Requirements and Solution Approaches to Personality-Adaptive Conversational Agents in Mental Health Care

Abstract

Artificial intelligence (AI) technologies enable Conversational Agents (CAs) to perform highly complex tasks in a human-like manner and may help people cope with anxiety to improve their mental health and well-being. To support patients with their mental well-being in an authentic way, CAs need to be imbued with human-like behavior, such as personality. In this paper we cover an innovative form of CA, so-called Personality-Adaptive Conversational Agents (PACAs) that automatically infer users’ personality traits and adapt accordingly to their personality. We empirically investigate their benefits and caveats in mental health care. The results of our study show that PACAs can be beneficial for mental health support, but they also raise concerns about trust and privacy issues. We present a set of relevant requirements for designing PACAs and provide solution approaches that can be followed when designing and implementing PACAs for mental health care.