Hi Benjamin

Sorry, right now the only multiple comparisons corrections implemented in lme are the original Benjamini and Hochberg (1995) FDR procedure (lme_mass_FDR) and a more recent and powerful two-stage FDR procedure (lme_mass_FDR2):

Benjamini, Y., Krieger, A.M., Yekutieli, D. (2006). Adaptive linear step-up procedures that control the false discovery rate. Biometrika, 93, 491-507.

In my experience, using cross-sectional data, this procedure is as powerful to detect effects in neuroimage data as alternative cluster-wise corrections with strong control of the family-wise error rate (FWE). As explained at the bottom of the wiki page you can use mri_surfcluster on the resulting sided p-value maps together with the FDR p-value threshold to obtain the detected clusters and their anatomical coordinates.

Having said that, it would be great if we could have an implementation of any multiple comparisons correction with strong control of the FWE for lme (FDR procedures only provide weak control). The problem is that Monte Carlo simulation-based corrections implemented in Freesurfer are not feasible here because the linear mixed effects model is too computationally expensive. Certainly, other powerful but computationally expensive statistical models are expected to become widely used by the neruoimaging community in the next few years. As with lme, simulation-based procedures will not be feasible for them. The only solution I see is the application of random field theory (RFT, cluster- and peak-wise) correction procedures. The problem is that the new lme spatiotemporal model has dramatic spatial variability for the degrees of freedom of the tests. It must be noted that transforming the F-statistics to Gaussian Z-scores is not a solution as Gaussian RFT is heavily based on the assumption that the entire random field follows a multivariate normal distribution. This assumption is violated by the previous transformation.

In addition, we need somebody (a mathematician) with the enthusiasm, time and support to go over the RFT theory for Riemannian manifolds (look at the attachment). In my case, I've been very busy implementing new statistical models and have no time right now.

Best regards
-Jorge