[Eeglablist] questions about ICA

[Scott Makeig]

Qiao -

1) What is plotted are the averaged ICA activations (EEG.icaact). These 
are proportional to uV, but differently for every channel (according to 
the component map in EEG.icawinv. To see the component contributions to 
the ERP in uV, use

Plot > Component ERPs > With component maps

Once you learn to use and interpret this function (envtopo()), it 
conveys the most information about the relation of component activities 
to the whole-signal ERP.

2) If the two datasets were collected in the same session, you may 
decompose them together. If not, then you must compare the two 
decompositions. Begin by matching the component maps using matcorr(). 
You may also correlate or cluster on correlations of the component 
spectra, or spectra plus maps, etc. This is a difficult topic that we 
are exploring in our research as well.

Scott Makeig


qiaozhimei wrote:

>Dear professor:
>  I have some questions about ICA. 
>  1).When I do plot>component ERPs>In rectangular array,I have no idea what is the real mean of the vertical axis of every little figure,that is,what informations are contained in the values of the vertical axis.
>  2).I have tow different datasets which are from two diffrent experimental situation of one subject,for example,pain and non-pain situaltion.Now I want to compare the ICA of the two dataset,how should I do? Because the two datasets decomposed different components, how to compare thoes components?
>
>Qiao 
>Thank you!
>
>-----------------------------------------------------------??126??????,????????????,http://www.126.com/?????????,260???12???,???????,?????
>_______________________________________________
>Eeglablist mailing list Eeglablist at sccn.ucsd.edu
>Eeglablist page: http://www.sccn.ucsd.edu/eeglab/eeglabmail.html
>To unsubscribe, send an empty email to eeglablist-unsub at sccn.ucsd.edu
>To switch to digest mode, send an empty email to eeglablist-digest at sccn.ucsd.edu
>  
>