FWHM blur estimate 0s

My out.ss_review text files show the blur estimates (FWHM) as all 0s for every subject. Here is an example below. Do you know if there is a problem?

Thank you!

subject ID : s2030
AFNI version : AFNI_19.2.04
AFNI package : linux_openmp_64
TR : 2.3
TRs removed (per run) : 2
num stim classes provided : 4
final anatomy dset : anat_final.s2030+tlrc.HEAD
final stats dset : stats.s2030_REML+tlrc.HEAD
final voxel resolution : 3.000000 3.000000 3.000000

motion limit : 0.5
num TRs above mot limit : 1
average motion (per TR) : 0.0822601
average censored motion : 0.0812067
max motion displacement : 2.18731
max censored displacement : 2.18731
outlier limit : 0.1
average outlier frac (TR) : 0.00232259
num TRs above out limit : 0

num runs found : 2
num TRs per run : 268 268
num TRs per run (applied) : 268 266
num TRs per run (censored): 0 2
fraction censored per run : 0 0.00746269
TRs total (uncensored) : 536
TRs total : 534
degrees of freedom used : 28
degrees of freedom left : 506

TRs censored : 2
censor fraction : 0.003731
num regs of interest : 4
num TRs per stim (orig) : 147 144 150 142
num TRs censored per stim : 0 2 2 0
fraction TRs censored : 0.000 0.014 0.013 0.000
ave mot per sresp (orig) : 0.074274 0.080722 0.105286 0.070972
ave mot per sresp (cens) : 0.074274 0.076754 0.101811 0.070972

TSNR average : 214.54
global correlation (GCOR) : 0.199103
anat/EPI mask Dice coef : 0.749513
maximum F-stat (masked) : 82.661
blur estimates (ACF) : 0.432719 5.31759 12.7936
blur estimates (FWHM) : 0 0 0

That is expected. Basically, those Gaussian FWHM parameters were considered obsolete, and so now 3dFWHMx outputs them as zeros (also to allow for some script backwards compatibility). The mixed-model auto-correlation function actually models the ACF of the data very well, while a Gaussian curve does not. So we highly recommend use of those mixed-model ACF parameters instead.

  • rick