Dear SPM experts,
The results from my second-level analysis (a paired samples T test based on 20 subjects) show that p-values for cluster-level inference are all NaN's for "corrected" and 0.000 for "uncorrected"; I have attached a screenshot of the table showing these. Could anyone please tell me what these exactly mean? I am also wondering why the p-values for voxel-level inference are identical across all significant clusters. I looked into the first-level con images, and they all look fine to me. It appears that there was a similar question posted on this forum a while ago, but I couldn't find definite answers to this. Many thanks in advance!
Cheers,
Yuta
|