Rational integration of noisy evidence and prior semantic expectations in sentence interpretation

Edited by John R. Anderson, Carnegie Mellon University, Pittsburgh, PA, and approved April 3, 2013 (received for review September 20, 2012)
May 1, 2013
110 (20) 8051-8056

Abstract

Sentence processing theories typically assume that the input to our language processing mechanisms is an error-free sequence of words. However, this assumption is an oversimplification because noise is present in typical language use (for instance, due to a noisy environment, producer errors, or perceiver errors). A complete theory of human sentence comprehension therefore needs to explain how humans understand language given imperfect input. Indeed, like many cognitive systems, language processing mechanisms may even be “well designed”–in this case for the task of recovering intended meaning from noisy utterances. In particular, comprehension mechanisms may be sensitive to the types of information that an idealized statistical comprehender would be sensitive to. Here, we evaluate four predictions about such a rational (Bayesian) noisy-channel language comprehender in a sentence comprehension task: (i) semantic cues should pull sentence interpretation towards plausible meanings, especially if the wording of the more plausible meaning is close to the observed utterance in terms of the number of edits; (ii) this process should asymmetrically treat insertions and deletions due to the Bayesian “size principle”; such nonliteral interpretation of sentences should (iii) increase with the perceived noise rate of the communicative situation and (iv) decrease if semantically anomalous meanings are more likely to be communicated. These predictions are borne out, strongly suggesting that human language relies on rational statistical inference over a noisy channel.

Continue Reading

Acknowledgments

Thanks to Roger Levy, Melissa Kline, Kyle Mahowald, Harry Tily, Nathaniel Smith, audiences at Architectures and Mechanisms for Language Processing 2011 in Paris, France, and the CUNY Conference on Human Sentence Processing 2012 in New York. Thanks also to Eunice Lim and Peter Graff who helped us extensively in running experiment 3 in our laboratory. And a special thanks to Ev Fedorenko, who gave us very detailed comments and suggestions on this work at multiple stages of this project. This work was supported by National Science Foundation Grant 0844472 from the Linguistics Program (to E.G.).

