Poster Session C, Sunday, March 24, 5:00 – 7:00 pm, Pacific Concourse
Is it possible to distinguish true and spurious cross-frequency coupling?
Felix Siebenhühner1, Sheng H Wang1, Gabriele Arnulfo1,2, Lino Nobili1,2, Matias Palva1,3, Satu Palva1,3; 1University of Helsinki, Finland, 2University of Genoa, Italy, 3University of Glasgow, United Kingdom
Neuronal activity is characterized by oscillations and phase synchronization at multiple frequencies. Cross-frequency coupling (CFC) mechanisms are assumed to be central for the regulation of neuronal processing among spectrally distributed oscillations and networks. Two forms of CFC, phase-amplitude coupling (PAC) and n:m-cross-frequency phase synchrony (CFS), have been shown to characterize resting- and task-state electrophysiogical activity and have been proposed to underlie integration and coordination of neuronal processing across frequencies. However, the validity of CFC observations has been questioned because the estimation of CFC may be confounded by filtering artefacts caused by non-sinusoidal waveforms. The core assumption behind CFC is that the interaction is observed between two distinct processes, whereas analyses of non-sinusoidality assume a single underlying process for the signal. While it is impossible to distinguish true CFC and spurious CFC caused by non-sinuisodalities when analyzing a single source (local CFC), CFC is necessarily true when found between two separable sources. We advance here a method to distinguish true from potentially spurious inter-areal CFC. We studied human resting state brain dynamics with stereo-electroencephalography (SEEG) and source-reconstructed magnetoencephalography (MEG) data and observed both CFS and PAC among a wide range of frequencies. We show that a significant fraction of the observed CFC was true and not explainable by non-sinusoidalities. Interestingly, the anatomical profiles of CFS and PAC differed from each other and were similar between SEEG and MEG. In conclusion, these results provide conclusive evidence for genuine inter-areal cross-frequency coupling in human resting state networks.
Topic Area: METHODS: Electrophysiology