I think we're probably going about this the wrong way. Imagine trying to figure out how two computer programs are different by looking at their compiled machin

I think we're probably going about this the wrong way. Imagine trying to figure out how two computer programs are different by looking at their compiled machine code. Small differences in input can produce massive differences in program output. So if our brains are at least partly "algorithmic" (as opposed to a smooth function that maps input to output), then superficial similarity in brain structure probably doesn't mean a whole lot. There have been some recent attempts in biology to understand the genome of different organisms using approximations to Kolmogorov complexity. The basic idea is to tweak the existing genome slightly in different ways to see if it "compresses" better or worse. I wouldn't be surprised if this approach eventually reveals a starker difference between organisms than trying to compare physical structure.