HI,
My experiment involved a single fMR run with 2 different task-block conditions
alternating in time in 6 occasions across the run such as
1 2 3 4 5 6
AB AB AB AB AB AB......
Behaviourally there is a training effect such that the difference in performance between
condition A and B is modulated by training and experience in the task
since it is a single MRI run I though that one straighforward possibility could be to derive 6 copes for the difference between A>B across the 6 pairs above in the first level analyses
and then perform a higher level contrast specifying a linear contrast reduction on those copes
such as 2 1 0 -1 -2
this contrast would include 5 copes (I guess I will have to leave the cope6 out of this analyses)
I am guessing this is fine though this linear assumption may not need to fully fit the behavioural performance
I also want to use ICA and dual regression in these data to assess changes in functional connectivity
the problem with the above model seems that I wont be able to use any contrast files for the randomise part?....
One other option perhaps is to specify the contrasts within the first level to test for changes
within pairs of blocks separated in time? and which in this way I could use for the dual regression
connectivity approach?
such as
1 2 3 4 5 6
A B A B A B A B A B A B
1 -1 0 0 -1 1 0 0 0 0 0 0
or
1 -1 0 0 0 0 0 0 0 0 -1 1
Thanks, David
|