Presentation
Search Abstracts | Symposia | Slide Sessions | Poster Sessions | Lightning Talks
Investigating the roles of acoustic-phonetic and lexico-semantic processing in resolving perceptual ambiguity in Spanish-accented English
There is a Poster PDF for this presentation, but you must be a current member or registered to attend SNL 2023 to view it. Please go to your Account Home page to register.
Poster E78 in Poster Session E, Thursday, October 26, 10:15 am - 12:00 pm CEST, Espace Vieux-Port
This poster is part of the Sandbox Series.
Holly Zaharchuk1, Janet van Hell1; 1The Pennsylvania State University
Listeners exhibit remarkable speed and plasticity in adapting to L2-accented speech. However, the neurocognitive mechanisms that drive these rapid adaptation effects are currently unknown. To investigate these mechanisms, we designed an EEG/ERP study focusing on voice onset time (VOT), the primary acoustic-phonetic cue that distinguishes voiceless stop consonants /ptk/ from their voiced counterparts /bdg/. Short-lag VOTs characterize /ptk/ in Spanish and /bdg/ in English. Since L1 Spanish bilinguals tend to transfer these short-lag VOTs for /ptk/ into their L2 English, Spanish-accented English /ptk/ are equally likely tokens of /ptk/ or /bdg/ from a listener’s perspective. Thus, listeners must learn new mappings between VOT and /ptk/ in order to achieve successful comprehension of Spanish-accented English. We wanted to track the time-course of this adaptation process with EEG. Seventy monolingual English speakers will participate in this study during Summer 2023. EEG will be recorded during a cross-modal priming task, in which participants indicate whether an auditory target matches a visual prime. Auditory targets are English nouns of two types: multisyllabic words without onset competitors (e.g., pencil/*bencil), to expose participants to voiceless stops in unambiguous lexical contexts, and monosyllabic words with relevant onset competitors (e.g., park/bark), to test learning over time in ambiguous lexical contexts. A Spanish-accented bilingual female speaker from Mexico City, Mexico, recorded the stimuli (432). Critical targets (72 each mono/multisyllabic) have /ptk/ onsets, while filler targets (144 each mono/multisyllabic) have /mnɹlhw/ onsets. Each target is proceeded by one of three Prime Types: Identity (e.g., park), Competitor (e.g., bark), or Control (e.g., wand). ERPs will be time-locked to the onset of each auditory target. We will conduct mixed-effects models on trial-level ERPs for critical monosyllabic targets in three time windows: 150-350 ms, 350-600 ms (N400), and 600-900 ms (late N400). We predict an interaction between Prime Type and Duration at 150-350 ms. As the experiment progresses, participants will learn to relate the Spanish-accented talker’s VOT distributions to the appropriate phonemes. To the extent that adaptation is similar to rule-learning, the difference between Identity and Competitor primes should emerge on the P2/P200 component. If this difference is observed on the N200/PMN, this would suggest that adaptation involves updating expectations about the speech signal. The emergence of neurocognitive differences between Identity and Competitor primes should also be reflected behaviorally, with decreasing RTs and increasing accuracy for Competitor primes over time. If the behavioral results do not demonstrate learning of Spanish-accented VOTs, this would suggest that online sensitivity to accent-shifted VOTs develops before offline performance improves. Control primes should elicit larger effects than Identity or Competitor primes throughout the experiment. On the N400, participants should show consistently larger effects for Competitor and Control primes than for Identity primes. If an interaction between Prime Type and Duration is observed on the N400 but not on the P2/P200 or N200/PMN, this would suggest a lexico-semantic mechanism for adaptation to L2-accented speech. Overall, this work unites behavioral measures of perceptual adaptation with neurocognitive measures of online processing to gain a deeper understanding of speech recognition.
Topic Areas: Speech Perception, Phonology