Representations for Phonotactic Learning in Infancy

Abstract

Infants rapidly learn novel phonotactic constraints from brief listening experience. Four experiments explored the nature of the representations underlying this learning. 16.5- and 10.5-month-old infants heard training syllables in which particular consonants were restricted to particular syllable positions (first-order constraints) or to syllable positions depending on the identity of the adjacent vowel (second-order constraints). Later, in a headturn listening-preference task, infants were presented with new syllables that either followed the experimental constraints or violated them. Infants at both ages learned first- and second-order constraints on consonant position (Experiments 1 and 2) but found second-order constraints more difficult to learn (Experiment 2). Infants also spontaneously generalized first-order constraints to syllables containing a new, transfer vowel; they did so whether the transfer vowel was similar to the familiarization vowels (Experiment 3), or dissimilar from them (Experiment 4). These findings suggest that infants recruit representations of individuated segments during phonological learning. Furthermore, like adults, they represent phonological sequences in a flexible manner that allows them to detect patterns at multiple levels of phonological analysis.

ACKNOWLEDGMENTS

This research was supported by funds from the NIH (HD 44458), from the Natural Sciences and Engineering Research Council of Canada (NSERC) and Fonds québécois de la recherche sur la société et la culture (FQRSC), and training grants from the NIH (T32-MH19990 and T32-DC00035). We thank Renée Baillargeon, Gary Dell, Lisa Octigan, Megan Monday, and Adam Smith for helpful discussions. We also thank colleagues in the Language Acquisition Lab at the University of Illinois, particularly Amélie Bernard and Ye Xu, and colleagues in the Rochester Infant Lab at the University of Rochester, particularly Julie Markant.

Notes

¹One infant who met our criteria for prematurity was inadvertently included. Excluding this infant did not change the results.

²Some syllables used in these experiments are English words for adults. Specifically, in Experiment 1, 31% of the syllables were words. In subsequent experiments, the percentages were 42% (Experiment 2: 16.5 mos), 45% (Experiment 2: 10.5 mos), 26% (Experiment 3), and 22% (Experiment 4). Since each test trial could contain both words and nonwords, we were unable to test for effects of lexical status. However, in previous experiments on adult phonotactic learning in perception (Chambers et al., 2010; Onishi et al., 2002), lexical status has not been found to influence phonotactic learning; this suggests that when the task does not require lexical access, lexical status does not strongly govern performance (e.g., Mirman, McClelland, Holt, & Magnuson, 2008).

³Due to a programming error, the set assignment of /m/ and /n/ differed from Experiment 1 (set 1: /b, k, n, f/; set 2: /p, g, m, s/). The consonant assignment in Experiment 1 was arrived at to minimize the within set similarity of consonants. This consonant assignment was mistakenly not applied to the second-order materials for 10.5-month-olds in Experiment 2. Notice that this difference between the experiments would, if anything, make it easier for infants to learn the second-order constraints in Experiment 2, because the consonants within one of the two sets might be perceived as more cohesive: three of the four consonants in set 1, /b, m, f/, were labials.

⁴There was one procedural difference between Experiment 2 in Chambers et al. (2003) and the present Experiment 2. In the previous experiment, the delay between familiarization and test lasted 2 minutes, while in the present Experiment 2 it lasted 1 minute. However, this procedural difference is in the wrong direction to independently promote the observed preference difference between Chambers et al.’s Experiment 2 and the present experiment: A longer delay should promote forgetting, and therefore increase the probability of a familiarity preference at test (e.g., Hunter et al., 1983).

⁵The 10.5-month-olds in Experiments 3 and 4 were tested at the University of Rochester, with the generous support of Richard Aslin.