Research

Research Interest

My research revolves around the nature of the cognitive representations of human speech sounds and the relationship of these representations to their physical aspects.  Given this focus, I use empirical data to study the consequences of speaker-listeners’ linguistic experiences for their phonological (cognitive) knowledge, and their perception and production of speech.  My research has included investigating the effects of linguistic experience both on a macro level (e.g., learning a—first or additional—language) and on a micro level (e.g., exposure to speech stimuli in a lab for two minutes). 



Recent Projects
  • Cross-linguistic perception of consonant clusters 

I am currently working on a collaborative project with Prof. Ioana Chitoran at Université Paris Diderot, exploring cross-linguistic perception of onset consonant clusters. In this project, we explore the relation among  articulatory timing between consonants, prosody of native language, and perception/production of consonant clusters. This study compares perception of consonant clusters between speakers of Georgian, German, French, and Japanese. 



  • Dissertation follow-up: Spontaneous imitation in L2

I have long-standing interests in how language experience influences learners' phonological systems as well as their perception of speech. These interests lead me to a new question that builds on the results of my dissertation study: do bilingual speakers employ the same imitation strategy in their second language when it differs substantially from their native language in terms of the association of phonetic cues with corresponding phonological categories? To answer this question, I tested how stop VOT and post-stop F0 (the two cues for stop aspiration in Seoul Korean) operate when bilingual speakers of Seoul Korean and English imitate English voiceless stops, and I found that the bilingual participants imitated English /t/ with extended VOT not by raising post-/t/ F0 but by lengthening VOT, and F0-raised English /t/ did not induce imitative changes in the two acoustic properties measured. These findings suggest that bilingual speakers do not draw on their native cue primacy in performing imitation tasks in their second language, and adjust the phonetic properties that are relevant to the second language.



  • Dissertation: Cue Primacy and Spontaneous Imitation: Is Imitation Phonetic or Phonological?
My dissertation examines the difference between a primary cue and a non-primary cue in spontaneous speech imitation. By examining how two cues for aspirated stops in Seoul Korean operate in spontaneous imitation, this study aims to reveal whether the cognitive representations that are responsible for speech imitation are detailed phonetic properties (e.g., long VOT or high post-stop F0) or phonological categories (e.g., stop aspiration). Through a series of experiments,
I claim that speech imitation is mediated by language-specific associations between phonetic properties and phonological categories. 


  • Language experience, speech perception and loanword adaptation: Variable adaptation of English word-final plosives into Korean (Kwon, 2017)

This study, which started as my Qualifying Research Paper, investigated how Korean native listener-borrowers' experience with English influences their perception of English sounds, and how these influences lead to variable loanword adaptation. The loanword adaptation pattern under investigation is variable vowel insertion after English post-vocalic word-final plosives, such as 'cake' being adapted into 케익 /kʰɛik/ or 케이크 /kʰɛikʰɨ/. I propose that the seemingly random variation in loanword adaptation pattern is not a mere idiosyncrasy; it rather systematically reflects the perception of English sounds by different individuals with varying degrees of English experience. Showing that the variation originates in the different English experience of the listener-borrowers, this study aims to enhance our understanding of the role of speech perception in loanword adaptation and L2 perceptual learning.