Hi AFNI experts,
I need to get a voxel-wise assessment of temporal lag relative to a reference time course. People have done this by temporally shifting the reference signal and seeing which of the shifted version correlated best with the voxel signal (https://www.ncbi.nlm.nih.gov/pubmed/23378326). I came across the AFNI tool of "3ddelay", which I intuitively thought to be appropriate for this purpose; but my test showed a weird result.
I simply used the time course of a particular voxel as the reference and ran 3ddelay with "-nodtrnd -uS -nobias -nodsamp". Presumably, the "delay" estimated for that particular voxel should be 0 sec for its time course being identical to the reference. However, the output of 3ddelay indicated a ~2 sec delay on that voxel.
So am I understanding/using the 3ddelay correctly? If not, would you please point me to the right direction?

Thank you very much for your help!