[Eeglablist] Testing Time-Frequency for ROI (average of channels)

[Cowley, Ben]

Hello EEGLAB,

I would like to calculate the ERSP of a number of channels, i.e. a region of interest (ROI). One way is to average the time series, but that seems suboptimal since one might have a mix of phases across channels and end up with low power.
It seems a better way is to obtain the TFR matrices from each channel and take the average in each time x frequency cell. This is easily done by calling newtimef() in sequence for each channel, and average the results. So far so good. The question is, has anyone already provided a handy function to test these average TFR matrices? Because the testing part of newtimef() seems to boil down to a call to bootstat() (line 1279), but there are several steps before and after, that make things more complex. This seems like a problem someone would have solved already. Any idea?

best


Benjamin Cowley PhD
Docent of Cognitive Science
Associate Professor at Faculty of Education, University of Helsinki, Finland.