A Therapeutic Evaluation of a Creative Reaching Game in Immersive Virtual Reality

Baron, Lauren
Journal Title
Journal ISSN
Volume Title
University of Delaware
Virtual Reality (VR) has several applications beyond entertainment, e.g., architectural planning, surgical procedure assistance, and physical therapy. These implications are important and require the most accuracy for users, especially when their healthcare is involved. Users can lose accuracy immersed in a virtual environment (VE) from inaccurate depth perception. Often, people underestimate distant objects and overestimate close objects in VR, which concerns software developers and healthcare providers. VR Therapy systems also rely on information from VR hand controllers, which do not fully capture the movement from the rest of the limb. While VR games have shown much potential for rehabilitation, research on creative virtual therapy is still growing. Considering many possibilities for therapeutic interventions in VR, my goal is to create activities with an appropriate balance between the intensity level of therapy intervention with enjoyment and entertainment. I will also capture the limb's movement both from the arm and hands with a noninvasive elbow sleeve sensor. I propose a creative line art drawing game in an immersive VR environment as a tool for both upper extremity therapy and vision therapy using enjoyable multi-dimensional reaching tasks. To examine the validity of the proposed virtual therapy system, I conducted two preliminary human-subjects experiments: a mixed design varying the drawing content (Easy vs. Hard; a between-subjects factor) and the user’s position (Seated vs. Standing; a within-subjects factor) on 16 non-clinical participants; and a withinsubjects design varying the drawing content dimension (2D vs. 3D) and the content’s orientation (Vertical vs. Horizontal) on 12 non-clinical participants. My results of the first experiment (SUI 2021) show that the change of drawing content objectively influenced participants’ drawing performance, e.g., the completion time and the number of mistakes, while they did not feel the difference in the difficulty level between the contents subjectively. Interestingly, participants reported more enjoyment from drawing the Hard Chicken content than the Easy Fish content and more substantial body stretches in the Seated setting than the Standing setting. The results of the second experiment (submitted to ACM/IEEE CHASE 2023) show that for all levels, there was no significant difference between subjective easiness, comfortability, and enjoyment and between objective measures for task completion time and the number of mistakes. This finding suggests that all versions are at the same therapeutic intensity level, with no model being more prone to longer time or more mistakes and are all usable/feasible. This leads to the customization of therapy to the user with any of these configurations and orientations while keeping the same level of intensity; for example, if a patient has restricted lower limb mobility and requires to be seated, they can use the horizontal orientation interchangeably. However, there was significance with elbow resistance change, which shows how data collection from just the hands and not throughout the arm is insufficient for VR rehabilitation, particularly for the upper extremities. There is a need to improve depth perception, visual cues, and reaching capabilities in VEs, especially for 3D objects.
Virtual reality , Therapy , Rehabilitation