Contributions of phonetic token variability and word-type frequency to phonological representations

Abstract

The experiments here build on the widely reported finding that children are most accurate when producing phonotactic sequences with high ambient-language frequency. What remains controversial is a description of the input that children must be tracking for this effect to arise. We present a series of experiments that compare two ambient-language properties, token and type frequency, as they contribute to phonotactic learning. Token frequency is the raw number of exposures children have to a particular pattern; type frequency refers to a count of abstract entities, such as unique words. Our results suggest that children's production accuracy is most sensitive to a combination of type and token frequency: children were able to generalize a target phonotactic sequence to a new word when familiarized with multiple word-types across tokens from multiple talkers, but not when presented with either word-types with no talker variability or multiple talker-tokens of a single word.

A computational approach to morphonotactics: evidence from German

“…This aspect is often left aside or explicitly rejected in the study of phonotactic regularities in lexical acquisition, both in computational research, which is mostly interested in modeling phonotactic contrasts and wordlikeness effects as type contrasts and types' acceptability ratings (Albright and Hayes, 2003;Albright, 2009), and in psycholinguistic research on phonotactic extraction from artificial lexicons, where the role of word tokens on familiarity ratings has sometimes been questioned (Richtsmeier, 2011). On the contrary, our approach is more similar to those stating that both word type and word token frequencies are relevant in language processing and acquisition (e.g., Saffran et al, 1997;Hay et al, 2003;Coady and Aslin, 2004;Richtsmeier et al, 2010), as they shape the speakers' phonotactic knowledge according to fine-grained representations.…”

Impossible to _gnore: Word-Form Inconsistency Slows Preschool Children's Word-Learning

“…Further, studies of phonological-rule learning, though they do not examine word-meaning mapping, suggest that learners can extract consistent phonological patterns. For instance, Richtsmeier, Gerken, and Ohala (2011) found that 4-year-olds generalized phonotactic patterns to new environments when provided with both word-form variability and talker variability (see also Chambers, Onishi, & Fisher, 2011;Newport & Aslin, 2004). To do this, listeners must be ignoring or collapsing across variable segments.…”

“…It is still possible that, with lengthier exposure, learners might be able to disregard accent variability, or that children might succeed with greater within-talker variability (see Richtsmeier et al, 2011) instead of talkers using internally consistent accents. However, for monolingual English speaking children, brief experimental exposure appears insufficient to generate a merged percept.…”

Input frequency and the acquisition of syllable structure in Polish

“…Manipulating frequency of exposure experimentally, Richtsmeier et al (2011) show that both are A c c e p t e d M a n u s c r i p t 16 necessary for children to form generalizations about non-word acceptability. However, no studies we are aware of systematically compare predictions of type and token frequency for spontaneous production accuracy in children.…”

“…Finally, a number of authors recognize the potential for a combined role and interaction of type and token frequency (Bybee 1995;Pierrehumbert 2003;Richtsmeier, Gerken, & Ohala 2011). For example, Pierrehumbert (2003) posits that both phonetic token variability and type variability are critical to forming phonological abstractions at various levels.…”