2018 Sun Session C 1000

Sunday, November 4, 2018 | Session C, Terrace Lounge | 10am

The real-time dynamics of child-directed speech: Using pupillometry to evaluate children’s processing of natural pitch contours
M. Nencheva, E. Piazza, C. Lew-Williams

Young children have an overall preference for child-directed speech (CDS) over adult-directed speech (ADS)(1), and it is known that structural and prosodic features of CDS facilitate learning(2,3). However, little is known about the moment-to-moment features of CDS that drive such engagement. Parents use great variation in pitch in the course of a single word, and in fact introduce new words on pitch peaks(4,5), but how do word-level prosody dynamics shape young children’s processing? Here, we extracted 4 common word-level pitch contours from natural CDS, and then used pupillometry to quantify children’s engagement with pitch contours in CDS. Synchrony in pupil size across individuals has been used in previous adult studies to assess engagement with speech(6); when adults are not engaged with a common external stimulus, pupil dilations are out of sync, but when engaged with particular moments in a stimulus, pupil dilations synchronize(7). If pupil size synchrony measures engagement in toddlers, then we expect greater synchrony for CDS over ADS.

In Experiment 1, using CHILDES corpora, we automatically extracted pitch contours from natural CDS to an infant (6-12-m.o.) and a child (24-30-m.o.)(8,9). Hierarchical clustering(10) of noun pitch contours yielded 4 clusters (Figure 1): rises (blue), falls (green), hills (cyan), and valleys (red).

In Experiment 2, we used an eye-tracking paradigm to examine pupil dilation synchrony for CDS vs. ADS, and to compare synchrony for the 4 word-level contours. 24-30-month-olds (n=12) listened to the same children’s story twice, once in CDS and once in ADS (counterbalanced). Intermixed with stories were 20 trials of individual sentences that were recorded to follow the above-described 4 contours, plus a flat baseline contour. For each trial, we calculated the pairwise dynamic time-warping distance(11) between the pupil size time-series of the participants. Valleys and flats elicited the least synchrony, hills and falls elicited the most synchrony, and rises fell in-between (Figure 2B). We validated these results in the CDS story by clustering the pitch contours of words into the contour types described above and quantifying the pairwise synchrony during each word in the story (Figure 2C). A likelihood ratio test showed that contour type improved model fit both in sentence trials (p<.005) and natural stories (p<0.05) compared to a null model. Adding the interaction between contour type and source type did not improve model fit (p~0.2), suggesting that the synchrony findings were the same in natural speech and in controlled sentences. Synchrony was likely driven, in part, by naturalness contours (see naturalness ratings for valleys and hills in Figure 2D).

This two-part investigation yields a new, subsecond framework for understanding how young children engage with a signal known to support language learning. We identified the 4 most common pitch contours in CDS, and revealed a physiological response that is sensitive to their real-time dynamics. In particular, we observed high synchrony for hills, which likely reflects parents’ pervasive use of this contour when referring to key words in CDS(4,5). In current research, we are examining young children’s novel word learning from for high-synchrony contours.

References

  1. Cooper, R. P., & Aslin, R. N. (1990). Preference for infant‐directed speech in the first month after birth. Child development, 61(5), 1584-1595.
  2. Fernald, A., & Mazzie, C. (1991). Prosody and focus in speech to infants and adults. Developmental psychology, 27(2),
  3. Aslin, N. (1993). Segmentation of fluent speech into words: Learning models and the role of maternal input. In Developmental neurocognition: Speech and face processing in the first year of life (pp. 305-315). Springer, Dordrecht.
  4. Kang, , & Wheatley, T. (2017). Pupil dilation patterns spontaneously synchronize across individuals during shared attention. Journal of Experimental Psychology: General, 146(4), 569.
  5. Kang, O., & Wheatley, T. (2015). Pupil dilation patterns reflect the contents of consciousness. Consciousness and cognition, 35, 128-135.
  6. Soderstrom, , Blossom, M., Foygel, R., & Morgan, J. L. (2008). Acoustical cues and grammatical units in speech to two preverbal infants. Journal of Child Language, 35(4), 869-902.
  7. Weist, R. M., & Zevenbergen, A. (2008). Autobiographical memory and past time reference. Language Learning and Development, 4(4), 291-308.
  8. Pablo Montero, José Vilar (2014). TSclust: An R Package for Time Series Clustering. Journal of Statistical Software, 62(1), 1-43. URL http://www.jstatsoft.org/v62/i01/.
  9. Tormene P, Giorgino T, Quaglini S and Stefanelli M (2008). “Matching Incomplete Time Series with Dynamic Time Warping: An Algorithm and an Application to Post-Stroke Rehabilitation.” _Artificial Intelligence in Medicine_, *45*(1), pp. 11-34. doi: 10.1016/j.artmed.2008.11.007 (URL:http://doi.org/10.1016/j.artmed.2008.11.007).