On 2016 Aug 01, Joaquim Radua commented:

Re: the previous comment, I think there may be some unfortunate confusion. Raw p-values of current voxelwise meta-analyses have not the same meaning as usual p-values because they are not derived from the usual null hypothesis (“there are no differences between groups”), but from another null hypothesis (“all voxels show the same difference between groups”). Thus, up to the moment one of the only ways to "approximately" know if the results of a voxelwise meta-analysis are neither too liberal nor too conservative is to compare them with the results of a mega-analysis of the same data, and that's what it was done.

On 2016 Jul 20, Christopher Tench commented:

Here the authors have swapped a method that employs disciplined control of the type 1 error rate (FDR) for a method (SDM) that employs no disciplined control in order to demonstrate there are regions of true positive atrophy in narcolepsy. It does not, however, provide any evidence of this. Arbitrary p-value thresholds do not work in neuroscience. They most certainly do not make a meta-analysis, which demands rigorous statistics. There is no evidence of consistent reporting of regional GM atrophy from this study,

On 2016 Jul 20, Christopher Tench commented:

Here the authors have swapped a method that employs disciplined control of the type 1 error rate (FDR) for a method (SDM) that employs no disciplined control in order to demonstrate there are regions of true positive atrophy in narcolepsy. It does not, however, provide any evidence of this. Arbitrary p-value thresholds do not work in neuroscience. They most certainly do not make a meta-analysis, which demands rigorous statistics. There is no evidence of consistent reporting of regional GM atrophy from this study,

On 2016 Aug 01, Joaquim Radua commented:

Re: the previous comment, I think there may be some unfortunate confusion. Raw p-values of current voxelwise meta-analyses have not the same meaning as usual p-values because they are not derived from the usual null hypothesis (“there are no differences between groups”), but from another null hypothesis (“all voxels show the same difference between groups”). Thus, up to the moment one of the only ways to "approximately" know if the results of a voxelwise meta-analysis are neither too liberal nor too conservative is to compare them with the results of a mega-analysis of the same data, and that's what it was done.