Proceedings of the Annual Meeting of the Cognitive Science Society

Kehler and Rohde (2013) proposed a Bayesian theory of pro-noun interpretation where the influence of world knowledgeemerges as effects on the prior and the influence of informationstructure as effects on the likelihood: P(referent|pronoun) μP(pronoun|referent)P(referent). Here we present two experi-ments on Mandarin Chinese that allow us to test the generalityof the theory for a language with different syntactic-semanticassociations than English. Manipulations involving two dif-ferent classes of implicit-causality verbs and passive vs. activevoice confirmed key predictions of the Bayesian theory: effectsof these manipulations on the prior and likelihood in produc-tion were consistently reflected in pronoun interpretation pref-erences. Quantitative analysis shows that the Bayesian modelis the best fit for Mandarin compared to two competing anal-yses. These results lend both qualitative and quantitative sup-port to a cross linguistically general Bayesian theory of pro-noun interpretation.