Hello SPMers,
I have an fMRI study,
2X2 factorial design(1 2 by A B), 2 groups (Control10 and Patient11) with 4
conditions:
1A 2A 1B 2B
I would like to look at:
(1A >1B)>(2A>2B)
First, in the fixed effects model for each subject I would run the contrast:
(1A >1B)>(2A>2B)= (1A-1B)-(2A-2B)= 1A-1B-2A+2B) then
In the group model I would use a anova and feed in the resulting contrast
(con.img) for each subject and use the contrast: 10x1 11x-1 to show where
control>Patients for (1A >1B)>(2A>2B).
Is this correct or in the individual stats should I use a (1A >1B) and an
exclusive mask (2A>2B)? My problem here is that SPM does not create a resulting
con.img subject file that I would need to feed into the Random Effects analysis
to show the group differences.
Please let me know which method would be most acceptable since I have run both
at a single subject level and find more activation for the masking procedure, as
the following suggest although the mailing was discussing PET.
http://www.jiscmail.ac.uk/cgi-bin/wa.exe?A2=ind00&L=SPM&P=R244137&D=0&I=-3
Thanks you for your time.
Maria
|