This study examines the effects of lexical stress levels and information structure on vowel dispersion in American English, indicating significant interactions.
This study investigates how lexical stress levels ([StrLev]) and information structure, i.e., focus versus post-focus given, affect the spectral characteristics of four full-vowel categories (/æ, ɛ, ɔ, oʊ/), and whether the vowel categories ([VCat]) interact with the two factors. The stress levels considered are primary stress (P) as local, secondary stress (S) as localization, and unstressed unreduced (U) as locality. Speech data were collected from 25 American English native speakers, divided into two based on the presence/absence of post-focus pitch accent (the deaccenting versus the hyper-accenting group). Vowel dispersion, measured as Euclidean distances from the F1–F2 center, was analyzed by ANOVA separately for the two groups. Regardless of the groups, significant main effects of [StrLev] and [InfoStruc] were obtained: vowel dispersion was greatest for P and smallest for U, and post-focus vowels were less dispersed (more reduced) than focused ones. This indicates that the post-focus reduction is partly independent of deaccenting. However, a significant interaction of [StrLev] × [InfoStruc] was only present for the deaccenting group. A significant interaction of [VCat] × [StrLev] was obtained for both groups: spectral reduction induced by U was more drastic for /æ/ than for the other vowels, which indicates that stress-related reduction is not uniform across vowels.
No takes yet. Share an insight, caveat, or question.
Mariko Sugahara (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: