Hello,

We wanted to do a 2 x 2 seed to voxel analysis  in conn and have the groups suicide, non-suicide, genotype1 and genotype2. To look at whether having genotype1 vs genotype2 has any effect on interaction between suicide and non-suicide.
listed in order as covariates:
suicide, nonsuicide, genotype1, genotype2
in the subject contrast,
what is the difference between using [1, -1, 1, -1] and [1, -1, -1, 1] because they come out with 2 different results that aren't symmetrical. additionally, is this the correct way to setup a 2 x2 analysis?

Thank you,
Tien