Speech processing under effortful listening conditions in a first and second language in younger and older bilinguals

尽管人口老龄化和语音理解的重要性，但影响双语者言语知觉的因素还没有得到充分的研究，而且对老年人的第二语言(L2)语音处理几乎一无所知。在第一语言或母语(L1)中，语音理解需要一套复杂的过程，将退化的声学输入映射到存储的语音和词汇信息表示上，并将这些信息与长期知识相结合。当听到不完美的信号时，需要额外的认知资源(言语工作记忆、认知控制、高阶语言信息)来理解言语。*对于一个双语听者来说，这项任务更加困难，因为他们的语音能力较差，词汇和句法知识较低，所以知识并不完善。双语者的二语能力各不相同，其中二语习得年龄和熟练程度等因素起着重要作用。*我的NSERC DG计划将填补文献中的关键空白。我将在大样本中考察年轻和年长的英法双语者对二语句子的理解。我将使用阻碍(语音掩蔽)或促进(语境，视觉提示)听力的条件，我将考虑一整套自上而下认知资源的单独测量。我将测量行为、神经活动的处理阶段，以及倾听努力的评级。*研究主题1将使用事件相关脑电位(ERPs；N400)和由不同语境限制的句子诱发的时频脑电来研究噪声中的言语感知。我将确定听者使用语境线索在二语中整合单词的程度，这如何受到听力条件的影响，这是否会随着年龄的不同而不同，以及模式是否受到认知能力和听力努力的个体差异的影响。研究主题2将考察视听语音线索在多大程度上促进第二语言中的句子处理。我将研究自下而上(视觉/听力敏锐度、视位和音素知觉)和自上而下的因素(语境、工作记忆、处理速度)。研究一将考察双语者在L1和L2的单峰(视觉和听觉)和双峰(AV)条件下感知句子时的这些因素。我将确定双语者在第二语言中从AV言语线索中受益的程度，以及他们在第一语言中利用这些线索的能力。然后，研究2将测试AV语音提示是否有益于二语口头工作记忆。*这项研究将产生一个全面的二语语音理解模型，该模型考虑了复杂的听力条件和个体差异。鉴于影响听力的因素很多(噪音、通讯过程中的语音质量下降、老年性耳聋等)，这项工作将与加拿大人的日常生活有相当大的相关性。

Despite an aging population and the importance of speech understanding, the factors that influence speech perception in bilinguals are understudied and virtually nothing is known about second language (L2) speech processing in older adults. In a first or native language (L1), speech understanding requires a complex set of processes that map degraded acoustic inputs onto stored representations of phonological and lexical information and integrate that information with long-term knowledge. When listening to an imperfect signal, additional cognitive resources (verbal working memory [WM], cognitive control, higher-order language information) are needed to understand speech. ******This task is even harder for a bilingual listener who has imperfect knowledge due to poorer phonological competence and lower vocabulary and syntactic knowledge in L2. Bilinguals vary in L2 competence, with factors such as L2 age of acquisition and proficiency playing important roles. ******My NSERC DG program will close critical gaps in the literature. I will examine L2 sentence understanding in large samples of younger and older English-French bilinguals with a range of L2 competence. I will use conditions that impede (phonological masking) or facilitate (context, visual cues) listening, and I will consider a full set of individual measures of top-down cognitive resources. I will measure behaviour, stages of processing using neural activity, and ratings of listening effort. ******Research Theme 1 will examine speech perception in noise using event-related brain potentials (ERPs; N400) and time-frequency EEG evoked by sentences varying in contextual constraint. I will determine the degree to which listeners use contextual cues to integrate words in L2, how this is influenced by listening condition, whether this differs with age, and whether patterns are moderated by individual differences in cognitive ability and listening effort.******Research Theme 2 will examine the extent to which audiovisual (AV) speech cues facilitate sentence processing in L2. I will examine bottom-up (visual/hearing acuity, viseme and phoneme perception) and top-down factors (context, WM, processing speed). Study 1 will examine these factors as bilinguals perceive sentences in unimodal (visual- and auditory-only) and bimodal (AV) conditions in L1 and L2. I will determine the extent to which bilinguals benefit from AV speech cues in L2, over and above their ability to exploit these cues in L1. Study 2 will then test if AV speech cues benefit L2 verbal WM.******This research will have a transformative impact by producing a comprehensive model of L2 speech understanding that accounts for complex listening conditions and individual differences. Given that many factors influence listening (noise, degraded speech during telecommunication, presbycusis, etc.), this work will have considerable relevance to the lives of everyday Canadians.