Hi all,
I ran the FSLVBM pipeline and have noticed the following. I wanted to understand the effect of Age on observed differences and so I ran two analyses, one controlling for age by specifying it in the design matrix and another without age in the design matrix. Interestingly, in spite of using age in the matrix, I did not choose to set up the contrast for age and so my contrast map was looking at just regions where controls>patients. So when we looked at the same contrast map from the two analyses, they both differed in the amount of differences.
Our understanding is that FSL is calculating the influence of age on the matrix in spite of not choosing to see it in the contrasts. Is this correct? If so, how do we know what is the amount of variance that this variable is contributing to the analysis? If we wanted to look at just how the effect of age on the observed differences, what would be a good way to go about it? Any help will be appreciated.
Thank you.
P.S: I am attaching the design.png files to this message.
|