Abstract
We investigated the recognition units of single-kanji and two-kanji-compound words in lexical access. In Experiment 1, we examined if the word recognition is facilitated in two conditions: (1) when the prime and the target words are identical and (2) when the prime and the target words contain an identical kanji. The priming effect was found only when the compound targets were identical to the prime. In Re-experiment 1, facilitation among kanji words was re-examined. The priming effect was obtained in the identical targets in both single and compound words. In Experiment 2, facilitation of word recognition among single kanji words was investigated. The results showed that facilitation was observed when the target was identical to the prime. Taken together, these results indicated that the kanji word recognition in lexical access is accomplished by word-sized units rather than by character-sized units. The findings were discussed with reference to the logogen model (Morton, 1977).
