[Eeglablist] best way to plot marked data

[Tarik S Bel-Bahar]

Hi Nicole, hoping all's well in Eugene! Some quick thoughts below that may
be of use...

Regarding plotting times that were marked for rejection. James' solution
seems cool, this is certainly something many researchers do, both for
checking data and for publication/visualization. Overall, note that one
need only have worst/baddest periods removed before ICA, and then have a
round of artifact-detection after ICA-cleaning.

1. After markup, but without rejection, one can go into the eeglab reject
variable and get info about the times that have been marked. That can be
saved out, and my understanding is that should be same as the info in the
TMPREJ you referred to.

2. Within eegplot, there are ways to plot various time windows, see the
specs on the function. essentially, as you know, when you make some marked
periods in continuous data via eegplot, save and close the marks, and then
re-open eegplot, you can see the marks there. So the idea is to re-open
eegplot and "feed it" the specific marked times you want to show.

3. One way is to simply do something like the following, based on examining
eegh output, help for functions you are using, and some matlab coding.
A. Get the start and end times that were marked into a a list. (by saving
out of the eeglab structure the marked periods after they are marked)
B. Plot the full time of one or more channels using normal plotting
function in matlab. (or for example, time-frequency plot of the data)
C. Plot lines on the image that show the start and end of the marked
periods.
D. Repeat with different colors for different coders, artifacts, etc..

4. Caveats, make sure your "marked times" are directly related to the "time
in continuous data" for accurate plotting of the marks.

5. You may also be interested in a function like "add new events"
(previously mentioned on the list, googlable, in eeglab distribution. This
would allow you to make a series of events which are plotted in eegplot,
and the events you make could be built from the marked times your
artifact-detectors have developed.

6. If you haven't had a chance to, try googling your topic + eeglablist,
you may find some interesting past discussions of same in past eeglablist
posts.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://sccn.ucsd.edu/pipermail/eeglablist/attachments/20180323/28eb312f/attachment.html>