[Eeglablist] Continuing old threads on ERP and ICA

[Tarik S Bel-Bahar]

Hello Emmanuelle,
There were several more recent discussions about this. See also Makoto'
recommended pipeline (googlable). Please check them out and see if that
resolves your current question. Note that there are some different
processing preferences in pure-erp camp and ICA-for-ERP camps.

My understanding is that it's okay to apply an ICA decomposition from 1Hz
file to a 0.1 hz file, for ERP purposes.

[Eeglablist] Applying ICA weights from 1Hz filtered data to 0.1Hz ...
https://sccn.ucsd.edu/pipermail/eeglablist/2015/010145.html
Sep 2, 2015 - [*Eeglablist*] *Applying ICA* weights from 1Hz filtered data
to *0.1Hz* ... *Makoto* On Fri, Aug 21, 2015 at 9:50 PM, chao wang
<hnchaowang at ...
*ICA* running very slowly - Swartz Center for *...*
Feb 1, 2017
*ICA* analysis and Interpolation
Mar 15, 2016
Data Processing Steps for *ICA*? - Swartz Center for *...*
Nov 5, 2014
the correct EEG data processing pipeline - Swartz *...*
Oct 28, 2012
More results from sccn.ucsd.edu


Makoto's preprocessing pipeline - SCCNhttps://sccn.ucsd.edu/wiki/
Makoto's_preprocessing_pipeline





On Wed, Jul 19, 2017 at 12:10 PM, Emmanuelle Renauld <
emmanuelle.renauld.1 at ulaval.ca> wrote:

> ​Hi list,
>
>
> I was reading on ICA and ERPs. I was thinking on following the suggestion
> in the thread "filters, ICA and erp" from 2011, and also discussed here:
>
> https://github.com/CSC-UW/csc-eeg-tools/wiki/Filtering-and-ICA.
> <https://github.com/CSC-UW/csc-eeg-tools/wiki/Filtering-and-ICA> They
> propose that the best solution as a tradeoff between filtering at 1 Hz to
> have good ICA or filtering at 0.1 Hz to have good ERP would be filtering at
> 1 Hz, computing ICA, and then applying the result to a 0.1 Hz filtered data
> set.
>
>
> It seems easy to do. However, I also found another thread on this
> subject, "Applying ICA weight matrix on another dataset​" from 2013, but
> the discussion has not been finished. People were wondering if the sphering
> matrix had to be recomputed from the new data set or not. It seems to me
> that the last message was this one : https://sccn.ucsd.edu/piperm
> ail/eeglablist/2013/006317.html. I haven't found a lot of information
> about this on Google. Does anyone know if a decision has been made on this
> subject since then?
>
>
> Also, can anyone tell me if, 6 years later, this is still considered the
> best solution?
>
>
>
> Thank you very much,
>
>
>
>
> Emmanuelle Renauld
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc
> sd.edu
> For digest mode, send an email with the subject "set digest mime" to
> eeglablist-request at sccn.ucsd.edu
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20170720/bea810c0/attachment.html>