[Eeglablist] epoching before ICA for continuous data?

[Tarik S Bel-Bahar]

Hello, Many EEG ICA researchers just get a great ICA decomposition and then
re-apply it to full continuous data, and then epoch that data. The first
epoching step before ICA, however meaningless, can benefit you by allowing
for a cleaner ICA decomposition. It's non-trivial how short or long the
fake epochs are, as this will influence how many epochs are detected and
dropped due to artifact before you do the ICA. Cheers!



> On Tue, Oct 6, 2015 at 1:27 PM, Makoto Miyakoshi <mmiyakoshi at ucsd.edu>
> wrote:
>
>> Dear Amy,
>>
>> ICA shuffles all the datapoints as a preprocessing step. So it does not
>> matter whether your data are continuous or chopped up into pieces. So
>> Michael is absolutely right!
>>
>> Makoto
>>
>> On Tue, Sep 22, 2015 at 3:17 PM, Amy Cao <amycao723 at gmail.com> wrote:
>>
>>> Hi All,
>>>
>>> I have one question:
>>>
>>> I am dealing with resting-state EEG data. So I am wondering whether I
>>> need to epoch my data into small segments before ICA or I could directly
>>> use the continuous data for ICA.
>>>
>>> Thank you,
>>>
>>> Jianwei
>>>
>>> _______________________________________________
>>> 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
>>>
>>
>>
>>
>> --
>> Makoto Miyakoshi
>> Swartz Center for Computational Neuroscience
>> Institute for Neural Computation, University of California San Diego
>>
>> _______________________________________________
>> 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
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20151006/212e4899/attachment.html>