Thanks for your help!
Out of curiosity, I ran both of the following contrasts on two data sets:
[.25 .25 .25 .25] and [1 1 1 1]
The resulting maps were exactly the same, with the same z-values. This leads
me to believe that the actual values in the contrasts do not matter, that it is the
relationship between the elements of each contrast that determine the results.
Which begs the question..... I've heard that [-.5 -.5 .5 .5] was not the optimal
way in which to see if the average activation of A&B differs from the average
activation of C&D...... But I don't understand this if [11] is a contrast yielding
the mean activation of two groups....
Any thoughts on this?
Thanks!
|