Effect of robotic guidance on motor learning of a timing task

Abstract

While robotic guidance can improve ongoing performance of a motor task, several studies have found that it ultimately impairs motor learning. Here, we studied whether robotic guidance improved short and long term learning of a task for which timing is the critical performance determinant (a pinball-like hitting task). Training with guidance allowed subjects to learn the trained task, but had no benefit compared to training without guidance. In fact subjects who trained with guidance reacted as if the assistance provided on assisted trials was a perturbation rather than following its example, as they showed significant aftereffects that increased timing error on the trials following assisted trials. The results also suggest that training with robotic guidance slightly reduced generalization to untrained targets, perhaps because it reduced the range of errors that subjects explored during training.