I encountered an error in while running contrast manager during 1st-level analysis. Turns out that in one of the sessions (I have 6 sessions total), this one particular subject's performance accuracy was 0 for all 4 trials belonging to one of the conditions (there are 3 conditions). Other sessions do not have this issue.

For incorrect trials, I model them as a separate condition. However, since there are no correct trials for this condition in the aforementioned session, the onsets were all blanks. Naturally, SPM gave me an error. I thought I'd deal with this by replacing these missing onsets with time = 1TR before the end of the run. Apparently, SPM doesn't like this either and won't run the contrast manager. Does anyone have any suggestion how to handle this?