A computational model of social attitude effects on the nonverbal behavior for a relational agent

Brian Ravenet (1) , Magalie Ochs (1) , Catherine Pelachaud (1)

1 LTCI - Laboratoire Traitement et Communication de l'Information

The relations we have with others influence in a very subtle way our body gestures. These are cues that are used in an interaction in an unconscious process to communicate and understand an attitude. Depending on the gestures and facial signals one is displaying, a social attitude can be perceived. In this paper, we propose to model socio-emotional agents and how they can express such an attitude in a dyadic interaction. The focus will be on generating the nonverbal behavior of the virtual agent given the social attitude it wants to convey. Depending on the nature of the relation an agent has with someone else, its role and desires, it should display a different attitude and therefore it should display different nonverbal behaviors. In this paper, we propose a computational model based on the findings in Human and Social Sciences on the correspondence between social attitudes and nonverbal behaviors. This computational model is used to select the proper behavior of an agent during an interaction depending on the social attitudes it wants to convey and depending on its gender.

Mots clés: virtual agent; relational agent; social attitude; nonverbal behavior

Domaines: Interface homme-machine [cs.HC]; Intelligence artificielle [cs.AI]

Brian Ravenet, Magalie Ochs, Catherine Pelachaud. A computational model of social attitude effects on the nonverbal behavior for a relational agent. Workshop Affect Compagnon Artificiel Interaction, Nov 2012, Grenoble, France. ⟨hal-01131395⟩