A Comparison of Transposed-Character Effect Between Chinese and Japanese Speakers: Processing Position Information of Hanzi and Kanji Characters.

Transposed-letter (TL) effect implies that TL pseudowords (e.g., mnokey) are more likely to be recognized as their base words (e.g., monkey) than substituted-letter nonwords (e.g., markey). Compared to TL effect studies, transposed-character (TC) effect studies on logograms, such as Chinese Hanzi and Japanese Kanji, remain limited. This study compared the TC effect in Chinese and Japanese speakers through two experiments. Experiment 1 is a single-word lexical decision task, in which 20 Chinese and 21 Japanese speakers were presented the same two-character stimuli, including TC pseudowords (e.g., -), control nonwords with a substituted character (e.g., ), and real words as fillers. Chinese participants exhibited a TC effect: they took longer to reject TC pseudowords and misidentified them more frequently than they did control nonwords. Conversely, the Japanese participants showed no TC effect, and their reaction times and error rates showed no significant differences between the two conditions. In Experiment 2, 63 Chinese and 63 Japanese speakers participated in a masked-priming lexical decision task. The TC pseudowords (e.g., -) and control nonwords (e.g., ) used in Experiment 1 and their base words (e.g., ) were used as the three primes and prime times (i.e., 60 ms, 150 ms and 300 ms). Although the results show that TC effects were observed in both Chinese and Japanese speakers, the TC effect size was larger for Chinese speakers than for Japanese speakers. In conclusion, this study found that there is a difference in the strictness of Hanzi/Kanji character position information processing between Chinese and Japanese speakers. These findings suggest that Chinese speakers process Hanzi/Kanji characters more flexibly than do Japanese speakers.