Brain-computer interface training that uses real-time feedback from electrical brain activity can improve cognitive function in patients with Parkinson’s disease, multiple sclerosis, dementia, and traumatic brain injury, according to a study published in Wiley Advanced Science.
Researchers at the University of Texas at Austin used EEG to track error-related potential (ErrP) — an electrical signature the brain emits when recognizing an error. By decoding the error positivity component in real time and feeding it back to participants, the system helped the brain amplify its own marker of conscious error detection.
“That lets us drive learning gains for exactly the small errors that behavioral training alone couldn’t touch,” said senior author Jose del R. Millan.
Over five days of training, participants using a joystick to correct cursor trajectories showed improved perception of small visuomotor errors. The approach proved more effective than conventional behavioral training and safer than pharmacological options. Researchers see potential applications in neuropsychiatric rehabilitation and even high-performance fields like motorsports.
