Dear Matthew,
I'm reading this old post and I'd like to ask you further details to better understand the FWE correction applied in fsl randomise
What do you mean by "all family-wise errors are corrected by using the null-distribution of the maximum test statistic across the image"?
I'm familiar with permutations and tstats, but I don't really understand how you use the null distribution of the max tstat (why the max tstat?) in this correction.
thanks for your support,
Marco
|