Dr. Christopher Hammerly

Colloquium Series

A Unified Model of Person/Animacy Effects in Incremental Sentence Comprehension

Abstract; One central fact about language comprehension is we make incremental commitments: we do not wait until the end of a clause or sentence to parse and interpret what is being communicated, but rather actively predict how things will turn out as information unfolds. As a result, a central question for psycholinguists has been what information we use to make predictions, and what the nature of these predictions might be. In this talk, I present two visual world eye tracking studies, one with speakers of Ojibwe (Algonquian) and one with speakers of English (Germanic), that shed light on how person/animacy information is used to predictively parse and interpret argument structure relationships. I then outline a model of incremental processing commitments based on a universal set of constraints. These constraints are derived from “prominence hierarchies” related to person/animacy, thematic role, and syntactic position, providing a single set of principles that can explain processing phenomena related to person/animacy information in a wide range of typologically distinct languages.