Listeners’ Normalization of Vowel Quality Is Influenced by ‘Restored’ Consonantal Context

Abstract

When listeners’ identifications of speech sounds are influenced by adjacent sounds, is it only the quantitative phonetic characteristics of these neighboring sounds that matter, or could their qualitative linguistic identity play a role? We tested this by inducing subjects to ‘restore’ a noise-obliterated medial consonant in VCe utterances by first presenting them with several prior utterances where this medial consonant could be heard clearly and was consistently the same, either a /b/ or a /d/. Included as V were synthetic steady-state vowels from the /i-u/ continuum. More /u/’s were identified out of this continuum in the environment of physically present /d/’s than /b/’s. Restored /d/’s had the same effect, thus indicating that the influence of context need not operate only via physical phonetic features. These results suggest that strict phonetic invariance of phonological units may not be necessary.

The effect of prosodic weakening on the production and perception of trans-consonantal vowel coarticulation in German

“…Listeners were evidently sensitive to these prosodic influences because these categorization differences derived from fitting Gaussians to their production data were also in evidence in their perceptual responses: thus, in the unstressed context in which V 1 had no pitch-accent, listeners' category boundaries were shifted towards /Y/ (i.e., there were more /U/ responses), their psychometric curves were flatter, and the influence of V 2 on their categorizations towards the /Y/-end of the continuum was significantly diminished compared with their responses in the stressed context. In general, these results provide further evidence for the considerable amount of phonetic variation to which listeners are sensitive (Local, 2003): Listeners evidently have knowledge not only of segmental context effects, as numerous studies on the compensation for coarticulation have shown (Lindblom and Studdert-Kennedy, 1967;Mann and Repp, 1980;Ohala and Feder, 1994), but in addition and as the results of the present study show, listeners' responses are also biased by the expected influences of prosodic weakening on vowels in speech production. These results demonstrate furthermore the very close connection between speech production and perception (Fowler et al, 2003;Fowler, 2005) as far as contextual influences, both segmental and prosodic, are concerned.…”

“…If listeners do not compensate for this greater coarticulatory influence, then just such a mismatch between the production and perception of coarticulation would arise that could, according to Ohala's (1993) model, make weak constituents unstable and prone to change. So far, this issue is unresolved if only because listener compensation for coarticulation has been extensively studied in relation to segmental (Mann and Repp, 1980;Fujisaki and Kunisaki, 1976;Ohala and Feder, 1994) but not prosodic variation (although see Fowler, 1981, for an analysis of the perception of coarticulation in unstressed vowels).…”

Inter- and intra-dimensional dependencies in implicit phonotactic learning

“…Phonotactic patterns can also be acquired in the lab: Participants familiarized with stimuli conforming to a particular pattern come to distinguish in performance between novel pattern-conforming and pattern-nonconforming stimuli. Such effects have been observed in learners as young as four months (Chambers et al, 2003;Saffran and Thiessen, 2003;Seidl and Buckley, 2005;Cristià et al, 2011), and in paradigms as diverse as phoneme restoration (Ohala and Feder, 1994), explicit categorization (Pycha et al, 2003;Wilson, 2003;Endress et al, 2005), allomorph selection (Peperkamp et al, 2006), speeded repetition (Onishi et al, 2002), induced speech errors (Dell et al, 2000;Goldrick, 2004;Warker and Dell, 2006), language-game responses (Wilson, 2006), and immediate recall (Majerus et al, 2004). These experiments are essentially conceptformation tasks in which participants learn to categorize stimuli, explicitly or implicitly, according to whether they conform to the target phonotactic pattern.…”

Linking Cognitive and Social Aspects of Sound Change Using Agent‐Based Modeling

“…Various studies have shown that categorization in perception is strongly influenced by different kinds of context. For example, the category boundary between /i‐u/ is closer to /i/ in a fronting context such as /Vdə/ (Ohala & Feder, ) or /jist‐just/, yeast‐used (Harrington et al., ) than in a non‐fronting context such as /Vbə/ or /swip‐swup/, sweep‐swoop . This effect of context on category boundaries is a consequence in the IP‐model of sub‐phonemic abstraction.…”