That's going to cause problems eventually. The easiest thing to do is to simply redefine some of the non-error trials to be errors. If you have a lot of trials, then losing 3 will not be a big deal. You only need to redefine one of each type across all runs. It's a little hokey, but it solves the booking problem.
doug
Dan Dillon wrote:
Dear FreeSurfers,
We are running a paradigm that occasionally elicits erroneous responses from subjects. In our mkanalysis step, we have 13 conditions--10 are legit, and codes 11, 12, 13 correspond to various kinds of errors. Thus, we only use the first 10 codes in mkcontrast, etc. However, occasionally we have a subject who makes no errors. How should we account for this? Is it okay to just change the # of conditions in the mkanalysis .info file (from 13 to 10) and then run selxavg on these folks? Can we then average their data in with people who have all 13 conditions?
Thanks!
Dan Dillon
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer