Conceptual Processing in Chinese-English Bilinguals: An fMRI Study of Cross-Language Conceptual Priming (Brief Report)

In recent years, functional neuroimaging and cortical stimulation methods have been used to explore neural organization of the bilingual brain, and one of the areas that has received considerable attention is the bilingual conceptual system. While some studies have provided evidence for distinct, non-overlapping representations between the first language (L1) and the second language (L2) (e.g., Kim, Relkin, Lee, & Hirsch, 1997), other studies have provided support for different, yet overlapping representations between the two conceptual systems (e.g., Illes et al., 1999; Klein, Milner, Zatorre, Meyer, & Evans, 1995; Klein, Milner, Zatorre, Zhao, & Nikelski, 1999). Factors such as age of L2 acquisition (e.g., Chee, Hon, Lee, & Soon, 2001), L2 proficiency (e.g., Perani et al., 1998), L1 literacy experience (e.g., Wang, Koda, & Perfetti, 2003), and L1 and L2 similarity (e.g., Gandour et al., 2000) have been proposed to explain these discrepant findings (see Fabbro, 2001 for a review). In this study, we explored the degree of overlap between the neural representations of Chinese and English, two languages that are different on orthographic, syntactic, and acoustic levels (Wang, Inhoff, & Chen, 1999). We examined conceptual processing in Chinese-English early bilinguals using a repetition priming procedure during functional magnetic resonance imaging (fMRI). Repetition priming refers to an individual’s enhanced performance on a task as a result of repeated exposure to the same stimulus. On a behavioral level, this improvement is manifested as a reduction in response times and error rates. The underlying mechanisms of repetition priming effects have been well characterized by the theory of transfer appropriate processing (Morris, Bransford, & Franks, 1977). Briefly, the logic is that the processing involved during the first encounter of a stimulus may be transferred to the second encounter of the same stimulus, and the amount of benefit experienced during the second exposure is proportional to the degree of overlap in processing that is required by both exposures (see Roediger, Weldon, & Challis, 1989 for a review). In other words, the magnitude of the priming effect can be interpreted as an index of shared processes between the two encounters—the greater the overlap, the bigger the priming effect. In a series of behavioral experiments with monolingual English subjects, we have previously established that repetition of conceptual processing is both necessary and sufficient to produce a facilitation effect on a verb generation task (ThompsonSchill & Kan, 2001; see Zeelenberg & Pecher, 2003 for a bilingual study using a similar paradigm). In a verb generation task, subjects are asked to generate verbs (e.g., “eat”) in response to concrete nouns (e.g., “apple”). Furthermore, we have also observed a physiological priming effect (i.e., decreased blood flow) in an fMRI experiment using the same paradigm (Thompson-Schill, D'esposito, & Kan, 1999). Thus, this methodology seems ideal in the exploration of conceptual processing in bilingual individuals. If the concept “apple” were represented by different conceptual systems as a function of language, we would not expect activation of the concept in one language (e.g., Chinese) to facilitate processing of the concept in the second language (e.g., English). On the other hand, if the concept “apple” shared a common representation between the two languages, we would expect a facilitation effect in subsequent processing of the same concept, even when the processing occurred in a different language. We hypothesized that if Chinese and English do in fact share a common conceptual system, we should observe a behavioral cross language priming effect. Furthermore, if the neural