I have a bunch of thickness files in fsaverage space (already smoothed at fwhm). 

I was wondering whether is possible to compute quickly an asymmetry overlay for each one, but it's turning more difficult than I thought. Vertex indices are different between lh and rh. So I started looking at fsaverage_sym, which has already registrations lh-rh. I thought this would be a nice shortcut to avoid registering every subject again. But there are a couple of confusing points:

1. Why isn't lh.inflated well aligned between fsaverage and fsaverage_sym (see figure below) ?

2. I found an old post in 2012, in which Donald advises to not run analyses on a simple left-right registration because one hemisphere would be more processed than the other. Since I might want to run some analyses, can you please explain what is the best way to get asymmetry surfaces (possibly by avoiding new registrations)? Why wouldn't a simple L-R registration of fsaverage be ok?

I run the analyses in R and don't necessarily need the glm tools within FS.

Dorian