Individual listeners’ perceptual cue weight and the time course of cue integration
Much of my research is focused on further elucidating individual listeners’ time course of acoustic cue integration in lexical access and the relationship between cue weighting and cue integration strategy. As a test case, my dissertation project investigates the time course of acoustic cue integration in the processing of the Korean three-way laryngeal stop contrasts by native Korean listeners and English L2 learners of Korean. Using cue-weighting speech perception tasks and the visual-world eye-tracking paradigm, the project seeks to understand how L1 and L2 listeners weight and integrate fine-grained acoustic information in spoken word recognition. Specifically, the project explores three critical research questions in speech perception and spoken word recognition:
Are acoustic cues integrated continuously when the weight of the cue available later is greater than that of the cue available earlier?
Do listeners’ cue weighting and cue integration strategies in the L1 affect how they weight and integrate acoustic cues that signal a similar phonological contrast in the L2?
Does an individual’s cue weighting predict their cue integration strategy?
My dissertation project aims to address these questions by investigating: (a) how individual native Korean listeners weight and integrate Voice Onset Time (early cue) and onset fundamental frequency of the following vowel (later cue) when hearing individual Korean words that begin with a laryngeal stop; (b) how individual English L2 learners of Korean weight and integrate VOT and onset F0 of the following vowel when hearing individual Korean words that begin with a laryngeal stop.
Effect of L1 lexical encoding of acoustic cues in L2 speech perception
I have been carrying out research framed within the cue-weighting theory of speech perception that seeks to test cue-weighting transfer in L2 learners’ processing of suprasegmental lexical contrasts (Kim & Tremblay, 2021, 2022; Tremblay, Broersma, Zeng, Kim, Lee, & Shin, 2021). The cue-weighting theory emphasizes that speech perception is multidimensional, and acoustic cues are weighted differently not only across categories but also across languages: Even if speakers of different language backgrounds hear the same acoustic stimuli, their weighting of acoustic cues varies depending on their L1. Accordingly, the cue-weighting theory stipulates that the contribution of individual acoustic cues to distinguish among phonological categories transfers from the L1 to the L2. The main question explored in my research is whether listeners’ use of prosodic cues to lexical contrasts can transfer from one phonological phenomenon in the L1 to another in the L2, and what might be the scope of this cue-weighting transfer.
Speech production and sound change
Another important research area in which I have been invested is speech production and sound change. Particularly, I have been interested in the linguistic and non-linguistic factors that drive sound change in a speech community, inferred from individual speakers’ speech production. One of my research projects investigated the acoustic correlates of the Korean three-way laryngeal stop distinction in Gyeongsang Korean speakers living in Seoul (Kim & Jongman, 2022). The study aimed to examine how and to what extent inter-dialect contact with Seoul Korean affects Gyeongsang speakers’ use of acoustic cues to the laryngeal stop contrast in speech production, focusing on VOT and onset F0. Comparisons between the dialectal groups with different degrees of exposure to Seoul Korean suggest that the recent change in the Korean stop contrasts observed in the Gyeongsang speech community is mainly due to inter-dialect contact.
Loanword Phonology
Words are often borrowed from one language to another. These borrowed words are often called loanwords and undergo modification processes to conform to the allowable sound pattern of the borrowing language. Such modification processes make loanword adaptation interesting, as the phonological patterns in loanwords often provide insight into native speakers’ knowledge that may not be necessarily obvious in the native data. As such, loanword adaptation can be considered a real-life wug test (Kang, 2011). There have been a variety of theoretical and empirical approaches to describe loanword phonology and the input representation of the adaptation process. Such approaches include phonological adaptation (e.g., LaCharité & Paradis, 2005; Paradis & LaCharité, 2008), perception-based adaptation (e.g., Peperkamp & Dupoux, 2003; Peperkamp, 2005), orthography-based adaptation (e.g., Daland et al. 2015; Kim, 2020), and statistical tendencies in the lexicon of the borrowing language (e.g., Zuraw, 2000). I am mainly interested in further elucidating the role of speech perception in loanword adaptation to better understand how and to what extent listeners’ perceptual patterns modulate the loanword adaptation.