|
Abstract:
|
Speech sounds are typically perceived categorically . The acoustic information in speech sounds is perceptually grouped into phonetic categories . It is widely known that language influences the way speech sounds are categorized . That is , one's native language influences where category boundaries are placed . However , it is less understood how bilingual listeners categorize speech sounds . There is evidence showing that bilinguals have different category boundaries from monolinguals , but there is also evidence suggesting that bilinguals have different category boundaries depending on the language they are using at the moment . This phenomenon has been referred as the double phonetic boundary . The goal of this investigation was to verify the existence of the double phonemic boundary in bilingual listeners . As has been done in other studies , bilingual speakers of Spanish and English were asked to identify the speech sound /ta / from a 10 -token speech continuum ranging in VOT from /da / to /ta / in two language contexts . In this study , however , two additional procedures were carried out . First , English monolinguals were asked to identify the continuum in two language contexts . It was expected that bilinguals , but not monolinguals , would show a double phonetic boundary . Second , while participants' behavioral measures were assessed , electrophysiological measures [event -related potentials , (ERPs )] also were recorded . This was done in order to observe how speech sounds are represented in the brain . It as expected that bilinguals , but not monolinguals , would show different ERP amplitudes across language contexts . The behavioral results showed that phonemic boundaries did not differ across language contexts for either bilinguals or monolinguals . Further analyses showed bilinguals , but not monolinguals , perceived specific speech sounds - -in the "ambiguous zone" - -differently across language contexts . The electrophysiological results showed that the ERPs of bilinguals , but not monolinguals , differed across language contexts . Interestingly , behavioral measures correlated significantly with electrophysiological measures only in bilinguals . This result showed that the ERP amplitude was in accordance with the number of sounds perceived as 'ta' across language contexts . The challenges of testing the double phonemic boundary are discussed , along with the limitations of the methodology used in this study . |