If you heard "sweatshops are bad" or "styrophone cups are bad" you would first look for its antecedents. So maybe the cognitive algorithm goes something like this: If X in "X is bad" is not associated with unfavorable antecedents, then examin X's consquences by default.