Camille Maumet

This paper evaluates eight meta-analytic approaches for image-based fMRI meta-analysis using simulated and real data, finding that combining contrast estimates with random-effects models or non-parametric statistics provides the best accuracy.