Enhancing motor imagery through TMS-Based neurofeedback: A randomized controlled study in healthy adults

Background:

Motor imagery (MI) engages neural networks that overlap extensively with actual motor execution and is widely used in motor learning and neurorehabilitation. However, the effectiveness of MI depends strongly on an individual’s ability to generate vivid imagery, which varies across participants and can limit training benefits. Neurofeedback (NF) has been proposed as a method to enhance MI quality by providing real-time information about neural activity. While NF using EEG, fMRI, and NIRS has shown promise, the long-term effects of transcranial magnetic stimulation (TMS)-based NF using motor evoked potentials (MEPs), a direct physiological index of corticospinal excitability, remain unclear. This study examined whether TMS-based NF can modulate corticospinal excitability and improve subjective MI vividness over multiple training sessions.

Methods:

Twenty-two healthy adults were randomly assigned to a feedback (FB) group or a control (CON) group (n = 11 each). Participants performed MI-based mental practice three times per week for two weeks, completing 60 MI trials in each session. During a coincidence-timing task, TMS was delivered at a precise moment to elicit MEPs. The FB group received visual NF based on trial-by-trial normalized MEP amplitudes, whereas the CON group practiced without feedback. Corticospinal excitability was quantified as %MEP relative to resting values, and MI vividness was evaluated using a visual analogue scale (VAS) before and after each session.

Results:

A significant interaction between group and training day was observed for %MEP. The FB group demonstrated consistent increases in corticospinal excitability from Day 2 to Day 6, whereas the CON group showed a significant increase only on Day 5. For MI vividness, VAS scores increased significantly on Days 5 and 6 in the FB group, while no significant changes were observed in the CON group. These findings suggest that physiological modulation of corticospinal excitability may occur earlier than improvements in subjective MI vividness.

Conclusion:

TMS-based NF enhanced corticospinal excitability and improved MI vividness across training sessions. These results support the feasibility of MEP-guided NF as a method for optimizing MI-based training. Given that MEP reflects corticospinal output, this approach may also provide a meaningful foundation for developing brain–computer interface (BCI) applications and future neurorehabilitation strategies. Further research should clarify whether these physiological improvements translate to behavioral performance gains and evaluate applicability in clinical populations.

Daiki Matsuda is a researcher specializing in motor imagery, neurofeedback, and neurorehabilitation. He holds academic positions at Fukuoka International University of Health and Welfare. His work focuses on TMS-based neurophysiological assessment and the development of neurofeedback approaches to enhance motor learning. He has published research on MEP-based neurofeedback and continues to explore its applications in brain–computer interfaces and rehabilitation sciences.