Avatar-Mediated Telepsychotherapy in a Metaverse-Informed Virtual Environment: Qualitative Study of User Experiences and Design Requirements.

Background: Face-to-face psychotherapy remains the gold standard for mental health treatment; however, it faces accessibility barriers, including geographical constraints, costs, and stigma. Although videoconferencing psychotherapy addresses some limitations, challenges with engagement and therapeutic alliance persist. Avatar-mediated telepsychotherapy (AMT) in a metaverse-informed virtual environment has been proposed as a possible extension of digitally delivered mental health care, combining digital accessibility with enhanced presence through anonymous avatar-mediated communication. Objective: This exploratory qualitative study examined how participants experienced avatar-mediated psychotherapy within a virtual environment, with particular attention to perceived design features that may facilitate psychological engagement, self-disclosure, and sustained participation. Methods: We conducted a qualitative study using a minimum viable product platform among adults aged 19-70 years who self-reported psychological health management needs but had not been diagnosed with a major mental disorder, recruited from Korea University Anam Hospital and Korea University (n=60). The participants attended 4 weekly sessions (50 minutes each) using anonymous avatars reflecting real-time facial expressions and nonverbal cues. After completion, 30 participants (16 women and 14 men; 15 hospital employees and 15 university students) were purposively selected for semistructured interviews (30-50 minutes). Data were analyzed by 4 researchers using thematic analysis with ATLAS.ti software (Lumivero). Results: From 508 initial codes, 4 main themes emerged through iterative analysis. (1) Balancing anonymity and self-disclosure through avatars: anonymous avatars appeared to facilitate greater self-expression, with a sense of psychological safety throughout therapy sessions while maintaining emotional connection with therapists through synchronized nonverbal expressions. The synchronization of nonverbal expressions appeared to contribute to self-identification. (2) The holistic role of metaverse spaces in psychotherapy: pretherapy spaces were perceived as a psychological transitional area, alleviating tension and anticipatory anxiety. Therapy spaces required focused design that accommodated user preferences regarding the visibility of their own avatars. Posttherapy spaces were perceived as supporting self-reflection and anonymous community connection. (3) Consideration for meta-connection: participants described a need for cyclic feedback mechanisms that could help connect session experiences with everyday reflection and intended behavior. Metaverse settings demanded distinct therapist protocols to confirm therapists' attentiveness through avatar-mediated verbal and nonverbal communication to maintain therapeutic alliance. (4) Consideration for metaverse-informed psychotherapy space design: avatar customization may serve not only as a means of self-expression but also as a source of social connection or a sense of community. Technical stability, cross-platform accessibility, and user guidance were identified as key design considerations for developing psychologically engaging and technically reliable avatar-mediated psychotherapy environments. Conclusions: This study yields a set of qualitative design requirements for AMT, including anonymous avatars with behavioral realism rather than visual similarity, multistage spaces supporting the therapeutic journey, cyclic feedback mechanisms connecting virtual and real-world experiences, and integrated therapist competency development. These findings provide preliminary design implications for accessible and engaging AMT environments.