JMIR Publications

ABSTRACT

Background:

The role of avatars within virtual reality is increasingly important. Avatars have been introduced with increasing quality mainly due to historical technological limitations with respect to computer graphics and hardware. The VR community may assume that with the development of higher quality and therefore more 'lifelike avatars', this implies an advance in avatar performance, yielding ever better therapeutic outcomes in clinical applications, such as self-compassion therapy. But is that true? If not then what can be done to optimize human-avatar interactions for therapeutic benefit?

Objective:

Considering the use of avatars in VR for self-compassion therapy, we conduct a study which focuses on improvements to user experience via avoidance of 'excessive realism' with the introduction of low-fidelity avatars, increased individualization and enhanced embodiment.

Methods:

This study has acquired the original VR environment of Halim et al.’s [1] such that work can be done to continue an investigation into design changes and the effect on user experience and the therapeutic outcomes. This study used the User Experience Questionnaire (UEQ), Self-Compassion Scale (SCS), and Patient Health Questionnaire depression scale (PHQ-8) to measure experimental changes.

Results:

The results show an improvement to self-compassion outcomes following introduction of greater avatar customisation, SCS p = 0.03. An increase in user experience and therapeutic outcomes for depression was observed after introducing mirror techniques to enhance body ownership over a personalized aesthetically low-fidelity avatar; PHQ-8 p = 0.01 and UEQ all categories p < 0.04 with exception ‘Efficiency’ category. Also, with respect to user experience, significant improvements in areas pertaining to feelings of dependability suggests avoidance of Uncanny Valley phenomenon.

Conclusions:

Our work explores important questions to strengthen the community's knowledge of design decisions surrounding avatars that lead to optimal therapeutic outcomes when engaging VR therapy. This paper has demonstrated the harmlessness of low-fidelity avatars in VR therapy, reinforced the importance of individualizing aspects of virtual reality therapy and demonstrated the importance of virtual mirrors with respect to user experience in such VR therapy applications. Clinical Trial: This study is not RCT.