[Eeglablist] Double Dipping with ICA

[Scott Makeig]

Nicholas - This is not at all a case of 'double dipping' ... In fact, ICA
should return the same components the 2nd time, since they will remain
(maximally) independent in the data, although because you will have removed
some dimensions from the data (in removing some ICs), you will have to
reduce the expected dimension of the data to compensate (e.g., if you
remove 2 ICs from 64-channel data, then the 2nd ICA decomposition should
use the PCA option to reduce it to 62 dimensions.  This is nearly the only
place in which using PCA dimension reduction before ICA is  justified - see
Fiorenzo Artoni's paper on PCA and ICA.

Scott Makeig

On Sun, Nov 8, 2020 at 11:22 AM Nicholas Dogris <
drdogris at neurofieldneurotherapy.com> wrote:

> Will my data be distorted if I run ICA, remove IC's such as eye blinks, and
> then run ICA again on the same dataset?  Are there reasons not to do this
> or to do it?
>
> Cheers,
>
> Nick
> _______________________________________________
> 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
>


-- 
Scott Makeig, Research Scientist and Director, Swartz Center for
Computational Neuroscience, Institute for Neural Computation, University of
California San Diego, La Jolla CA 92093-0559, http://sccn.ucsd.edu/~scott