Two Strategies for Learning a Visually Guided Motor Task

Abstract

The principal question asked is whether in a visually-guided motor task, a subject tracking a known target employs a different strategy of movement to that used when tracking an unknown target. 22 subjects performed a series of 150 visual tracking tasks each 5 sec. long. The target-move-ment patterns used for the first 50 trials were all different, but for the remaining 100 trials they were identical. Subjects, however, were not informed of the repetition until the final 50 trials. When the task was made repetitive, even though the subjects were unaware of the repetition, learning occurred as evidenced by a progressive reduction in tracking error, although tracking lag remained above the mean reaction-time. Once subjects were aware of the repetition, tracking lags often reached zero or even negative values and tracking error dropped even further. It is argued that the former learning is confined to subconscious improvement in the intermittent response to visual inspection of tracking error, whereas the latter is achieved by adopting a truly predictive mode of tracking. Further experiments were devised to evaluate the role of visual information in movement control when using the predictive strategy. The main finding was that even when moving predictively, visual information was used to regulate motor output, largely to modify the timing of the predictive response to synchronize with the stimulus.