Hi SPMers,
I did an fMRI study that involved a group of subjects doing a task that
that varied along factors A and B and along factors 1 and 2, and I'm
interested in the interaction effect. To address this, I ran individual
t-tests of the type [(A1-A2)-(B1-B2)] and [(B1-B2)-(A1-A2)] and
then did a second level random effects analysis.
However, the person reviewing my paper is insisting that I do an ANOVA,
and claims that the t-tests I ran are not true interaction maps, and that
they will identify brain regions showing two main effects and no
interaction. Here is his/her example:
"... if one were to apply the following values to each condition
(3-2)-(2-1), one might conclude that this region showed a main effect (A>B), a
main effect (1>2), and no interaction. Alternatively, if the values of
a voxel were (3-1)-(1-1), this might yield main effect (A>B), main
effect (1>2), and an interaction. In this second example, the voxel only
shows a A vs B difference for the stimuli of type 1 (which is the goal of
the study). This is not true in the first example. In the first example,
one sees the same A vs B difference for stimuli of type 2. The
interaction/subtraction image reported in the paper will capture and
confound both patterns of activation."
If this person is right in demanding an ANOVA, could you please explain
how to perform one so I can see interaction effects? (I can look at the
F-values, but I don't see how I can isolate interaction from main
effects).
If the person is wrong (as far as I can tell, s/he is), how do I convince
him or her that what I've done is valid? This is the second go-round with
this reviewer and s/he won't okay the paper until I've done the ANOVA
even though I've already explained that the t-tests I've run address the
same concepts.
Thanks!
Karli
|