The variant of set theory and logic that recognizes degrees of applicability of predicates. Thus although classical logic regards ‘this room is hot’ as either true or false, it may better represent the way we actually reason (or the way we ought to reason) to say that it is pretty much true, or truer than saying that the room is cold. In fuzzy logic a determinate quantity of merit (represented by 1) is shared between a proposition and its negation: if it becomes pretty much true that the room is hot (e.g. to degree 1−n) it becomes pretty much false (degree n) that it is not hot. Fuzzy logic has extensive application in artificial intelligence, and in the design of systems that control real events and which need to be responsive to changes of degree in important features. The first mathematical treatment was the paper ‘Fuzzy Sets’, in Information and Control (1965), by L. A. Zadeh.