Figure 7.

(A) Electrode positions used to create each ROI, and (B) the mean classification accuracies of the four ROIs with that obtained using all electrodes (‘Scalp’), excluding the six ear electrodes. Each ROI was individually re-referenced, where a CAR was used for the scalp ROIs (‘Scalp’, ‘Frontal’, ‘Central’, and ‘Occipital’) while the mean of three electrodes attached on an opposite ear area was used as a reference signal for ear ROI (‘Ear’). Error bars indicate standard deviations of the estimated classification accuracies of each ROI. There is no significant difference between the four ROIs (Friedman test; p = 0.63).