Mechanism, modulation, and expression of empathy in a virtual human

Abstract

Empathy is believed to play a prominent role in contributing to an efficient and satisfying cooperative social interaction by adjusting one's own behavior to that of others. Thus, endowing virtual humans with the ability to empathize not only enhances their cooperative social skills, but also makes them more likeable, trustworthy, and caring. Supported by psychological models of empathy, we propose an approach to model empathy for EMMA — an Empathic MultiModal Agent — based on three processing steps: First, the Empathy Mechanism consists of an internal simulation of perceived emotional facial expressions and results in an internal emotional feedback that represents the empathic emotion. Second, the Empathy Modulation consists of modulating the empathic emotion through different predefined modulation factors. Third, the Expression of Empathy consists of triggering EMMA's multiple modalities like facial and verbal behaviors. In a conversational agent scenario involving the virtual humans MAX and EMMA, we illustrate our proposed model of empathy and we introduce a planned empirical evaluation of EMMA's empathic behavior.