Beyond Humanoid Emotions: Incorporating Traits, Attitudes and Moods

Abstract

No longer does the idea of robot emotions seem far-fetched; not their experiential side, of course, but rather those manifestations of emotion, especially in robots created in human likeness, which would be beneficial for successful interaction with people. Nonetheless, the concept of robot emotions is still a new one, with a myriad of questions to be answered, not the least of which is: What is emotion? In robotics, it is often used as an umbrella term for all things affective, but based on our previous work (see [1] for a summary), we believe that it would be more beneficial to model each affective phenomenon explicitly. Going beyond emotions brings the entire spectrum of affect into play, providing a comprehensive framework with which human-robot interaction could be improved. The robotic framework we propose that combines a number of different phenomena and emphasizes their interconnectedness and synergy is called TAME (Traits, Attitudes, Moods, Emotions). By using TAME, in this paper we’d like to address some of the open questions that arise in the area of implementing and testing humanoid affect.