Ruben,
When you use SPMG1, there is only basis function. So, I assume your 3dDeconvolve script contains the following:
-stim_times 1 ${global_path}/timings/${i}.baseline.1D 'SPMG1(12)' \
-stim_label 1 B \
-stim_times 2 ${global_path}/timings/${i}.instructions.1D 'SPMG1(40)' \
-stim_label 2 I \
-stim_times_AM2 3 ${global_path}/timings/parametric_modulation/${i}_scenarios.parametric_modulation.1D 'SPMG1(90)' \
-stim_label 3 S \
> 'SYM: S[1] -I[1]'
This contrast does not make sense. There is only one regression coefficient for I (which is I[0]; thus, I[1] does not exist. And you have the same problem with S[1] -B[1] because you don't have B[1].
Similarly for SPMG2, you only have two regression coefficients for B and I. Therefore, B[2], B[3], I[2] and I[3] do not exist.
Gang
Edited 2 time(s). Last edit at 11/01/2021 08:52AM by Gang.