Attentional focus of feedback for improving performance of reach-to-grasp after stroke : a randomised crossover study

Objective To investigate whether feedback inducing an external focus (EF) of attention (about movement effects) was more effective for retraining reach-to-grasp after stroke compared with feedback inducing an internal focus (IF) of attention (about body movement). It was predicted that inducing an EF of attention would be more beneficial to motor performance. Design Crossover trial where participants were assigned at random to two feedback order groups: IF followed by EF or EF followed by IF. Setting Research laboratory. Participants Forty-two people with upper limb impairment after stroke. Intervention Participants performed three reaching tasks: (A) reaching to grasp a jar; (B) placing a jar forwards on to a table; and (C) placing a jar on to a shelf. Ninety-six reaches were performed in total over one training session. Main outcome measures Kinematic measures were collected using motion analysis. Primary outcome measures were movement duration, peak velocity of the wrist, size of peak aperture and peak elbow extension. Results Feedback inducing an EF of attention produced shorter movement durations {first feedback order group: IF mean 2.53 seconds [standard deviation (SD) 1.85]; EF mean 2.12 seconds (SD 1.63), mean difference 0.41 seconds; 95% confidence interval -0.68 to 1.5; P = 0.008}, an increased percentage time to peak deceleration (P = 0.01) when performing Task B, and an increased percentage time to peak velocity (P = 0.039) when performing Task A compared with feedback inducing an IF of attention. However, an order effect was present whereby performance was improved if an EF of attention was preceded by an IF of attention. Conclusions Feedback inducing an EF of attention may be of some benefit for improving motor performance of reaching in people with stroke in the short term; however, these results should be interpreted with caution. Further research using a randomised design is recommended to enable effects on motor learning to be assessed