Detection of highly motivated time segments of motor imagery brain computer interface signals

Motivation of subject who associated electroencephalogram based brain computer interface experiments is one of the most important parameters which affect this kind of research´s performance. Researchers can not able to measure how the motivation of subject exactly changes during the experiment. The proposed method was successfully applied to the BCI Competition 2003 Data Set III by using discrete Fourier Transform features with linear discriminant analysis classifier. The results showed that while the motivation of subject was very low level at first 0.5 sec, it increased after that time rapidly and again decreased after 3.25 sec on 6 second long of a trial.