Search

This study investigates substantive bias, a phenomenon wherein learners exhibit a preference for phonetically motivated patterns in language acquisition. The study presents evidence that variable input, rather than categorical input, can activate substantive bias. Native speakers of Hong Kong Cantonese were randomly assigned to categorical or variable training conditions for vowel backness harmony or disharmony, or to a no-training control condition. Results indicate that participants in the categorical and control conditions did not exhibit a bias towards either pattern. However, participants in the variable conditions displayed a bias towards vowel harmony, suggesting that input variability can strengthen the effect of substantive bias. These findings contribute to our comprehension of the role of input variability in phonological learning and the mechanisms involved in the acquisition of phonetically motivated and unmotivated phonological patterns.

• Samoan paradigms have been reanalyzed over time in a way that is frequency-matching. • At the same time, reanalysis also obeys OCP-place phonotactic constraints. • Paradigm learning is argued to be sensitive to phonetically motivated phonotactics. • To test this, consonant similarity is quantified using spectral distance measures. • Consonant similarity is found to correlate with the strength of OCP-place in Samoan. Paradigms with conflicting data patterns can be difficult to learn, resulting in a type of language change called reanalysis. Existing models of morphophonology predict reanalysis to occur in a way that matches frequency distributions within the paradigm. Using evidence from Samoan, this paper argues that in addition, reanalysis may be constrained by phonotactics (global distributional regularities in the lexicon) and phonetic substance. More concretely, I find that reanalysis of Samoan thematic consonants generally matches distributional patterns within the paradigm. However, reanalysis is also modulated by a phonotactic dispreference against sequences of homorganic consonants, analyzed here in Optimality Theoretic terms by OCP-place. These results are supported by an iterated learning model that is based in MaxEnt (Goldwater and Johnson, 2003). In a study where phonetic similarity is measured as the spectral distance between two phones, I find that similarity of consonants is closely correlated with the strength of OCP-place effects in Samoan; this suggests that OCP-place is rooted in phonetic similarity avoidance, and more generally that in reanalysis, speakers preferentially utilize phonetically-motivated phonotactics. [ABSTRACT FROM AUTHOR]

The role of phonetic naturalness in biasing the learning of sound patterns remains an unsettled question in phonology. The present study investigates naturalness bias in phonotactic learning using a novel experimental paradigm that tests whether learners reproduce a phonetically-motivated phonotactic implicational about the distribution of major place contrasts in stops. Stops differing in place of articulation are easier to distinguish word-initially than word-finally, so place contrasts in word-final position should also exist in word-initial position. The reverse is not necessarily true. This implicational is typologically supported as well as motivated by perceptual naturalness. In two artificial grammar learning experiments, I exposed participants to place contrasts in stops either word-initially or word-finally and tested whether they extended the contrasts to the other word-edge position. Participants successfully learned to recognize novel words that fit the phonotactic pattern they had been trained on, but they were equally willing to extend the place contrasts in both directions, yielding no evidence for naturalness bias. These results contrast with those of a similar study that found asymmetric extension of the stop voicing contrast, supporting an effect of naturalness bias. Confusion data suggests that the reduction in perceptibility from word-initial to word-final position may be greater for stop voicing than for stop place of articulation. This difference may underlie the divergent results of the two studies, leading to the hypothesis that the strength of a substantive bias depends on the magnitude of its phonetic precursor.

This article examines whether children alter a variable phonological pattern in an artificial language towards a phonetically-natural form. We address acquisition of a variable rounding harmony pattern through the use of two artificial languages; one with dominant harmony pattern, and another with dominant non-harmony pattern. Overall, children favor harmony pattern in their production of the languages. In the language where harmony is non-dominant, children's subsequent production entirely reverses the pattern so that harmony predominates. This differs starkly from adults. Our results compare to the regularization found in child learning of morphosyntactic variation, suggesting a role for naturalness in variable phonological learning. [ABSTRACT FROM AUTHOR]

General vowel harmony and disharmony rules have comparable formal complexity but differ dramatically in typological frequency and phonetic motivation. Previous studies found no difference in learning between vowel harmony and disharmony; this putative equivalence has been used to discount the view that learners are influenced by substantive learning biases. In the current study, we use a more nuanced test to show that there is a clear difference in learning between vowel harmony and disharmony: learners readily infer a vowel harmony pattern, but not a disharmony pattern. The findings suggest that vowel disharmony is in fact strongly disfavored during learning. [ABSTRACT FROM AUTHOR]

Saltatory alternations occur when two sounds alternate with each other, excluding a third sound that is phonetically intermediate between the two alternating sounds (e.g. [p] alternates with [β], with nonakemating, phonetically intermediate [b]). Such alternations are attested in natural language, so they must be leamable; however, experimental work suggests that they are dispreferred by language learners. This article presents a computationally implemented phonological framework that can account for both the existence and the dispreferred status of saltatory alternations. The framework is implemented in a maximum entropy learning model (Goldwater & Johnson 2003) with two significant components. The first is a set of constraints penalizing correspondence between specific segments, formalized as *MAP constraints (Zuraw 2007, 2013), which enables the model to learn saltatory alternations at all. The second is a substantive bias based on the P-map (Steriade 2009 [2001]), implemented via the model's prior probability distribution, which favors alternations between perceptually similar sounds. Comparing the model's predictions to results from artificial language experiments, the substantively biased model outperforms control models that do not have a substantive bias, providing support for the role of substantive bias in phonological learning.* [ABSTRACT FROM AUTHOR]

Highlights: [•] Adults learned novel phonological alternations in an artificial language. [•] Participants learning alternations between dissimilar sounds assumed that more similar sounds also alternated. [•] Learning saltatory alternations (e.g., /p/ becomes [v], /b/ remains [b]) was difficult, even with explicit training. [•] It is argued that learners are biased to prefer alternations between perceptually similar sounds. [Copyright &y& Elsevier]