[Eeglablist] Using Automatic Artifact Detection

[Cedric Cannard]

Dear Daniel,

Yes, ASR requires your EEG data to be continuous. Are you planning to do ERP/ERSP analysis with these epoched data? 
You could try to convert it back to continuous, run ASR with the reconstruction method, and then convert back to epoched:
EEG.data = EEG.data(:,:); % convert back to continuous
% Run ASR (reconstruction method)
EEG.data = reshape(EEG.data, EEG.nbchan, EEG.pnts, EEG.trials); % convert back to epoched

But it's probably not a good idea. You definitely can't reject bad segments within epochs because your signal will no longer be time-locked to the events, and your epochs will end up with different lengths, which cannot be analyzed. 
The typical approach is to run ASR before segmentation, or removing whole trials that contain artifacts. If it's just eye blinks, run ICA and ICALbel to remove them without losing brain signals. 


Cedric


Sent with Proton Mail secure email.

------- Original Message -------
On Tuesday, June 27th, 2023 at 9:10 PM, Daniel Gnajewski via eeglablist <eeglablist at sccn.ucsd.edu> wrote:


> Good Morning EEGLAB Newslist,
> 
> I have a question about the criteria for using automatic artifact rejection in EEGLAB. So the meta file I'm using has been segmented and EEGLAB will not let me use it for Automatic Artifact Rejection. Is there a way to get around this or is the fact the data isn't continuous why it's not allowing me to use AAR on it.
> 
> Very Respectfully,
> Daniel Gnajewski
> _______________________________________________
> 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