Hi,
I have a question regarding higher-level analysis with added covariates of
interest. I have a group of 13 scanned before and after a clinical
intervention, and I want to know if the difference in activation is related
to the reduction in pain scores at follow up. My query relates to how to
input the covariates and how to interpret the maps.
I have two events in my lower level design:
A 1 0 0 0
B 0 1 0 0
A-B 0 0 -1 1
B-A 0 0 1 -1
At higher level, I want to look at the within group difference (pre- minus
post intervention, and post minus pre-intervention):
1 1 1 0 0 0...
1 1 0 1 0 0...
...
1 -1 0 0 1 0...
1 -1 0 0 0 1...
I understand that the contrasts here would be 1 -1 and -1 1.
I am interested now in what activation co-varies with the pain scores.
Should I simply demean the scores or do I need to orthogonalise them? Would
the model look like this for example:
1 1 1 0 0 0... 2
1 1 0 1 0 0... 3
...
1 -1 0 0 1 0... 3
1 -1 0 0 0 1... -2
with the following contrasts:
Pre-Post 1 0 0 0... 0 0
Post-Pre -1 0 0 0... 0 0
Pre pain 0 0 0 0... 0 1
Post pain 0 0 0 0... 0 1
Pre-post pain 0 0 0 0... 1 -1
Post-pre pain 0 0 0 0... -1 1
Are theses contrasts going to tell me if there is any covariance between
changes in activation and changes in pain scores at follow up?
Thanks,
Kate MacIver
Pain Research Institute
Liverpool
|