2018 Sun Session C 0930

Sunday, November 4, 2018 | Session C, Terrace Lounge | 9:30am

Increased spectral variability in the vowels of infant-directed speech is not universal
E. McClay, S. Cebioglu, T. Broesch, H. Yeung

Acoustic differences between infant-directed (IDS) and adult-directed speech (ADS) are well-understood. Prosodically, IDS is slower with higher, more variable pitch [1-3] that enhances non-verbal communication and phonological learning [4-5]. Spectrally, IDS is ‘hyperarticulated’ [6-8], and vowels in IDS are spoken with increased variability [6], but the function of these spectral features is controversial [7-9]. Moreover, there are isolated reports of languages where IDS vowels are not fully hyperarticulated [1, 10]. No study has reported a case where IDS has the same or less intra-vowel spectral variation than ADS.

Increased spectral variation is central to debates about the ‘function’ of IDS [7-9]; we show that this feature of IDS vowels is not universal. We investigated IDS in two very different populations: speakers of dialects of Lenakel from small-scale villages in Vanuatu and English speakers from urban North America. Higher and more variable pitch in IDS has been found in both groups [3, 11], but the spectral qualities of Lenakel are unexplored.

We elicited speech from mothers (n=37) on Tanna, an island of Vanuatu, with a local research assistant (RA), and are presently collecting data in a city in Canada (n=8) using the same methodology. First, a mother, her 6- to 18-month-old child, and an adult peer were invited to participate. The peer then left the immediate area, and the local RA played three videos, showing an English speaker manipulating a toy and repeating its label. The three labels were nonce words (/tisisi, kususu, pasasa/) designed to be phonologically stable in both English and Lenakel, and to sample from extremes of the vowel space. This stage ended once the mother could successfully label the toys herself.

Data collection involved recording the mother speaking to her child about the target toys for several minutes. An ADS sample was recorded directly afterwards, with the mother speaking to her peer about the just-completed task. During these interactions, a researcher tracked when the mother said tokens of each nonce word, until each had been uttered at least 8 times in both IDS and ADS. Vowels from target words were segmented and measured mechanically with [12]. All values were normed to Barks, and vowel spaces were calculated using the phonR statistical package [13].

Prosodically, mothers in Tanna and Canada had higher and somewhat more variable pitch in the IDS compared to ADS target vowels (Figure 1), replicating [3]’s results. Spectrally, Figure 2 illustrates the mean values of /i/, /a/, and /u/ tokens in ADS and IDS, and ‘ellipses’ of token variability whose axes are one standard deviation from the mean. Data suggested that Vanuatuan mothers did not show more within-category variation in IDS target vowels than ADS vowels. Preliminary Canadian data indicate that the Vanuatu finding is not an artefact of our methodology, since Canadian mothers so far display the expected increase in spectral variability of IDS vowels. Results reinforce the idea that while IDS prosody seems universal, IDS may not be universally more spectrally variable than ADS, and rather varies with socio/linguistic factors.

References

  1. Rattanasone, Nan Xu, Burnham, D., & Reilly, R. G. (2013). Tone and vowel enhancement in Cantonese infant-directed speech at 3, 6, 9 & 12 months of Journal of Phonetics 41, 332- 343.
  2. Fernald, A., Taeschner, T., Dunn, J., Papousek, M., de Boysson-Bardies, B. & Fukui, I. (1989). A cross-language study of prosodic modifications in mothers’ and fathers’ speech to preverbal infants. Journal of child language, 16(3), 477-501.
  3. Broesch, L., & Bryant, G. A. (2015). Prosody in infant-directed speech is similar across Western and traditional cultures. Journal of Cognition and Development, 16(1), 31-43.
  4. Fernald, A., & Mazzie, C. (1991). Prosody and focus in speech to infants and Developmental psychology, 27(2), 209.
  5. Adriaans, , & Swingley, D. (2017). Prosodic exaggeration within infant-directed speech: consequences for vowel learnability. The Journal of the Acoustical Society of America, 141(5), 3070-3078.
  6. Kuhl, P. K. , Andruski, J. E., Chistovich, I. A., Chistovich, L. A., Kozhevnikova, E. V., Ryskina, L., Stolyarova, E. I., Sundberg, U., & Lacerda, F. (1997). Cross-language analysis of phonetic units in language addressed to infants. Science, 277(5326), 684–686.
  7. Martin, , Schatz, T., Versteegh, M., Miyazawa, K., Mazuka, R., Dupoux, E., & Cristia, A. (2015). Mothers speak less clearly to infants than to adults: A comprehensive test of the hyperarticulation hypothesis. Psychological Science, 26(3), 341–347.
  8. McMurray, , Kovack-Lesh, K. A., Goodwin, D., & McEchron, W. (2013). Infant directed speech and the development of speech perception: Enhancing development or an unintended consequence? Cognition, 129, 362–378
  9. Eaves, S., Feldman, N. H., Griffiths, T. L., & Shafto, P. (2016). Infant-directed speech is consistent with teaching. Psychological Review, 123(6), 748–771.
  10. Englund, K. T., & Behne, D. M. (2005). Infant directed speech in natural interaction–Norwegian vowel quantity and quality. Journal of psycholinguistic research, 34(3), 259-280.
  11. Broesch, T. L, & Bryant, G. A. (2018). Fathers’ Infant‐Directed Speech in a Small‐Scale Society. Child Development, 89(2), e29-e41.
  12. McAuliffe, , Socolof, M., Mihuc, S., Wagner, M. & Sonderegger, M. (2017). Montreal Forced Aligner [Computer program]. Version 1.0.0.
  13. McCloy, D. R. (2016). “phonR: Tools for phoneticians and phonologists.” R package.