Supporting Information

Supporting Information (PDF)
Supporting Information

References

1
L Frazier, JD Fodor, The sausage machine: A new two-stage parsing model. Cognition 6, 291–325 (1978).
2
E Gibson, A computational theory of human linguistic processing: Memory limitations and processing breakdown. PhD thesis (Carnegie Mellon University, Pittsburgh). (1991).
3
E Gibson, Linguistic complexity: Locality of syntactic dependencies. Cognition 68, 1–76 (1998).
4
D Jurafsky, A probabilistic model of lexical and syntactic access and disambiguation. Cogn Sci 20, 137–194 (1996).
5
J Hale, A probabilistic Earley parser as a psycholinguistic model. Proc NAACL 2:159–166. (2001).
6
R Levy, Expectation-based syntactic comprehension. Cognition 106, 1126–1177 (2008).
7
C Shannon, A mathematical theory of communication. Bell Syst Tech J 27:623–656. (1948).
8
F Jelinek, Continuous speech recognition by statistical methods. Proc IEEE 64, 532–556 (1976).
9
M Aylett, A Turk, The smooth signal redundancy hypothesis: A functional explanation for relationships between redundancy, prosodic prominence, and duration in spontaneous speech. Lang Speech 47, 31–56 (2004).
10
M Clayards, MK Tanenhaus, RN Aslin, RA Jacobs, Perception of speech reflects optimal use of probabilistic speech cues. Cognition 108, 804–809 (2008).
11
LC Dilley, MA Pitt, Altering context speech rate can cause words to appear or disappear. Psychol Sci 21, 1664–1670 (2010).
12
E Bates, JL McDonald, BM MacWhinney, M Appelbaum, A maximum likelihood procedure for the analysis of group and individual data in aphasia research. Brain Lang 40, 231–265 (1991).
13
B MacWhinney, J Osmán-Sági, DI Slobin, Sentence comprehension in aphasia in two clear case-marking languages. Brain Lang 41, 234–249 (1991).
14
A Blackwell, E Bates, Inducing agrammatic profiles in normals: Evidence for the selective vulnerability of morphology under cognitive resource limitation. J Cogn Neurosci 7, 228–257 (1995).
15
F Ferreira, KGD Bailey, V Ferraro, Good-enough representations in language comprehension. Curr Dir Psychol Sci 11, 11–15 (2002).
16
F Ferreira, K Christianson, A Hollingworth, Misinterpretations of garden-path sentences: Implications for models of sentence processing and reanalysis. J Psycholinguist Res 30, 3–20 (2001).
17
R Levy, A noisy-channel model of rational human sentence comprehension under uncertain input. Proceedings of the 13th Conference on Empirical Methods in Natural Language Processing (Association for Computational Linguistics, Stroudsburg, PA), pp 234–243. (2008).
18
R Levy, K Bicknell, T Slattery, K Rayner, Eye movement evidence that readers maintain and act on uncertainty about past linguistic input. Proc Natl Acad Sci USA 106, 21086–21090 (2009).
19
R Levy, Integrating surprisal and uncertain-input models in online sentence comprehension: Formal techniques and empirical results. Proceedings of the 49th Annual Meeting of the Association for Computational Linguistics (Association for Computational Linguistics, Portland, OR), pp 1055-1065. (2011).
20
D Marr Vision (Freeman, New York, 1982).
21
WS Geisler, Sequential ideal-observer analysis of visual discriminations. Psychol Rev 96, 267–314 (1989).
22
WS Geisler, RL Diehl, A Bayesian approach to the evolution of perceptual and cognitive systems. Cogn Sci 27, 379–402 (2003).
23
B Levin English Verb Classes and Alternations: A Preliminary Investigation (Univ of Chicago Press, Chicago, 1993).
24
F Xu, JB Tenenbaum, Word learning as Bayesian inference. Psychol Rev 114, 245–272 (2007).
25
DJC MacKay Information Theory, Inference and Learning Algorithms (Cambridge Univ Press, Cambridge, UK, 2003).
26
V Levenshtein, Binary codes capable of correcting deletions, insertions, and reversals. Sov Phys Dokl 10, 707–710 (1966).
27
JC Trueswell, MK Tanenhaus, SM Garnsey, Semantic influences on parsing: Use of thematic role information in syntactic disambiguation. J Mem Lang 33, 285–318 (1994).
28
MK Tanenhaus, MJ Spivey-Knowlton, KM Eberhard, JC Sedivy, Integration of visual and linguistic information in spoken language comprehension. Science 268, 1632–1634 (1995).
29
B MacWhinney, E Bates, R Kliegl, Cue validity and sentence interpretation in English, German, and Italian. J Verbal Learn Verbal Behav 23, 127–150 (1984).
30
F Ferreira, The misinterpretation of noncanonical sentences. Cognit Psychol 47, 164–203 (2003).
31
M Kutas, SA Hillyard, Reading senseless sentences: Brain potentials reflect semantic incongruity. Science 207, 203–205 (1980).
32
L Osterhout, PJ Holcomb, Event-related brain potentials elicited by syntactic anomaly. J Mem Lang 31, 785–806 (1992).
33
P Hagoort, C Brown, J Groothusen, The syntactic positive shift as an ERP measure of sentence processing. Lang Cogn Process 8, 439–483 (1993).
34
GR Kuperberg, D Caplan, T Sitnikova, M Eddy, P Holcomb, Neural correlates of processing syntactic, semantic and thematic relationships in sentences. Lang Cogn Process 21, 489–530 (2006).
35
A Kim, L Osterhout, The independence of combinatory semantic processing: Evidence from event-related potentials. J Mem Lang 52, 205–225 (2005).
36
GR Kuperberg, Neural mechanisms of language comprehension: Challenges to syntax. Brain Res 1146, 23–49 (2007).
37
TF Münte, HJ Heinze, M Matzke, BM Wieringa, S Johannes, Brain potentials and syntactic violations revisited: No evidence for specificity of the syntactic positive shift. Neuropsychologia 36, 217–226 (1998).
38
S Coulson, JW King, M Kutas, Expect the unexpected: Event-related brain response to morphosyntactic violations. Lang Cogn Process 13, 21–58 (1998).
39
M van Herten, DJ Chwilla, HHJ Kolk, When heuristics clash with parsing routines: ERP evidence for conflict monitoring in sentence perception. J Cogn Neurosci 18, 1181–1197 (2006).
40
L Osterhout, LA Mobley, Event-related brain potentials elicited by failure to agree. J Mem Lang 34, 739–773 (1995).
41
Y Yamada, HJ Neville, An ERP study of syntactic processing in English and nonsense sentences. Brain Res 1130, 167–180 (2007).
42
HHJ Kolk, DJ Chwilla, M van Herten, PJW Oor, Structure and limited capacity in verbal working memory: A study with event-related potentials. Brain Lang 85, 1–36 (2003).
43
JR Anderson, LJ Schooler, Reflections of the environment in memory. Psychol Sci 2, 396–408 (1991).
44
N Chater, M Oaksford, Ten years of the rational analysis of cognition. Trends Cogn Sci 3, 57–65 (1999).
45
MC Frank, H Tily, I Arnon, S Goldwater, Beyond transitional probabilities: Human learners impose a parsimony bias in statistical word segmentation. Proceedings of the Cognitive Science Society (Portland, OR), pp 760–765. (2010).
46
Gibson E, Fedorenko E (2013) The need for quantitative methods in syntax and semantics research. Lang Cogn Process 28(1–2):88–124.
47
R Munro, et al., Crowdsourcing and language studies: The new generation of linguistic data. Proceedings of the NAACL HLT 2010 Workshop on Creating Speech and Language Data with Amazon’s Mechanical Turk (Association for Computational Linguistics, Stroudsburg, PA), pp 122–130. (2010).

