Automatic Activation of Linguistic Information in Chinese Character Recognition

This chapter presents an experiment that examinee fluent Chinese readers' character recognition performance with a modified Stroop interference paradigm (that is, picture-character interference) to see the cues that were effective and the ways in which they were integrated during reading. Depending on the relationship between the object depicted in each picture and the printed character, seven experimental conditions were created. Several interesting results were observed. First, the subjects' picture-naming performance was disrupted more when the printed character and the character representing the name of the picture shared neither graphic nor phonological information. Second, little disruption was observed when the distracting symbol was a pseudo-character, suggesting that its non-lexical status prevented the activation of other linguistic information (for example, phonological and graphic clues) embedded in the stimulus. Third, when the disrupting symbol was a real character, the magnitude of interference was the same for lack of graphic similarity as for lack of phonological similarity, supporting the notion that both graphic and phonological information were automatically activated in Chinese character recognition. Moreover, the integration of these two types of information was found to be additive in nature.