Many of the 780,000 people affected by stroke each year are left with slow, asymmetric walking patterns. The proposed project will evaluate the effectiveness of two competing motor learning approaches to restore symmetric gait for faster, more efficient, and safer walking.

Walking after stroke is characterized by reduced gait speed and the presence of interlimb spatiotemporal asymmetry. These step length and stance time asymmetries can be energy inefficient, challenge balance control, increase the risk of falls and injury, and limit functional mobility. Current rehabilitation to improve gait is based on one of two competing motor learning strategies: minimizing or augmenting symmetry errors during training. Conventional rehabilitation often involves walking on a treadmill while therapists attempt to minimize symmetry errors during training. Although this approach can successfully improve gait speed, it does not produce long-term changes in symmetry. Conversely, augmenting or amplifying symmetry errors has been produced by walking on a split belt treadmill with the belts set at different fixed speeds. While this approach produced an 'after-effect' resulting in step length symmetry for short periods of time, with some evidence of long term learning in people with stroke, it had no influence on stance time asymmetry. The investigators propose that patients need real-time proprioceptive feedback of symmetry errors so that they are actively engaged in the learning process. For this project, the investigators developed and validated a novel, responsive, 'closed loop' control system, using a split-belt instrumented treadmill that continuously adjusts the difference in belt speeds to be proportional to the patient's current asymmetry. Using this system, the investigators can either augment or minimize asymmetry on a step-by-step basis to determine which motor learning strategy produces the largest improvement in overground spatiotemporal symmetry.

Inclusion Criteria:

• ability to walk >10 m overground without physical assistance
• overground comfortable gait speed (CGS) < 1.0 m/s (using assistive devices and bracing below the knee as needed)
• able to walk independently on the treadmill at >80% CGS
• exhibits stance time and/or step length asymmetry during CGS

Exclusion Criteria:

• cerebellar lesion
• uncontrolled cardiorespiratory/metabolic disease (cardiac arrhythmia, uncontrolled hypertension or diabetes, orthostatic hypertension, chronic emphysema)or other neurological or orthopedic disorders that may affect gait training
• botulinum toxin to the lower limb in the past 6 months
• a history of balance deficits or unexplained falls not related to the stroke
• uncontrolled seizures
• concurrent physical therapy
• Mini-Mental Status Exam (MMSE) < 24
• communication impairments which could impede understanding of the purpose or procedures of the study or an inability to comply with experimental procedures