Excellent! Worked great.

Okay, so now I'd like to do the same thing but using the CSPLIN basis function set.

Here's deconvolve:
#!/bin/sh

3dDeconvolve \
-input all_float.nii.gz \
-bucket stats \
-concat concat.1D \
-polort 3 \
-num_stimts 14 \
-stim_times 1 congruent_center.1D 'CSPLIN(0,18,7)' \
-stim_label 1 con_center \
-stim_times 2 congruent_cue.1D 'CSPLIN(0,18,7)' \
-stim_label 2 con_cue \
-stim_times 3 congruent_none.1D 'CSPLIN(0,18,7)' \
-stim_label 3 con_none \
-stim_times 4 incongruent_center.1D 'CSPLIN(0,18,7)' \
-stim_label 4 incon_center \
-stim_times 5 incongruent_cue.1D 'CSPLIN(0,18,7)' \
-stim_label 5 incon_cue \
-stim_times 6 incongruent_none.1D 'CSPLIN(0,18,7)' \
-stim_label 6 incon_none \
-stim_times 7 errors.1D 'CSPLIN(0,18,7)' \
-stim_label 7 errors \
-stim_times 8 drops.1D 'CSPLIN(0,18,7)' \
-stim_label 8 drops \
-stim_file 9 'all_mcf.par[0]' -stim_label 9 rx \
-stim_file 10 'all_mcf.par[1]' -stim_label 10 ry \
-stim_file 11 'all_mcf.par[2]' -stim_label 11 rz \
-stim_file 12 'all_mcf.par[3]' -stim_label 12 dx \
-stim_file 13 'all_mcf.par[4]' -stim_label 13 dy \
-stim_file 14 'all_mcf.par[5]' -stim_label 14 dz \
-iresp 1 iresp_congruent_center \
-iresp 2 iresp_congruent_cue \
-iresp 3 iresp_congruent_none \
-iresp 4 iresp_incongruent_center \
-iresp 5 iresp_incongruent_cue \
-iresp 6 iresp_incongruent_none \
-glt 1 alerting.mat -glt_label 1 alerting \
-glt 1 orienting.mat -glt_label 2 orienting \
-glt 1 conflict.mat -glt_label 3 conflict \
-fout \
-rout \
-jobs 2 \
-mask mask.nii.gz \
-GOFORIT 100

##############
error message:
Matrix error: In matrix file: alerting.mat Expected: 1 x 86 Actual: 1 x 38

##############

So, I'm using 7 basis functions. Do I need a column for each basis function?