Hi, peter
1) What's different between your acquisition and that of your colleague?
My colleague's experiment was watching movies and my experiment was purely auditory. We all use a few minutes of natural stimulation, our data analysis methods are the same (do correlation). We were surprised by the bright ventricles, which my colleague's experiment did not show.
2) Have you checked the alignment of your data to the anatomical?
I'm sure we've checked and found no problem. We also tried the nonlinear alignment you recommended, and the ventricles are still bright. We did several versions of proc script, but the results were the same, including in original space, add giant_move ..
-align_opts_aea -cost lpc+ZZ -giant_move
-regress_censor_motion 0.2
-regress_censor_outliers 0.1
But censor in my experiment is not very appropriate, because I do voxel-by-voxel correlation based on task, the TR value is 0 which was censored out.
3) Does your data have the slice timing built in correctly? Can you post the output log?
I didn't do this before...But I did it today, and it didn't change. (The slice timing file is attached)
Thanks a lot,
Dan