Creating Rapport with Virtual Agents

Abstract

Recent research has established the potential for virtual characters to establish rapport with humans through simple contingent nonverbal behaviors. We hypothesized that the contingency, not just the frequency of positive feedback is crucial when it comes to creating rapport. The primary goal in this study was evaluative: can an agent generate behavior that engenders feelings of rapport in human speakers and how does this compare to human generated feedback? A secondary goal was to answer the question: Is contingency (as opposed to frequency) of agent feedback crucial when it comes to creating feelings of rapport? Results suggest that contingency matters when it comes to creating rapport and that agent generated behavior was as good as human listeners in creating rapport. A "virtual human listener" condition performed worse than other conditions.

Real feelings for virtual people: Emotional attachments and interpersonal attraction in video games.

“…That is, likeable people look good, are friendly, and are useful, and all three contribute to an overall sense of attraction. Previous research on virtual characters has identified a number of other potentially important factors, including personality (Arellano, Verona & Perales, 2008), politeness and impoliteness (Campano & Sabouret, 2009), rapport (Gratch, Wang, Gerten, Fast & Duffy, 2007), directional attention (Mojzisch, Schilbach, Helmert, Pannasch, Velichovsky & Vogeley, 2006), the social dynamics of display (Ochs & Prendinger, 2010), and similarity/homophily (Jones, Pelham, Carvallo & Mirenberg, 2004). …”

Promoting motivation with virtual agents and avatars: role of visual presence and appearance

“…Other design features include providing an appropriate voice (Nass & Brave 2005), engaging non-verbal communication (e.g. deictic gestures and emotional expressions) (Baylor & Kim 2009) and establishing and maintaining rapport between the agent and the participant (Gratch et al 2007). For additional functionality, the agent could also respond and interact adaptively through cognitive, emotional and motivational models of the user (e.g.…”

“…For additional functionality, the agent could also respond and interact adaptively through cognitive, emotional and motivational models of the user (e.g. Swartout et al 2006;Gratch et al 2007;Arellano et al 2008;McQuiggan et al 2008), or interact with natural language, all of which can add to its believability. The advantages of accessibility in real time and customization are significant, but not if considerable time is required to construct these personalized agent social models.…”

Social Signal Processing: The Research Agenda

“…Similarly, facial expressions influence a human user's evaluation of an Embodied Conversational Agent (Ruttkay & Pelachaud, 2004). Contingency of signals has a key role in creating rapport between human user and virtual agent (Gratch et al, 2007). Politeness cues (Wang et al, 2005) and empathic expressions (Niewiadomski et al, 2008) are perceived as more appropriate in interactive scenarios.…”