Motor performance with a simulated artificial limb.

The present study was conducted to examine performance differences on a reaching and grasping task related to an activity of daily living. This involved either the anatomical limb or a simulated artificial limb. College-aged volunteers (2 men and 4 women), one of whom was left-handed, performed the reaching and grasping task. The apparatus, placed on a table before the seated participant, was a square wooden board which contained a starting key and holes for the insertion and removal of a small Fiberglas dowel. At the beginning of the trial the participant depressed the start key, reached forward and grasped the dowel, and then returned the dowel to a finishing hole located directly in front of the start key. The results of 2 (side) x 2 (type of limb) repeated-measures multivariate analyses of variance on the mean and standard deviation of the movement times showed a significant main effect for type of limb (Wilks lambda 3,3 = .047 and .079, respectively, p < .05). Analyses of variance on mean total transport time, extension time, flexion time, and their standard deviations showed that times were slower and less consistent with the prosthesis for all measures. These results and those of later research should be focused on the development of training principles for both therapists and individuals with an amputation. In addition, the simulated prosthesis is an excellent experimental model for basic and clinical research in the control and acquisition of coordinated movement.