Effects of Cognition on Explicit Learning of a Novel Walking Task in Individuals Post-Stroke: Preliminary Results

• Matthew Carr, Human Physiology, University of Delaware

• Darcy Reisman, Physical Therapy, University of Delaware

Introduction and Purpose

Over 795,000 people in the United States alone experience a stroke each year, making it a leading cause of long term disability. Stroke survivors identify the ability to walk as one of the most important goals for rehabilitation. Current practice in physical therapy utilizes explicit learning techniques such as verbal cueing and visual feedback to help patients learn to walk again. The ability to use these techniques, specifically visual feedback, has been shown to be strongly related to one’s cognition. We therefore hypothesize that the ability to use visual feedback in an explicit learning paradigm can be explained partly by cognition, above and beyond the effects of age.

Methodology

Feedback and cognitive data for 27 participants with chronic (>6 months) stroke was collected in two separate sessions. Feedback data collections involved walking on a split-belt treadmill and cognitive collections included tests from the Cognitive Battery of the NIH Toolbox for Assessment of Neurological and Behavioral Function (NIHTB-CB) and the Wechsler Memory Scale-IV (WMS-IV). We examined the Fluid Cognition Composite Score (FCCS) and the Flanker Inhibitory Control and Attention Test from the NIHTB-CB and the Designs I from the WMS-IV. The explicit learning component was calculated using the subjects’ step length (SL) asymmetry defined as (LongerSL – ShorterSL) / ( LongerSL+ShorterSL) and taking the difference of the means of asymmetry of the last five strides with visual feedback minus the first five strides without visual feedback. Separate hierarchical regression models were used with the explicit learning component as the dependent variable and age and one cognitive score (Flanker Uncorrected standard score, FCCS uncorrected standard score, or Designs I total score) as the predictor variables.

Results

After accounting for age, none of the cognitive tests accounted for significant variability in the explicit learning component (ΔR2 =0.013 , P=0.570 (Flanker); ΔR2 =0.001 , P=0.902 (FCCS); ΔR2 =0.118 , P=0.084 (Designs I)).

Conclusion

None of the cognitive tests accounted for significant variability in the ability to use the visual feedback to learn a new walking pattern in these participants with chronic stroke. However these preliminary results suggest that visuospatial memory, as represented by the Designs I score, may be important for explicit locomotor learning in people with chronic stroke.