Dear SPMers,
I want to compute first level contrasts for a repeated measures ANOVA
(partitioned errors). The fMRI paradigm (block design) consists of two
conditions (0-back/control vs 2-back/ task) – each participant was measured
twice (t1 / t2)
On the first level I compute the following t-contrasts:
(Conditions: 0-back_t1 / 2-back_t1 / 0-back_t2 / 2-back_t2)
-1 -1 1 1 (main effect time)
-1 1 -1 1 (main effect task)
Is it correct to compute session specific task-effects in the same design
matrix by adding the t-contrasts:
-1 1 0 0 (task at t1)
0 0 -1 1 (task at t2) ?
Would this lead to different results than analyzing t1 and t2 in separate
designs? (conditions: 0back_t1 / 2back_t1; contrast: -1 1)
Thanks in advance,
Ben Becker
|