The prosodic bootstrapping of phrases: Evidence from prelinguistic infants

https://doi.org/10.1016/S0749-596X(03)00024-XGet rights and content

Abstract

The current study explores infants’ use of prosodic cues coincident with phrases in processing fluent speech. After familiarization with two versions of the same word sequence, both 6- and 9-month-olds showed a preference for a passage containing the sequence as a noun phrase over a passage with the same sequence as a syntactic non-unit. However, this result was found only in one of the two groups, the group exposed to a stronger prosodic difference between the syntactic and non-syntactic sequences. Six month olds were tested in the same way on passages containing verb phrases. In this case, both groups preferred the passage with the verb phrase to the passage with the same word sequence as a syntactic non-unit. These results provide the first evidence that infants as young as 6 months old are sensitive to prosodic markers of syntactic units smaller than the clause, and, in addition, that they use this sensitivity to recognize phrasal units, both noun and verb phrases, in fluent speech. This ability to use phrase-level prosodic cues is variable, however, and appears to depend on the strength or number of cues associated with these syntactic units.

Section snippets

Experiment 1

Following Jusczyk et al.’s (1992) success in demonstrating sensitivity to cues marking major phrase boundaries in 9 month olds, Experiment 1 explores whether 9 month olds actually deploy this sensitivity to cues internal to the clause in their processing of fluent speech. This effort is encouraged by Nazzi et al.’s (2000) suggestion that even 6 month olds make use of prosodic cues correlated with clauses in speech processing. Still, there are reasons to question whether a parallel effect will

Experiment 2

In Experiment 1, 9 month olds demonstrated a preference for the passage with the well-formed version of a noun-phrase (NP) word sequence over a phonologically matching non-unit (NU) version after familiarization with both. This suggests that 9-month-old infants sometimes use phrase-level prosodic cues in segmenting continuous speech. If 6 month olds also use prosodic cues in this way, then they too should show a preference for the passage containing a familiarized NP target sequence over a

General discussion

The results reported in this paper constitute the first evidence that syntactically influenced prosodic cues may affect infants’ recognition of major phrasal units embedded within continuous passages of speech. In several significant instances, infants preferred passages containing a familiar phonological sequence when this sequence constituted a well-formed phrasal unit (NP or VP) than when it crossed the subject/VP boundary and constituted a syntactic non-unit. Furthermore, while previous

Acknowledgements

This work was supported by a Research Grant from NICHD (15795) and a Senior Scientist Award from NIMH (01490) to P.W.J., a National Science Foundation Graduate Research Fellowship to M.S., an IGERT postdoctoral fellowship to A.S., and a Swarthmore College sabbatical award to D.G.K.N. Thanks are due to Ann Marie Jusczyk and Natasha Henline for their help and support in recruiting and running subjects, to Stefanie Shattuck-Hufnagel for advice on the presentation of the intonation data and

References (60)

  • D. Mandel et al.

    Does sentential prosody help infants organize and remember speech information?

    Cognition

    (1994)
  • D. Mandel et al.

    Infants remember the order of words in a spoken sentence

    Cognitive Development

    (1996)
  • J. Mehler et al.

    A precursor of language acquisition in young infants

    Cognition

    (1988)
  • J.L. Morgan et al.

    Structural packaging in the input to language learning: Contributions of prosodic and morphological marking of phrases to the acquisition of language

    Cognitive Psychology

    (1987)
  • J.L. Morgan et al.

    The role of constituent structure in the induction of an artificial language

    Journal of Verbal Learning and Verbal Behavior

    (1981)
  • P.A. Morse

    The discrimination of speech and nonspeech stimuli in early infancy

    Journal of Experimental Child Psychology

    (1972)
  • J. Werker et al.

    Cross-language speech perception: Evidence for perceptual reorganization during the first year of life

    Infant Behavior and Development

    (1984)
  • M.E. Beckman et al.

