Statistical learning by 8-month-old infants

Science

Volumn 274, Issue 5294, 1996, Pages 1926-1928

  Saffran, Jenny R ^a   Aslin, Richard N ^a   Newport, Elissa L ^a  

^a UNIVERSITY OF ROCHESTER   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ARTICLE; AUDITORY STIMULATION; HUMAN; HUMAN EXPERIMENT; INFANT; LANGUAGE; LEARNING; NORMAL HUMAN; PRIORITY JOURNAL; SPEECH; SPEECH DISCRIMINATION; STATISTICS;

DISCRIMINATION LEARNING; HUMANS; INFANT; LANGUAGE DEVELOPMENT; LEARNING; SPEECH PERCEPTION;

EID: 0030451408     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.274.5294.1926     Document Type: Article

References (39)

1. Certain species-specific skills develop without any experiential input, including bat echolocation [E. Gould, Dev. Psychobiol. 8, 33 (1975)] and cricket song [R. Hoy, Am. Zool. 14, 1067 (1974)].
2. Certain species-specific skills develop without any experiential input, including bat echolocation [E. Gould, Dev. Psychobiol. 8, 33 (1975)] and cricket song [R. Hoy, Am. Zool. 14, 1067 (1974)].
3. Examples of behaviors mediated by early experience are imprinting [E. Hess, Imprinting (Van Nostrand, New York, 1973); M. Leon, Physiol. Behav. 14, 311 (1975)] and suckling responses in newborn rats [M. H. Teicher and E. M. Glass, Science 198, 635 (1977)].
4. Examples of behaviors mediated by early experience are imprinting [E. Hess, Imprinting (Van Nostrand, New York, 1973); M. Leon, Physiol. Behav. 14, 311 (1975)] and suckling responses in newborn rats [M. H. Teicher and E. M. Glass, Science 198, 635 (1977)].
5. Examples of behaviors mediated by early experience are imprinting [E. Hess, Imprinting (Van Nostrand, New York, 1973); M. Leon, Physiol. Behav. 14, 311 (1975)] and suckling responses in newborn rats [M. H. Teicher and E. M. Glass, Science 198, 635 (1977)].
6. These milestones have been well-documented both in English [for example, R. Brown, A First Language (Harvard Univ. Press, Cambridge, MA, 1973)] and cross-linguistically [for example, E. Lenneberg, Biological Foundations of Language (Wiley, New York, 1967); D. Slobin, Ed., vols. 1 to 3 of The Crosslinguistic Study of Language Acquisition (Erlbaum, Hillsdale, NJ, 1985, 1987, 1992)].
7. These milestones have been well-documented both in English [for example, R. Brown, A First Language (Harvard Univ. Press, Cambridge, MA, 1973)] and cross-linguistically [for example, E. Lenneberg, Biological Foundations of Language (Wiley, New York, 1967); D. Slobin, Ed., vols. 1 to 3 of The Crosslinguistic Study of Language Acquisition (Erlbaum, Hillsdale, NJ, 1985, 1987, 1992)].
8. These milestones have been well-documented both in English [for example, R. Brown, A First Language (Harvard Univ. Press, Cambridge, MA, 1973)] and cross-linguistically [for example, E. Lenneberg, Biological Foundations of Language (Wiley, New York, 1967); D. Slobin, Ed., vols. 1 to 3 of The Crosslinguistic Study of Language Acquisition (Erlbaum, Hillsdale, NJ, 1985, 1987, 1992)].
9. This "argument from the poverty of the stimulus" remains widely accepted [for example, N. Chomsky, Aspects of the Theory of Syntax (MIT Press, Cambridge, MA, 1965); S. Crain, Behav. Brain Sci. 14, 597 (1991)].
10. This "argument from the poverty of the stimulus" remains widely accepted [for example, N. Chomsky, Aspects of the Theory of Syntax (MIT Press, Cambridge, MA, 1965); S. Crain, Behav. Brain Sci. 14, 597 (1991)].
11. D. Bickerton, Behav. Brain Sci. 7, 173 (1984); N. Chomsky, Rules and Representations (Columbia Univ. Press, New York, 1981); J. Fodor, Modularity of Mind (MIT Press, Cambridge, MA, 1983); L. Gleitman and E. Newport, in Language: An Invitation to Cognitive Science, L Gleitman and M. Liberman, Eds. (MIT Press, Cambridge, MA, 1995), pp. 1-24
12. D. Bickerton, Behav. Brain Sci. 7, 173 (1984); N. Chomsky, Rules and Representations (Columbia Univ. Press, New York, 1981); J. Fodor, Modularity of Mind (MIT Press, Cambridge, MA, 1983); L. Gleitman and E. Newport, in Language: An Invitation to Cognitive Science, L Gleitman and M. Liberman, Eds. (MIT Press, Cambridge, MA, 1995), pp. 1-24
13. D. Bickerton, Behav. Brain Sci. 7, 173 (1984); N. Chomsky, Rules and Representations (Columbia Univ. Press, New York, 1981); J. Fodor, Modularity of Mind (MIT Press, Cambridge, MA, 1983); L. Gleitman and E. Newport, in Language: An Invitation to Cognitive Science, L Gleitman and M. Liberman, Eds. (MIT Press, Cambridge, MA, 1995), pp. 1-24
