Information processing is a process of uncertainty resolution. Information-theoretic constructs such as surprisal and entropy reflect the fine-grained probabilistic knowledge which people have accumulated over time. The information-theoretic constructs explain the extent of processing difficulty that people encounter, for example when comprehending language. Processing difficulty and cognitive effort in turn are a direct reflection of predictability.

Satiation refers to an increase, over time, in the willingness of a native-speaker consultant to agree that a given syntactic structure is grammatically well-formed. Studies show that satiation can be induced under laboratory conditions, within a single testing session; that the effect is restricted to a small number of sentence types (chiefly those involving wh-extraction from wh-islands, subjects, and certain complex NPs); that experimentally induced satiation can persist for at least four weeks; and that satiation sometimes ”carries over” to syntactically related sentence types. Tables are provided showing the methods and findings of satiation studies on seven different types of syntactic violation. Larger issues include (i) whether the satiable sentence types form a natural class within generative syntax; (ii) whether satiation is a unitary phenomenon, or merely a family of similar phenomena; and (iii) how, in principle, satiation can serve as a tool for language research.