BCIwiki (bciwiki.com) — A machine learning model based on the Random Forest algorithm can classify EEG signals from subjects performing real and imagined motor activities, and it works on consumer-grade EEG hardware as well as research-grade systems, according to a preprint posted to arXiv on September 10, 2026. The work is by Tarciana C de Brito Guerra, Taline Nóbrega, Edgard Morya, Allan de M. Martins and Vicente A de Sousa, and has not been peer reviewed.
The team evaluated its Random Forest classifier on both a consumer-grade and a research-grade EEG system. The algorithm efficiently distinguished imagery from real activities and identified the related body part, even with consumer-grade EEG. The study also found that interpersonal variability in EEG signals negatively affects the classification process.
EEG is a fundamental tool for understanding the brain's electrical activity related to human motor activities, and brain-computer interfaces use that activity to build assistive technologies, especially for people with physical disabilities. Extracting signal features and patterns remains complex, which is why the task is often delegated to machine learning algorithms. The researchers say correct interpretation and classification of EEG signals is a prerequisite for tools controlled by cognitive processes.