Integrative mechanisms in real-time spoken language processing

口语是以惊人的速度产生和理解的。例如，流利的语音中的单词通常以每秒2到3个的速度产生，而许多核心语言差异(如区分不同语音的物理特性)取决于仅涉及数十毫秒的计时差异。人类的大脑如何应对语言的动态变化呢？例如，在听的过程中，我们的心理机器是如何跟上信息在发声时到达的速度的？ 我的研究项目正在进行的目标是了解使语言处理快速、渐进、高效和准确地发生的认知机制。到目前为止，这项工作强调了稳健的解释在一定程度上取决于听者对接下来要说的内容的形式和内容的隐含期望。此外，这项工作还为听话者使用发生口语交际的物理语境中的信息线索提供了确凿的证据。 展望未来，一个重要的目标是确保对语言处理的实证研究反映自然语言的许多复杂和错综复杂的特征，以及不同类型的情景环境，这些情景可能会使听者或多或少地注意言语，或者使听力或多或少变得困难，例如环境噪音或听者作为母语或非母语使用者的地位。 这项拟议的研究旨在使用眼球跟踪技术来探索在这些类型的情景中的语言理解，这些技术已经在我的实验室成功地用于研究语言过程的毫秒级的时刻性质。这项研究的结果有望为支持核心语言能力的认知机制提供重要的见解。这项研究还将有助于更深入地理解交际障碍的性质，以及在涉及人机交互的情况下应强调的自然交际行为的特征。

Spoken language is produced and understood at a striking pace. For example, words in fluent speech are often produced at a rate of 2 or 3 per second, and many core linguistic distinctions (such as the physical properties that distinguish different speech sounds) hinge on timing differences involving just tens of milliseconds. How does the human mind cope with the dynamics of language? For example, during listening, how does our mental machinery manage to keep up with the pace at which information arrives in an unfolding utterance? The ongoing goals of my research program are to understand the cognitive mechanisms that allow language processing to occur rapidly, incrementally, and with a high degree of efficiency and accuracy. To date this work has highlighted the way in which robust interpretation depends in part on a listener's implicit expectations about the form and content of what will be said next. In addition, this work has provided solid evidence for listeners' use of informational cues from the physical context in which spoken communication occurs. Going forward, one critical objective is to ensure that empirical investigations of language processing reflect the many complex and intricate features of natural language as well as the different kinds of situational circumstances that can make listeners more or less attentive to speech, or that make listening more or less difficult such as environmental noise or a listener's status as a native or non-native user of the language. The proposed research aims to explore language comprehension in these types of scenarios using eye-tracking techniques that have been successfully employed in my laboratory to study the moment-by-moment nature of language processes at the millisecond level. The outcomes of this research are expected to provide important insights into the cognitive mechanisms that support core language abilities. This research will also contribute to a deeper understanding of the nature of communicative impairments as well as the features of natural communicative behaviour that should be emphasized in situations involving human-computer interaction.