Role of Syllable and Syllable Subunits in Korean Word Recognition: Word Megabase Analysis

The present study is to investigate the major representational units involved in Korean word recognition in reading by examining the frequency effects of syllables and their subsyllabic components, specifically antibody and rime units.In the field of psycholinguistics, traditional research has typically relied on lexical decision experiments in which researchers pre-select target stimuli according to the conditions they aim to manipulate and then analyze the reaction times elicited from participants.Recently, however, large-scale word mega-databases constructed with extensive vocabularies and broad participant samples across various countries have enabled researchers to investigate lexical-variable effects without conducting new experiments.Using the Korean word mega-database, the present study examined whether the previously reported effects of syllable and antibody units observed in controlled laboratory settings would converge with the patterns derived from large-scale corpus-based data.Compared with traditional experiment-based approaches, mega-databases provide frequency information for a vast range of words and word forms, offering a more ecologically valid reflection of reading performance.The results showed that higher syllable token frequency inhibited word recognition, consistent with earlier findings, whereas syllable type frequency did not yield any significant effect.In comparisons between antibody and rime frequencies, increased antibody frequency was associated with slower word-recognition performance, again replicating prior observations.By contrast, rime frequency showed no detectable influence, revealing a Korean-specific pattern in which subsyllabic competition appears to operate primarily at the antibody level rather than at the rime level.