Somato-Motor Network Neural Connectivity Correlates with Visuomotor Adaptation

Abstract

Humans show tremendous individual differences in learning motor skills. Prediction of individual differences in motor learning is of theoretical significance and practical relevance. Brain as the central neural apparatus supporting motor control and learning has attracted considerable research efforts on examining brain functional and structural predictors of individual differences in motor learning. Many previous studies applied a resting state (RS) approach in recording brain activity, selected the contralateral primary motor cortex (M1) a priori as a seed or region of interest (ROI), and reported that functional connectivity between M1 and other brain regions predicted individual differences in tasks mostly on motor sequence learning. However, task-evoked brain neural activity and connectivity that represent different brain dynamics from spontaneous brain dynamics have not been studied in this line of research. This study presents findings on RS and finger-tapping evoked cortical somato-motor network (SMN) neural connectivity correlates with individual performances in a visuomotor adaptation (VMA) task, based on analysis of multimodal data from the Cam-CAN study. SMN nodes were initially localized through finger-tapping evoked brain magnetic fields as sampled with magnetoencephalography (MEG). It was found that SMN node specific neural connectivity strength at the contralateral primary somatosensory cortex (S1), supplementary motor area (SMA), and dorsal premotor cortex (PMd) during different stages around finger tapping but not in RS were significantly correlated with individual mean target errors in the final adaptation stage. The results imply critical roles of somatosensory and secondary motor areas in human motor learning.