Linguistic Prominence of Pitch withinthe Native Language Determines Accuracy of Tone Processing
Linguistic Prominence of Pitch withinthe Native Language Determines Accuracy of Tone Processing
复制标题
母语中音高的语言重要性决定了声调处理的准确性
DOI：
--
复制
发表时间：
2014
期刊：
影响因子：
--
通讯作者：
Isabelle Darcy
中科院分区：
文献类型：
--
作者：
Vance Schaefer;Isabelle Darcy
我们可以区分四组不同的语言，它们使用基音(或基频，缩写为f0，即语音高度)来区分单词(从专用到无使用)。我们问，使用音高来区分一个人的第一语言(L1)中的单词，无论是声调、重音还是音调重音，是否有助于感知泰语等未知语言的音调对比。我们使用一种方法来研究这个问题，比较了四种不同的第一语言在音高突出(音高被用来区分第一语言中词汇项的程度，从排他性到无)的使用上的不同类型。这项研究具体考察了说另一种声调语言(普通话)、重音语言(日语)、重音语言(英语)的天真听者(=非泰语学习者)如何感知泰语声调，即使用基音来区分单词的意义(韩语)。此外，泰语被采用是因为它具有三个水平和两个轮廓声调，与普通话等语言相比，可以更好地理解声调高度和方向。普通话只有一个水平声调，但有三个轮廓声调，可以比较特定类型的声调对比的表现。目前的二语音系学模型的理论框架主要集中在非母语的音节感知上，即元音和辅音(感知同化模型：Best，1995；言语学习模型：Fge，1995)。这类研究也适用于对超音段的感知(即声调、音调、重音、语调)，因为它可能遵循已提出的针对音段的相同机制。例如，非母语音节的母语塑造知觉的经典案例是日语母语人士在英语中的/L/~/r/区分。由于/L/和/r/在母语中缺乏对比，使得日本听者在感觉上要么会联想到日语中这两个音，要么会在英语母语水平上区分这两个音(Goto，1971；Miywaki，Strange，Verbrugge，Liberman，Jenkins，&Fujimura，1975；Sheldon&Strange，1982)。然而，我们不能从对节段的感知中自动推断超节段的感知是如何工作的。也就是说，知觉理论在超音段领域的应用并不简单，主要是因为语言音高至少以语调的形式存在于所有的第一语言中，如果不是为了对比单词的话。
We can distinguish four groups of languages that differ in the use they make of pitch (or fundamental frequency, abbreviated as f0, i.e., the height of the voice) to distinguish words (ranging from exclusive use to none). We ask whether using pitch to distinguish words in a person’s first language (L1), either as tone, stress or pitch-accent, would help perceive tonal contrasts in an unknown language such as Thai. We investigate this question using one methodology to compare four different L1s varying typologically in the usage of pitch prominence (defined as the degree to which pitch is used to distinguish lexical items in the L1, ranging from exclusive to none). This study specifically examines how Thai tones, i.e., the use of pitch to distinguish words, are perceived by naive listeners (= non-learners of Thai) who speak another tone language (Mandarin Chinese), a pitch accent language (Japanese), a stress language (English), and a language which does not employ pitch in any way to distinguish the meaning of words (Korean). Additionally, Thai was adopted as it features three level and two contour tones, allowing for a better understanding of tone height and direction as compared to a language like Mandarin Chinese which features only one level tone, but three contour tones, allowing for a comparison of performance on specific types of tonal contrasts. The current theoretical framework of second language (L2) phonology models tends to largely focus on the non-native perception of segmentals, i.e., vowels and consonants (Perceptual Assimilation Model (PAM): Best, 1995; Speech Learning Model (SLM): Flege, 1995). Such research may be applicable as well to the perception of suprasegmentals (i.e., tone, pitch accent, stress, intonation) in that it may follow the same mechanisms that have been put forward for segmentals. For example, the classic case of the L1 shaping perception of non-native segmentals is the /l/~/r/ distinction in English for L1 speakers of Japanese. The lack of a contrast between /l/ and /r/ in their L1 causes Japanese listeners to associate either to the flap in Japanese in perception despite being able to produce the two sounds or to distinguish the two in isolated cases at the level of L1 English speakers (Goto, 1971; Miyawaki, Strange, Verbrugge, Liberman, Jenkins, & Fujimura, 1975; Sheldon & Strange, 1982). Yet, we cannot automatically infer from the perception of segmentals how the perception of suprasegmentals works. That is, the application of perception theory to suprasegmental domains is not straightforward, mostly because linguistic pitch is present in all the L1s in at least the form of intonation, if not to contrast words.
DOI：
--
发表时间：
2006
期刊：
Journal of Physics G
影响因子：
--
作者：
Connie K. So
通讯作者：
Connie K. So