A performance study of the wavelet-phase stability in the quantification of neural correlates of auditory selective attention

Large-scale neural correlates of auditory selective attention reflected in the electroencephalogram (EEG) have been identified by using the complex wavelet-phase stability measure (WPS). In this paper, we study the feasibility of using the WPS in extracting the correlates of selective attention by comparing its performance to the widely used linear interdependency measures, i.e., the wavelet coherence and the correlation coefficient. The outcome reveals that the phase measure outperforms the others in discriminating the attended and unattended single sweep auditory late responses (ALRs). Particularly, the number of response sweeps that are needed to perform the differentiation is largely reduced by using the proposed measure. It is concluded that a faster (in terms of using fewer sweeps) and more robust objective quantification of selective attention can be achieved by using the phase stability measure.