Perceptual Limits for a Robotic Rehabilitation Environment Using Visual Feedback Distortion

Abstract

Imperceptible visual distortion, in the form of a disguised progression of performance goals, may be a helpful addition to rehabilitation after stroke and other brain injuries. This paper describes work that has been done to lay the groundwork for testing this hypothesis. We have constructed and validated an experimental environment that provides controllable visual distortion and allows precise force and position measurements. To estimate the amount of visual distortion that should be imperceptible, we measured the limits for force and distance/position perception in our rehabilitation environment for young and elderly unimpaired subjects and for a single traumatic brain injury (TBI) patient. We found the Just Noticeable Difference (JND) for produced force to be 19.7% (0.296 N) and the JND for movement distance/finger position to be 13.0% (3.99 mm) for young subjects (ages 18-35). For elderly subjects (ages 61-80), the JND for force was measured to be 31.0% (0.619 N) and the JND for distance/position was 16.1% (5.01 mm). JNDs of 46.0% (0.920 N) and 45.0% (14.8 mm) were found for the motor-impaired individual. In addition, a subject's rating of effort was found to be profoundly influenced by visual feedback concerning the force magnitude. Even when this feedback was distorted, it accounted for 99% of the variance of the effort rating. These results indicate that substantial visual distortions should be imperceptible to the subject, and that visual feedback can be used to influence the subject's perceived experience in our robotic environment. This means that we should be able to use imperceptible visual distortion to alter a patient's perception of therapeutic exercise in a robotic environment.