Effects of visual feedback distortion on gait adaptation Seung-Jae Kim Biomedical Engineering California Baptist University Riverside, CA 92504 1-951-552-8635 sjkim@calbaptist.edu Joon-Ho Shin Dept. of Stroke Rehabilitation National Rehabilitation Center Seoul, Korea 82-2-901-1884 asfreelyas@gmail.com ABSTRACT Gait rehabilitation following neurological disorders often utilizes the correction of stepping movements, which entails adaptive changes through practice that eventually leads to permanent gait adaptation. To enable gait training to be more effective, providing visual feedback on subjects’ movements during rehabilitation training can help engage subjects’ participation beyond their specified task. Our previous experiments [1-2], which used the implicit and explicit distortion of visual feedback, showed the effect of visual feedback distortion on changes in gait symmetry and suggested that such an effect involves non-volitional resources. We have further explored the process of gait adaptation induced from visual feedback distortion with the following question: How would the changes in gait symmetry induced by implicit visual feedback distortion, versus conscious control of gait symmetry, modify adaptation? Our results showed that subjects trained with visual feedback distortion retained aftereffects longer than with conscious correction of stepping. This study suggests that a therapeutic program employing visual feedback distortion paradigm could provide an effective way to help patients correct gait patterns, thereby improving the outcome of rehabilitation. Keywords Gait rehabilitation, Visual feedback distortion, Modulation of locomotion, Step length symmetry, Gait adaptation. 1. INTRODUCTION Gait rehabilitation following neurological disorders often utilizes correction of stepping movements. Such rehabilitation approaches require adaptive changes through practice that eventually leads to permanent motor adaptation of human locomotion [3]. Therefore, gait rehabilitation needs to emphasize techniques that drive recalibrations of motor commands for locomotion rather than just evoke a reactive type of adjustment during training [4]. Providing visual feedback on subject’s movements during gait rehabilitation may enhance the process of such gait adaptation by engaging subjects’ participation beyond their voluntary effort [5-6]. We have previously demonstrated that an implicit and also an explicit distortion of visual feedback regarding step length entail unintentional modulations in gait spatial pattern to some degree [1-2]. In this paradigm, the visual feedback of step length symmetry was displayed on a computer screen and was also distorted so that subjects perceived their step length as being asymmetric during treadmill walking. We found that a gradual distortion of visual feedback systematically modulated gait step length away from symmetry. Since walking is normally an implicit process, it is possible that gait adaptation is driven using more implicit and automatic processes than voluntary control of movement. Thus, the aim of this study was to explore the process of gait adaptation resulted from visual feedback distortion. Here we tested whether the changes in gait pattern induced by implicit visual feedback distortion, versus conscious correction of walking, modifies adaptation. Two different conditions were tested: 1) implicit visual distortion condition was given no specific instruction with on-line visual distortion feedback, and 2) conscious correction condition was given voluntary control of gait symmetry with on-line visual-guided feedback. In treadmill walking trials, the visual feedback display was removed during adaptation while subjects continued walking. Then, retention of aftereffects (de-adaptation) was assessed in all subjects. Our study showed that subjects trained with visual feedback distortion retained aftereffects longer than the conscious correction group, suggesting a potential effect of visual feedback distortion on recalibrations of motor commands for locomotion. In the context of gait rehabilitation, a therapeutic program involving visual feedback distortion, in the context of gait rehabilitation can provide an effective way to help subjects correct gait patterns, thereby improving the outcome of rehabilitation. 2. MATERIALS & METHODS Seven healthy volunteers participated in this study. All subjects gave informed written consent before participating and were familiar with walking on a treadmill. Subjects were first habituated to walk on the treadmill and determined a comfortable walking speed (range 1.8~2.2 mile/hour) after 10 to 15 minutes of habituation. Subjects continued to walk comfortably on the treadmill. The on-line visual feedback showed step length information represented by bar graphs provided via a computer screen in front of a treadmill (Fig. 1). A computer screen (24” LCD monitor) was placed in front of the treadmill to display visual feedback information (the distance between subjects and the screen was about 4 feet). The step length is defined as the distance between each foot when heel-strike occurs for one leg. The visual feedback consisted of bar heights representing the instantaneous distance between each foot, so that the bars were gradually increased during the swing phase (two feet were getting further apart). The maximum height of the bar occurred when heel strike occurred on the corresponding side, and the maximally displayed bar trace stayed on until the following swing phase began. When distorting the visual feedback, we increased the range of the bar for only one side (the right side). The distortion increments used in this study was 2%. For example, 2% of distortion changed the scaling factor