This is a complex design matrix in which I have three control stimuli
presented followed by three task stimuli presented. This is done six times,
as session one. There are then session 2 & 3, which are the same by with
different task stimuli. The results for each subject are fine, and even a
conjunction of four subjects is fine. However, when I do a conjunction of
eight subjects, it states that there are 'no suprathreshold cluster', even
if I select one subject from the contrast table. Here is the steps I am
taking, can you please let me know, if there is something that I doing
wrong, or if there is a limit to the information that can be computed.
FMRI models Specify a model
Interscan interval (sec) - 3
Scans per session = 132 132 132 …… Repeated 3sessions x No. of
subjects
Are conditions replicated -yes
Are timing/parameters the same -yes
Number of conditions or trials - 2
Condition 1 – control
Condition 2 – task
Stochastic design - no
SOA - Variable
Vector of onset:control - 0 3.7 7.3 22 25.7 29.3 44
47.7 51.3 66 69.7 73.3 88 91.7 95.3 110 113.7 117.3
Variable duration - no
Vector of onset:task - 11 14.7 18.3 33 36.7 40.3 55
58.7 62.3 77 80.7 84.3 99 102.7 106.3 121 124.7 128.3
Variable duration -no
Parametric modulation - none
These trials - events
Basis set - hrf (with time & dispersion derivatives)
Interactions among trials (Volterra) - no
User specified regressors - 0
Estimate a specified model Select scans – snrmat01_*.img,
snrref01_*.img, snrrot01_*.img
Remove global effects - None
High-pass filter - 88 88 88
Low-pass filter - hrf
Setup trial-specific F-contrast - No
Estimate - Now
Results Set up the contrasts
task01_subj01 - -1 0 0 1 0 0....
task02_subj01 - 0 0 0 0 0 0 –1 0 0 1 0 0.....
task03_subj01 - 0 0 0 0 0 0 0 0 0 0 0 0 –1 0 0 1 0 0
task01_subj02 - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 -1 0 0
1 0 0 ....
task02_subj02 - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 –1 0 0 1 0 0....
task03_subj02 - 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 –1 0 0 1 0 0
Etc.
To select for conjunction - hold control and select contrasts for
conjunction.
|