Abstract:
Measures of cross-frequency coupling, which reveal nonlinear and non-Gaussian dynamics within time series, are increasingly common in applications to neurophysiology.  Nevertheless, the existence and usefulness of longstanding techniques related to higher-order spectra (HOS) is still not widely appreciated among researchers. This situation may be attributed in part to the reputedly obscure interpretation of HOS. In this talk, I will first give a brief overview of HOS; second, suggest some new strategies for interpreting them; and third, describe a novel decomposition which promises a general-purpose tool for pattern identification and separation in time series.  The application of this new technique is demonstrated through the blind recovery of auditory evoked responses in human intracranial recordings. 

 References:

1. C. K. Kovach, H. Oya, and H. Kawasaki, “The bispectrum and its relationship to phase-amplitude coupling,” NeuroImage, vol. 173, pp. 518 – 539, 2018. 
2.  C.K. Kovach and M.A. Howard, “Decomposition of higher-order spectra for blind multiple-input deconvolution, pattern identification and separation,” Signal Processing, vol. 165, pp. 357 – 379, 2019.