華語聲調知覺處理之時序問題初探

Abstract

聲調與⾳音素在聲調語⾔言之⼝口語感知均具舉⾜足輕重之地位，但前⼈人研究對於聲調(超⾳音段訊息) 及⼦子⾳音⺟母⾳音(⾳音段訊息)的整合與競存以及何種實驗模式較適合研究聲調感知並無定論。 本研 究提出三年之提案，計畫使⽤用同樣的刺激調查研究此課題。第⼀一年進⾏行柵欄典範(Gating paradigm) 實驗，以精準控制受試者所接收之語⾳音表徵，了解四個聲調被提取的時間是否有 差異，並探討鼻⾳音韻尾對聲調辨別速度可能產⽣生的影響。第⼆二年以⾳音素監測(Phoneme monitoring)作為範本，進⾏行⺟母⾳音及聲調的監測，⽐比較各聲調監測的正確率及反應時間，並對 其訊息提取速度進⾏行推測，第三年則使⽤用眼動儀，以視覺字典範(Visual Word Paradigm) 探 測四個聲調辨別的速度，以及辨別過程中與其他聲調競爭者間的競存現象。 本研究的研究 重點有四: 1)研究華語四個聲調的處理之時間差異 2) 聲調和⾳音段訊息在訊號處理的過程中被 激化的時序，3) 超⾳音段訊息之處理是否與⾳音段訊息存在交互作⽤用，以及4)⽐比較三種實驗典範 之結果差異並討論其優劣。本研究主要⺫⽬目的為提出在⽐比較聲調與⾳音段的處理時序時，必須考慮四個聲調之間內在的處理時差，以及⾳音段和⾳音節結構對聲調處理可能產⽣生的影響。

Word recognition in tone languages has long been a fascinating subject in the field of linguistics, yet the temporal unfolding of tonal information relative to segmental information is still unclear and the results from different experimental settings yield conflicting results. We propose a three-year project utilizing the same sets of stimuli in three types of tasks to examine the time course of Mandarin tone recognition. Three experiments will be conducted: 1) Gating paradigm- an quasi-offline task 2) Vowel and Tone monitoring paradigm- offline task and 3) Visual word paradigm - online task. Specific issues the present study aims to investigate are 1) the intrinsic temporal divergence in Mandarin tone processing, 2) how tonal and segmental information are accessed when processing spoken words in Mandarin 3) whether there are interactions between tonal and segmental informations during the processing and 4) how the experimental settings may affect the results in tone processing. We hope to demonstrate that both intrinsic acoustic features of tones as well as the interaction between segmental and tonal information need to be taken into consideration when constructing models for spoken word recognition in Mandarin.