Ya, I run another subject, and find there is no meaningful pattern when I choose target-to-seed.
But there is only positive prediction when I choose seed-to-target, is it normal?
And another question, I seperate my 3 condiotns: face, object, and flobject, but I find the 3dGC.R results(activation pattern) are quit similar. It back to the conclusion that 3dSynthesize can't truly seperate those conditions....
I think, If I run 3dDdeconvolve and use only 1 stim_file(I put commands below), then the output contains only 1 condition, right? Can I use this method?

3dDeconvolve \
-input xxx+orig \
-mask xx_mask+orig \
-censor censor.1D \
-polort 3 \
-concat start_points.1D \
-num_stimts 7 \
-stim_file 1 Face_wav.1D \
-stim_label 1 Face \
-stim_file 2 xx.1D'[0]' -stim_base 2 -stim_label 2 roll \
-stim_file 3 xx.1D'[1]' -stim_base 3 -stim_label 3 pitch \
-stim_file 4 xx.1D'[2]' -stim_base 4 -stim_label 4 yaw \
-stim_file 5 xx.1D'[3]' -stim_base 5 -stim_label 5 dS \
-stim_file 6 xx.1D'[4]' -stim_base 6 -stim_label 6 dL \
-stim_file 7 xx.1D'[5]' -stim_base 7 -stim_label 7 dP \
-fitts xx_face_fitts \
-bucket xx_face_bucket \
-cbucket xx_face_cbucket \
-GOFORIT 3 \
-tout \
-xjpeg xmat_xx.jpg \
-x1D xmat_xx.x1D


Jingle