Doctoral Dissertation Research: Why adapt? Phonotactic learning as non-native language adaptation

博士论文研究：为什么要适应？

• 批准号: 1728173
• 负责人: Matthew Goldrick
• 金额: $ 1.01万
• 依托单位: Northwestern University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-08-01 至 2019-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1728173&HistoricalAwards=false
• 关键词: Doctoral; Dissertation; Research; Why; adapt

In order to successfully communicate, listeners must continuously adapt to systematic differences between speakers (e.g., changing expectations about what you will hear based on a speaker's dialect). At the same time, listeners need to ignore incidental variation (e.g., if a speaker happens to be talking with their mouth full, you should not change how you expect they will speak after they have finished eating; Kraljic, Brennan, & Samuel, 2008). In this study, a series of experiments examine the limits of adaptation to novel phonotactics, or constraints on syllable structure (e.g. in English, but not Spanish, the sequence /sk/ can begin syllables, such as in the word school). This project tests the hypothesis that speakers use their past experience of phonotactic variation to distinguish systematic from incidental variation. Speakers naturally encounter variation in phonotactics between talkers of different languages, but do not encounter such variation between talkers of the same language; this predicts that speakers will only adapt to talker-specific syllable structures when exposed to talkers who differ in their language backgrounds (e.g., a native French talker and a native English talker). This hypothesis is explored in three experiments. In study 1, listeners are exposed to two talkers, each with a different phonotactic constraint, in a perception experiment. Adaptation to talker-specific constraints is predicted only when talkers differ in language background. Study 2 explores the role of talker language background on adaptation to phonotactics in speech production, using a tongue twister paradigm. Study 3 investigates the structure and granularity of listeners' prior knowledge. Can listeners hold models of only a native vs. non-native language, or can they maintain models for two separate non-native languages?

为了成功地沟通，听众必须不断适应说话者之间的系统差异（例如，根据说话者的方言改变对你将听到的内容的期望）。同时，听众需要忽略偶然的变化（例如，如果说话者碰巧在嘴里塞满东西的情况下说话，你不应该改变你期望他们吃完饭后说话的方式；Kraljic，Brennan，＆Samuel，2008）。在这项研究中，一系列实验检验了适应新语音学的局限性，或对音节结构的限制（例如，在英语中，但不是西班牙语，序列 /sk/ 可以开始音节，例如在单词 school 中）。该项目测试了这样的假设：说话者利用他们过去的音位变异经验来区分系统变异和偶然变异。说话者自然会遇到不同语言的说话者之间语音策略的差异，但不会遇到相同语言的说话者之间的这种差异；这表明，当说话者接触到不同语言背景的说话者（例如，以法语为母语的人和以英语为母语的人）时，说话者只会适应特定于说话者的音节结构。这个假设通过三个实验进行了探讨。在研究 1 中，在感知实验中，听众接触到两个说话者，每个人都有不同的语音约束。仅当说话者的语言背景不同时，才能预测对特定于说话者的约束的适应。研究 2 使用绕口令范式探讨了说话者语言背景对语音生成中语音策略适应的作用。研究 3 研究了听众先验知识的结构和粒度。听众是否可以仅持有母语与非母语的模型，或者他们可以维护两种单独的非母语的模型吗？