[Eeglablist] Added fields to events, which are suddenly gone

[Tarik S Bel-Bahar]

Hello Nils, not sure if the below will help, best wishes!

..so clearly something is happening between the time you pick epochs based
on the new event value, and when you go back to the "original" data set and
see the values you put in are gone. Or alternatively, something is
happening when the original data set is refreshed, after you go back to it
to get the next "event".

If you are indeed able to pull out new epochs with your new events for one
value, perhaps the dirty solution is to simply repopulate the event codes
again in the "original file", and then pull the event codes with 2,
etc...repeating the process for each one. Not elegant....

try to see what specific functions get run that update/modify event info

I assume you are saving the original data after inserting the new event
info into it, and then when you re-open the file, the extra info you put is
gone.

try to talk directly into the event info structure if you are not already,
or make a fascimile in matlab to draw from

try to change the event code info to all String or all Integers

try doing it from the gui, for just a few events, and see if you can make
the events remain. There may be some small not-documented expectation that
you're missing.

I remember having to "repopulate" event fields on the fly as necessary from
a larger index of values. This seemed to work ok through a quick loop, but
had to be done again and again.

I remember that this was due to something about eeglab refreshing the
events according to some consistency it needed.












On Thu, Jan 19, 2017 at 4:30 AM, Nils Hachmeister <
nils.hachmeister at uni-bielefeld.de> wrote:

> Hello everyone,
>
> I have an odd eeglab behaviour, which I don't understand: I added an
> additional field to the events in my data set in order to do some
> additional analyses. This works fine. After adding the field to each event
> in the MATLAB commandline and calling eeglab redraw I can, for instance,
> generate an Channel ERP Image using the gui, sorting trials by the new
> field.
>
> When I want to use this field to select epochs (via the corresponding
> menu), for instance, selecting only those epoch for which the time-locking
> event's new field-value is 0, I get a second data set and all the events
> still have the new field. That's just fine. However, going back to the
> original data set to select all epochs for which the new field's value is
> 2, I notice that the new field is gone for all the events.
>
> First I suspected eeglab would do some kind of consistency test, notice
> that the events have a field which is not present in the urevents and
> correct that. That's why in a second attempt I added the new field to the
> urevents (rather than the events) and then copied the new field to each
> event from its corresponding urevent.
>
> Again trying to select a subset of the epochs, the new field is removed
> from the original data set for both, events and urevents. Since I cannot
> see, how this might be sensible behaviour I suspect that I do something
> wrong, due to my limited understanding of the structure spanning between
> event, urevent and epoch.
>
> Can someone help me?!?
>
> Best
>
> Nils Hachmeister
>
>
> _______________________________________________
> Eeglablist page: http://sccn.ucsd.edu/eeglab/eeglabmail.html
> To unsubscribe, send an empty email to eeglablist-unsubscribe at sccn.uc
> sd.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/20170120/2eb3113e/attachment.html>