| BACKGROUND: Language is the important tool for human communication and the important feature that makes us a human. A phoneme is the smallest unit of speech sounds in any given language that serves to distinguish one word from another. Almost all languages have their local phonological variation called dialects. One aspect of dialects is the variation of phonemes. Phonemic merger is a common phenomenon of language variation in which two phonemes in one dialect of a language are pronounced contrast, but are merged, or not distinguished in pronunciation, in another dialect of this language. For example: the pin-pen merger and bin-ben merger occur in English, and the /ε/–/e/ merger occurs in French and so on. In Chinese language, phonemic merger also occurs. For example, in the main dialects in the south of the Yangtze River in Anhui Province, the two constants /n/ and /l/ are often merged as /n/ or /l/. The people from this area often pronounces both niu1 (cattle) and liu1 (slip) either as liu1 or as niu1. However, in the main dialects of the north of the Yangtze River in Anhui Province, the two constants are clearly distinguished in pronunciation. More interestingly, behavioral studies have shown that subjects come from the merged area also have difficulty to perceive the two merged phonemes. For example, the subjects could not distinguish the two sounds niu1 and liu1 in auditory perception. Currently, only few electrophysiological studies have been carried out for exploring the neural basis of phonemic merger. The other question related to phonemic merger is what would happen when people encounter the same merged phonemes during the foreign language learning ? This thesis aims to to explore the neural basis of phonemic merger at the early stage of audiotry processing and examine whether this phenomenon occurs during the foreign language learning.METHODS: Fifteen subjects come from /n/-/l/ merged area as the Merged group, and the other fifteen subjects come from /n/-/l/ unmerged area as the Unmerged group. All of them are undergraduate or graduate students who are native Chinese-speakers, right-handed with normal hearing and without professional music training experience. Subjects were presented with Chinese auditory words (liu1 and niu1) and English words (line and nine) that contained phonemes /n/ and /l/ using a passive oddball paradigm. The mismatch negativity (MMN), a component of the auditory event-related potential (ERP), was recorded. Subjects were also behaviorally tested for their auditory ability to discriminate the two phonemes.RESULTS: (1) MMN amplitude recorded from the Merged group was significantly smaller than that from the Unmerged group when using Chinese words as auditory stimulation (-0.586±0.65 vs. -1.560±0.70μV, n=30, P < 0.000); (2) MMN amplitude of the Merged group was also significantly smaller than that of the Unmerged group when using Chinese words as auditory stimulation (-0.638±0.50 vs. -1.43±0.69μV, n=30, P < 0.001); (3) in the behavioural test, not only the ability to distinguish phonemes /n/ and /l/ of the Merged group was significantly poorer than that of the Merged group when Chinese words were used (performance accuracy: 72.10%±19.71 vs. 98.67%±1.55, n=30, P< 0.000) but also poorer when English words (performance accuracy: 68.88%±19.71 vs. 98.67%±2.45, n=30, P< 0.000) were used.CONCLUSIONS: A lower MMN amplitude in the Merged group than that in the Unmerged group indicates that phonemic merger occurs because lack of activation of memory traces at the early stage of auditory processing and these memory traces are dialect-specific. The Merged group cannot distinguish the phonemes /n/ and /l/ in Engish, suggesting that the formation of a foreign language phonological system is determined by the phonological system of native language and phonemic merger could affect the phoneme perception during foreign language learning. |
PDF Full Text Request