A Method for Using Neurofeedback to Guide Mental Imagery for Improving Motor Skill

Abstract

Mental imagery (MI) is gaining attention as a strategy towards endogenous brain stimulation for improving motor skill. Neurofeedback (NF) is commonly used to guide MI in order to activate the relevant brain networks. The current study investigates an individualized EEG-based method for NF through broad consideration of interactions between different brain networks. We selected the change in brain functional connectivity (FC) as an objective neurophysiological measure of change in motor skill during a longitudinal physical training (PT) program. Digital tracing tasks were developed for skill training and the spatial error in tracing was used to gauge the change in skill. We used partial least squares algorithms to find the most robust contributing networks towards correlation between the resting state FC and the acquired motor skill. We used the network with the largest margin for increasing FC as the candidate for NF training while experimenting with MI during a neurofeedback training program. The participant was informed of the changes in instantaneous FC through real-time audio feedback to help guide the MI. We showed over 20% reduction in tracing error through neurofeedback training alone, without any additional PT. We also showed retention of improvement in skill for several days after the completion of neurofeedback training. Our proposed methodology shows promise for a highly individualized approach towards improvement in motor skill. Given that EEG is an accessible health and wellness technology, such a method could provide a practical complementary option towards personalized therapeutic strategies to improve motor function.

Keywords

• Functional connectivity
• partial least squares
• mental imagery
• neurofeedback
• EEG
• motor skill