[Eeglablist] ICA of the evoked gamma?

[Nenad Polomac]

Hi everybody,

I am interested in evoked gamma band response. I would like to extract the
independent components from my EEG data which represent this effect. My
idea was to calculate the ICA on my 60 channel EEG single trials and then
to average the time-courses each of the components over trials. Then on
this evoked trial I would calculate time frequency with the Morlet wavelet
and obtain evoke power each of components (freq wise baseline correction as
well). Finally, those components that have in a particular time window
increase in power relative to baseline would be considered as the one that
represent gamma band response. I would use these info for the rejecting of
all other components from the original (non averaged) ICA data. After that
I would back project remaining (gamma representing) components into the
channels. My question would be, is this ok to do from the mathematical
point of view? I am particularly asking about the step where time-courses
each of the components is averaged over trials.
I would really appreciate some opinions and experiences.

All the best!

Nenad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20140707/3072445a/attachment.html>