[Eeglablist] Min. data length for ICA analysis?

David Groppe dgroppe at cogsci.ucsd.edu
Tue Sep 21 21:59:28 PDT 2010


Hi Rajeev,
      In the following article:

Onton,J., Westerfield, M., Townsend, J.,Makeig, S. (2006) Imaging human
EEG dynamics using independent component analysis. Neurosci. Biobehav.
Rev. 30(6), 808–822.

Onton et al. recommend a minimum of 20 time points per channel^2 while
noting that more data than that generally helps.  By evaluating the
reliability of ICA, we've found that 20 time points per channel^2 is
indeed sufficient to produce a usable proportion of reliable ICs, but
that you can do much better with more data.  See Figure 7 of:

Groppe, D.M., Makeig, S., &  Kutas, M. (2009) Identifying reliable
independent components via split-half comparisons. NeuroImage, 45
pp.1199-1211.
(http://www.cogsci.ucsd.edu/~dgroppe/PUBLICATIONS/Groppe2009.pdf)

hope this helps,
     -David

On Mon, Sep 20, 2010 at 3:25 PM, Rajeev Yadav <rxy2009 at yahoo.com> wrote:
> Hi,
>
>
> I have a basic question for ICA analysis.
> What should be the minimum length for data to perform ICA analysis?
>
> Regards
> Rajeev Yadav
> ~~~~~~~~~~~~~~~~~~~~~~~
> "As many branches, so many trees."
> ~~~~~~~~~~~~~~~~~~~~~~~
>
>
> _______________________________________________
> 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
>



-- 
David Groppe, Ph.D.
dgroppe at cogsci.ucsd.edu
http://www.cogsci.ucsd.edu/~dgroppe/



More information about the eeglablist mailing list