JMIR Publications

ABSTRACT

Background:

Background:

The COVID-19 pandemic accelerated the adoption of technology-mediated mental health services, yet questions remain about whether immersive digital platforms can match the therapeutic effectiveness of traditional face-to-face therapy. Virtual Reality (VR) offers unique affordances beyond conventional telehealth by providing embodied presence and shared virtual spaces, potentially addressing limitations of video-based teletherapy. However, empirical evidence directly comparing VR-mediated therapy with in-person sessions using both subjective and objective measures remains scarce.

Objective:

Objective:

This study aimed to compare therapeutic engagement, self-disclosure, and emotional arousal between VR-mediated and face-to-face counselling sessions using a mixed-methods approach combining self-report measures with continuous physiological monitoring.

Methods:

Methods:

We conducted a within-subjects experimental study with 30 adult participants (19 male, 11 female; mean age 32.5 years, SD 12.1) who each completed one VR-based and one face-to-face counselling session with licensed clinical psychologists. Sessions were counterbalanced and followed a semi-structured protocol. The VR condition used Oculus Quest 2 headsets with avatar-mediated interaction in a virtual counselling environment. We collected subjective data via validated instruments including the Session Evaluation Questionnaire (SEQ), Working Alliance Inventory-Short Form (WAI-SF), and custom engagement scales. Physiological data included continuous heart rate variability (HRV) measured via photoplethysmography and electrodermal activity (EDA) recorded throughout each session. Statistical analyses employed paired t-tests and Wilcoxon signed-rank tests for within-subject comparisons.

Results:

Results:

While participants initially rated face-to-face sessions as more appropriate (mean 7.63, SD 1.42 vs mean 6.80, SD 1.98; Z=2.541, P<.05) and reported feeling better immediately after in-person sessions (mean 7.00, SD 1.83 vs mean 6.03, SD 2.31; Z=2.585, P<.05), there were no significant differences in willingness to continue therapy (P>.05) or recommendation likelihood (P>.05) between modalities. Notably, 73% of participants reported greater self-disclosure in VR sessions, with qualitative data revealing that avatar-mediated interaction reduced social anxiety and facilitated openness. Physiological measures showed no significant differences in heart rate (mean 85.67, SD 12.85 vs mean 83.12, SD 12.45; P>.05) or skin conductance levels (mean 9.47, SD 3.67 vs mean 9.18, SD 3.54; P>.05) between conditions, indicating comparable emotional arousal. Therapist-rated alliance scores were equivalent across modalities (P>.05).

Conclusions:

Results:

While participants initially rated face-to-face sessions as more appropriate (mean 7.63, SD 1.42 vs mean 6.80, SD 1.98; Z=2.541, P<.05) and reported feeling better immediately after in-person sessions (mean 7.00, SD 1.83 vs mean 6.03, SD 2.31; Z=2.585, P<.05), there were no significant differences in willingness to continue therapy (P>.05) or recommendation likelihood (P>.05) between modalities. Notably, 73% of participants reported greater self-disclosure in VR sessions, with qualitative data revealing that avatar-mediated interaction reduced social anxiety and facilitated openness. Physiological measures showed no significant differences in heart rate (mean 85.67, SD 12.85 vs mean 83.12, SD 12.45; P>.05) or skin conductance levels (mean 9.47, SD 3.67 vs mean 9.18, SD 3.54; P>.05) between conditions, indicating comparable emotional arousal. Therapist-rated alliance scores were equivalent across modalities (P>.05).