Hello fellow SPMers,
In my study I have 14 subjects who underwent two fMRI sessions in which
the same condition was compared to baseline. In SPM2 I designated the
two sessions as replications and used the contrasts [1,0] for session 1
and [0,1] for session 2, for each individual first-level analysis. I
subsequently did a 2nd-level group analysis for each session and
observed less activation in the second session. How can I best
determine if the difference between sessions is statistically
significant? Is is correct to do this either with a paired t-test or a
one-way (within-subjects) ANOVA with two conditions (ie. the [1,0] and
[0,1] contrast images)? And then I would enter either a T- or
F-contrast of [1,-1] to identify areas where session1 activation was
greater than session 2 activation? Which is more appropriate, a
T-contrast or an F-contrast?
Thanks for your help,
Daron
___________________________
Daron Owen
Graduate Student
Medical Biophysics
University of Western Ontario
London, Ontario, CANADA
|