Engagement and Disclosures in LLM-Powered Cognitive Behavioral Therapy Exercises: A Factorial Design Comparing the Influence of a Robot vs. Chatbot Over Time

TL;DR

This study compares how embodiment in robots versus chatbots affects user engagement and disclosures over time during CBT exercises, revealing that physical robots enhance engagement and intimacy as time progresses, unlike chatbots.

Contribution

It provides empirical evidence on the differential impact of embodiment in therapeutic LLM applications over time, highlighting the benefits of robots in sustained engagement.

Findings

Engagement increased over time with robots but decreased with chatbots.

Intimacy of disclosures grew in robot condition, declined in chatbot condition.

Embodiment influences long-term user engagement and emotional sharing.

Abstract

Many researchers are working to address the worldwide mental health crisis by developing therapeutic technologies that increase the accessibility of care, including leveraging large language model (LLM) capabilities in chatbots and socially assistive robots (SARs) used for therapeutic applications. Yet, the effects of these technologies over time remain unexplored. In this study, we use a factorial design to assess the impact of embodiment and time spent engaging in therapeutic exercises on participant disclosures. We assessed transcripts gathered from a two-week study in which 26 university student participants completed daily interactive Cognitive Behavioral Therapy (CBT) exercises in their residences using either an LLM-powered SAR or a disembodied chatbot. We evaluated the levels of active engagement and high intimacy of their disclosures (opinions, judgments, and emotions) during each session and over time. Our findings show significant interactions between time and embodiment for both outcome measures: participant engagement and intimacy increased over time in the physical robot condition, while both measures decreased in the chatbot condition.