Does Real-Time Feedback Affect Sensorimotor EEG Patterns in Routine Motor Imagery Practice?

Area of Science:

• Neuroscience
• Rehabilitation Medicine
• Brain-Computer Interfaces

Background:

• Motor imagery (MI) activates similar neural circuits as actual movement, making it valuable for training and neurorehabilitation.
• Electroencephalogram (EEG)-based event-related desynchronization (ERD) of sensorimotor rhythms can mark brain activity during MI.
• Brain-computer interfaces (BCIs) can provide real-time feedback on brain activity during MI, but their effectiveness compared to open-loop training is unclear.

Purpose of the Study:

• To investigate whether BCI-enabled real-time feedback enhances brain responses during motor imagery compared to open-loop training.
• To determine the stability of brain activation patterns during motor imagery over time and across different feedback conditions.

Main Methods:

• Seven healthy volunteers with prior motor imagery experience underwent six EEG sessions over 4.5 months.
• Participants performed kinesthetic imagery of arm and finger movements.
• Real-time EEG feedback (closed-loop) was provided in the first and last sessions; other sessions used no feedback (open-loop).
• Spatiotemporal and amplitude features of ERD patterns were analyzed using linear mixed-effects modeling.

Main Results:

• Spatial sources of ERD were highly stable across sessions for most participants (ρ > 0.94).
• Only one subject showed a statistically significant difference in activation patterns between feedback conditions (p = 0.009).
• Real-time visual feedback via BCI did not significantly increase ERD strength.

Conclusions:

• BCI-enabled feedback did not significantly enhance brain responses during motor imagery in this study.
• Well-habituated individuals may achieve motor imagery benefits using simplified, open-loop training methods.
• At-home self-practice with simplified setups could be a viable option for motor imagery training.