Emotion Model for Life-Like Agent and Its Evaluation

This paper proposes an emotion model for life-like agents with emotions and motivations. This model consists of reactive and deliberative mechanisms. The former generates low-level instantaneous responses to external stimuli that come from the real world and virtual worlds. The latter mechanism especially focuses on emotions. A basic idea of the model comes from a psychological theory, called the cogniti ve appraisal theory. In the model, cognitive and emotional processes interact with each other based on the theory. A multi-module architecture is employed in order to carry out the interactions. The model also has a learning mechanism to diversify behavioral patterns. These features are effective in giving users the illusion of life. We applied the proposed model to characters in a virtual world and show the results obtained from three experiments with users.