Abstract

Romanization is a common, yet sometimes controversial, tool used to make Korean and Japanese more accessible to beginning learners, but its effects on orthographic processing and phoneme discrimination ability are not well understood. The participants consisted of 34 university students (21 female, 13 male; mean age = 20.97) enrolled in 100- and 200-level Korean and Japanese courses at Brigham Young University. Using a quasi-experimental, counterbalanced design, participants completed phonological judgement tasks in three orthographies per language while accuracy, reaction time, and cortical activation were measured using functional near-infrared spectroscopy (fNIRS); post-task surveys contextualized learner responses. Orthography significantly affected accuracy in Japanese, F(2, 48) = 10.69, p < .001, ηp² = .31, and Korean, F(2, 48) = 4.50, p = .016, ηp² = .16, with Modified Hepburn (94%) and Revised Romanization (88%) yielding the highest accuracy. Reaction time differed by orthography in Japanese, F(2, 48) = 12.27, p < .001, ηp² = .34, but not in Korean, where course level predicted reaction time, F(1, 49) = 12.87, p < .001, ηp² = .21. Romanized orthographies primarily engaged the left inferior parietal lobule and left precentral gyrus, whereas native scripts recruited additional temporal, frontal, and visual regions. These findings indicate that orthographic familiarity and proficiency shape behavioral performance and neural processing, with implications for script sequencing in Korean and Japanese instruction.

Degree

MA

College and Department

Humanities; Center for Language Studies

Rights

https://lib.byu.edu/about/copyright/

Date Submitted

2026-06-26

Document Type

Thesis

Keywords

Japanese, Korean, language processing, neurolinguistics, second language learning

Language

english

Download

Share

COinS