AFNI Gurus,
We're seeing that different groups are smoothing their data at different points along the analysis pipeline. Some are blurring prior to regression and standardization to TLRC. Others are waiting until after regression, but prior to TLRC standardization. Another group smoothed the data at two points, first a 3mm FWHM prior to any other preprocessing and then a subsequent 6mm FWHM following regression and TLRC standardization (voxel size = 3x3x3)
We're wondering what are the respective pros/cons of blurring at these different points in the analysis path, and do you have a recommendation on when is the most appropriate time to perform the smoothing?
