Doctoral Dissertation Research: The perception and production of prominence in Spanish by heritage speakers and L2 learners

西班牙语是美国第二大通用语言，仅次于英语，17%的美国人自认为是西语裔或拉丁裔。有相当大比例的美裔拉美裔人会不同程度地使用双语。大多数在美国出生的拉美裔人以西班牙语为传统语言，但英语占主导地位。鉴于这一群体在人口统计学上的重要性，以及研究界对双语益处的日益认识，确定这些讲西班牙语的遗产使用者用他们的两种语言的语言能力是一个与社会相关的研究课题。尽管遗产语言习得最近得到了更多的关注，但这一领域的研究仍然非常稀少。本论文将通过考察美国传统西班牙语使用者的语音，更具体地说，在词汇和句子层面的重音线索的使用，来更好地理解传统使用者的语言系统。尽管单词重音在西班牙语和英语中都具有对比性，但在这两种语言中它的认识是不同的。在英语中，音段提示(即辅音和元音等个别发音)在区分重读音节和非重读音节方面起着关键作用，而在西班牙语中，这种区别更多地出现在超音段水平上(即音高、幅度和时长)。因此，研究表明，英语听众在感知西班牙语重音方面遇到了很大的困难。通过一系列心理语言学实验，本文考察了西班牙传统语言使用者在感知西班牙语重音时是否成功地注意到了超音段线索。此外，将对讲西班牙语的人的发音进行声学分析，以调查他们在产生西班牙语重音时是否有效地使用了超音段线索。他们的知觉和生产之间的结果将进行比较，看看这两个领域是否存在不同的模式。由于有报道称，说遗产的人在成长过程中听到遗产语言的频率比说遗产语言的频率更高，因此很可能会发生感知和产出的不对称习得。西班牙语传统使用者的表现将与西班牙语单一语言者和西班牙语英语第二语言学习者的表现进行比较，以了解西班牙传统使用者是否与这两个群体不同，如果不同，在哪些方面和程度上不同。分析传统语言使用者与单语使用者和第二语言学习者的不同之处，将有助于确定实现两种语言能力平衡的最佳方法。

Spanish is the second most-spoken language in the U.S., after English, and 17% of the country's population self-identify as Hispanic or Latino/a. A significant percentage of U.S.-Latinos are bilingual to different degrees. Most U.S.-born Latinos have Spanish as their heritage language, but are dominant in English. Given the demographic importance of this population and an increased awareness in the research community of the benefits of bilingualism, determining the linguistic competence of these heritage speakers of Spanish in their two languages is a socially-relevant topic of research.Although heritage language acquisition is recently gaining more attention, research in this area is still very scarce. This dissertation will contribute to a better understanding of heritage speakers' linguistic system by examining their phonology, more specifically, the use of word- and sentence-level stress cues in the perception and production of Spanish by US Spanish heritage speakers. Although word stress is contrastive in both Spanish and English, it is realized differently in the two languages. Unlike English in which segmental cues (i.e., individual sounds like consonants and vowels) play a critical role in the distinction between stressed and unstressed syllables, in Spanish the distinction occurs more on the suprasegmental level (i.e., pitch, amplitude, and duration). Therefore, studies have shown that English listeners experience great difficulty in perceiving stress in Spanish. Using a set of psycholinguistic experiments, this dissertation examines whether Spanish heritage speakers successfully attend to suprasegmental cues when they perceive Spanish stress. Moreover, Spanish heritage speakers' production will be acoustically analyzed to investigate whether they effectively use suprasegmental cues when they produce Spanish stress. Results between their perception and production will be compared to see whether different patterns are found in these two domains. Since it has been reported that heritage speakers grow up hearing the heritage language more frequently than speaking it, asymmetrical acquisition of perception and production is likely to occur. Spanish heritage speakers' performance will be compared to that of Spanish monolinguals and English second language learners of Spanish, in order to see whether Spanish heritage speakers differ from these two groups, and, if so, in which aspects and to what extent. Analyzing the ways in which heritage speakers may differ from monolingual speakers and second language learners will help determine the best methodologies to achieve balanced competence in two languages.