The role of action information feedback in the acquisition of simple motor skills.

The role of visual action information feedback (AIF) in the control and retention of movements has been studied using three types of movement; bar pressing with a large gain on the visual display, timed movements with a display which showed time elapsed, and untimed movements of considerable extent with a visual display of the same size. In all three tasks control subjects received terminal information feedback (TIF), either visually or verbally, and in the third task a movement to a stop was also used. The presence of a visual cue which was larger than the actual movement led to overestimation of the target pressure when the feedback was removed, even after extended practice. Subjects tended to believe that they had been inaccurate and that the bar had become stiffer when the feedback wets removed. Halving the size of the visual display had no effect on test performance of subjective error. Inaccurate test performance was also found in the lateral displacement task, even when the visual cue was not necessary for the acquisition of the task, and subjects' estimations of error indicated that they felt they had overestimated in test. In the timed movement task there were no differences between AIF and TIF trained subjects in either objective or subjective error. When the possibility of counting was removed by the addition of a shadowing task, AIF trained subjects tended to be less accurate after a delay than were TIF trained subjects. These results are interpreted as evidence of visual dominance of other sensory systems in a long term learning situation, provided vision is used to code the movement early in learning, and are discussed in terms of the part played by visual feedback in the formation of motor programmes.