Preview

PDF, English Download (3MB) | Terms of use

Abstract

Motor learning is a fundamental ability and one of the most robust models to study neural plasticity. The majority of human motor learning imaging studies focused on either short-term or long-term learning using one single imaging modality. These studies were thus not able to systematically investigate the dynamic process of motor learning from a multimodal perspective. The current project combined both short-term and long-term motor learning to comprehensively characterize neural plasticity at multiple phenotypic levels of the brain: functional activation, functional connectivity, grey matter volume, and glutamate concentration. To this end, this project involved a cross-sectional and a longitudinal study with multimodal brain imaging techniques (task fMRI, resting-state fMRI, gray matter structural fMRI, pharmacological fMRI, and MRS). Short-term motor learning was significantly correlated with brain network features related to network efficiency. It was also associated with a highly reliable cerebellum-centered network which was significantly modulated by the NMDA antagonist ketamine. Long-term motor learning was associated with increased activation in premotor / SMA and parietal regions and with increased gray matter volume of the SMA and the hippocampus. In addition, long-term motor learning was accompanied by a decrease in the functional connectivity of a network centered on the sensorimotor cortex which was related to handknob glutamate concentration levels and which involved regions that were highlighted by our activation and structural analyses. Taken together, this thesis contributes important evidence to the neurofunctional and neurostructural underpinnings of motor learning and points to the critical roles of the cerebellum, the hippocampus and the relevance of glutamate for motor learning in humans.