Mutual adaptation to mind mapping in human-agent interaction

This paper describes a human-agent interaction framework in which a user and a life-like agent mutually acquire the other´s mind mapping through a mutual mind reading game. Through development of various life-like agents and an internal state (mind) of an agent like emotion, processing load has been recognized to play an important role in making them believable to a user. For establishing effective and natural communication between an agent and a user, they need to read the other´s mind from expressions and we call the mapping from expressions to mind states mind mapping. If an agent and a user do not obtain these mind mappings, they can not utilize behavior which significantly depends on the other´s mind. We formalize such mutual mind reading and propose a framework in which a user and a life-like agent mutually acquire mind mappings. In our framework, a user plays a mutual mind reading game with an agent and they gradually learn to read the other´s mind. We implement our framework and perform experiments to investigate its effectiveness.