Phonological preparation units in Korean speech production remain a topic of debate. Some studies support segmental (onset-based) units (Witzel et al., 2013; Han & Verdonschot, 2019), while others argue for syllable-level planning (Verdonschot et al., 2021) or propose an intermediate CV-body unit (Li et al., 2022). This study introduces the nucleus as a potential preparation unit, not independently tested in prior work. We employed a picture naming task with priming based on shared phonological units. Participants viewed a prime picture followed by a target picture, which they were instructed to name aloud as quickly as possible. The task was conducted in both Korean (n = 70) and English (n = 101) with native Korean speakers. Trials with response times below 150 ms or above 3 s were excluded from analysis. Among six priming conditions—no overlap, onset, nucleus, body (CV), rime (VC), and syllable—nucleus overlap yielded the strongest facilitation effect in both L1 and L2 word production, suggesting that the nucleus may function as a core preparation unit across languages. This effect was consistent across L2 proficiency levels, though the small group of advanced learners (n = 6) showed a relatively weaker effect. These findings offer new cross-linguistic evidence for nucleus-based encoding and underscore the value of testing sub-syllabic components in speech production models.
Hayeun Jang (Wed,) studied this question.