HI again

first of all thank you for such a swift answer.

At this point i have run 3ddeconvolve on all my individuals subjects and then i have run an anova for a group analysis. However the results of this analysis are a little puzzling so i was hoping that you could tell me whether i am misusing the stim_times_AM option

Here is my 3ddeconvolve script:

3dDeconvolve \
-input "${subject}/BRIK/${subject}_allrunsTMS+tlrc" \
-mask "${exp_path}/Group/16_subjects_mask+tlrc" \
-polort 2 \
-concat concat.1D \
-censor "${exp_path}/QandAdir/${subject}_censor.1D" \
-num_stimts 8 \
-stim_times_AM1 1 "${exp_path}/newest_corrected_timeseries/${subject}_married.1D" 'GAM' \
-stim_times 2 "${exp_path}/newest_corrected_timeseries/${subject}_all_nc.txt" 'GAM' \
-stim_file 3 "${subject}/BRIK/${subject}_reg_params.1D[0]" \
-stim_file 4 "${subject}/BRIK/${subject}_reg_params.1D[1]" \
-stim_file 5 "${subject}/BRIK/${subject}_reg_params.1D[2]" \
-stim_file 6 "${subject}/BRIK/${subject}_reg_params.1D[3]" \
-stim_file 7 "${subject}/BRIK/${subject}_reg_params.1D[4]" \
-stim_file 8 "${subject}/BRIK/${subject}_reg_params.1D[5]" \
-stim_label 1 Comp \
-stim_label 2 Noncomp \
-fout -rout -tout -bout \
-bucket ${subject}/BRIK/${subject}_decon_witems\

as you can see i am using the AM option only once as only one of my two conditions has amplitudes associated with it. Is that an issue?

After the deconvolution i then normaliza both my conditions with respect to the baseline in order to convert them to percent signal change.

After that i run an anova comparing the two across subjects.

Could you please let me know if there is anything i am doing incorrectly?

thanks again
giulia