The Affective Social Computing Group HomePage

We are interested in research on affect, emotion, and personality at the intersection of research in Artificial Intelligence, Human-Computer Interaction, and Robotics. Emotional systems in humans influence important cognitive processes such as salience determination, focus and attention, priority determination, interruption in emergency situation, memorization and recall, goal generation, goal attribution, categorization, and preference. All these processes are important for intelligent systems with limited resources evolving in an unpredictable environment, including artificial ones (W. Clancey; N. Frijda; M. Minsky, D. Norman, A. Ortony, R. Picard, D. Rumelhart, H. Simon; A. Sloman; R. Zajonc).

We currently focus on two aspects of the role of emotions in (1) communication (human-computer interaction, and computer-mediated communication); (2) decision-making; (3) study of the co-evolution of social structures and people with information technologies. My research therefore involves:

1. designing and developing computer systems and interfaces with increased "awareness" of their user's states and communication patterns in order to adapt appropriately to their users: In particular, we focus on Affect, emotions, and personality. Even though there exists a multitude of applications where human Affect would gain to be acknowledged and responded to, we currently specialize on telemedicine, driver's safety, and training. As our ideas progress, we will explore other applications.
2. studying and importing some of the important functions of human emotions and social interactions to the development of Artificially intelligent systems, both software agents and robots: Indeed, pure reasoning and logic have proven to be insufficient to account for true intelligence in real life situations, in which there simply is no time to determine which action to take out of an infinite number of possible ones, given a set of premises.
   
   Our view is not that computers could actually feel, but that it is possible to extract some major functions of emotional systems in humans and simulate them to render artificial systems more intelligent as well as aware of human affective and/or social processes. We can learn from the evolution of emotions in human intelligence, and its role in our development as a species. Furthermore, with the explosion of multi-agent systems, in which several agents interact and must resolve conflicts, simulating the social dimension of emotional intelligence can lead to the development of increasingly intelligent artificial artifacts.
3. Evaluating our systems and intelligent user interfaces concurrently as we design them to ensure that the technologies of the future are truly accepted and enhance human lives.