Conflicting results in EEG alpha feedback studies: why amplitude integration should replace percent time

Success or failure of EEG feedback training for alpha enhancement can depend on how alpha activity is quantified and fed back. Alpha-enhancement failures usually employ a percent time (%) technique; successes typically use amplitude integration (variation of). To dramatize the differences between percent and integration techniques, we derived both measures simultaneously from left occipital (O1) and left central (C3) sites for 16 male subjects who were given 5.6 hours of integrated alpha feedback from the midline occipital (Ozeta) site. At both the O1 and C3 sites the integrated and percent measures were not equivalent and not linearly related. Statistically significant differences in the (integrated, percent) correlation coefficients (z-transpormed) were observed under the different recording conditions: alpha enhancement, alpha suppression, and baselines. Theoretical discussion of integration and percent techniques is given and the adoption of amplitude integration measures and feedback stimuli is strongly advocated.