[Eeglablist] pop_rejcont bugs

[Natalie "Tasha" Poppa]

Greetings,

I am using EEGLAB version 14.1.2 with MATLAB 2016b on a Mac. I find a few
issues with the pop_rejcont function.

First, from the GUI pop_rejcont will remove any pre-existing event markers
and replace them only with boundary markers.
I found that someone had described a similar issue in 2014:
https://sccn.ucsd.edu/pipermail/eeglablist/2014/008717.html

Second, from the command line, I do not have this event-marker deletion
issue, but pop_rejcont doesn't actually remove the bad segments that it
identifies either. The function will run apparently as normal, and list the
number of bad segments for removal in the command window, etc. but then
doesn't actually remove them, even though a new dataset is created (i.e., I
end up with the same number of trials and visual inspection of the data
shows obviously bad segments remain).

Under different circumstances I would just use pop_rejepochs to clean bad
segments, but this is not a good solution for my particular analysis
because my epochs are 1 second and ICA must be applied to these data.

If there is no quick solution to the pop_rejcont issue, can anyone tell me
if it would be possible to epoch the data for cleaning, then un-epoch it
prior to ICA (and how)? I tried to figure this out but did not have success.

Thank you!

-Tasha


Tasha Poppa
Doctoral Student
Psychology - Brain and Cognitive Sciences
University of Southern California