Human-Inspired Socially-Aware Interfaces

Social interactions shape our human life and are inherently emotional. Human conversational partners usually try to interpret - consciously or unconsciously - the speaker's or listener's affective cues and respond to them accordingly. With the objective to contribute to more natural and intuitive ways of communicating with machines, an increasing number of research projects has started to investigate how to simulate similar affective behaviors in socially-interactive agents. In this paper we present an overview of the state of the art in social-interactive agents that expose a socially-aware interface including mechanisms to recognize a user's emotional state, to respond to it appropriately and to continuously learn how to adapt to the needs and preferences of a human user. To this end, we focus on three essential properties of socially-aware interfaces: Social Perception, Socially-Aware Behavior Synthesis, and Learning Socially-Aware Behaviors. We also analyze the limitations of current approaches and discuss directions for future development.