EEG-Based BCI System Using Adaptive Features Extraction and Classification Procedures

Figure 4

In this work, each participant took part in a total of 6 sessions (a). (b) shows 3 possible compositions of the session. In the top example, module is repeated because the average accuracy in the previous 3 repetitions was lower than 40%. At the end of the session, phase is not even reached because the average accuracy in the last 6 repetitions was <70%. The middle example is similar to the top one ( module is indeed repeated), but this time phase is reached at the end of the session. Finally, the bottom example shows the case of a shorter session.