Resting state fast brain dynamics predict interindividual variability in motor performance

Abstract

Motor learning features rapid enhancement during practice then offline post-practice gains with the reorganization of related brain networks. We hypothesised that fast transient, sub-second variations in magnetoencephalographic (MEG) network activity during the resting-state (RS) reflect early learning-related plasticity mechanisms and/or interindividual motor variability in performance. MEG RS activity was recorded before and 20 min after motor learning. Hidden Markov modelling (HMM) of MEG power envelope signals highlighted 8 recurrent topographical states. For two states, motor performance levels were associated with HMM temporal parameters both in pre- and post-learning resting-state sessions. However, no association emerged with offline changes in performance. These results suggest a trait-like relationship between spontaneous transient neural dynamics at rest and interindividual variations in motor abilities. On the other hand, transient RS dynamics seem not to be state-dependent, i.e., modulated by learning experience and reflect neural plasticity, at least on the short timescale.