[Eeglablist] error: pop_readegi

William McGeown william.mcgeown at strath.ac.uk
Tue Oct 12 04:23:52 PDT 2021

Dear All,
I was hoping that someone can help.
I am trying to load EGI data into EEGLAB.  I can't use later versions of NetStation (to export in additional formats), so as far as I am aware, calling pop_readegi on the raw binary files is the method that I should use.
I have exported the data as a raw binary file (that contains the events and I need to have these included).

I have downloaded the GSN-HydroCel-129.sfp channel location file and have placed it in the sample_locs folder.

I have tried to use the following command:
EEG = pop_readegi(filename, [],[],'GSN-HydroCel-129.sfp')

But I get the following error messages:
Warning: This function can only import continuous files or
         epoch files with only one length for data epochs
Importing binary EGI data file ...
Error using readegi (line 199)
Number of values read not equal to the number expected.

Error in pop_readegi (line 114)
    [Head EEG.data Eventdata SegCatIndex] = readegi( filename, datachunks);

The method works for other datasets that I have that includes markers, but doesn't seem to work for this one.
Can someone help?
It may be something to do with the way the paradigm was coded, with an unusually high number of markers that are recorded in separate channels.  There are around 1900 channels, that are mostly empty but that contain markers. I didn't create the paradigm, but it could be that a separate channel has been created for every marker.
I have 128GB of RAM (and I don't receive any error message from Matlab that the matrix is too large), so I don't think that is the issue.
Is there any way to load the data in with the markers (even by tweaking the pop_readegi function)?  I'm not sure how to do this myself.
I can share a datafile if need be.

My plan was to load 2 .raw files per person using this method, then to merge them afterwards.
Maybe I'll try to change the paradigm for future participants, but I have a number of participants with data using the paradigm as it currently is that I would like to retain, so any help would be really very much appreciated.



More information about the eeglablist mailing list