Dear SPMers,
We have administered an fMRI task in a single patient to examine pre- post-treatment perceptual differences. We applied a perceptual reception task (including happy, neutral, fearful faces + houses as non social stimuli) twice (pre- post-treatment). Comparing the between-session differences for all stimuli (all faces & houses) in the patient I found some interesting activity increases & decreases at pre-treatment. Now I want to use PPI to test for between-session differences in connectivity of the regions which show decreased / increased activity at pretreatment. I’ve performed a PPI analysis according to the SPM5 manual (used SPM5 to analyze the data) – that means I’ve extracted VOIs for the regions with between-session differences & created PPIs for these regions.
My main questions are:
1. For each region I created the PPI variable & included all 4 conditions and weighted them with ¼ each. Is this valid?
2. In the PPI GLM analysis I included both sessions which gives me a design-matrix with 6 regressors:
PPI.ppi session 1 / PPI.y session 1 / PPI.P session 1 / PPI.ppi session 2 / PPI.y session 2 / PPI.P session 2
I’ve computed between group differences in connectivity using the contrasts
a) 0 0 1 0 0 -1 (stronger connectivity pre-treatment)
b) 0 0 -1 0 0 1 (stronger connectivity post-treatment)
Are these the correct contrasts to test for between-session differences in connectivity?
Thanks in advance & best regards
Ben
|