Dear FSL group,
I am comparing activity associated with verbal and spatial working memory. In FEAT, I ran a first-level analysis of each run, and a session-level analysis of all runs for each subject. I now want to compare verbal and spatial associated activity on a group level.
Which of the following approaches is preferred from the standpoint of statistical validity:
#1. Create contrasts (i.e., Spatial > Verbal, Verbal > Spatial) at the subject level and then simply run a group average, or
#2. Create group average of Spatial and a group average of Verbal and then contrast them?
Any feedback would be much appreciated. Thanks.
Cheers,
Pete Fried
Dept. of Anatomy & Neurobiology
Boston University School of Medicine
|