Lexical retrieval and error recovery in reading: A model based on eye fixations

Abstract

This paper presents a model of reading that accounts for the oral interpretation of an ambiguous word, the time it takes to derive and integrate the interpretation, and the time it takes to detect a subsequent inconsistency. The model's predictions are compared to the sequence and duration of the readers' eye fixations on “garden path” passages such as: Cinderella was sad because she couldn't go to the dance that night. There were big tears in her brown dress. The model predicts that the interpretation of an ambiguous word (such as tears) depends on the contextual priming and the interpretation's relative frequency. The duration of the eye fixations on a disambiguating word (such as dress) depends on how consistent it is with the reader's prior interpretation of the text. The eye fixations and question-answering data also indicate different ways of recovering from the initial misinterpretation.