Dear AFNI experts,

My data involves each participant doing a picture-viewing task after exercise, and after rest. After preprocessing my 31 subjects using afni_proc.py, I have found that I have some subjects with lots of head motion:

AES101 Exercise condition - 41.0% TRs censored, Rest condition: 34.8% TRs censored
AES104 Exercise condition - 70.9% TRs censored, Rest condition: 66.8% TRs censored
AES110 Exercise condition - 43.9% TRs censored, Rest condition: 29.1% TRs censored
AES111 Exercise condition - 19.7% TRs censored, Rest condition: 44.3% TRs censored
AES118 Exercise condition - 50.8% TRs censored, Rest condition: 38.1% TRs censored
AES131 Exercise condition - 88.1% TRs censored, Rest condition: 57.0% TRs censored
AES136 Exercise condition - 84.8% TRs censored, Rest condition: 78.3% TRs censored
AES138 Exercise condition - 57.4% TRs censored, Rest condition: 78.7% TRs censored

My pipeline consisted of dcm2niix to convert structural and functional files from .dcm to .nii.gz, @Align_Centers to align the structural and functional .nii.gz files to the MNI base, @SSWarper, and finally, afni_proc.py. The afni_proc.py script I used for subject "AES101" after the exercise condition is below:

afni_proc.py \
-subj_id AES101_Exercise \
-copy_anat /Volumes/LaCie/AES101/EXERCISE/anatSS.AES101.nii \
-anat_has_skull no \
-anat_follower anat_w_skull anat /Volumes/LaCie/AES101/EXERCISE/101EXanat_shft.nii.gz \
-dsets /Volumes/LaCie/AES101/EXERCISE/101EXpicview_shft.nii.gz \
-blocks tshift align tlrc volreg blur mask scale regress \
-radial_correlate_blocks tcat volreg \
-tcat_remove_first_trs 2 \
-align_opts_aea -cost lpc+ZZ -giant_move -check_flip \
-tlrc_base MNI152_2009_template_SSW.nii.gz \
-tlrc_NL_warp \
-tlrc_NL_warped_dsets /Volumes/LaCie/AES101/EXERCISE/anatQQ.AES101.nii \
/Volumes/LaCie/AES101/EXERCISE/anatQQ.AES101.aff12.1D \
/Volumes/LaCie/AES101/EXERCISE/anatQQ.AES101_WARP.nii \
-volreg_align_to MIN_OUTLIER \
-volreg_align_e2a \
-volreg_tlrc_warp \
-mask_epi_anat yes \
-blur_size 4.0 \
-regress_stim_times /Volumes/LaCie/stimtimes/neutral_adjusted.1D /Volumes/LaCie/stimtimes/pleasant_adjusted.1D /Volumes/LaCie/stimtimes/unpleasant_adjusted.1D \
-regress_stim_labels neutral pleasant unpleasant \
-regress_basis 'BLOCK(20,1)' \
-regress_opts_3dD \
-jobs 4 \
-gltsym 'SYM: pleasant -neutral' -glt_label 1 P-N \
-gltsym 'SYM: unpleasant -neutral' -glt_label 2 U-N \
-regress_motion_per_run \
-regress_censor_motion 0.3 \
-regress_censor_outliers 0.05 \
-regress_3dD_stop \
-regress_reml_exec \
-regress_compute_fitts \
-regress_make_ideal_sum sum_ideal.1D \
-regress_est_blur_epits \
-regress_est_blur_errts \
-regress_run_clustsim no \
-html_review_style pythonic \
-execute

I have two questions -
1. Is there any other preprocessing step or afni_proc.py script, etc. that you would suggest for me to try to help reduce the % of TRs censored?
2. My mentor has suggested me to use slomoco. I have tried slomoco for one of these subjects, and while many error messages were returned, an output file was generated (this output file is for some reason much smaller in size than the input epi file I put in, as well). This slomoco output file, "101EXpicview_shft.steadystate.slicemocoxy_afni+orig", is in BRIK/HEAD format, and already has had its first 4 TRs removed, since I removed the first 4 TRs before running slomoco. The slomoco developers suggest to not run 3dvolreg as a preprocessing step after slomoco - so I was wondering what my afni_proc.py script should then look like? I know my dset should be specified as 101EXpicview_shft.steadystate.slicemocoxy_afni+orig, and that my stimtiming files should be adjusted to reflect the removal of the first four TRs.

Thank you very much for any feedback you can provide.