Hi Martin

there is no easy answer to this question. Basically, the optimal smoothing filter depends on the size of the effect you are looking for (the so-called "matched-filter theorem"). IN general, if you have bigger groups of subjects and/or expect focal differences use a smaller kernel.

Sorry I don't have better advice to give than that cheers Bruce

On Wed, 16 Mar 2016, Martin Juneja wrote:

Hello everyone, I am very new to FreeSurfer. Although I know what smoothing means etc. but I am not sure what smoothing level is the best to choose while comparing cortical thinning for patients vs. controls. Qdec gives several options to choose from among 0, 5, 10, 15, 20 and 25. Literature says 10 mm is the typical value to choose but when I calculate clusters, I don't see any significant clusters (at FDR p < 0.05) but at smoothing level 20 mm, I can see several clusters at the same threshold i.e. at FDR p < 0.05.

I would really appreciate if someone, expert in this field could guide me regarding this. Also, may be if someone could recommend me a few papers from literature discussing this issue e.g. effect of smoothing on cortical thickness etc., that would be very useful.

Thanks, MJ