Dear all,

I conducted an experiment and analyzed the data by SPM, but an unexpected problem happened.

The detailed processing are as follows.

1. The preprocessing are establised as the manual.

2. In the experiment, there are three conditions, they are a, b, c. When I define the GLM, I re-organise them as e=[a], f=[a b], g=[a b c].

   Then I set three T-test as 1 0 0, 0 1 0, 0 0 1 to estimate the effects of e, f and g.  

   It's expected more activation areas in the second contrast (f, 0 1 0) than that in the first one (e, 1 0 0 ), and the most in the third contrast.

   But the results seem like nonsense. Because we get the most in the first contrast and then the second one and the third one.

 

I think there's no problem in the analysis procedure, but the results are unexpected. Would you mind give me some commets and suggestions on the unexpected results and the data analysis procedure?