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.

A RCT to Reduce Cardiopulmonary Re-hospitalizations

This five-year, NIH/NHLBI-funded project, in collaboration with Boston Medical Center, will develop procedures and technologies for educating hospital patients about their conditions and post-discharge self-care, medications, and follow-up appointments, in an effort to reduce re-hospitalizations due to poor patient communication and patient non-compliance. We are developing a conversational agent that will conduct a bedside dialogue with patients immediately prior to hospital discharge. This intervention will be compared with a standard-of-care control condition in a randomized clinical trial involving 750 patients actually being discharged from Boston Medical Center.

In a follow-on project funded by the Agency for Healthcare Research and Quality ("Virtual Patient Advocate to Reduce Adverse Drug Events"), we will also be developing an intervention that follows patients home from the hospital through to their first visit with their Primary Care Physician, using a PDA-based relational agent.