Relational Agents are computer agents designed to form long-term, social-emotional relationships with their users. We are investigating the use of these agents in task domains in which human-agent relationships actually improve task outcomes, such as in coaching, counseling, psychotherapy and healthcare.

Tinker: A Relational Guide for the Boston Museum of Science

People go to museums as much to be entertained and to socialize as they do to be educated. Contemporary museums use interactive exhibits, multimedia, games, automated mobile guides, and other mechanisms for entertaining individuals and groups of visitors. The incorporation of principles from the social psychology of human personal relationships represents a promising and important direction of research to further engage and entertain museum visitors.

In this project, we have developed a virtual museum guide agent named "Tinker" that uses human relationship-building behaviors to engage museum visitors. The agent appears in the form of a human-sized anthropomorphic robot, and uses nonverbal conversational behavior, empathy, social dialogue, reciprocal self-disclosure and other relational behavior to establish social bonds with museum visitors. The agent also uses a biometric identification system (based on hand image features) so that it can re-identify visitors it has already talked to, and maintains persistent discourse and relational models, so that prior conversations can be seamlessly continued. Tinker can provide visitors with information on and directions to a range of exhibits in the museum, as well as discuss the theory and implementation underlying her own creation.

This effort is in collaboration with the ComputerPlace exhibit in the Boston Museum of Science. Over 125,000 visitors have interacted with Tinker since the exhibit became operational in September, 2007.