Are second level GLM p-value images one- or two-tailed in Nilearn?


I can’t seem to figure out whether the p-values in the Second_Level_Model.compute_contrast() nilearn function are one- or two-tailed. I think they are be one-tailed based on an example tutorial I found, but can’t seem to definitively confirm this in the docs. Does anyone here happen to know?

Apologies if I missed this elsewhere (either here or in Nilearn’s documentation).

Thanks in advance!

my guess would be that it is one tailed (as most things in neuroimaging) but I need to double check that

if that helps: those 2 functions have a “two sided” boolean arguments

1 Like

Thanks for your help!

1 Like