It will all come down to whether the reviewers let you include non-sig results
On 08/18/2017 04:37 PM, Martin Juneja wrote:
Hello FS experts,
I am correlating some of the behavioral measures with cortical thickness (CT) and cortical gyrification (CG) measures.
- For CT-behavioral: I have some expected results after correcting for
multiple comparisons using mri_glmfit-sim
- For CG-behavioral: Again, I have some expected results (more
expected that CT-behavioral results) - only 3 clusters (sig.mgh result) but none of those survived after correcting for multiple comparisons using PALM, I can't use mri_glmfit-sim because my design is non-orthogonal.
I also learnt from FS discussion forum and literature that CG is already extremely smooth so it would require large area to survive multiple comparisons correction.
I was wondering if I can report my CG results based on this argument. If yes, then any more suggestions/references which I can use as arguments to justify my CG findings, that would really help.
Thanks.
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer