[Eeglablist] Component spectra and maps question

[ida miokovic]

Dear List,

I have been plotting "Component spectra and maps" in EEGLAB of an
independent components (after ICA procedure) and having some dilemmas
regarding the components obtained as largest contributing ones.

1st setup in the pop-up window (Plot-->Component spectra and maps") - in
text box "Electrode number to analyze" I put 0 in order to get the whole
scalp analyzed. And "Compute comp spectra" was checked. In Matlab command
window, percent relative variance is obtained.

2nd setup in the pop-up window - in "Electrode number to analyze" I put []
in order to get the electrode with the max power at the frequency of my
interest. "Compute comp spectra" was checked. In Matlab command window,
percent relative variance is obtained.

3rd setup in pop-up window - in "Electrode number to analyze" I put "the
number of electrode obtained in 2nd setup" in order to get the components
accounts to this electrode at the frequency of interest. "Compute comp
spectra" was *unchecked*. In Matlab command window, percent variance
accounted for is obtained.

I went through tutorial and understand that 3rd setup gives results as
components accounts to the electrode at the frequency of interest. But now
I am confused with what is the interpretation of components obtained in 1st
and 2nd setup. In my case, components obtained in 2nd and 3rd step in some
cases are identical and in some 4 of 5 are the same.

Thank you in advance for any help,

Kind regards

Ida
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20130813/91d5213c/attachment.html>