Dear SPM'ers,

We are interested in looking at linear trend within and between groups. I was able to set up the SPM parametric modulation with 1st polynomial expansion on one of the regressors of interest and obtain contrast for linear increase or decrease. It was straightforward to use that at single group analyses to understand the pattern of activation. However, I am confused on how to compare groups. We have three groups and even comparing groups pairwise would go a long way. Is there a way to set this design up in SPM? The contrast I received from SPM parametric modulation are intrinsically directional and so setting up a T contrast of 1 -1 or an F contrast of [1 -1 0; 0 1 -1] doesn't adhere in this case.

Kindly advice!!!

Best,

Manish Dalwani

Senior Instructor

Dept. of Pscyhiatry

Univ. of CO