Abstract
The processing of Japanese written words has been supposed to depend only on their script, namely Kanji or Kana. There are, however, Kanji-Kana-combined-words (KKW; e.g., _??__??__??_), which signify their meanings as a whole. The purpose of this study was to examine whether the recognition unit corresponding to visual form information of the KKW is used in the lexical decision task. Experiment I showed that the more the number of the Kana letters was, the longer the recognition time of the KKW with unfamiliar script (e.g., _??__??__??_) was, while the effect was not obtained for the KKW with familiar script (e.g., _??__??__??_). In Experiment II, the lexical decision latency tended to be reduced by the existence of blank (s) between letters in the KKW with unfamiliar script. However, this effect was not obtained for the KKW with unfamiliar script. All these results suggest that the KKW is processed using the recognition unit which corresponds to visual form information of the whole-word.
