What People Share With a Robot When Feeling Lonely and Stressed and How It Helps Over Time

TL;DR

The paper addresses loneliness and perceived stress among young adults by evaluating a five-session, LLM-powered QTrobot intervention designed for cognitive reappraisal. Using linear mixed-effects models, it shows significant reductions in loneliness ($$\beta=-0.15, p<0.001$$) and perceived stress ($$\beta=-0.06, p<0.001$$) over time. It also analyzes 560 user disclosures with 384-dimensional embeddings, identifying six themes via $K$-means clustering and GPT-4o-mini labeling, and finds that socially oriented disclosures relate to higher loneliness and stress while introspective/academic topics relate to lower distress. The findings advance understanding of how wellbeing and disclosure content dynamically interact in human–robot interaction, informing the design of socio-emotionally aware, adaptive robots for emotional support.

Abstract

Loneliness and stress are prevalent among young adults and are linked to significant psychological and health-related consequences. Social robots may offer a promising avenue for emotional support, especially when considering the ongoing advancements in conversational AI. This study investigates how repeated interactions with a social robot influence feelings of loneliness and perceived stress, and how such feelings are reflected in the themes of user disclosures towards the robot. Participants engaged in a five-session robot-led intervention, where a large language model powered QTrobot facilitated structured conversations designed to support cognitive reappraisal. Results from linear mixed-effects models show significant reductions in both loneliness and perceived stress over time. Additionally, semantic clustering of 560 user disclosures towards the robot revealed six distinct conversational themes. Results from a Kruskal-Wallis H-test demonstrate that participants reporting higher loneliness and stress more frequently engaged in socially focused disclosures, such as friendship and connection, whereas lower distress was associated with introspective and goal-oriented themes (e.g., academic ambitions). By exploring both how the intervention affects well-being, as well as how well-being shapes the content of robot-directed conversations, we aim to capture the dynamic nature of emotional support in huma-robot interaction.

Figures (2)

• Figure 1: The deployment settings. Image from Laban2025AReappraisal.
• Figure 2: From left to right: (1) Mean loneliness score by session. (2) Mean perceived stress scores by session.