Hi,
I think I am experiencing a similar problem with 3dFourier. The attached plot shows data from a single run (170 TR's @ 2 sec) in an event-related design. The blue trace is the data after despiking only, and the red trace is the same data afte despiking plus 3dFourier (-lopass 0.1667 -hipass 0.0250 -retrend). I noticed that the trend that should be added back in after running 3dFourier seems quite different from the trend that is present in the raw (despiked) data. Also, if I filter the same data using a simple digitat filter (same cutoff frequencies) and then add the mean and trend back in, the waveform looks quite different. Something seems funny. Would the truncation artifact effect things in this way?
Thanks
Aaron