The Role of Orthographic, Phonetic, and Semantic Information in L2 Chinese Character Processing: A Comparison of Alphabetic, Japanese, and Korean Learners

Previous studies indicated that word processing in English as a second language (L2) is affected by learners' native language. However, little is known about L2 Chinese character processing. The present study aims to fill this gap by investigating the role of orthographic, phonetic, and semantic information in L2 Chinese character processing by learners from different native language backgrounds.

In this study, 41 Japanese learners, 66 Korean learners, and 69 learners using different alphabetic languages (e.g., English, Russian, and Turkman) were recruited. Based on the participants' native language writing systems and exposure to Chinese characters, we categorized them into three groups: alphabetic (sound-based writing system), Korean (sound-based writing system and belonging to Sinosphere), and Japanese (using Chinese characters belonging to the meaning-based writing system). A picture-pseudo character matching experiment was conducted to explore learners' reliance on semantic and phonetic radicals when processing a compound character. In each experiment trial, the participants were provided with a picture and an audio description indicating the meaning and pronunciation cues of a newly created object. They were required to rank five pseudo characters to show how well each character matched the object. The positions of phonetic and semantic radicals in the pseudo characters and the radicals' relatedness to the meaning or pronunciation cues provided were manipulated.

The results indicate that firstly L2 Chinese learners were generally sensitive to radicals' phonetic and semantic information and radicals' conventional positions. Pseudo-characters with cue-related radicals received significantly higher scores than pseudo-control characters without cue-related radicals. Pseudo-characters with radicals in conventional positions received significantly higher scores than those in unconventional positions. Secondly, learners' native languages influenced their radical reliance on character processing. When the radicals were in conventional positions, alphabetic learners significantly relied more on phonetic than semantic information. Korean learners showed a preference for phonetic information and Japanese for semantic information. When the radicals were in unconventional positions, alphabetic and Korean learners relied equally on semantic and phonetic information, while Japanese learners significantly relied more on semantic information than phonetic information. Thirdly, Japanese learners were more sensitive to radicals' conventional position than Korean and alphabetic learners. No significant difference was found in Japanese learners' ranking between pseudo-control characters (the radicals were in conventional positions but unrelated to semantic and phonetic cues) and characters with related phonetic radicals in unconventional positions. In contrast, Korean and alphabetic learners significantly ranked the pseudo-control characters as the least matching labels for the objects.

The findings of this study delineate that alphabetic learners relied more on phonetic information. In contrast, Japanese learners relied more on semantic information in character processing, while Korean learners relied equally on phonetic and semantic information. These findings shed light on L2 Chinese character processing by revealing learners' sensitivity and reliance on radicals' orthographic, phonetic, and semantic information. The pedagogical suggestions for teaching Chinese characters to L2 Chinese learners of different native language backgrounds are provided.