Impact of syllabic constraints on error production in Korean speech: a study of serial order control with practice

This study tested whether serial order in Korean speech production is shaped by a language-specific, hierarchical syllabic organization. Using a speech-error induction paradigm, native Korean speakers produced four-syllable CVC nonwords while hearing auditory nonword distractors that targeted specific within-syllable positions (onset, vowel, coda). We hypothesized tighter coupling of onset–vowel (CV/body) than vowel–coda, predicting greater distractor-induced errors at codas. The results supported this prediction, with coda positions showing reliably higher error rates than onsets and position-specific memory effects concentrated in coda slots, indicating an interaction between syllabic status and serial memory. With repetition, errors declined selectively for onset–vowel pairings, whereas coda vulnerability persisted, showing that practice strengthens serial stability primarily where the hierarchical linkage is already strong. Together, these findings demonstrate that the ordering of segments in Korean is not governed by a flat, segment-by-segment sequence; rather, it reflects a language-specific hierarchy in which CV/body functions as a more cohesive unit than VC. The finding thus provides convergent evidence that serial order control in Korean is constrained by its syllabic architecture, with implications for cross-linguistic models of phonological encoding and for theories that integrate hierarchical structure into serial planning mechanisms.