QUESTION: We know studies of whole-brain analyses with no statistically significant coordinates can be enrolled to conduct an ES-SDM analysis. Could you please so kind to tell me how does the ES-SDM take null findings into account?

Findings from studies with null findings are recreated with effect size and variance maps as any other study, with the only difference that all voxels in the effect size map are estimated to have a null effect size. These null effect sizes are included in the random-effects meta-analytic models as any other effect size, thus modifying the meta-analytic effect size and variance.

So would you just enter a blank file for that study?

Yes  ;-)

Dear Dr. Radua,

Thank you for this incredibly helpful software and extensive documentation and tutorials. 

I am running a meta-analysis with 5 studies, and 1 study has null results so I created a blank text file as directed. I noticed that when I run all 5 studies, and then when I run only 4 studies excluding the null study, the results are identical. Am I doing something wrong? How are null results handled, exactly? I expected the Z values to decrease when the null study was included, but all output is exactly the same.

Many thanks for your help. 

Best,
Caitlin

Dear Caitlin,

This is very strange, it is hard to believe that z-values are identical. Can you send me the text files by email?

Joaquim