social robotics, agent modeling, and cognitive science.

Return to previous page