Theses Doctoral

The Integration of Principles of Motor Learning to Reduce Gait Asymmetry Using a Novel Robotic Device in Individuals Chronically Post-Stroke

Bishop, Lauri

Unilateral deficits resulting from stroke manifest as reduced velocity, decreased cadence and asymmetries in temporal, spatial and force parameters during ambulation. Gait asymmetries and compensatory strategies employed during gait result in a higher mechanical energy cost that limits activity and community participation. Despite conventional rehabilitation efforts, individuals often remain with chronic gait deficits after stroke. Robotic-based therapies have been developed as an alternative to conventional rehabilitation. These therapies offer the means to provide task-specific training at an intensity greater than that of conventional approaches; however, to date outcomes have been similar to that of conventional training. One factor potentially contributing to the limited efficacy of robotic training is the active-assist control strategy that is often employed. This type of training strategy reduces the users’ engagement in the learning process and limits skilled learning.
The tethered pelvic assist device (TPAD) is a robotic device that employs actuated tethers at the pelvis to guide the user along a pre-set movement trajectory. While other robotic devices restrict movement to a fixed trajectory, the TPAD promotes shifting weight onto the paretic limb, but permits users to freely move the limb to navigate spatiotemporal aspects of training independently. This allows individuals to participate in the problem-solving process required for motor learning to occur, facilitating a more active role in the motor task itself, and thus promoting learning.
Earlier work utilized the TPAD to reduce gait asymmetry in a population of individuals in the chronic phase after stroke in a single training session (Bishop et al., 2015; Vashista, 2015). Results demonstrated an increase in propulsive forces of the affected limb as a result of the intervention, but these gains did not transfer to overground gait. A follow up study explored the feasibility and efficacy of two different training strategies using the TPAD (Bishop et al., 2017). Both training strategies proved feasible and similarly efficacious. The current work examines the feasibility and preliminary efficacy of a five-day intervention using the TPAD with faded visual feedback and a short bout of task-specific overground training to reduce gait asymmetry in a population of individuals at least six months after stroke.
Participants underwent a series of three Pre Test assessments within a one-week interval prior to initiating the intervention. Training occurred over five consecutive days, with a Post Test assessment administered on conclusion of Day 5 of training. A one-week Follow Up assessment was also recorded. Results demonstrated this intervention coupling TPAD training with additional tenets of motor learning including visual feedback and salient task-specific overground training was feasible in terms of safety, tolerance and adherence. Further, while participant’s load asymmetry was not significantly reduced on the treadmill from Baseline to Post Training (p >0.05), there was a significant improvement in stance symmetry during overground gait (F = 8.498, p = 0.002). These results suggest that the integration of motor learning tenets with robotic TPAD training was useful in facilitating gains to overground walking. Implications to the broader scope of robotic training suggest that creating an environment in which the user plays a more active role is useful at maximizing effects of robotic training. Future work should include comparison groups (TPAD treadmill training, overground training, and combined TPAD and overground training) with a more robust sample size for a longer duration of training to parse out contributing factors to overground gains. Future work should also consider a longer training and follow up interval in an effort to determine whether individuals are able to maintain improvements longer than the immediate post training period.


  • thumnail for Bishop_columbia_0054D_14711.pdf Bishop_columbia_0054D_14711.pdf application/pdf 8.68 MB Download File

More About This Work

Academic Units
Thesis Advisors
Quinn, Lori
Ph.D., Columbia University
Published Here
June 10, 2018