Hello,

I am conducting a seed-to-voxel functional connectivity analysis in CONN (resting-state fMRI, group comparison). My significant clusters are large (e.g., 1727 voxels) and the CONN output lists anatomical sub-regions from the Harvard-Oxford atlas.

Example result of that cluster:

346 voxels (20%) covering 61% of PT l 
227 voxels (13%) covering 58% of pSTG l 
158 voxels (9%) covering 16% of CO l 
...
59 voxels (3%) covering 19% of HG l 
51 voxels (3%) covering 14% of PP l 
17 voxels (1%) covering 4% of aMTG l 
10 voxels (1%) covering 0% of PostCG l 
9 voxels (1%) covering 1% of aSMG l 
2 voxels (0%) covering 0% of Thalamus l 

My question is: Is there a recommended minimum threshold for reporting these sub-regions in a results table? For example, should I only list sub-regions above a certain percentage of the cluster (e.g., ≥3% or ≥5%), or above a minimum voxel count?

Do you have suggestion should I report regions comprising spesific percentage (for example (>5) of the cluster or with substantial anatomical coverage? Or should I just report the all regions in the cluster?

Thank you in advance!