INVITED PAPER Visual-Feedback Distortion in a Robotic Rehabilitation Environment In a finger-motion rehabilitation trial, where a patient presses against robot resistance, progress can be aided by visual feedback that distorts the force the patient believes is being exerted. By Bambi R. Brewer , Roberta L. Klatzky, and Yoky Matsuoka ABSTRACT | We create a robotic rehabilitation environment that distorts the visual feedback of a movement representation to restore lost movements. The use of visual-feedback distor- tion produces a perceptual gap between the perceived (visual) and actual somatosensory experiences, in which movements can be manipulated without patients’ knowledge. Previously, we reported the smallest amount of distortion that is imper- ceptible by measuring just noticeable difference (JND), the distortion size that is tolerable without detection, and the in- variance of perceived physical effort under visual-feedback distortion. In this paper, we report the performance changes under visual-feedback distortion in a robotic rehabilitation environment. By interleaving trials with no visual feedback, we showed that the internal somatosensory representation of the movement goal changed along with visual distortion for un- impaired and motor-impaired subjects. In addition, a gradual change of visual-feedback distortion within one experiment allowed movement changes significantly larger than the JND. These changes were robust under sudden lack of visual feedback and even when subjects were informed of possible distortion ahead of time. Finally, preliminary results of a six- week rehabilitation paradigm are reported. Improvements in the patient’s hand condition were found, particularly in the range of motion of the index and middle fingers. KEYWORDS | Feedback distortion; hand; rehabilitation robotics; stroke; traumatic brain injury I. INTRODUCTION Almost 1 million strokes and traumatic brain injuries occur each year in the United States alone [1], [2]. Most sur- vivors require rehabilitation to address resulting motor deficits; for instance, 50% of stroke survivors exhibit some hemiparesis six months poststroke [1]. Currently, thera- pists use techniques such as passive movement of joints and task practice to restore function. They also teach compensatory techniques to increase functional indepen- dence. Because robots can repeat movements accurately without fatigue, robot-assisted therapy has been proposed to enable new types of rehabilitation exercise and to increase the amount of therapy available to each patient [3], [4]. It has been shown that robot-assisted rehabilita- tion improves patients’ mobility and strength to the point where it is at least equal to that which is achieved by human-assisted therapy [5]. Beyond robots’ ability to repeat movements without fatigue, a robotic environment provides an opportunity to manipulate the feedback given to the user. This manipula- tion could create an artificial environment that is ben- eficial in rehabilitation. For example, Patton et al. [6] have shown corrections on impaired arm trajectories when the error in somatosensory and visual feedback was enhanced by force perturbation. While these changes are known to wash out for unimpaired subjects, Patton et al. have shown that the learned movement under force perturbation does not wash out as easily for stroke patients. In our robotic paradigm, we distort the visual feed- back with respect to the real somatosensory feedback. Manuscript received July 26, 2005; revised February 17, 2006. This work was supported in part by the National Science Foundation (NSF) Presidential Early Career Award for Scientists and Engineers (PECASE) 0238204 and the National Institute on Disability and Rehabilitation Research (NIDRR) under Project H133A020502. B. R. Brewer is with the Robotics Institute, Pittsburgh, PA 15213 USA (e-mail: bambi@andrew.cmu.edu). R. L. Klatzky is with the Psychology Department, Carnegie Mellon University, Pittsburgh, PA 15213 USA (e-mail: klatzky@cmu.edu). Y. Matsuoka is with the Robotics Institute, Pittsburgh, PA 15213 USA and also with the Mechanical Engineering Department, Carnegie Mellon University, Pittsburgh, PA 15213 USA (e-mail: yoky@cs.cmu.edu). Digital Object Identifier: 10.1109/JPROC.2006.880715 Vol. 94, No. 9, September 2006 | Proceedings of the IEEE 1739 0018-9219/$20.00 Ó2006 IEEE