INTRODUCTION: Successfully learned motor skills can generalize to untrained contexts. Specifically, some recent work has showed that newly learned motor skills can symmetrically generalize to the untrained arms (from right to left arm and vice versa), both immediately as well as after a period of 24 hours of learning (Yadav & Mutha, 2019;2020). However, the neural substrate underlying this form of generalization after skill memory has stabilized (following 24hrs) is unclear. Contralateral primary motor cortex (M1) is considered a key site for skill learning-related neuroplastic changes and skill memory stabilization (Kantak et al., 2010; Christiansen et al., 2020).

OBJECTIVE: We probed whether skill generalization is causally mediated by contralateral M1. We hypothesized that this key area for skill memory stabilization following learning plays critical role (as compared to ipsilateral M1) for interlimb skill generalization.

METHODS: In an ongoing study (in accordance with University Ethics Committee and principles of the Declaration of Helsinki) spanning over two days, we tested young healthy right-handed participants (currently n=10, target sample size n=30, planned completion Dec-2023) who learned a novel motor skill on Day-1 with right arm. Following learning they received offline 1Hz rTMS (1800 pulses delivered using Neuronavigation) over either contraM1 (left M1, Group1) or ipsiM1 (Right M1, Group2). The skill task involved tracking movements (10 blocks of 16 trials each) from a start circle (green-color, 1cm diameter) to one of the eight target circles (blue-color, 1cm diameter) that appeared on the screen at a distance of 15 cm, connected via a tunnel (white parallel lines, 1cm width). Participants were instructed to make fast and accurate movements to the target while moving within the tunnel. After each movement, participants received performance feedback along with a numeric score (Motor Skill Error). In addition, we measured Motor Evoked Potentials (MEPs) using surface electrodes over first dorsal interosseus muscle (right hand for Group1 and left hand for Group2), pre and post learning, as well as immediately, at 5 min and 10 mins post rTMS to assess change in corticospinal excitability. On Day-2 after 24 hrs, all participants were tested for interlimb skill generalization by performing the task with their untrained (left arm, 10 blocks), as well as the trained arm (right hand, 1 block).

RESULTS: Our preliminary findings indicate that both groups learned the skill task on Day-1 (reduction in Skill Error over 10 Blocks in all current subjects). As expected, 1 Hz rTMS also led to a decrease in contralateral MEPs for both groups (most pronounced at 5min post rTMS). Next, when untrained arm was tested on Day-2, we observed higher skill error on Block1 in Group1 (contra-M1 rTMS) as compared to Group2 (ipsi-M1 rTMS). However, both groups reached similar performance with untrained arm at the end (Block 10) and exhibited comparable performance when the trained arm was tested in a subsequent (and last) Block on Day-2.

CONCLUSION: Hence, our early results suggest that 1Hz rTMS over contra-M1 but not ipsi-M1 can impair skill generalization to untrained arm without affecting its learning ability and follow-up performance of the trained arm.