Faking It: How performed versus authentic southern accents impact word recognition

Abstract

While listeners are not necessarily good at distinguishing performed versus real regional dialects (e.g., Heaton, 2019), there is also evidence that performed dialects can be inaccurate, limited to salient variables, and/or phonetically exaggerated (e.g., Schilling-Estes, 1998; Zetterholm, 2003; Bell & Gibson, 2011). In this study we investigate how performed speech impacts word recognition, comparing listener responses to performed Southern dialects in one auditory lexical decision study versus native Southern dialects in another. In both studies, stimuli included monosyllabic words with vowels from the PRIZE, KIT, DRESS, THOUGHT, STRUT and FACE lexical sets, which have been shown to be strong markers of Southern US English (e.g., Gunter et al., 2017). 123 native speakers of US English (46 Southern) heard Southern tokens from 4 women from Southwest Virginia, and 32 (15 Southern) heard Southern tokens from 6 actresses who did not identify as Southern. Interactions between the dialect of the speaker (real, performed), the dialect of the listener (Southern, Not-Southern), and the vowel in the word will be analyzed in terms of both accuracy and responses times.