Background and Purpose—Training of the lower extremity (LE) using a robot coupled with virtual environments has shown to transfer to improved overground locomotion. The purpose of this study was to determine whether the transfer of training of LE movements to locomotion was greater using a virtual environment coupled with a robot or with the robot alone.

Methods—A single, blind, randomized clinical trial was conducted. Eighteen individuals poststroke participated in a 4-week training protocol. One group trained with the robot virtual reality (VR) system and the other group trained with the robot alone. Outcome measures were temporal features of gait measured in a laboratory setting and the community.

Results—Greater changes in velocity and distance walked were demonstrated for the group trained with the robotic device coupled with the VR than training with the robot alone. Similarly, significantly greater improvements in the distance walked and number of steps taken in the community were measured for the group that trained with robot coupled with the VR. These differences were maintained at 3 months' follow-up.

Conclusions—The study is the first to demonstrate that LE training of individuals with chronic hemiparesis using a robotic device coupled with VR improved walking ability in the laboratory and the community better than robot training alone.