14. D. Bickerton, Behav. Brain Sci. 7, 173 (1984); N. Chomsky, Rules and Representations (Columbia Univ. Press, New York, 1981); J. Fodor, Modularity of Mind (MIT Press, Cambridge, MA, 1983); L. Gleitman and E. Newport, in Language: An Invitation to Cognitive Science, L Gleitman and M. Liberman, Eds. (MIT Press, Cambridge, MA, 1995), pp. 1-24
15. Examples include vowel structure [P. K. Kuhl, K. A. Williams, F. Lacerda, K. N. Stevens, B. Lindblom, Science 255, 606 (1992)], phonotactics [P. Jusczyk, A. Friederici, J. Wessels, V. Svenkerud, A. Jusczyk, J. Mem. Lang. 32, 401 (1993)], and prosodic structure [P. Jusczyk, A. Cutler, N. Redanz, Child Dev. 64, 675 (1993)].
16. Examples include vowel structure [P. K. Kuhl, K. A. Williams, F. Lacerda, K. N. Stevens, B. Lindblom, Science 255, 606 (1992)], phonotactics [P. Jusczyk, A. Friederici, J. Wessels, V. Svenkerud, A. Jusczyk, J. Mem. Lang. 32, 401 (1993)], and prosodic structure [P. Jusczyk, A. Cutler, N. Redanz, Child Dev. 64, 675 (1993)].
17. Examples include vowel structure [P. K. Kuhl, K. A. Williams, F. Lacerda, K. N. Stevens, B. Lindblom, Science 255, 606 (1992)], phonotactics [P. Jusczyk, A. Friederici, J. Wessels, V. Svenkerud, A. Jusczyk, J. Mem. Lang. 32, 401 (1993)], and prosodic structure [P. Jusczyk, A. Cutler, N. Redanz, Child Dev. 64, 675 (1993)].
18. Exceptions include research on prenatal exposure to maternal speech (A. DeCasper, J.-P. Lecanuet, M.-C. Busnel, C. Granier-Deferre, R. Maugeais, Infant Behav. Dev. 17, 159 (1994)] and early postnatal preferences [J. Mehler et al., Cognition 29, 149 (1988)].
19. Exceptions include research on prenatal exposure to maternal speech (A. DeCasper, J.-P. Lecanuet, M.-C. Busnel, C. Granier-Deferre, R. Maugeais, Infant Behav. Dev. 17, 159 (1994)] and early postnatal preferences [J. Mehler et al., Cognition 29, 149 (1988)].
20. R. Cole and J. Jakimik, in Perception and Production of Fluent Speech, R, Cole, Ed. (Erlbaum, Hillsdale, NJ, 1980), pp. 133-163.
21. P. Jusczyk and R. Aslin, Cognitive Psychol. 29, 1 (1995).
22. A. Christophe, E. Dupoux, J. Bertoncini, J. Mehler, J. Acousf. Soc. Am. 95, 1570 (1994); A. Cutler and D. Carter, Comput. Speech Lang. 2, 133 (1987).
23. A. Christophe, E. Dupoux, J. Bertoncini, J. Mehler, J. Acousf. Soc. Am. 95, 1570 (1994); A. Cutler and D. Carter, Comput. Speech Lang. 2, 133 (1987).
24. Z. Harris, Language 31, 190 (1955); J. Hayes and H. Clark, in Cognition and the Development of Language, J. Hayes, Ed. (Wiley, New York, 1970). See M. Brent and T. Cartwright [Cognition 61, 93 (1996)] for a discussion of related statistical cues to word boundaries.
25. Z. Harris, Language 31, 190 (1955); J. Hayes and H. Clark, in Cognition and the Development of Language, J. Hayes, Ed. (Wiley, New York, 1970). See M. Brent and T. Cartwright [Cognition 61, 93 (1996)] for a discussion of related statistical cues to word boundaries.
26. Z. Harris, Language 31, 190 (1955); J. Hayes and H. Clark, in Cognition and the Development of Language, J. Hayes, Ed. (Wiley, New York, 1970). See M. Brent and T. Cartwright [Cognition 61, 93 (1996)] for a discussion of related statistical cues to word boundaries.
27. The transitional probability of Y|X = frequency of XY/frequency of X
28. J. Saffran, E. Newport, R. Aslin, J. Mem. Lang. 35, 606(1996); _, R. Tunick, S, Barrueco, Psychol. Sci., in press.
29. J. Saffran, E. Newport, R. Aslin, J. Mem. Lang. 35, 606(1996); _, R. Tunick, S, Barrueco, Psychol. Sci., in press.
30. Each infant was tested individually while seated on the parent's lap in a sound-attenuated booth. Synthetic speech was generated off-line by the Macin-Talk system and stored on disk at a sampling rate of 22 kHz for on-line playback through an Audiomedia board in an Apple Quadra 650 computer. An observer outside the testing booth monitored the infant's looking behavior with the use of a color video system, using a buttonbox connected to the computer to initiate trials and score head-turn responses. Both the parent and the observer listened to masking music over headphones to eliminate bias. During the 2-min familiarization phase, the infant's gaze was first directed to a blinking light located on the front wall of the testing booth, and then the sound sequence was presented