You are saying that we know humans cannot be bred for behaviour the way a dog can, when no one has ever TRIED to breed humans for 10 or 20 generations, let alone 100.
You are making a leap of faith again!
Closed societies with much multigenerational inbreeding (demonstrated by polydactily and certain dieseases) do not demonstrate specific behavior patters further attesting that human behavior is not 'inbred.' Lots of other things are. So, there goes your hypothesis.
I think it's time for you to let go of that strawman.