Modeling a Social Brain for Interactive Agents: Integrating Mirroring and Mentalizing

Abstract

Abstract. Human interaction has a distinct collaborative quality based on the attribution of communicative intentionality. Two networks in the human brain are often described as part of the "social brain": the mirror system for recognizing intentional behavior and the mentalizing (theory of mind) system for processing it. We equip virtual agents with both systems and model their interaction during embodied communication. Results of simulation experiments demonstrate how higher orders of theory of mind lead to more robustness of communication by enabling interactive grounding processes.

Revisiting Human-Agent Communication: The Importance of Joint Co-construction and Understanding Mental States

“…How these processes play out in social behavior, different modalities, or at different points of a communicative interaction still needs to be understood and modeled computationally. Particularly important questions relate to how low-level perceptual processing interacts with mentalizing, and how predictive processes come to interact and couple with each other across interactants ( Kahl and Kopp, 2015 ).…”

Understanding and Predicting Bonding in Conversations Using Thin Slices of Facial Expressions and Body Language

“…Agents that can mirror whole body movements (e.g. [8]) may be highly effective at facilitating bonding.…”

“…[7]) or mirroring (e.g. [8]) between the VA's behavior and the user. Detailed models of rapport have also been developed [9].…”

Personality, Attitudes, and Bonding in Conversations

“…The importance of attentiveness may suggest that designing agents around mirroring (e.g. [3]) and contingent nonverbal cues (e.g. [4]) may be the most promising approaches.…”

“…[2]), mirroring (e.g. [3]), and the contingency of their non-verbal responses [4]. Detailed models of bonding and rapport [5], and interpersonal emotions in conversations [6], have also been developed.…”