What do users think about Virtual Reality relaxation applications? A mixed methods study of online user reviews using natural language processing

Abstract The advent of affordable Virtual Reality (VR) technology has spurred consumer and commercial interest in VR relaxation applications, which has quickly grown into a popular non-gaming genre on digital marketplaces. While laboratory studies have demonstrated efficacy of VR relaxation for mental health purposes, little is known about how users experience this type of intervention and no study has examined the reception of consumer versions among regular users in everyday life. Studying published user reviews offers a unique window into naturalistic user experiences that complements traditional qualitative methods by circumventing the sampling bias of interview studies, and allowing analyses on full samples, unconstrained by coding resources. Using an innovative, semi-automated Natural Language Processing technique, the current study analyzed 1379 published reviews (including star ratings) of 30 different VR relaxation applications available for the Oculus Go and Gear VR. The uncovered topic structure and sentiment analysis thereof suggests that users have an overall positive view of VR relaxation applications, describing them as successful in inducing immersion and relaxation, and having appreciated gamification elements. However, perceived quality varied substantially between applications that explained more variance in star ratings than specific features. Critical issues raised were both technical (e.g. “overheating”) in nature and related to specific design elements and use. Implications for the design of consumer VR applications and future research are discussed.