Social, Affective Virtual Characters: Augmenting Rapport and Customizability through Personality
'Personality' is a characteristic, stable pattern of thought, affect, and behavior that a person shows when interacting in the physical and social world. In our work we design, build, and evaluate autonomous, humanoid 3D characters that are capable of multi-modal interaction with humans. One of the key features is that the characters have a consistent personality that influences their perception, emotions, and behavior. This consistency is an important factor in making the agents more engaging, relatable, and recognizable. In a counseling context, virtual assistants with distinguishable personalities can adapt to the personality of the human they are interacting with, while in a social skill training application they can serve as diverse social counterparts.
Ulysses Bernardet is currently a postdoctoral researcher at the Simon Fraser University in Vancouver, Canada. He holds a doctorate in psychology from the University of Zurich and has a background in psychology, computer science, and neurobiology. Ulysses’ research follows an interdisciplinary approach that brings together psychological and neurobiological models of emotion, motivation, and behavior regulation with mixed and virtual reality. He is the main author of the large-scale neural systems simulator iqr, and has conceptualized and built the mixed-reality space “eXperience Induction Machine”, as well as over 10 large-scale, complex real-time interactive systems. At the core of his research is the development of models of personality and nonverbal communication that drive the behavior of virtual humans. Ulysses likes to refer to this approach of “understanding humans by building them” as Synthetic Psychology.