Files in this item



application/pdfBULKES-DISSERTATION-2017.pdf (1MB)
(no description provided)PDF


Title:Reading between the lines: psycholinguistic indices of prediction and formulaicity in language comprehension
Author(s):Bulkes, Nyssa Z
Director of Research:Tanner, Darren
Doctoral Committee Chair(s):Tanner, Darren
Doctoral Committee Member(s):Christianson, Kiel; Garnsey, Susan; Ionin, Tania
Department / Program:Linguistics
Degree Granting Institution:University of Illinois at Urbana-Champaign
Figurative language
Formulaic language
Sentence processing
Second language acquisition
Abstract:A comprehensive model of language processing must account for not only how people process literal language, but also how nonliteral language is processed. Further, of theoretical interest to psycholinguists is the role that prediction plays in language processing, namely the conditions under which anticipating linguistic forms and structures can facilitate language comprehension. L1 research has underscored prediction as facilitative; namely, the more informative the surrounding context, the more readers anticipate upcoming information. Research using the transposed-letter (TL) effect shows that a target with transposed letters (cholocate) are read faster than targets containing substitutions (choeotate), as letter position/identity are encoded separately (Perea & Lupker, 2003, 2004). Luke and Christianson (2012) demonstrated that higher semantic constraints lead to specific expectations for letter position/identity, showing that TL effects index prediction. While L2 research has investigated prediction in L2 processing, this research primarily addresses comprehension of literal language. In cases of semantically opaque—or idiomatic—language, it is unclear whether phrase literality affects predictive mechanisms in L1 or L2 processing. Finally, it is also unclear whether semantic opacity differentiates how expressions—literal or nonliteral—are stored and retrieved from the lexicon, namely in cases where dimension such as whole-string or substring frequency are controlled for. Results from three experiments in this dissertation support a dual-route model of language processing, where the mode of processing that is employed is ultimately determined by context.
Issue Date:2017-06-13
Rights Information:Copyright 2017 Nyssa Bulkes
Date Available in IDEALS:2017-09-29
Date Deposited:2017-08

This item appears in the following Collection(s)

Item Statistics