Claire,
For your first question, 3dMEMA is unable to handle your complicated situation with quite a few explanatory variables. Unless the number of trials is dramatically different between the two conditions (e.g., one condition is 3 times more than the other), I would not worry too much about the reliability difference between the conditions.
Your second question: are these your explanatory variables?
Group – 2 levels, between-subjects factor
Condition – 2 levels, within-subject factor
Time (scan 1 vs scan 2) – 2 levels, within-subject factor
Voxel-wise covariate ASL: between-subjects variable?
Component - number of beta values from TENT
I still recommend 3dMVM: Treat Component as a within-subject factor, and use -vVars to code the voxel-wise covariate ASL. One question about centering: Do the two groups have roughly the same ASL value at each voxel? We can discuss the issue as to how to interpret the results once you have the output.
Gang
Edited 1 time(s). Last edit at 11/26/2014 01:01PM by Gang.