Why is that? Structural prediction and ambiguity resolution in a very large corpus of English sentences

Douglas William Roland, Jeffrey L. Elman, Victor S. Ferreira

Research output: Contribution to journalArticle

42 Citations (Scopus)

Abstract

Previous psycholinguistic research has shown that a variety of contextual factors can influence the interpretation of syntactically ambiguous structures, but psycholinguistic experimentation inherently does not allow for the investigation of the role that these factors play in natural (uncontrolled) language use. We use regression modeling in conjunction with data from the British National Corpus to measure the amount and specificity of the information available for disambiguation in natural language use. We examine the Direct Object/Sentential Complement ambiguity and the closely related issue of complementizer use in sentential complements, and find that both ambiguity resolution and complementizer use can be predicted from contextual information.

Original languageEnglish
Pages (from-to)245-272
Number of pages28
JournalCognition
Volume98
Issue number3
DOIs
Publication statusPublished - 2006 Jan 1
Externally publishedYes

Fingerprint

Psycholinguistics
psycholinguistics
Language
language
available information
regression
interpretation
Research
Language Use
Ambiguity Resolution
Natural Language
Prediction
Complementizer
Modeling
Contextual Factors
Syntax
Experimentation
Specificity
Direct Object
Contextual

ASJC Scopus subject areas

  • Language and Linguistics
  • Experimental and Cognitive Psychology
  • Developmental and Educational Psychology
  • Linguistics and Language
  • Cognitive Neuroscience

Cite this

Why is that? Structural prediction and ambiguity resolution in a very large corpus of English sentences. / Roland, Douglas William; Elman, Jeffrey L.; Ferreira, Victor S.

In: Cognition, Vol. 98, No. 3, 01.01.2006, p. 245-272.

Research output: Contribution to journalArticle

@article{cf47501d2d09481cb438650676eca118,
title = "Why is that? Structural prediction and ambiguity resolution in a very large corpus of English sentences",
abstract = "Previous psycholinguistic research has shown that a variety of contextual factors can influence the interpretation of syntactically ambiguous structures, but psycholinguistic experimentation inherently does not allow for the investigation of the role that these factors play in natural (uncontrolled) language use. We use regression modeling in conjunction with data from the British National Corpus to measure the amount and specificity of the information available for disambiguation in natural language use. We examine the Direct Object/Sentential Complement ambiguity and the closely related issue of complementizer use in sentential complements, and find that both ambiguity resolution and complementizer use can be predicted from contextual information.",
author = "Roland, {Douglas William} and Elman, {Jeffrey L.} and Ferreira, {Victor S.}",
year = "2006",
month = "1",
day = "1",
doi = "10.1016/j.cognition.2004.11.008",
language = "English",
volume = "98",
pages = "245--272",
journal = "Cognition",
issn = "0010-0277",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Why is that? Structural prediction and ambiguity resolution in a very large corpus of English sentences

AU - Roland, Douglas William

AU - Elman, Jeffrey L.

AU - Ferreira, Victor S.

PY - 2006/1/1

Y1 - 2006/1/1

N2 - Previous psycholinguistic research has shown that a variety of contextual factors can influence the interpretation of syntactically ambiguous structures, but psycholinguistic experimentation inherently does not allow for the investigation of the role that these factors play in natural (uncontrolled) language use. We use regression modeling in conjunction with data from the British National Corpus to measure the amount and specificity of the information available for disambiguation in natural language use. We examine the Direct Object/Sentential Complement ambiguity and the closely related issue of complementizer use in sentential complements, and find that both ambiguity resolution and complementizer use can be predicted from contextual information.

AB - Previous psycholinguistic research has shown that a variety of contextual factors can influence the interpretation of syntactically ambiguous structures, but psycholinguistic experimentation inherently does not allow for the investigation of the role that these factors play in natural (uncontrolled) language use. We use regression modeling in conjunction with data from the British National Corpus to measure the amount and specificity of the information available for disambiguation in natural language use. We examine the Direct Object/Sentential Complement ambiguity and the closely related issue of complementizer use in sentential complements, and find that both ambiguity resolution and complementizer use can be predicted from contextual information.

UR - http://www.scopus.com/inward/record.url?scp=30044437820&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=30044437820&partnerID=8YFLogxK

U2 - 10.1016/j.cognition.2004.11.008

DO - 10.1016/j.cognition.2004.11.008

M3 - Article

C2 - 16399264

AN - SCOPUS:30044437820

VL - 98

SP - 245

EP - 272

JO - Cognition

JF - Cognition

SN - 0010-0277

IS - 3

ER -