I would like to identify voxels that are NOT activated by a given contrast (in individual subjects). I do not see any way out, except resorting to a Bayesian approach. We moved from SPM (which features 1st level Bayesian tools, whatever they are worth) to Nilearn, and if possible would prefer to stay in the Python world. Would anybody know
1/ whether there is an alternative to the full reanalysis of the raw data with Bayesian methods
2/ whether Pyton tools exist that could do the trick (with minimal effort…)
Thank you in advance for any suggestion
Laurent