| dc.description.abstract | In bilingual language processing, the parallel activation hypothesis suggests that bilinguals activate their two languages simultaneously during language processing. Support for the parallel activation mainly comes from studies of lexical (word-form) processing, with relatively less attention to phonological (sound) processing. According to studies of monolingual phonological processing, phonotactic probability, the likelihood of occurrence of a sound sequence, influences both word recognition and production. Specifically, common sound sequences are recognized and/or produced more quickly and more accurately than rare sound sequences (Edwards, Beckman & Munson, 2004; Vitevitch, Luce, Charles-Luce, & Kemmerer, 1997; Vitevitch & Luce, 1998; 1999). The goal of this research was to examine the influence of phonotactic probability on phonological processing when phonotactic probability was matched (Experiment 1) versus mismatched (Experiment 2) across the bilingual speakers' two languages. In Experiment 1, three groups of children participated: English monolinguals, Korean monolinguals, and Korean-English bilinguals. A same-different task with nonword stimuli was used. The nonwords were matched in phonotactic probability across the two languages (i.e., English-low/Korean-low versus English-high/Korean-high). Results showed that all three groups responded more accurately and quickly to English-high/Korean-high than English-low/Korean-low nonwords. This replicates past findings of the facilitative effects of phonotactic probability for English monolinguals and extends it to Korean monolinguals and Korean-English bilinguals. In Experiment 2, only bilingual children participated in a same-different task with nonword stimuli mismatched in phonotactic probability. Specifically, phonological processing of English-low/Korean-high versus English-high/Korean-low nonwords was examined across two phonetic contexts (i.e., English-phonetic and Korean-phonetic). Phonetic context was based on the speaker who recorded the stimuli (i.e., native English versus native Korean speaker). Results showed a significant interaction between phonotactic probability and phonetic context. In the English-phonetic context, English-low/Korean-high nonwords were responded to more accurately and quickly than English-high/Korean-low nonwords. In contrast, in the Korean-phonetic context, English-high/Korean-low nonwords tended to be responded to more accurately and quickly than English-low/Korean-high nonwords. The results are interpreted as bilinguals encounter competition effects of phonotactic probability on phonological processing when the probability was mismatched across the two languages. This competition effect from mismatched probability suggests the presence of parallel activation of both languages in phonological processing. Combined with the results of Experiment 1, the magnitude of parallel activation was found to vary across the native and non-native languages. Specifically, both facilitation and competition effects were significant in non-native (English) language processing, while the effects were not significant in native (Korean) language processing. Such an asymmetry in the magnitude of parallel activation between the native and non-native languages is consistent with previous findings of parallel activation in bilingual lexical processing. Taken together, the findings of the current study suggest that phonological representations of the two languages are activated simultaneously and language status may be a factor that mediates the magnitude of parallel activation. | |
