[Eeglablist] Run ICA on continuous data or epoch data

[Arnaud Delorme]

Dear Fengi,

to answer your two questions. The fact that you are using overlapping epochs should not impair ICA. You can run ICA on data epoch or on continuous data (I prefer continuous but that's my personal choice).

Best,

Arno

On Jun 22, 2011, at 9:45 AM, Fengji Geng wrote:

> Hi folks,
> I have two questions about ICA. I would appreciate very much if anybody can help me with these two questions.
>  
> I use 128-channel NetStation to record data. My data is about 16 minutes long. My experiment is like several pictures present regularly and the interval between pictures is always 2 s. Before running ICA, I first extract epochs from continuous data. The time range for each epoch is from -1 s to 2.5 s. As the interval between picture is only 2 s, the neighbour epochs have overlaps. Thus, my question is if this overlap could impair the ICA.
>  
> ICA can be run on both continuous data and epoch data. Another question I have is if I only want to run ICA to detect artifact, is it better to run ICA on continuous data or epoch data?
>  
> Thanks a lot!
> 
> 
> -- 
> Fengji 
> 
> _______________________________________________
> 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