Volume Conduction Influences Scalp-Based Connectivity Estimates

• Published in: Frontiers in computational neuroscience Vol. 10; p. 121
• Main Authors: Brunner, Clemens, Billinger, Martin, Seeber, Martin, Mullen, Timothy R, Makeig, Scott
• Format: Journal Article
• Language: English
• Published: Switzerland Frontiers Research Foundation 22.11.2016; Frontiers Media S.A
• Subjects: Conduction; Cortex; EEG; Electroencephalography; Fourier transforms; Neural networks; Neuroscience; Time series; United States > US; Austria; source analysis; inverse problem; volume conduction; Electroencephalographic; directed transfer function; connectivity
• ISSN: 1662-5188; 1662-5188
• DOI: 10.3389/fncom.2016.00121

[...]in general direct application of connectivity measures to scalp EEG signals produces less than accurate results and also does not allow their clear interpretation in terms of underlying source dynamics. [...]the observed changes in the DTF can be explained by the fact that adding a sinusoid to all EEG channels with equal strength does not reflect volume conduction of a realistic brain source. Simulation Although the analytic derivation can be extended to more than two sources, computing and notating the individual elements of the mixed transfer matrix becomes quite unwieldy. [...]we simulated three cortical sources and used a realistic forward model of volume conduction from brain to scalp to numerically estimate their summed activity at three scalp EEG channels. [...]while DTF and other connectivity estimators can be applied to either scalp channels or to (estimated) source signals, results are highly likely to be more accurate when the analysis is based on source activities.