A Model of Emotion for Empathic Communication

Abstract

Most people believe that robots have no emotions, and nor do they need them. However, we strongly believe that having emotions is essential for robots to understand and sympathize with the feelings of people, thereby allowing them to be accepted into the human society. In this paper, we propose a model of emotion based on some neurological and psychological findings concerning empathic communication between humans and robots. Then, we examine a method for generating affect for given visual stimuli using a recurrent neural network as a first step.