from two loudspeakers located on the side walls. The infant's gaze was directed to one of two blinking lights on these side walls during familiarization, but there was no relation between lights and sound. Immediately after familiarization, 12 test trials were presented (six words and six nonwords). Each test trial began with the central blinking light. When the observer signaled with a button press that the infant had fixated on the central light, one of the two side blinking lights was turned on and the center light was extinguished. When the infant faced the side light (a head turn of at least 30° in the direction of the light), the three-syllable test string was played and repeated until the infant looked away from the light for 2 s or until 15 s of looking had occurred. The observer simply recorded the direction of the infant's head turn, and the computer measured looking times, determined when the 2-s lookaway criterion had been met, and controlled the randomization and presentation of stimuli. Cumulative looking time across each of the two types of test trials provided the measure of preference.
31. The direction of the fixation preference depends on the degree of familiarity with the stimuli. If the infants have become highly familiar with the stimuli, they show dishabituation behavior, preferring the novel stimuli.
32. Two counterbalanced stimulus conditions were generated. For each condition, 45 tokens of each of four trisyllabic nonsense words (condition A: tupiro, golabu, bidaku, and padoti; condition B: dapiku, tilado, burobi, and pagotu) were spoken in random order to create a 2-min speech stream, with the stipulation that the same word never occurred twice in a row.
33. Test stimuli: tupiro, golabu, dapiku, and tilado. In condition A, the first two strings were words and the last two strings were nonwords (the transitional probabilities between the syllables in the nonwords were all zero relative to the exposure corpus, as these syllable pairs had never occurred during familiarization). In condition B, the first two strings were nonwords and the last two strings were words. This between-subjects counterbalanced design ensured that any observed preferences for words or nonwords across both conditions would not be artifacts of any general preferences for certain syllable strings. Each of the four test strings were presented (repeated win a 500-ms interval between test strings) on three different trials, resulting in a total of 12 test trials per infant.
34. There were no significant differences between the infants in condition A and condition B: f(22) = 0.31. The data from the two groups were thus combined for the other analyses.
35. Condition A words: pabiku, tibudo, golatu, and daropi; condition B words: tudaro, pigola, bikuti, and budops.
36. Test stimuli: pabiku, tibudo, tudaro, and pigola. In condition A, the first two strings were words and the second two strings were part-words. For example, the part-word pigola spanned the word boundary between daropi#golatu and thus was heard during exposure. In condition B, the first two strings were part-words and the second two strings were words. The part-words were thus three-syllable sequences that the infants had heard during the course of the exposure period. The difficulty of this test discrimination can be seen by comparing the transitional probabilities between the syllables in the words (1.0 between syllables 1 and 2 and between syllables 2 and 3) to the transitional probabilities between the syllables in the part-words (0.33 between syllables 1 and 2 and 1.0 between syllables 2 and 3).
37. There were no significant differences between the infants in condition A and condition B: t(22) = 0.49. The data from the two groups were thus combined for the other analyses.
38. For example, this same general mechanism could be used to find an object, such as a human face, in the environment.
39. We thank J. Gallipeau, J. Hooker, P. Jusczyk, A. Jusczyk, T. Mintz, K. Ruppert, and J. Sawusch for their help with various aspects of this research, and P. Jusczyk, S. Pollak, M. Spivey-Knowlton, and M. Tanenhaus for their helpful comments on a previous draft. Supported by an NSF predoctoral fellowship (J.R.S.), NSF grant SBR9421064 (R.N.A.), and NIH grant DC00167 (E.L.N.). The parents of all participants gave informed consent.