[Eeglablist] Feature Request

Bradley Voytek bradley.voytek at gmail.com
Wed Dec 9 12:24:04 PST 2009


Hey EEGLABers:

Normally I wouldn't do this, but I've had this idea floating around
for a while that seems like it would be very useful but I don't have
the time to learn how to make it happen because it requires some
interactive GUI programming in matlab which I have no experience with.

Here is a mock-up of how I think the GUI would look:
http://darb.ketyov.com/files/Voytek-EEGLAB-FeatureRequest.pdf

This would operate on epoched data, and each trial would be plotted
individually. There would be three rows of boxes aligned to each
channel to allow the user to define the VEOG and HEOG channels. There
is a separate plot of the VEOG and HEOG channels at the bottom, and
this plot only appears after the relevant channels are defined. If two
VEOG or HEOG channels are defined then it plots the difference under
the assumption that the channels are diametrically opposed (i.e., left
and right eye external canthi for HEOG or inferior and superior
channels for VEOG). This lets the user see the time-course of the eye
movements.

Similar to the timtopo function, if the user clicks on the main plot
of the channel data it updates the scalp topography for that time
point (the x coordinate where the user clicked on the data plot). This
will let the user see how eye movements or artifacts might be
contaminating their data.

The first row of boxes will allow the user to selectively interpolate
one or a few channels in each trial separately. This ability becomes
important when you already have few trials per condition and don't
want to throw out a lot of data. Once any interpolation boxes are
check the data plot updates with a preview of how the interpolated
data will look, and the second scalp topography also updates so the
user can compare the data pre/post interpolation.

The user can then click one of three buttons to move to the trial:
make no changes, save changes and continue, or reject trial.

Anyone out there feeling ambitious and willing to help out the EEGLAB community?

::brad


More information about the eeglablist mailing list