Hi there,
I recently submitted a paper where I ran static ICA on resting state dataset and I selected 40 components. The resultant components (plus their variability from the temporal properties tab) were correlated with behavioral measurements. As an example, I had a significant result in the FP within one component (picture attached). This map was cluster thresholded at p≤0.05, voxel thresholded at p<0.001 (FDR-corrected) using Gaussian Random Field Theory. The reviewer is concerned I did not properly correct for the # of multiple comparisons, can you please advise?

Reviewer comment: "The authors also chose 40 ICs in their fastICA pipeline. It's indicated that "for each network" a voxel and cluster threshold was used. This is broadly appropriate – but again the concern about multiple comparisons comes up. To illustrate, my understanding is that the authors have 40 ICs, 4 IRI subscales and 5 MAIA subscales, for a total of 800 potential effect size evaluations (not including any interactive effects). Plus, they evaluated both connectivity and variability – so 1600 potential effect size evaluations. Plus they evaluated IRI-Total as a separate measure (so 1800 potential effect size evaluations). If I'm wrong about this, I'd be happy to have the authors point this out to me."