THE EFFECT OF TALKER ACCENT ON PERCEPTUAL CUE WEIGHTING IN ENGLISH TENSE-LAX VOWELS: NATIVE SPEAKERS AND KOREAN L2 LEARNERS | Hyunjung So | ResearchPod

Transcript: THE EFFECT OF TALKER ACCENT ON PERCEPTUAL CUE WEIGHTING IN ENGLISH TENSE-LAX VOWELS: NATIVE SPEAKERS AND KOREAN L2 LEARNERS

Sam: Native English listeners may change how they hear a vowel just because the speaker has a Korean accent. In this study, they leaned harder on vowel length to tell "seat" from "sit", while advanced Korean learners didn't budge.

Alex: Before the details, place it for me. Whose work is this, and what kind of study?

Sam: It's a perception experiment by Hyunjung So at Pusan National University, in the journal Phonetics and Speech Sciences. It grew out of a master's thesis. Sixteen native English listeners took part, along with fourteen advanced Korean learners of English.

Alex: I work near this field, not in it. Why should someone like me care whether listeners shift a vowel judgment?

Sam: Because it tests whether an accent works as social information on its own. Nobody tells you who the speaker is. If the accent alone retunes how you parse sounds, perception is pulling on what you know about groups of talkers. It also bears on how we teach advanced learners. We'll cover the background, the one trick in the design, what they found, and how far to trust it.

Alex: Start with the background. What makes "seat" versus "sit" a good test case?

Sam: English has what phoneticians call tense–lax pairs, vowels like the ones in seat and sit. Native speakers mainly separate them by vowel quality. That's the spectral cue, set by where the tongue sits. Vowel length is a secondary cue. Korean has no such pair. So Korean learners tend to lean on length, the vowel in seat just being longer.

Alex: So the two groups weigh the same two cues differently. Where does accent come in?

Sam: Through Exemplar Theory, the idea that we store each heard utterance as a rich memory, voice and speaker included. Over many encounters, those memories cluster. So hearing a Korean accent could activate stored knowledge about Korean talkers. That includes the fact that they tend to mark this contrast by length. If so, a native listener should give length more weight after hearing that accent.

Alex: Has anyone shown that before, or is this the first try?

Sam: The direct predecessor is Zhang, from 2022, with Chinese learners and Chinese-accented English. Native listeners there reduced their reliance on vowel quality when they heard the Chinese accent. The learners didn't shift at all. This study asks whether that generalizes to a new accent and a new learner group.

Alex: Walk me through the design. How do you get accent in without changing the target word itself?

Sam: There were two experiments. The first was a baseline. Listeners heard a synthetic grid of seat-to-sit tokens, with vowel quality varied in seven steps and vowel length in seven steps, crossed. They just picked seat or sit. That shows how much each group relies on each cue. The paper calls this cue weighting.

THE EFFECT OF TALKER ACCENT ON PERCEPTUAL CUE WEIGHTING IN ENGLISH TENSE-LAX VOWELS: NATIVE SPEAKERS AND KOREAN L2 LEARNERS

Hyunjung So

Oct 11, 2026·7:59·Original Paper

Perceptual cue weightingExemplar TheoryTense-lax vowel distinctionSecond-language (L2) speech processingSocial-indexical informationSpeech Perception

Listen

8 min

Summary

Research Question and Approach

This study investigates how implicit sociolinguistic cues—specifically, a talker's accent—influence the perceptual strategies listeners use to distinguish English tense–lax vowel pairs (e.g., /i/ versus /ɪ/). Drawing on Exemplar Theory, the researcher hypothesized that listeners would associate a Korean accent with the tendency of Korean talkers to rely on vowel duration rather than spectral cues (formant frequencies) for this distinction. The study compared native English speakers and advanced Korean learners of English across two experiments, measuring whether participants dynamically shifted their perceptual weights for spectrum and duration when exposed to American-accented versus Korean-accented carrier sentences.

Key Findings

Experiment 1 established that native speakers rely on both spectral and duration cues, while advanced Korean learners demonstrate a heavy, near-exclusive reliance on duration. In the main experiment (Experiment 2), native speakers exhibited significant perceptual flexibility: when primed with Korean-accented English, they increased their reliance on duration cues while maintaining their spectral weighting. This suggests native listeners use accent as a social-indexical cue to update their listening strategy. Conversely, the Korean learner group showed no significant shift in perceptual weighting across accent conditions. The author suggests this may be due to the cognitive demands of second-language processing or a "ceiling effect," as these learners were already relying on duration cues almost exclusively, leaving little room for further adjustment.

Why It Matters

These findings suggest that speech perception is not merely a static decoding process but is deeply integrated with a listener's social knowledge and accumulated linguistic experience. For educators, the results highlight that high standardized test scores do not necessarily equate to native-like perceptual flexibility. This suggests that advanced learners might benefit from training that explicitly targets the use of spectral cues and exposes them to a diverse range of talker accents to build a more robust, flexible set of perceptual exemplars.

Key Terms and Definitions

Exemplar Theory — A cognitive model proposing that listeners store individual speech experiences (exemplars) that include both acoustic information and social context, which are then used to predict and categorize future speech input.
Perceptual Cue Weighting — The relative importance a listener assigns to different acoustic features (such as vowel duration versus spectral formants) when identifying speech sounds.
Tense-Lax Vowel Contrast — An English phonological distinction where vowels differ in tongue height, muscle tension, and duration; Korean lacks this contrast, often leading learners to rely primarily on duration.
Syllable-Timed Language — A rhythmic classification (like Korean) where syllables are roughly equal in duration, contrasting with English's stress-timed rhythm; the study manipulated this prosodic feature to create authentic-sounding Korean-accented stimuli.

THE EFFECT OF TALKER ACCENT ON PERCEPTUAL CUE WEIGHTING IN ENGLISH TENSE-LAX VOWELS: NATIVE SPEAKERS AND KOREAN L2 LEARNERS: Research Paper Podcast