Successful language understanding requires the comprehender to resolve uncertainty in language. One source of potential uncertainty emerges from pronouns (e.g. he, she) since pronouns carry little information and usually do not fully specify the intended referent. Nevertheless, humans interpret pronouns rapidly and efficiently despite having limited cognitive resources. Here we report three pronoun interpretation experiments showing that listeners reverse-engineer a speaker’s referential intentions based on Bayesian principles (see Kehler & Rohde, 2013 for results in English): the influence of semantics inference emerges as effects on next-mention bias, whereas the influence of syntactic prominence emerges as effects on the likelihood of a particular referent being pronominalized. Using Mandarin Chinese, we were able to test the generality of the Bayesian pronoun interpretation theory crosslingusitically, as well as to further evaluate the predictions of the theory in ways that are not possible in English. Our results lend both qualitative and quantitative support to a crosslinguistically general Bayesian theory of pronoun interpretation.

UPCOMING COG LUNCH TALKS

• 3/21/17 Kevin Woods, McDermott Lab
• 4/4/17 Wiktor Mlynarski, McDermott Lab
• 4/11/17 Devika Narain, Jazayeri Lab
• 4/25/17 Kelsey Allen, Tenenbaum Lab
• 5/2/17 Josh Rule, Tenenbaum Lab
• 5/9/17 Julia Leonard, Schulz/Gabrieli Labs
• 5/16/17 Max Kleiman-Weiner, Tenenbaum Lab