Hi,
I have 3 groups (A,B,C), each with 10 subjects (A1,A2,...A10; B1,B2,...B10; C1, C2, ...C10). I get the 1st-level smoothness of each subject. I want to use 3dClustsim to calculate cluster confined within a gray matter (GM) mask for higher-level analysis. Each subject's GM was obtained from "segment" using T1 images as input.
For 1-sample t-test of group A, the average GM mask is calculated by averaging the GM of (A1,A2,...A10)(and then binarized), and I can use the average smoothness of (A1,A2,...A10).
For 2-sample t-test of groups A and B, the average GM mask is calculated by averaging the GM of (A1,A2,...A10,B1,B2,...B10)(and then binarized), and I can use the average smoothness of (A1,A2,...A10, B1,B2,...B10).
For an one-way ANOVA to see the effect of group, the average GM mask is calculated by averaging the GM of all 30 subjects (and then binarized), and I can use the average smoothness of all 30 subjects.
For a conjunction analysis to see activation in group A as well as activation that is higher in group A than in group B (i.e. conjunction of contrasts "A" and "A>B"), the average GM mask is calculated by averaging the GM of (A1,A2,...A10, B1,B2,...B10)(and then binarized), and I can use the average smoothness of (A1,A2,...A10, B1,B2,...B10).
Are these statements all correct?
Thanks in advance.
Mark
|