Imo we do this all the time :) But, perhaps it is important to recognize that those models are simplifications. It's easy, probably tempting, to project a group of humans onto one axis. But it's wrong, you don't know someone until you know someone. A good rule to live by (and imo a correct assumption) is that, by far, most people are kind.
You say most. If someone consistently acts in ways that are not compatible with kindness, but are compatible with bloodthirstiness, may we model them as bloodthirsty?
