Virtual reality training for stroke rehabilitation

Abstract

Applications and evaluations of an advanced form of computer technology - virtual reality (VR), have begun to appear in studies on the assessment, treatment and functional outcomes of persons with cognitive impairments (Thomas et al, 1996; Rizzo et al, 2000; Lengenfelder et al., 2002). VR can create the illusion that a person is in, and interacting with, an artificial world (Burdea, Richard & Coffiet, 1996; Popescu et al., 2002). VR can emulate real-life situations and may enhance the learning and transfer of skills to everyday circumstances. There have been preliminary findings that there is a clear, positive transfer effect from virtual and real training, and suggestions that the elements of cognitive strategy and cognitive loads of the training are broadly equivalent (Rose et al, 2000). The present study evaluated the effectiveness of a 2D virtual reality (2DVR) programme to train people with stroke on how to access and use the station facilities of Mass Transit Railway (MTR) - the most popular public transport means in Hong Kong. A flat-screen 2DVR based street survival skills training program and a corresponding psycho-educational programme with video-modeling were developed for the present study. The training contents included road crossing, use of MTR and money management. Effectiveness of the 2DVR and the psycho-educational programmes were evaluated and compared through a randomized control group pre-test and post-test research design. Eleven subjects participated in a pilot study to verify the usability and instructional strategies of the newly development 2DVR programme. Pilot training results and feedback from participants were obtained to design the main study. In the main study, 58 subjects were randomly assigned into 3 treatment groups. Twenty subjects received a 10-session street survival skills training using the 2DVR strategy. Another 16 subjects received the same skills training using video-based psycho-educational programme like demonstration, role-play, and immediate feedback with verbal reinforcement. Another 22 subjects formed the control group who received no MTR skills training during the study period. Assessed by a behavioural checklist on MTR skills and a newly validated MTR self-efficacy scale, subjects of both training groups showed significant improvement in their knowledge, skills and self-efficacy in using the MTR (P<0.01). While MTR skills and MTR self-efficacy of the control group remained stable over a 4-week interval. The results evidenced both training programmes were effective in training the stroke patients in MTR skills. Data analyses showed the 2 training programmes affected differential improvements in MTR skills and MTR self-efficacy. For example, the 2DVR group showed significant better improvement in way finding skills such as identifying the MTR entry, exit and the gate (P< 0.05). The subjects of the psycho-educational programme showed better improvement in MTR self-efficacy than the 2DVR group. They showed statistically significant better improvement in 4 road-crossing self-efficacy items (P< 0.05). Moreover, MTR skill training outcomes of the subjects who participated in both groups were significantly correlated with their pre-training self-efficacy. It might show that self-efficacy is an important psychological construct to mediate successful training outcome. Additional studies are recommended to identify the most effective training procedures for maintaining these skills and the best transfer ratio in the training of VR-based community living skills.