Depression is one of the leading causes of disability in the world, effecting more than 350 million people. Unlike other treatable epidemics that affect our society. However, less than half of those suffering receive treatment. To address this issue, several automated computer-based cognitive behavioral therapy systems have been developed that enable patients to receive therapeutic care at home. While effective, these automated interventions are still inferior to traditional face-to-face therapy sessions with human counselors, due to poor retention and lack of interactivity and adaptability.
We are exploring the use of relational agents to both increase retention and dynamically adapt to user affective state in the context of a cognitive-behavioral therapy-based intervention for individuals with depressive symptoms.