> 2. Contrast maps. I have two conditions in the experiment and I want to contrast the brain
> activation across them. What should be the appropriate extent threshold to apply to the
> contrast? Should I first subtract the residual of one condition from the other's, then use that
> difference timeseries as input to FWHMx and proceed as usual?

->Even though not ideal, the current practice is to estimate the FWHM values based on the residuals from the individual subject ->regression analysis. And the FWHM values might be automatically provided if you use afni_proc.py

Thanks Gang for your response. I use afni_proc.py and try to stick with guidelines provided by you guys. Please bear with me for a moment as I try to explain myself.

My understanding is that the errts file contains that part of the BOLD signal that cannot be explained by the model regressors. This residual is what is used to estimate the cluster extent threshold (through FWHMx and 3dClustSim). If I am contrasting two conditions to find statistically significant differences between them then I should also contrast the residuals in the two conditions. It may well be that the residuals in the individual conditions are large but the difference between them is small. And since I'm interested in the contrast anyway, I will use the difference in residuals as the residual for the contrast.

Does this sound like a reasonable thing to do? For the purpose of contrasting conditions, is it not a better approach than the current practice is to estimate the FWHM values based on the residuals from the individual condition regression analysis? Or am I complete off the mark?

Best,
Gaurav