This study examined the interaction between semantic and orthographic processing in Korean learners of English as they visually recognize English words. Fifty Korean learners of English performed a visual lexical decision task consisting of 200 stimuli: 50 loanwords, including 25 with a single phonological competitor in Korean and 25 with multiple semantic competitors, 50 control words, and 100 nonwords. Response times were analyzed using linear mixed-effects models, and accuracy was analyzed using logistic mixed-effects models. The results showed that loanwords elicited significantly faster responses and higher accuracy than control words. However, the number of semantic competitors and the frequency of use in Korean were not significant predictors. Instead, English lexical frequency significantly influenced both response times and accuracy, and the degree of orthographic (spelling-to-pronunciation) opacity, the extent to which spelling and pronunciation diverge, emerged as an additional key predictor of performance. Specifically, words with higher orthographic opacity tended to yield slower responses and lower accuracy. These results indicate that L2 lexical processing in Korean learners is primarily driven by within-language factors, including English lexical frequency and orthographic opacity, rather than by L1-based familiarity. This finding suggests that cross-linguistic influences on L2 word recognition are constrained by the organization of orthographic and phonological representations in L2 and the strength of orthography-phonology mappings.
Lee et al. (Fri,) studied this question.