Semantics Research Group Meeting, April 23, 2010

[Japanese | English]

4:30pm, April 23, 2010
National Institute of Informatics (National Center of Sciences Bldg.),
12th floor, Classroom 2 (1213)
Oleg Kiselyov (FNMOC) and Chung-chieh Shan (Rutgers)
Bounded-rational theory of mind for conversational implicature
Game- and relevance-theoretic accounts of pragmatic reasoning often implement Grice's maxim of Manner by penalizing utterances for their syntactic complexity or processing effort. We propose to explain this penalty as the risk of misinterpretation in a bounded-rational theory of mind. The core of this explanation is to represent and execute probabilistic models (of felicity and truth) and inference algorithms (for resolving underspecification) in the same programming language. The payoff is that the same cognitive machinery that the hearer uses to decide what to do, by interpreting utterances, can also be used by the speaker to decide what to say, by predicting how potential utterances would be interpreted or misinterpreted by the hearer.

Semantics Research Group

Last modified: 2010-04-01 10:31:27 JST