ABSTRACT: Spatio-temporal patterns observed in scalp EEG during preparation for two specific voluntary movements are presented. Discrimination of these spatio-temporal patterns is proposed as an enhanced framework for the development of a Brain-Computer Interface. The feasibility of single-trial discrimination is shown through a simple spatio-temporal template matching (correlation) classifier.
For several years now, there has been an increasing interest in classifying EEG manifestations associated with voluntary actions , particularly with the preparation for specific movements . Reliable classification of EEG patterns voluntarily generated by a subject could (in the long run) be applied towards the development of new mechanisms for man-computer communication. Even early developments of this kind could be extremely useful to improve the communication of certain individuals with disabilities (e.g., quadriplegics) with computers and, through them, with their environment.
In this study, we investigated the existence of specific spatio-temporal patterns in the scalp EEG that appear in association with the preparation for the movement of the index finger and the toe.