Hi Rajani,
It's difficult to say anything with great confidence about the
picture without knowing more about your study, but it doesn't look
unreasonable for a semantic vs size decisions. Other than the
expected fronto-temporal areas, you're getting some anterior and
posterior cingulate activation as well as some IPL -- that's normal.
I'm not entirely sure I understood the analysis, though -- is this
based on a 3rd level RFX (or mixed effects) analysis where the inputs
are the cope1.feat/ dirs from the fixed-effects 2nd level
(subject-level) analyses? If you only did a 2nd level analysis and
included all 1st level runs from each subject, then the activations
are inflated by artificially increasing the DOFs and decreasing the
variance (FFX). Also, your choice of a thresholding method will tend
towards large, spread out activation clusters. Try either using a
higher z-value for the cluster analysis (3.1 or above) or using a
voxel-wise FWE method rather than cluster-wise. But regardless, the
results look plausible.
hope this helps.
Joe
------------------------
Joseph T. Devlin, Ph. D.
Dept. of Psychology
UCL
Gower Street
London, WC1E 6BT, UK
email: [log in to unmask]
|