請用此 Handle URI 來引用此文件:
http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/74816
標題: | 修正性回饋在非母語語音對立下之感知成效:以臺灣閩南語清濁音為例 Effects of Corrective Feedback on Perception of Non-native Contrasts: A Training Study of Voicing Contrast in Taiwan Southern Min |
作者: | Yin-Ching Chang 張殷綮 |
指導教授: | 邱振豪(Chen-Hao Chiu) |
關鍵字: | 修正性回饋,語言感知,語音感知訓練,臺灣閩南語,清濁音, corrective feedback,speech perception,high-variability perception training,voicing,Taiwan Southern Min, |
出版年 : | 2021 |
學位: | 碩士 |
摘要: | 以往有關修正性回饋(corrective feedback) 的文獻探討多著重於發音上的修正性回饋(pronunciation-focused corrective feedback),並證實其在矯正發音上帶來一定的功效。相對而言,著重於語音感知層面的修正性回饋(perception-focused corrective feedback) 較不受重視,尤其是對可能會造成不同認知處理過程作為學習策略,因而再細分為不同類別的語音感知修正性回饋,相關的文獻探討可謂寥寥無幾、屈指可數。語音感知上的修正性回饋宣稱能重塑、確認並強化學習者對目標語言中逐漸發展的語音心理表徵,從而使學習者能正確分辨出語音間的細微差別。本文旨在探討不同型態的語音感知修正性回饋是如何影響臺灣華語母語使用者判別臺灣閩南語的清濁音。在訓練階段,受試者會隨機分配至不同組別,接收三種不同形式的聽覺回饋與視覺刺激,分別為回答錯誤的警示加上目標音節、非目標音節,抑或是目標音節與非目標音節的組合(組合式),來當作受試者判別臺灣閩南語清濁音的基準。此系列實驗共包括兩個子實驗,第一個實驗主要討論不同形式的聽覺回饋與視覺刺激是否會強化學習者的學習;第二個實驗將探討由不同發音人錄製而成的聽覺刺激材料是否會造成不同的語音感知訓練成果。研究結果證實了語音感知修正性回饋的成效。而三種不同型態的聽覺回饋中,尤以組合式的聽覺回饋,透過提供音素最小辨義詞對(phonemic minimal pair) 的方式,帶來最佳的訓練功效。然而,將訓練而來的語音辨識能力歸納至其他尚未訓練過的音節卻十分有限。研究成果亦顯示受試者可能會利用個人的語音特徵,來辨別臺語的清濁音,從而影響不同型態的語音感知修正性回饋所引起的表現。 Corrective feedback (CF) is largely concerned with pronunciation-focused feedback and has proven its positive effects on modifying articulation. In comparison, perception-focused CF is rather understudied, especially for different types of CF which may trigger different kinds of cognitive processing as learning strategies. Perception-focused CF is alleged to entrench learners’ mental representation of the target phonemes. The present study constitutes two experiments to investigate to what extent Taiwan Mandarin speakers can correctly identify voicing contrasts in Taiwan Southern Min using different types of perception-focused CF. Auditory and visual feedback is provided during the training session. For incorrectly identified responses, one of the three auditory CFs (a rejection followed by a target form, a rejection followed by a nontarget form, and a rejection followed by target and nontarget forms) was provided to direct their attention to related cues. While Experiment I examined whether different CF types would induce differential effects in acquiring a non-native contrast, Experiment II furthered the issue of high-variability perception training, in which stimuli were produced by multiple speakers. The results showed positive effects of all three types of auditory CF while combinatory CF which provided information with a minimal pair (a rejection followed by target and nontarget forms) was the most effective, in line with previous findings. However, generalizations based on corrective feedback (i.e., transfer from trained to untrained stimuli) were only limited. The comparison between the two experiments also suggest that participants may rely on speaker-specific characteristics to identify voicing differences. |
URI: | http://tdr.lib.ntu.edu.tw/jspui/handle/123456789/74816 |
DOI: | 10.6342/NTU202100314 |
全文授權: | 有償授權 |
顯示於系所單位: | 語言學研究所 |
文件中的檔案:
檔案 | 大小 | 格式 | |
---|---|---|---|
U0001-0102202110343000.pdf 目前未授權公開取用 | 4.25 MB | Adobe PDF |
系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。