Hi All,
I want to look at functional connectivity changes associated with taking a drug. I have 3 groups, each taking a different dose of the drug. I specifically want to use Conn to look at functional connectivity changes showing a dose-response effect, e.g., increasing global correlation with increasing dose of drug (Group A > Group B > Group C).

In my second level GUI, I have variables specifying:
AllSubjects
DoseGroupC
DoseGroupB
DoseGroupA

And the contrast weights I put in were:
0
1
2
3

(the zero was intended to set an intercept, and then "1, 2, 3" was intended to model progressively increasing connectivity with each dose group). I really don't think this is the right way to do it, though. Conn also gave me a message that it changed "0, 1, 2, 3" to "1.5, -0.5, 0.5, 1.5". I thought maybe Conn would mean-center the contrast weights, but that isn't mean-centered. Those weights don't make sense to me (if mean-centering, it should be -1.5 for the "All" variable).

Basically I know I didn't do this correctly but unsure what would be correct. Does anyone know if what I want to do is even possible, and if so, what kind of contrast weights I should be putting into the model at the second level?
Thanks very much!
Emily