3dFWHMx

AFNI Message Board
1

Hello,

I am attempting to use 3dFWHMx for the first time to get input for 3dClustSim for an analysis examining the effects of three conditions. I have run 3dMVM, masked my data, and now am attempting to use the 3dFWHMx -acf option. My question is - I have three comparisons within my 3dMVM (pulling in three seperate errts_GAM files), would I need a different -acf estimate for each of my three comparisons? I have run all three of my conditions separately through 3dFWHMx and the average smoothing estimates from my output (to go into 3dClustSim) are all slightly different. Why would this be? Would it be appropriate to use the results of my three separate smoothing estimates and average these together to use for 3dClustSim?

Here is the command and options I am using:
3dFWHMx -mask /avg_anat_masked_resampled+tlrc -acf NULL -input /${subj}_CONDITION_errts_GAM+tlrc >> FWHMx_smoothEST_RunxContext

Thank you in advance!

2

Would it be appropriate to use the results of my three separate smoothing estimates and
average these together to use for 3dClustSim?

Such an approach should be OK as an approximation since the cluster threshold from the ACF method with 3dClustSim is a rough estimate anyway.