Dear Ward and others,

Given that there were 9 subjects. Each subject underwent a scanning session, during which 2 types of stimuli (say, A and B) were given in event-related way. We want to test the difference between the two types of stimuli. Please tell me which of the following way is more reasonable.

(1) With 3dDeconvolution, we got the partial F values of each type of stimuli from each individual. Then we use the F value to perform a paired t-test across subjects. I think this is something like Doyon et al (PNAS, 2002, 1017-1022) did. In their study, the group fMRI results were obtained from each session by averaging the individual Z score maps. Then comparisons of the functional data between two sessions were assessed statistically (P < 0.05) by subtracting z score maps using the formula (session b - session a)/square root of 2.

(2) With 3dDeconvolvetion ¨Cglm, we could get the individual t map by comparing the area under curve of one kind of task with that of another. Then we can do a one sample t-test to test whether the mean t value across all the 9 subjects is equal to zero.

Thanks a lot.

Yufeng