[Eeglablist] rejecting single channel trials

Andreas Widmann widmann at uni-leipzig.de
Fri Mar 23 11:34:23 PDT 2007


Hi Ingmar,

attached you find a patched pop_loadbv version importing also the 
channel information for bad segment events into the EEGLAB *event* 
structure.

Subsequently, you have to map this information into the corresponding 
*reject* structure. Attached also a sample function how this could be 
done. Most likely you have to adjust at least the event marker type 
(e.g. 'bad amplitude').

When plotting the data make sure to hide the event duration ("Display > 
Hide event duration") otherwise the rejected epochs/channels might be 
covered.

Best,
Andreas

> Hi there,
> 
> following situation: I have a preprocessed data set, exported from
> Vision Analyzer (72PA.dat; 72PA.vhdr; 72PA.vmrk). Certain trials in
> certain channels where marked as rejected. E.g.: channel 5 trial 2
> was marked "Bad Amplitude", but all other channels were all right.
> 
> After importing the data set with EEGLAB:
> 
> EEG = pop_loadbv( dir, fileName, 1, [1 2 3 4 5 6 7 8 9 10 11 12 13 14
> 15 16 17 18 19 20] );
> 
> all channels in trial 2 are marked as "Bad Amplitude". For further 
> calculations I need to remove only the data from channel 5 trial 2.
> 
> Obviously the marker file "72PA.vmrk" contains this information, but
> isn't correctly represented in EEGLAB after the import. Plotting the
> data set using "Channel data (scroll)", shows that all channels are
> marked in trial 2.
> 
> How can I use the information in the marker file, to tell EEGLAB
> which trials in a single channel have to be rejected?
> 
> Thank you, Ingmar
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: pop_loadbv.m
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20070323/3134e8a4/attachment.ksh>
-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: eeg_bvbadseg2rejmanual2.m
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20070323/3134e8a4/attachment-0001.ksh>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3257 bytes
Desc: S/MIME Cryptographic Signature
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20070323/3134e8a4/attachment.bin>


More information about the eeglablist mailing list