<div>Dear EEGlab:</div>
<div> </div>
<div>supose I only use some goodIC to reconstrust data</div>
<div> </div>
<div>is this formula right?</div>
<div>data=eeg.icawinv(:,goodIC)*eeg.icaact(goodIC,:)</div>
<div> </div>
<div>the result looks right but how do I use EEG.icasphere in this formula?</div>
<div> </div>
<div>Appreciate your help</div>
<div> </div>
<div>Xian</div>
<div>Columbia fMRI research center</div>
<div> </div>
<div> </div>
<div>>> from an old EEGLAB email<br>Yes, it can be used. No, it cannot be used without error (meaning<br>expecting that (EEG.icawinv*EEG.icaweights*EEG.icasphere)*EEG.data<br>recovers the original data). The reason is that the number of dimension
<br>is reduced by PCA before running ICA and there is no way you can recover<br>the information lost when performing this transformation.<br><br>You can still visualize the component scalp topographies and they are<br>meaningful (you simply obtained them applying ICA not on the data but on
<br>a PCA reduction of the data).<br><br>Best,<br><br>Arno<br><br>_______________________________________________<br>eeglablist mailing list <a href="mailto:eeglablist@sccn.ucsd.edu">eeglablist@sccn.ucsd.edu</a><br>Eeglablist page:
<a href="http://sccn.ucsd.edu/eeglab/eeglabmail.html">http://sccn.ucsd.edu/eeglab/eeglabmail.html</a><br>To unsubscribe, send an empty email to <a href="mailto:eeglablist-unsubscribe@sccn.ucsd.edu">eeglablist-unsubscribe@sccn.ucsd.edu
</a><br> </div><br>