[Eeglablist] functional connectivity-coherence

[Wind, Johanna]

Hi everyone,

I measured functional connectivity (more precise: coherence) with the mathematical coherence function with its magnitude squared cross-power spectrum of the signals divided by the auto-spectrum of the signals (Golinska, 2011).
The power spectrum estimation was analyzed on the basis of Matlab-based pwelch-function with a sampling rate of Fs = 1024, a window-size of 2048 samples (2 seconds), an overlap of 50% and a frequency resolution of nFFT = 4096.

Is this a correct way to measure coherence (functional connectivity) and therefore to interpret interaction of brain regions? Or is there a better way to calculate coherence, e.g. in eeglab or based on Matlab?


Best regards,
Johanna


Johanna Wind
wissenschaftliche Mitarbeiterin
Johannes Gutenberg-Universität Mainz
Institut für Sportwissenschaft
Abt. Bewegungs- und Trainingswissenschaften
Raum: 00-113
Tel.: 06131-3927972
Albert Schweitzer-Straße 22
E-Mail: jwind01 at uni-mainz.de<mailto:jwind01 at uni-mainz.de>
Sprechstunde nach Vereinbarung