Abstract
BACKGROUND: Autism Spectrum Disorder (ASD) is characterized by social deficits and restricted, repetitive behaviors, with fewer than 10% achieving independent adulthood. Immersive virtual reality (IVR) provides a novel training approach through interactive and realistic environments. This study developed an IVR system to enhance adaptive skills in children and adolescents with high-functioning ASD. METHODS: Thirty-three individuals with high-functioning ASD (ages 8-18) were enrolled based on clinical diagnoses confirmed by Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) and Autism Diagnostic Interview-Revised (ADI-R), with an intelligence quotient ≥ 80. The study employed a single-arm, within-subject pre-post design. Participants received weekly one-hour IVR training sessions, requiring 6-10 sessions to complete 36 tasks twice. Primary outcomes included changes in IVR task scores and completion times, while secondary outcomes assessed parent-reported questionnaires - Adaptive Behavior Assessment System-Second Edition (ABAS-II), Autism Behavior Checklist (ABC), and Behavior Rating Inventory of Executive Function (BRIEF) - neuropsychological tests (Go/No-Go, 0-back, 1-back, and emotional face recognition tasks), and semi-structured interviews. Usability was evaluated via self-reported comfort levels, willingness to continue and device-related questions. Generalized Estimating Equation models analyzed changes across all measures. RESULTS: The IVR training demonstrated high usability, with an 87.9% completion rate and no severe adverse effects. Some participants reported mild discomforts, including dizziness (28.6%) and fatigue (25.0%), as well as device-related issues such as unsteady walking (34.5%) and headset discomfort (31.0%). However, comfort levels increased over time (adjusted P = 0.026), indicating gradual adaptation to the system. IVR task scores improved by 5.5% (adjusted P = 0.034), and completion times decreased by 29.59% (adjusted P < 0.001). Parent-reported measures showed a 43.22% reduction in ABC Relating subscale scores (adjusted P = 0.006) and moderate reductions in BRIEF Behavioral Regulation and Metacognition indices (adjusted P = 0.020, 0.019). Reaction times for the 1-back and emotional face recognition tasks decreased by 14.81% and 14.14% respectively (adjusted P = 0.012, 0.014). Qualitative feedback indicated improvements in social interaction, emotional regulation, repetitive behaviors, attention, and daily living skills. CONCLUSION: This pilot study supports IVR training as a feasible and potentially effective training method for improving adaptive skills in children and adolescents with high functioning ASD.