[Eeglablist] filters, ICA and erp

[Javier Lopez-Calderon]

I agree with you about removing alpha; I was intentionally extremizing the idea of applying ICA on filtered data in order to correct/clusterize unfiltered data, which I intuitively disagree. I'll need some time to review the arguments that are flying in my mind.
Thanks for your answer, Rey.
Javier


On Oct 14, 2011, at 12:41 PM, Rey Ramirez wrote:

> Javier,
> I would not recommend doing what you suggested. Alpha is extremely important and you should not remove it. Idle rhythms of the brain like alpha and mu are partially phase reset by stimuli, and this is in part what produces ERPs/ERFs. I think you are assuming a purely amplitude modulation theory for the generation of ERP/ERF, and even if that was in part true for some systems (e.g., auditory, since Tau rhythm is often absent in subjects), if you filter out 10Hz, you will not get the peaks at 100, 200, 300, etc, and completely destroy your data.
> Let's be clear the reason why ICA comes up better after 1Hz high pass filtering is that the very slow waves (<1Hz) seem to make all IC more dependent (i.e., they all go up and down together). That's not true for alpha.
> In any case, you can always do complex infomax ICA for single frequencies or Independent Vector Analysis for for spectral bands.
> Hope this makes some sense,
> Ret 
> 
> 
> On Thu, Oct 13, 2011 at 2:59 PM, Javier Lopez-Calderon <javlopez at uc.cl> wrote:
> If this would be true then this would certainly be a good way to remove alpha activity from bored subjects' EEG, for instance. We'd just need to band-pass filter the data between 8-12 Hz, run ICA, and do the trick with the unfiltered data...
> 
> There should be something wrong with this rationale...right?
> 
> Javier
> 
> 
> 
> >
> > BTW, someone else suggested using the 1-Hz high-pass cutoff, performing ICA, and then applying the component coefficients to the unfiltered data.  That sounds like a great suggestion, although I don't know if there is a technical reason why it wouldn't work.  Does anyone out there know if there would be a problem with this?
> >
> > Steve
> 
> 
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.ucsd.edu
> For digest mode, send an email with the subject "set digest mime" to eeglablist-request at sccn.ucsd.edu
> 
> 
> 
> -- 
> Rey R. Ramírez, Ph.D.
> MEG Brain Imaging Center
> Institute for Learning & Brain Sciences
> Portage Bay Building
> Box 357988
> University of Washington
> Seattle, WA 98195-7988
> e-mail: rrramir at uw.edu
> Tel: (206) 543-6484 
> Fax: (206) 221-6472
> Cell: (619) 709-3590
> 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20111014/f4346d4c/attachment.html>