Dear John and SPM experts,
I would like to know how to set parameters in spm_dbm.m
when I try to use it to examine hemispheric asymmetry.
My interests are:
(1) to examine whether the brains of a group show any
significant hemispheric asymmetry of shape.
and
(2) to examine whether the brains of two groups show
any significant differences in the tendency of
hemisperic asymmetry of shape.
The number(s) of subjects: case (1): 24
case (2): 8 and 16
(The subjects in the two cases are the same.)
Normalization: affine+nonlinear(11*13*10)
Using the xtract_shapes function in spm_dbm.m, I made
a 24subs_shape.mat file which contains 24 subjects'
normalization data.
Then I tried to use run_stats in spm_dbm.m. After choosing
the file 24subs_shape.mat, I was asked to set
'Interesting Group weights',
'Interesting Covariates',
'Uninteresting Group weights',
'Uninteresting Covariates', and
'Restrict to asymmetries?'.
I guess I should enter Yes in the last, but I am not sure what
the first four mean. Should I enter, say, all 1s for the case (1),
and eight 1s and sixteen -1s for the case (2)?
The function run_stats in spm_dbm.m seems to output a P value
based on chi-square test. Is it the probability of the test agaist
the hypothesis that there is no over-all(whole-hemisphere) asymmetry?
If so, how can we see local asymmetries? By using the function
show_difference in spm_dbm.m?
Sorry for many questions and misunderstandings if any.
Regards,
Kota KATANODA
Department of Cognitive Neuroscience,
Faculty of Medicine,
University of Tokyo
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|