Download the full-sized PDF of An analysis of the processing of multiword units in sentence reading and unit presentation using eye movement data: Implications for theories of MWUsDownload the full-sized PDF



Permanent link (DOI):


Export to: EndNote  |  Zotero  |  Mendeley


This file is in the following communities:

Graduate Studies and Research, Faculty of


This file is in the following collections:

Theses and Dissertations

An analysis of the processing of multiword units in sentence reading and unit presentation using eye movement data: Implications for theories of MWUs Open Access


Other title
eye movement
reading theory
exemplar-based language models
language processing
multiword units
Type of item
Degree grantor
University of Alberta
Author or creator
Columbus, Georgina C
Supervisor and department
R. Harald Baayen (Linguistics)
Patrick A. Bolger (Linguistics)
Examining committee member and department
Jeremy Caplan (Psychology)
R. Harald Baayen (Linguistics)
Patrick A. Bolger (Linguistics)
John Newman (Linguistics)
Inbal Arnon (Psychology, University of Haifa)
Department of Linguistics

Date accepted
Graduation date
Doctor of Philosophy
Degree level
Multiword units (MWUs) have been the subject of much research in psycholinguistics, due to their syntactic and semantic idiosyncrasies. While studies traditionally focused on idioms (a piece of cake), more recent work has focused on another type: lexical bundles (in the middle of). How MWUs are stored and retrieved remains a central question in the literature, the answer to which will add to our understanding of language processing. To date though, there have been few investigations comparing the processing of different types of MWUs. This dissertation aims to fill that gap through analyses of eye movement data during normal sentence reading and trigram reading. The sentence reading results suggest that the familiarity rating for the MWU types analysed here is a relevant predictor of MWU processing. Surprisingly however, individual word frequency has more predictive capacity for MWU reading times than does MWU frequency. Much of the variance is explained by individual word frequency instead. Overall, the three MWU types investigated here are distinguished from one another in fixation durations on words, particularly for idioms and lexical bundles. For sentence reading times, in contrast, the effects of the MWU types are cancelled out, suggesting that processing difficulties may have been resolved at the sentence level. The second study investigates MWU type effects while reading them without context. Each MWU in this study is a trigram taken from the Google Web1T n-gram corpus (Brants & Franz, 2006) using stratified sampling across n-gram frequencies. The trigrams were coded for MWU type based on the categories used in Chapter 1. The results show that MWU effects are visible at the trigram level even without context. Somewhat surprisingly, however, there is also evidence of MWU types affecting processing of the first word in the first fixation duration, and of the first bigram in the subgaze duration. The findings suggest the semantic composition of MWUs is apparent to the reader very early. Our results support a usage-based model of language access and storage, such as those put forward by Bybee (e.g., 2006), Pierrehumbert (2001) and Bod (1998), where individual and unit frequency both affect reading times.
Permission is hereby granted to the University of Alberta Libraries to reproduce single copies of this thesis and to lend or sell such copies for private, scholarly or scientific research purposes only. Where the thesis is converted to, or otherwise made available in digital form, the University of Alberta will advise potential users of the thesis of these terms. The author reserves all other publication and other rights in association with the copyright in the thesis and, except as herein before provided, neither the thesis nor any substantial portion thereof may be printed or otherwise reproduced in any material form whatsoever without the author's prior written permission.
Citation for previous publication

File Details

Date Uploaded
Date Modified
Audit Status
Audits have not yet been run on this file.
File format: pdf (Portable Document Format)
Mime type: application/pdf
File size: 1648115
Last modified: 2015:10:12 15:53:03-06:00
Filename: ColumbusPhDThesisFinal.pdf
Original checksum: 6611fde5f12cde5ef83ccf6bef06edda
Well formed: true
Valid: true
File title: Prefatory_Columbus.pdf
File title: Microsoft Word - Prefatory_Columbus.docx
File author: Georgie Columbus
Page count: 182
Activity of users you follow
User Activity Date