[Mne_analysis] .ave file handing of binary-coded triggers

Kanal Eliezer ekanal at cmu.edu
Mon Oct 5 12:32:48 EDT 2009
Search archives:

Hello folks -

I have an experiment in which I have a few trial-relevant variables I  
would like to store on the MEG data channel on each trial. The optimal  
way for me to store this would be to encode them using some sort of  
binary coding scheme, convert that to decimal, and store the decimal  
value as a voltage on the data channel. So, if I three yes/no  
variables, I could encode the first bit to the first var, the second  
to the second, and the third bit to the third. (I would leave the  
fourth always on so that I always have some non-zero value to write to  
the channel.) Given a trial of yes/no/yes, I would get the binary  
number 1101 = 13, and store a value of 13 on the channel.

My question is whether or not there is a way I can describe this sort  
of setup in an mne .ave (and .cov) file. The alternative method which  
I used previously of sending numerous consecutive triggers to the data  
channel has the significant drawback that each trigger takes ~15 ms.  
to transmit. Thus, any analysis of the third trigger will have to take  
a 45 ms delay into account from the onset of the actual event, which  
can be confusing to analyze and even more confusing to communicate.  
Thanks -

Elli Kanal

More information about the Mne_analysis mailing list