Hello!

I am wondering if someone could shed some light on the difference between conducting a meta-regression vs. a mean analysis with a covariate. The covariate variable is a 0/1 binary (calling it X).

I at first thought these would yield similar things, though they do not, and the meta-regression yields similar maps to conducting a subgroup analysis with the X variable as 1 (though not exactly the same). 

Is the meta-regression looking at an interaction between the 0/1 binary variable X and the main independent variable on brain regions (i.e., does X act as a moderator)? While the mean analysis with a covariate is just looking to see if the main independent variable relates to activation in certain brain regions while controlling for X (i.e., no interaction)? Hope this makes sense, happy to clarify. 

Thanks so much!
Anna