Dear Joe,
The montage function doesn't touch the events so what you describe
doesn't sound feasible. What I suggest is that after you change the
events and save, try clearing the workspace and re-loading the dataset
with spm_eeg_load. Then check that the events in this dataset are
modified. If this is the case and running the montage changes them
back, I'd really like to look at that. If not, then there might be
some problem with how you updated the events. You can send me the
relevant part of your script.
Best,
Vladimir
On Thu, Apr 1, 2010 at 2:01 PM, Joseph Brooks <[log in to unmask]> wrote:
> I've written a function to rename the event markers in my EEG data (BDF
> format loaded with spm_eeg_convert) by updating the values in the
> D.events.value field. I then save this (using D.save). I then pass the
> resulting file to spm_eeg_montage in order to rerefernce the data. When I
> check the event markers in the file that comes out of the montage function,
> the event markers have been returned to their original state (as if I hadn't
> even run my script to change them). Interestingly, when I check the file on
> which I ran my script, the changes ARE still there in markers in that file.
> So it looks like the montage function is recomputing the event markers? Is
> this supposed to be happening? Obviously I can just run my script after
> montaging. But, how is the montage function recomputing these values? Is it
> on the basis of the STATUS channel? In that case, if I change the data in
> the status channel (instead of in the D.events.value structure) would this
> cause the events to stay the same after running spm_eeg_montage?
>
> Thanks for any help that you can give.
>
> Joseph Brooks
> Institute of Cognitive Neuroscience
> University College London
>
|