Hi everybody,
I have a problem in estimating a second level analysis (2 sample t test)
without error message.
We have 2 groups (17 and 12 subjects) making the same 2 conditions (A &
B)and rest during 1 scan.
we use SPM2 on a Linux machine with matlab 6.1
we first made individual analyses having 3 contrasts (A ; B ; A - B)
then we design and estimate the second level (2 samples t test) in order
to compare the 2 populations for each of the 3 contrasts
only 1 analysis induce problem :
when we open the results section and select the SPM.mat, the SPM contrast
manager is empty and "??? error while evaluating uicontrol callback"
appear in the matlab window.
when we make a "review design" : files & factors are set normaly but the
design matrix is empty and an error appear in the matlab window " error
using ==> image image CData cannot be a complex error in ...
spm_DesRep.m line 622"
the same analysis was first done with 18 subjects in one group without any
problem but 1 subject was excluded (lefthanded) and now the analysis can't
be done.
If we use, as the second level analysis, an ANova, then it works fine.
Actually, the 2 groups made 5 scans each with different conditions (3
motor tasks, 2 cognitive tasks) so we make 5 times the analyses as
describe above and few of them (only with the motor tasks) induce problem
as above.
dis someone have an idea of what happen ???
it would we very helpful.
thank you in advance.
Isabelle Faillenot
Ingénieure hospitalière
INSERM EMI 342
services radiologie & neurologie
CHU de Saint-Etienne
|