In a few earlier studies, spatilly smoothing was performed before ReHo analysis. However, Mr. LONG Xiang-Yu found that, for some voxels, the ReHo value is very high because spatially smoothing certainly increase the local similarity. That is why we now recommend to perform spatially smoothing after ReHo calculation. Please note that we are not sure which is "better" for group level analysis, especially between-group studies. I suggest you can compare the results.
Forums
Indeed, generally smooth
Indeed, generally smooth procedure was done before any post-process, like correlation analysis or ICA, but in ReHo situation, smooth should be used very carefully. This procedure could be check in some papers, e.g.,
Cao Qingjiu, Zang Yufeng, Sun Li, Sui Manqiu, Long Xiangyu, Zou Qihong, Wang Yufeng. Abnormal neural activity in children with attention deficit hyperactivity disorder: a resting-state functional magnetic resonance imaging study. Neuroreport. 2006 Jul 17; 17(10): 1033-6.