単語表記の親近性が語構成文字の認知に及ぼす影響

抄録

Two experiments were conducted to investigate the effects of script familiarity on the recognition of component letters in Japanese words which were represented in katakana (a moraic script). The participants in Experiment 1 were briefly presented with either a masked katakana word (a familiar script word, マスコミ) or a masked katakana-hiragana mixed word (an unfamiliar script word, マすコみ), followed by a target letter (e.g., コ). In Experiment 2 the participants were presented with the target letter, followed by a brief presentation of either the familiar or unfamiliar script word. The task in both experiments was to decide whether the target letter was presented within the word. In both experiments the correct recognition rate for katakana was higher for familiar script words than for unfamiliar script words. However, in unfamiliar script words the correct recognition rate was higher for katakana than for hiragana in Experiment 1, and the reverse was observed in Experiment 2. These results are interpreted in terms of an interactive-activation model.