An inverse transform technique for the EEG phase reset analysis

We propose a signal processing tool for evaluating the electroencephalogram (EEG) phase reset due to auditory attention by utilizing an inverse analysis of the instantaneous phase for the very first time. EEGs were acquired through an auditory attention experiments with the highest entropy stimulation paradigm. We examined single sweeps of auditory late response (ALR) with the complex continuous wavelet transform. The phase in the frequency band that corresponds to auditory attention was reset to the mean phase of the averaged EEG. Then the inverse transform was applied to reconstruct the phase-modified signal. We found a significant enhancement of the N100 wave in the reconstructed signal. Analysis of the phase noise shows the effects of phase jittering on the generation of the N100 wave implying that a preferred phase is needed to generate the event related potential (ERP). Furthermore, by resetting the phase only at the theta border of no attention data to the mean phase of attention data yields a result that resembles the attention data. These results show direct connections between EEGs and ERPs and thus it is concluded that the phase reset of EEGs plays a very crucial role in the ERP generation.