Dear All,
in the light of recent discussions I am quite hesitant to ask this
question, but anyway: apart from running a Monte Carlo simulation on all
of the data, is there a way to estimate the likelihood of a cluster of a
given size (eg, 200 pixels) arising by chance if an uncorrected p-value
(eg, 0.001) is used in exploratory analyses ? I somehow hope that if you
know the search volume, there should be a way to estimate this.
Also, would it be justifiable to use the correlation of the extracted
voxel-timecourses with the parameter of interest as an additional
criterion ? IE, if the timecourse in the cluster and my parameter
correlate significantly, is this enough of a criterion to consider this
finding valid ?
Thanks for your input!
Marko
--
===============================================
Marko Wilke, MD
Research Fellow, Imaging Research Center
Children's Hospital Research Foundation
3333 Burnet Avenue - ML 5031
Cincinnati, OH 45229
Phone +1 (513) 636-0142 FAX +1 (513) 636-3754
[log in to unmask] http://www.irc.chmcc.org
|