Dear Mike,
The easiest thing to do is use a higher height threshold. There is
nothing magical about 3.1 -- it is a completely arbitrary decision --
so if it gives you such a large extent and you think the result is
reasonable, then try a higher value. Note that at some point
(normally in the 4-5 range), this becomes equivalent to a correct
voxel height (i.e. when the cluster size needs to only be a single
voxel to reach a corrected level of significance).
There is no way to guarantee that any given cluster will be limited
to a single brain region since there is no a priori method for
identifying a "single brain region." Within a cluster, you can get a
set of "local maxima" which may be useful to you for identifying
anatomically distinct peaks.
Hope this helps.
- Joe
-------------
Joseph T. Devlin, Ph. D.
FMRIB Centre, Dept. of Clinical Neurology
University of Oxford
Headley Way
Oxford OX9 3DU
Phone: +44-(0)1865-222-494
Email: [log in to unmask]
|