 |
 |
Dear
SPM list
We
have a question concerning the interpretation of PPI results in a between-
groups comparison.
We
evaluated 2 groups of subjects
(patients and controls) performing a fMRI study. The study included two cognitive tasks,
A and B, and a control task, C. We were interested in studying the
differences in PPI between task A and task B.
We
made individual PPI analysis to explore which regions showed a different
interaction with the amygdala consecutive to the change in cognitive factor (to
simplify “task A: cognitive factor present” to “task B: cognitive factor
absent”)
Then
we made second-level analysis to evaluate both individual patterns of PPI in
each group and between-groups differences in PPI.
Concerning
the individual patterns of PPI, we assumed that the PPI analysis would show, for
each group, regions which would have an interaction with the amygdala modulated
by the "cognitive factor A".
Our
problem is how to interpret the between-groups
results.
Here
is a specific example:
We
find that region "R" is not present in the pattern of the patient group, but is
present in the comparison Patients>Controls.
Does
this mean that "In the patient group, there is no change in the interaction
between region R and amygdala when factor A occurs (= "there is no difference in
the interaction of R and amygdala between tasks A and B"…?). However the
interaction between amygdala and R when factor A occurs is greater in patients
than in control subjects" ?
In
other words, patients, even though having no change in interaction between
region R and the amygdala when switching tasks, still have, compared to control,
a superior “amount” of interaction in these two regions in task
A.
We
would be very grateful if someone could tell us if this interpretation is
accurate and if not, explain mean the meaning of the between-group
results.
Thank
you in advance