Nonlinear error augmentation applied to upper limb movements enhances motor learning.

We investigated adaptation to a nonlinear, power-based error-augmentation feedback to assess the hypothesis that it can promote effective motor learning. Two groups of healthy adults (experimental and control groups) participated in a study involving a planar reaching task. Participants' hand was occluded from view. The hand held the puck of a graphics tablet, which tracked its movement, represented by a round cursor moving in a display placed in front of the participant. Canonical Familiarization, Baseline, Adaptation, and Washout phases characterize the experiment. In the control group, all four phases involved a 1:1 mapping between hand and cursor position, whereas, in the experimental group, real trajectories were distorted by a nonlinear function according to a power-based error-augmentation approach, during the Adaptation phase. The experimental group showed learning increase and average error reduction. The proposed approach extends the error augmentation techniques proposed so far, and may be useful for inducing motor learning in teleoperation, sports, and rehabilitation contexts.