Learning to Reach in a Mirror Reversed Environment

Abstract

Goal-directed reaching movements in a novel visual environment require humans to either learn new or adjust pre-existing visuomotor relationships. To study this in the laboratory, we introduced two visuomotor distortions: 1) a mirror reversed (MR) visuomotor distortion, in which cursor feedback is mirrored across the body midline (i.e., the y-axis) and 2) a visuomotor rotation (VR) distortion, in which the cursor trajectory is rotated relative to actual hand motion. Research thus far has focused on learning to reach with a VR distortion but less is known about learning to reach with an MR distortion. Specifically, previous research has not examined learning to reach with a small (20°) MR distortion. The primary goal of the current research was to determine implicit (i.e., unconscious) and explicit (i.e., conscious strategies) contributions to learning to reach with an MR distortion. The implicit and explicit contributions to learning to reach with an MR were then compared to those engaged when learning to reach with a small VR distortion, to establish similarities or differences in learning mechanisms engaged across MR and VR paradigms. Results revealed that implicit processes were not engaged when participants learned to reach with an MR distortion (Chapter 1). Instead, learning was driven by explicit processes, as evidenced by both perceptual and motor awareness (Chapter 2). Learning to reach with the MR distortion generalized to novel targets (Chapter 2) and led to faster re-learning when the same MR distortion was re-introduced (Chapter 3). In contrast, we found that learning to reach with a VR distortion did not engage explicit processes but rather was driven by implicit processes. Learning to reach with a VR distortion also did not generalize to novel targets. However, learning to reach with a VR distortion was influenced by learning to reach with an MR distortion beforehand. Specifically, early reaches with a VR distortion were more variable and took longer to initiate. Together, these findings demonstrate that learning to reach with an MR distortion engages different underlying processes than learning to reach with a VR distortion, suggesting that these distortions may reflect different forms of motor learning.