SurfSmooth t-maps

AFNI Message Board
1

Dear AFNI experts

I have t-maps resulting from 1st-level analysis (GLM) in the SPM package. I mapped the data on a surface, smoothed the t-maps using SufSmooth in AFNI. However, I am not sure whether it is correct to smooth t-maps (I did not smooth data during preprocessing) as it is common practice to smooth data during preprocessing. Do you think it is ok to smooth t-maps? Do you know any study that has smoothed t-maps?

I smooth the t-maps with the t-maps with the following command:

SurfSmooth - spec surf.spec
-surf_A smoothwm
-surf_B pial
-input t-map.niml.dset
-met HEAT_07
-target_fwhm 6
-Niter -1
-no_detrend_master
-output s_t-map.niml.dset

I did not use -blurmaster because I did not process the data in AFNI, and I did not have residual time-series (errts) from 3dDeconvolve. Also, I don’t know what blurmaster is doing. Is it correct to smooth the surface without blurmaster and with the above script?

2

Do you think it is ok to smooth t-maps?

If possible, it would be better to smooth the data before you perform the analysis. On the other hand, as approximation it would be roughly fine to directly smooth the t-statistic values if you run out of options.