Analysis of Brain Network Structure under Haptic Motor Learning

Motor learning is not limited to task execution but continues during post-training resting periods, potentially reflecting consolidation processes. While previous studies using fMRI have reported reactivation of task-related activity during resting-state, EEG-based network-level evidence remains limited. In this study, we investigate haptic motor learning using a tracking task under visuomotor rotation, and analyze resting-state EEG recorded before and after training. To capture dynamic brain network changes, we employ a tensor-based representation combined with time-varying graphical lasso. We quantify network reorganization using a metric based on the Frobenius norm of differences in Fisher z-transformed partial correlation matrices. The results show a significant positive correlation between resting-state network change and motor performance improvement. These findings suggest that post-training resting-state network reorganization reflects motor learning consolidation, highlighting the importance of spontaneous brain activity in haptic skill acquisition.