Value loading in the human brain: a worked example — LessWrong