Hello,
I am analyzing a data set with 24 participants who each have 4 runs (sessions). In the past, I have always run a first-level analysis for each session for each participant and then run an individual fixed effects second-level analysis for each participant. I then feed the results of each second-level analysis into a third-level Flame analysis that spans across participants.
I've noticed in my data that if I run the second-level analysis with all subjects in a single second level fixed effects model and a separate EV for each participant (as in the tutorial on the FSL site), I get identical second-level COPEs to the individual subject method. I also get identical results when I feed the results from both second-level approaches into FLAME 1 for the third-level analysis. However, when I feed the second-level results into FLAME 1+2, the thresholded z-stat results are highly similar, but slightly different depending on which second-level results I choose. In particular, the cope outputs are identical, but the tdof files differ for some voxels.
Is it normal that this second level choice should change the results of Flame 1+2? If so, could you please point me in the direction of why that is the case?
Thank you in advance for the help!
Best,
Anthony
Anthony Sali
Assistant Professor
Wake Forest University
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|