I have a fmri design with five test level in a run,A B C D E (there are six runs in a session ).and one subject take the task fmri five times,I only have 4 subjects and 4*5 sessions results.So I want to compare the difference of test effects( A B C D E)and session effects.I was wondering how one would conduct a 2-way repeated meassures ANOVA in such situation? Or there is a effective way to compare the effect of session and test(A B C D E)? Thanks for your help!
I have trouble understanding your data structure. You mentioned 5 tasks, 6 runs, unknown number of sessions, 4 subjects. What do you mean by “one subject take the task fmri five time”? Why are there “4*5 sessions”?
I am so sorry about my description,'one subject take the task fmri five times ’ means the number of sessions is five, what is more , the five task fmri is all the same,and we collected this same data structure on four people,so there is 4*5 session.
If I understand your data structure accurately, you have
Does it mean you have 25 effect estimates (beta values) per subject, leading to a 5 x 5 within-subject (repeated-measures) ANOVA? If so, you can use 3dMVM (or 3dANOVA3 -type 4). The number of subjects could be a concern.
Thanks for your suggestion.In fact ,I want to measure the extent of reliability of the bold response,So the effect estimates (beta values) of the five task fmri session were all the same.I have only 5 effect estimates (beta values) per subject.And the big problem is that the number of participants is four.If I calculate the mean 5 effect estimates (beta values) per subject for two way anova,it will lead to 5*4 two way anova. What is more,I want to compare the difference of effect of session and 5 task beta values between subjects. It is so bad that my repeated 5 task fmri is useless! how can I find a effective way to analyze this data structure? Thanks for your help very much!
I still have trouble understanding what you’re trying to achieve. If I have to guess, I would suggest that you obtain session-level effects for each subject by either (1) creating 5 separate regressors for each session, or (2) analyzing each session separately.