Back to the main page.

Bug 1960 - some 'rising' events are not found for NXE data

Reported 2013-01-26 17:39:00 +0100
Modified 2014-01-29 17:14:35 +0100
Product: FieldTrip
Component: fileio
Version: unspecified
Hardware: PC
Operating System: All
Importance: P3 normal
Assigned to: Eelke Spaak
Depends on:
See also:

Renate Rutiku - 2013-01-26 17:39:55 +0100

Nexstim data has square pulse triggers. So there are two events 'rising' and 'falling' and there should be an equal number of both events. But for a few datasets ft_definetrial() fails to find 1 or more 'rising' events. There is no equivalent problem for 'falling' events. An example dataset and the corresponding script can be downloaded here:

Renate Rutiku - 2013-01-27 10:21:39 +0100

Correction: now I also found the same problem for some 'falling' events.

Robert Oostenveld - 2013-01-28 11:32:10 +0100

I copied the data and example script from dropbox to our network drive. mac001> pwd /home/common/matlab/fieldtrip/data/test/bug1960 mac001> ls -l total 231912 -rw-r--r--@ 1 roboos 655 118734720 Jan 28 11:31 KI_12_1_4.nxe -rw-r--r--@ 1 roboos 655 932 Jan 28 11:31 example_script.m

Eelke Spaak - 2014-01-29 13:49:19 +0100

Hi Renate, Apologies for taking so long to get back to this. When looking at your data, I only see triggers in the TRIG1 channel. These all seem to have the same value. Could you give me some more information what the exact voltages in the trigger channel(s) mean, and how I can verify that the correct amount of events indeed is 90 per type? Thanks! Eelke

Renate Rutiku - 2014-01-29 17:14:35 +0100

Hi Eelke, thank You for getting back to me. The triggers do indeed all have the same value and the exact voltages do not have a meaning. Events in the experiment are marked by a square pulse in the voltage. In our experiments the rising edge of the pulse is usually aligned with the relevant event in the experiment (say stimulus onset). We then use a separate logfile of the experiment to match the triggers to different conditions and so on. This specific dataset should have 90 triggers because the corresponding experimental block had 90 trials and because the nexstim software counted 90 triggers during recording. Hope my answers are of some help. Please ask if anything is still blurry. Best, Renate