Associate Professor University of Florida Gainesville, Florida
Rationale: A fundamental feature of the EEG is that it appears qualitatively different over different head locations. Why this is so is less clear. We hypothesized that spatiotemporal variation of the EEG should be explicable in terms of some relevant features of the underlying and varying neurobiology. In this work we use intracranial EEG (iEEG) from human epilepsy patients to explore the resting state functional connectivity of the lateral frontal cortex as a player in determining its resting state dynamics. In particular, we asked whether primary and association brain areas – due to their sharply different functions – were distinguishable in ‘EEG space’. Methods: We studied wake and NREM sleep subdural iEEG recordings from the lateral frontal lobe in seven (N=7) human subjects with epilepsy undergoing presurgical monitoring. Functional connectivity between a pair of electrodes for a discrete data sequence x(n)– the amplitude cross-correlation – was computed as the Pearson correlation of their respective quadrature amplitudes , where y(n) was the Hilbert transform of x(n). Electrode connectivity (EC) was computed as the average of its functional connectivity with all the other electrodes in the grid. Electrode sample entropy (ESE) was computed from the Hilbert amplitude time-series of that electrode as –ln (A/B), where A was the number of vector pairs having |xm+1(i) - xm+1(j)| < r and B the number of template vector pairs having |xm(i) - xm(j)| < r, with m = 2 and r = 0.2. Mean electrode connectivity (mEC) and mean electrode sample entropy (mESE) were obtained by time-averaging all the ECs and ESEs corresponding to that electrode. Results: In general, mEC and mESE were in inverse proportion to each other, a finding that was strongest in sleep and for the whole signal and its low-frequency passbands. Extreme values of mEC and mESE occurred over the Rolandic region and were part of a more general rostro-caudal gradient observed in all patients, with larger (smaller) values of mEC (mESE) occurring anteriorly. Conclusions: Brain networks (at least partly) determine brain dynamics. mESE over a given brain location, at the spatial scale of subdural macro-electrodes, and at the temporal resolution of clinical recordings, is (inversely) related to that location’s connectivity to other brain regions. Over the lateral frontal lobe, these metrics demonstrate a rostro-caudal topography that clearly distinguish primary sensorimotor from frontal association cortex. These results are consistent with current notions regarding the structural and functional parcellation of the human frontal lobe. Our hypothesis-driven findings also effectively ‘diagnose’ Rolandic iEEG rhythms as separate from anterior frontal regions, and are independent of the classical mu rhythm associated with peri-Rolandic cortex. Funding: Please list any funding that was received in support of this abstract.: This work was supported by the Wilder family endowments to the University of Florida.