Dear all,
This question regards the poster entitled Generalisability, Random
Effects and Population Inference (AP Holmes and KJ Friston), which is
posted on the SPM Web site.
I would like to perform a similar analysis, but I am not sure that
I am setting it up properly. In this paper, a simple box-car type
visual fMRI experiment was performed in 6 subjects. The images of the
estimated activation parameter, termed alpha, were then written out as
images. My understanding of this activation parameter is that it
represents one of the betas in the BETA.mat file (presumably the first
beta, corresponding to the desired effect of interest).
In the next step of the analysis these subject activation images were
used as the 'raw' data for a one sample t-test. How is this setup in
SPM?
I tried to perform a similar analysis using a simple robust
finger_tapping motor paradigm. First I performed a simple box-car
analysis for each individual subject within SPM96 (including high-pass
filtering). I converted the first beta map for each subject to an
analyze type image (.img/.hdr) . So I now have 6 beta maps, 1 for each
subject in my 6-subject motor experiment. I tried using the following
models in the PET portion of SPM including:
mutisubject different conditions: 6 subjects, 1 condition per subject
multistudy different conditions: 6 studies, 1 subject per study, 1
condition per subject
I set the Contrast to a vector of ones and I used various combinations
of global scaling.
None of these produced any results, and the contrast appears to be
getting defaulted to the null set.
Does anyone have any insight as to how this should be setup within SPM?
Also, is there a way to perform a similar group analysis using either
the SPMt, or the SPMF from individual subject data to permit valid
population inferences? I have used the Random Effects kit, but I would
like to be able to run the experiment as described in the poster.
thanks
--
Best Regards,
Joseph Maldjian, MD
Assistant Professor of Radiology
Neuroradiology Section
Hospital of the University of Pennsylvania
office: (215) 349-5269
page: (215) 312-6813
fax: (215) 662-3283
[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|