Dear SPMers
I am revisiting some old data that was analysed using SPM99. One
significant change in SPM99 that I have noticed is that after a path was
installed sometime last year to enable the FDR and FWE options in
corrected height threshold, the results obtained by choosing the
uncorrected height threshold option is also SIGNIFICANTLY different than
the original version. The p values have become far less significant - is
this supposed to be the case or is there a bug? If the former is true,
what about all the results that have been published in the literature
using the old p-values?
Thanks
|