Dear Sukhi,
>i have data from an n-back task in several different groups of patients and
>healthy control subjects. The design and acquisition parameters of the
>n-back task are NOT identical between the groups. I would like to perform
>a conjunction analysis across some of these groups to extract commonly
>activated areas and to examine the response variation in these areas
>according to variation in genotype.
The simplest way to do this would be to perform a conjunction analysis at the
second level. This would entail creating subject-specific activation
contrasts
testing for the same component and looking for a conjunction over groups in a
second level one way ANOVA. I would not worry much about the variation in
each groups paradigm and the associated non-sphericity issues at the second
level.
To explore between or within group variation in genotype you would simply
enter the genotyping as subject-specific explanatory variables in a second
level
ANCOVA.
I hope this helps - Karl
|