Information & Authors

Information

Published in

Go to Proceedings of the National Academy of Sciences
Go to Proceedings of the National Academy of Sciences
Proceedings of the National Academy of Sciences
Vol. 110 | No. 20
May 14, 2013
PubMed: 23637344

Classifications

Submission history

Published online: May 1, 2013
Published in issue: May 14, 2013

Keywords

  1. communication
  2. psycholinguistics
  3. rational inference

Acknowledgments

Thanks to Roger Levy, Melissa Kline, Kyle Mahowald, Harry Tily, Nathaniel Smith, audiences at Architectures and Mechanisms for Language Processing 2011 in Paris, France, and the CUNY Conference on Human Sentence Processing 2012 in New York. Thanks also to Eunice Lim and Peter Graff who helped us extensively in running experiment 3 in our laboratory. And a special thanks to Ev Fedorenko, who gave us very detailed comments and suggestions on this work at multiple stages of this project. This work was supported by National Science Foundation Grant 0844472 from the Linguistics Program (to E.G.).

Notes

This article is a PNAS Direct Submission.

Authors

Affiliations

Edward Gibson1 [email protected]
Departments of aBrain and Cognitive Sciences and
Linguistics and Philosophy, Massachusetts Institute of Technology, Cambridge, MA 02139; and
Leon Bergen
Departments of aBrain and Cognitive Sciences and
Steven T. Piantadosi
Department of Brain & Cognitive Sciences, University of Rochester, Rochester, NY 14627-0268

Notes

1
To whom correspondence should be addressed. E-mail: [email protected].
Author contributions: E.G., L.B., and S.T.P. designed research; E.G. and L.B. performed research; E.G. and L.B. analyzed data; and E.G., L.B., and S.T.P. wrote the paper.

Competing Interests

The authors declare no conflict of interest.

Metrics & Citations

Metrics

Note: The article usage is presented with a three- to four-day delay and will update daily once available. Due to ths delay, usage data will not appear immediately following publication. Citation information is sourced from Crossref Cited-by service.


Citation statements

Altmetrics

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

Cited by

    Loading...

    View Options

    View options

    PDF format

    Download this article as a PDF file

    DOWNLOAD PDF

    Get Access

    Login options

    Check if you have access through your login credentials or your institution to get full access on this article.

    Personal login Institutional Login

    Recommend to a librarian

    Recommend PNAS to a Librarian

    Purchase options

    Purchase this article to get full access to it.

    Single Article Purchase

    Rational integration of noisy evidence and prior semantic expectations in sentence interpretation
    Proceedings of the National Academy of Sciences
    • Vol. 110
    • No. 20
    • pp. 7961-8313

    Media

    Figures

    Tables

    Other

    Share

    Share

    Share article link

    Share on social media