Emotional valence and gender effects on willingness to interact with virtual agents in people with depressive symptoms

Major Depressive Disorder (MDD) affects millions of people worldwide, but barely half have access to effective treatments. To cope with this problem efforts have been made to develop Information and Communication Technologies (ICT)-based interventions, which allow access to care when in-person therapy is unavailable. Among others, approaches involving virtual agents seem to be better in promoting adherence to these kinds of interventions. Although some studies have identified features influencing healthy people perception and acceptance towards assistive technologies embodied in virtual agents, some characteristics of depressive emotional and social functioning might be associated with a different attitude toward them. This work investigates the role played by the agent’s gender and the emotional valence (positive-happy, negative-sad, and neutral) of its facial expression, in establishing a potential long-term relationship with a user, with and without depressive symptoms.