Evidence for a learning bias against saltatory phonological alternations

Abstract

This study provides new experimental evidence that people learn phonological alternations in a biased way. Adult participants were exposed to alternations between phonetically dissimilar sounds (i.e., those differing in both voicing and manner, such as [p] and [v] By comparison, a control group of participants found it easier to learn the opposite pattern, where similar sounds alternated but dissimilar sounds did not. The results are taken as evidence that learners have a soft bias, considering alternations between perceptually similar sounds to be more likely.

Biased generalization of newly learned phonological alternations by 12-month-old infants

“…This asymmetry in generalization (i.e., from less similar sounds to more similar sounds, but not vice versa) is consistent with the proposal that learners are biased to prefer alternations between phonetically similar sounds (Steriade, 2001(Steriade, /2008Peperkamp et al, 2006;Wilson, 2006;J. White, 2013J. White, , 2014.…”

“…For instance, in a distributional learning model implemented by Peperkamp et al (2006) Although such a hard bias helps the model avoid erroneous phonological mappings, it would also rule out certain phonological patterns documented in existing natural languages (J. White, 2013White, , 2014. In contrast, other approaches have implemented the bias in maximum entropy learning models by assigning greater prior likelihoods to alternations involving small perceptual changes compared to those involving large perceptual changes (Wilson, 2006;J.…”

“…Evidence for such a bias has come from language typology Biased generalization of alternations 4 (Steriade 2001(Steriade /2008, artificial language experiments with adults (Wilson, 2006;J. White, 2014), and computational modeling (Peperkamp et al, 2006;Wilson, 2006;J.…”

“…

AbstractPrevious work has suggested that learners are sensitive to phonetic similarity when learning phonological patterns (e.g., Steriade, 2001Steriade, /2008White, 2014). We tested 12-month-old infants to see if their willingness to generalize newly learned phonological alternations depended on the phonetic similarity of the sounds involved.

…”

Can infants learn phonology in the lab? A meta-analytic answer

“…Two of them investigated effects of phonotactic learning on sensitivity to a sound contrast, such that infants come to be less sensitive to a pair of sounds when they are in complementary distribution than when they can occur in the same phonological positions (K. S. White, Peperkamp, Kirk, & Morgan, 2008;J. White, 2014).…”