sdm-help-list > assume error without being conservative Vs conservative: Conjunction analysis thresholding
Showing 1-1 of 1 posts
Display:
Results per page:
Nov 8, 2022  02:11 PM | Molly Rowlands
assume error without being conservative Vs conservative: Conjunction analysis thresholding
Hi, 

I wanted to understand more about the differences between applying the "conservative" or the "assume error without being conservative" options when thresholding a multimodal conjunction analysis.

I understand the simple overlap option simply retains all voxels that had significant p values in both meta-analyses. I also understand that "assuming error" means you assume that the said p value calculations contain some amount of error, and so you adjust for this likely amount of error in some way in the conjunction, instead of simply overlapping. I wondered if there is more information on this, and especially on the difference between "assuming error without being conservative", and "conservative". 

Thank you! 

Best, 
Molly