    Lengthenings and shortenings and the nature of prosodic constituency

  • M.E. Beckman et al.

    Intonational categories and the articulatory control of duration

  • Beckman, M. E., & Hirschberg, J. (1993). The ToBI annotation conventions. Unpublished manscript, retrieved from...
  • Beckman, M. E., & Jun, S.-A. (1996). K-ToBI (Korean) ToBI labeling convention, Version 2. Unpublished manuscript, Ohio...
  • M.E. Beckman et al.

    Intonational structure in Japanese and English

    Phonology Yearbook

    (1986)
  • Blenn, L., Seidl, A., & Hoehle, B. (in press). Recognition of phrases in early language acquisition: The role of...
  • R. Brown

    A first language: The early stages

    (1973)
  • T. Cho et al.

    Articulatory and acoustic studies of domain-initial strengthening in Korean

    UCLA Working Papers in Phonetics

    (1999)
  • Cinque

    A null theory of phrase and compound stress

    Linguistic Inquiry

    (1993)
  • W. Cooper et al.

    Syntax and speech

    (1980)
  • C. Fisher et al.

    Acoustic cues to grammatical structure in infant-directed speech: Cross-linguistic evidence

    Child Development

    (1996)
  • C. Fisher et al.

    Prosody in speech to infants: Direct and indirect cues to syntactic structure

  • J.E. Fricka et al.

    The temporal sequence of global–local processing in 3 month olds

    Infancy

    (2000)
  • Cited by (174)

    • How much does prosody help word segmentation? A simulation study on infant-directed speech

      2022, Cognition
      Citation Excerpt :

      Second, within-utterance prosodic breaks may be difficult to detect in the raw signal, yielding segmentation errors. Laboratory evidence suggests infants typically require multiple acoustic cues (pitch reset, preboundary lengthening, and pause) to detect a break (Seidl, 2007; Soderstrom, Seidl, Kemler Nelson, & Jusczyk, 2003; Wellmann et al., 2012), with the possible exception of long pauses (E. Johnson & Seidl, 2008). Also adults are better at discovering breaks when these are more strongly marked (e.g., discussions in E. Johnson & Seidl, 2008 and Wellmann et al., 2012).

    • Learning word order: early beginnings

      2021, Trends in Cognitive Sciences
    • “The tiger is hitting! the duck too!” 3-year-olds can use prosodic information to constrain their interpretation of ellipsis

      2021, Cognition
      Citation Excerpt :

      The work of Jusczyk et al. (1992) and Hirsh-Pasek et al. (1987), for instance, shows that by 7 months of age, infants can identify prosodic phrasal boundaries, and pay more attention to recorded stories in which pauses are inserted correctly at those boundaries, as opposed to within the prosodic phrases. Infants have also been shown to understand, from 6 months onwards, that words cannot span prosodic boundaries, i.e., that there cannot be a word which starts in a prosodic phrase but ends in a different phrase (e.g., Gout et al., 2004; Shukla et al., 2011; Soderstrom et al., 2003). Studies such as the ones mentioned above corroborate the Prosodic Bootstrapping hypothesis (e.g. Christophe et al., 2008; Christophe et al., 2016; Gutman, Dautriche, Crabbé, & Christophe, 2015; Höhle, Weissenborn, Schmitz, & Ischebeck, 2001; Morgan & Demuth, 1996), which states that children can use prosodic boundary information to construct a rudimentary syntactic structure that could help bootstrap lexical and syntactic acquisition (e.g. Christophe et al., 2016, 2008; Christophe, Gout, Peperkamp, & Morgan, 2003; de Carvalho, 2017; de Carvalho et al., 2018; Hirsh-Pasek et al., 1987; Morgan & Demuth, 1996).

    View all citing articles on Scopus
    1

    Although it was necessary to choose an order of authorship for publication, this work has been a true collaboration, and each author’s contribution to the project has been equally important.

    View full text