2018 Friday Poster 6539

Friday, November 2, 2018 | Poster Session I, Metcalf Small | 3pm

Early parental input, production skills and TP-based word segmentation
M. Hoareau, H. Yeung, T. Nazzi

Infants’ ability to recognize word forms in continuous speech, or word segmentation, develops during the first year of life. This ability is crucial for the acquisition of words, as single words are rarely presented in isolation. Moreover, since there is no deterministic cue in running speech that indicates the onsets of words, infants must be sensitive to a constellation of potential word boundary markers such as language-specific rhythmic units (stress vs. syllables) or familiar words. By their very nature, theses cues need to be learned, contrary to ‘domain-general’ cues, like statistical regularities (Newport & Aslin, 2000; Aslin, Saffran, & Newport, 1999), whose use is assumed to be independent of language development (Erickson & Thiessen, 2015). Yet, this issue is relatively unexplored, so we asked whether statistical cues to word segmentation are linked to well-known developmental factors that are highly variable between individuals, like quantity of early parental input and early language production abilities.

Thirty-three monolingual French-learning infants participated in the study. To measure speech input, we collected home recordings of speech input at 4 and 8 months, using the LENA™ system, and calculated the average number of adult words heard per hour. To evaluate production skills, we used parental reports of infants’ babbling inventory at both ages. For TP-based segmentation, infants were tested at 8 months using the headturn preference procedure, replicating Experiment 1 of Mersad & Nazzi (2012). They were familiarized to a 3-minute artificial language in which TPs were the only cue for word boundaries. The language was constructed by concatenating four trisyllabic “words”, 2 frequent and 2 infrequent ones. Then, they were tested with 4 lists of repeated words: 2 for the infrequent words and 2 for the part-words that occurred with the same frequency. The part-words were constructed by concatenating the last syllable of a frequent word and the first two syllables of the other frequent word. Thus, TPs within words were higher than within part-words (1 versus .75). We compared orientation times (OTs) for words and part-words to assess segmentation, and used their difference scores (words minus part-words) for the correlational analyses.

We first replicated the group preference for part-words over words (novelty effect p = .034) found in previous reports (Mersad & Nazzi, 2012; Saffran et al., 1996), showing again that French- learning 8-month-olds can use statistical cues to segment words (see Figure 1). Crucially, we found that this novelty effect increases with amount of adult words at 4 months (r = -.37, p = .035, see Figure 2), though not at 8 months (r = – .16, p = .372). It also increases with production abilities at the same age (r = -.42, p = .018, see Figure 2) but not at 4 months (r = -.08; p = .657). These findings establish for the first time that the ability to extract statistical information from speech is modulated by individual factors, like early speech experience and language production, suggesting it is more dependent on language experience than often assumed in the literature.

References

Aslin, R. N., Saffran, J. R., & Newport, E. L. (1999). Statistical learning in linguistic and nonlinguistic domains. In MacWhinney B. (Ed.), The emergence of language (pp. 359- 380). Mahwah, NJ: Lawrence Erlbaum Associates, Publishers.

Erickson, L. C., & Thiessen, E. D. (2015). Statistical learning of language: Theory, validity, and predictions of a statistical learning account of language acquisition. Developmental Review, 37, 66–108. https://doi.org/10.1016/j.dr.2015.05.002

Mersad, K., & Nazzi, T. (2012). When mommy comes to the rescue of statistics: Infants combine top-down and bottom-up cues to segment speech. Language Learning and Development, 8(3), 303–315. https://doi.org/10.1080/15475441.2011.609106

Newport, E. L., & Aslin, R. N. (2000). Innately constrained learning: Blending old and new approaches to language acquisition. Proceedings of the 24th Annual Boston University Conference on Language Development, 1–21.

Saffran, J. R., Aslin, R. N., & Newport, E. L. (1996). Statistical learning by 8-month-old infants. Science, 274(5294), 1926–1928. http://doi.org/10.1126/science.274.5294.1926