Thanks Rick!

2. I might have been confused when asking. I guess the coefficients of the main effects are the betas. Any contrast that we create will not be a beta per see, but, e.g., a subtraction of two betas (C = B2-B1 ). But please, do explain BLOCK(5) vs BLOCK(5,1) if you have the time - that is another thing I have been thinking about.

Writing this post made me think of another question. What is your take on this:

What is the difference/which method is "best" when you want to do a simple within group t-test to see if a contrast is significantly different from 0?

Is it to test the contrast (C = beta2 - beta1) that you get from glt using:

3dttest++ -setA \
-1 stats.sbj1.tlrc[contrast_sub_brick]
-2 stats.sbj2.tlrc[contrast_sub_brick]
...

Or using the sub bricks that you used ot create the contrast:

3dttest++ -AminusB -setA \
-1 stats.sbj1.tlrc[beta1_sub_brick]
-2 stats.sbj2.tlrc[beta1_sub_brick]
... \
-SetB \
-1 stats.sbj1.tlrc[beta2_sub_brick]
-2 stats.sbj2.tlrc[beta2_sub_brick]
... \

The latter pops out when using one of the afni functions to create the ttest scrip for you (don't know which one though - I never used it).

Thanks!