Dear all,
with reference to my last post (https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind2002&L=FSL&O=D&X=DB66B8466DAA8D8591&Y=anna.doll%40uni-bielefeld.de&P=159336), I would like to ask for advice again. I reported last time highly non significant F-test results (for the following F-test: [1 1 1 0 0 0 0] ) in comparison to highly significant (with large clusters) t-test results. The significant t-tests are definitly expected in this contrast as the contrast compares rest and viewing of pictures. Last time I got the advice that this behavior can be expected when using tfce. I now calcuted the same F and t-tests again without the tfce option. However the results are very similar.
I now realized that the critical value for the first, the problematic F-contrast is extremly high:
Critical Value for: randomise_vox_corrp_fstat1 is: 247.04
Critical Value for: randomise_vox_corrp_fstat2 is: 17.4541
This seems weird to me, too.
Did I do anything wrong in the design/contrast definition?
Are there any suggestions?
I would really appreciate any help!
Thanks a lot!
Anna
the randomise command: (I used nipype to create the command)
randomise -i /data/anna/memory/FSL_randomize/data/view_iaps_group_comparison.nii.gz -o "randomise" -d /data/anna/memory/notebooks/Annas_notebooks/design.mat -t /data/anna/memory/notebooks/Annas_notebooks/design.con -f /data/anna/memory/notebooks/Annas_notebooks/design.fts -m /data/anna/memory/FSL_randomize/data/group_mask_view_iaps_group_comparison.nii.gz -n 1000 -x
This are the first lines of my design.mat file:
/NumWaves 4
/NumPoints 60
/PPheights 4.300000e+01 1.000000e+00 1.000000e+00 1.000000e+00
/Matrix
5.266667e+00 0.000000e+00 1.000000e+00 0.000000e+00
1.326667e+01 1.000000e+00 0.000000e+00 0.000000e+00
1.326667e+01 0.000000e+00 1.000000e+00 0.000000e+00
-2.733333e+00 0.000000e+00 1.000000e+00 0.000000e+00
-1.373333e+01 0.000000e+00 1.000000e+00 0.000000e+00
... with a total of 60 lines
The design.con file:
/ContrastName1 group_1
/ContrastName2 group_2
/ContrastName3 group_3
/ContrastName4 con_1
/ContrastName5 con_2
/ContrastName6 con_3
/ContrastName7 con_4
/NumWaves 4
/NumContrasts 7
/PPheights 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
/RequiredEffect 100.000 100.000 100.000 100.000 100.000 100.000 100.000
/Matrix
0.000000e+00 1.000000e+00 0.000000e+00 0.000000e+00
0.000000e+00 0.000000e+00 1.000000e+00 0.000000e+00
0.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
0.000000e+00 1.000000e+00 0.000000e+00 -1.000000e+00
0.000000e+00 0.000000e+00 1.000000e+00 -1.000000e+00
0.000000e+00 -1.000000e+00 0.000000e+00 1.000000e+00
0.000000e+00 0.000000e+00 -1.000000e+00 1.000000e+00
And the design.fts file:
/NumWaves 7
/NumContrasts 2
/Matrix
1 1 1 0 0 0 0
0 0 0 1 1 0 0
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|