[Mne_analysis] Question about digital trigger channel for Neuromag-122 data

Matti Hamalainen msh at nmr.mgh.harvard.edu
Sun Jan 4 10:44:18 EST 2009
Search archives:

Hi Yang,

On Jan 4, 2009, at 8:58 AM, Yang Zhang wrote:

> I am looking for help on how to set up the digital trigger channel  
> correctly for analyzing Neuromag-122 raw data using MNE.
> I did trial-and-error in changing STI 001 ... 006, 015, 016,  to STI  
> 014 one at a time,  but each time, the event numbers were incorrect  
> (over 20,000 events were found) in doing mne_process_raw.  
> mne_fix_stim14 did not help remove the problem.
> I tried to senenv for trigger channel to be one of the STI channels  
> in the raw data, but the events were not detected at all.
>
> Your advice is greatly appreciated.
>
> ~ Yang

On Neuromag-122 systems the digital trigger channel does not exists  
but rather there are separate channels (STI 001 ... STI 006) which  
together give the trigger information. Furthermore, these channels are  
really analog channels and must be thresholded to detect the triggers  
correctly. This is the reason of getting 20,000 events when you  
experimented with different channels.

Fortunately, there is already a solution (a little bit cumbersome).  
Download the newest MNE Matlab toolbox (MNE-dist-matlab-2.6- 
beta-081102134414.tar.gz) from our download website. This new version  
contains a function called mne_make_combined_event_file which will  
combine STI 001...STI 006 and creates both a text and binary format  
event files. Once the binary event file (<name>-eve.fif) is in the  
same directory as your raw data file, it is consulted for trigger  
events and everything should work correctly (the --digtrig setting is  
irrelevant).

I hope this helps.

- Matti





More information about the Mne_analysis mailing list