hi Rick & all,
I see the issue discussed in this thread is now resolved: 3dDeconvolve happily saves out iresps when -input1D is used. It seems to me, though, that the "-sresp" option still shows this same incompatibility with -input1D (i.e., no file is saved, and a message appears: "** FATAL ERROR: Can't open dataset '(null)'"). If I'm correct about this, I wondered if a similar fix could easily be implemented! (recognizing that it's a matter of convenience, as the info could be computed from statistics printed to the screen).

Below is pasted a pared-down version of the relevant command. Many thanks,

Joe

3dDeconvolve \
-float \
-input1D ${ID}_clusterData.1D'['${CLUS}']' \
-TR_1D 2 \
-concat '1D: 0 180 360 540 720 900' \
-censor /jukebox/botvinick/joe/dots/${ID}/data/motioncensor_${ID}.1d \
-polort A \
-num_stimts 3 \
-stim_times 1 /jukebox/botvinick/joe/dots/${ID}/stimtimes/${ID}_st_motion_SignalOnly.txt 'TENT(2,12,6)' \
-stim_label 1 motion_SignalOnly \
-stim_times 2 /jukebox/botvinick/joe/dots/${ID}/stimtimes/${ID}_st_motion_Interference.txt 'TENT(2,12,6)' \
-stim_label 2 motion_Interference \
-stim_times 3 /jukebox/botvinick/joe/dots/${ID}/stimtimes/${ID}_st_motion_Arrow.txt 'TENT(2,12,6)' \
-stim_label 3 motion_Arrow \
-iresp 1 irespHRF_${ID}_cluster${CLUS}_SignalOnly \
-sresp 1 srespHRF_${ID}_cluster${CLUS}_SignalOnly