Hello!
As the title states, I am attempting a group-level analysis in FSL FEAT using repeated measures, and am not getting an expected output. I hope that someone has experience with something similar!
Our protocol involves two fMRI scans. In the 1st scan, subjects received multiple weighted punctate mechanical stimuli of 3 forces. In the 2nd scan, the same protocol was repeated, except participants received topical capsaicin in a small area before the scan. For both scans, subjects subjectively rated the pain intensity of each force on a numerical rating scale of 0 (no pain) - 100 (most intense pain possible) after the scan. Our goal was to quantify the development of mechanical punctate hyperalgesia: if subjects experienced an increase in pain in 2 out of the 3 stimulus intensities across pre-capsaicin and post-capsaicin scans, then they were labeled as hyperalgesic. All other subjects were grouped as nonhyperalgesic.
For our group-level analysis, we are attempting a hyperalgesia contrast and are only focusing on post-capsaicin scans; each input is a COPE image corresponding to one of the three stimulus intensities for each participant (3 COPE inputs per participant). In the GLM, we have 2 EVs: hyperalgesic and nonhyperalgesic. For each participant, all three of their stimulus intensities are unanimously labeled as one or the other. We have tried this analysis with a cluster Z-threshold of 2.3 and a corrected voxel P-threshold of 0.05.
When we run this analysis, we do not find any significant clusters as we would expect to. Previous analyses of this dataset have yielded significant results and clusters. Has anyone completed a similar analysis design and have experience with troubleshooting?
Thanks so much!