Doctoral dissertation research: The encoding, maintenance and retrieval of complex linguistic representations in working memory

该奖项的全部或部分资金来自《2021年美国救援计划法案》(公法117-2)。人类语言理解依赖于工作记忆(WM)的支持，但众所周知，WM的能力是有限的。为了克服这一限制，理解者形成结构化表示(块)，其中包含不同复杂程度的多个单词。WM编码、维护和检索过程如何支持不同复杂性的结构仍然知之甚少。本论文通过确定在工作流管理中支持复杂表征的成本和收益，研究了语言表征是如何在工作流管理中形成和存储的。虽然之前已经有了关于复杂性的工作，但这个项目旨在通过研究语言复杂性的不同方面来提供一个更细微的视角。通过对工作记忆和语言理解之间相互作用的工作的贡献，该项目还对语言相关疾病，如特定语言损伤和各种形式的失语症，具有潜在的临床意义。该项目调查了表征复杂性和三个工作记忆过程(编码、维持和提取)之间相对未被探索的相互作用。第一组研究调查形成具有更多特征的复杂表征是否可以促进或阻碍三个核心工作记忆过程中的一个或多个。第二组研究进一步考察了易化或干扰效应如何被组块表征中元素之间的相似性所调制。最后，第三组研究考察了层级组块结构的形成如何影响对其个别成分的提取。为了提供WM过程和语言复杂性之间相互作用的多维视角，该项目采用了行为(自定进度阅读)和电生理(EEG)方法。该奖项反映了NSF的法定使命，并通过使用基金会的智力优势和更广泛的影响审查标准进行评估，被认为值得支持。

This award is funded in whole or in part under the American Rescue Plan Act of 2021 (Public Law 117-2).Human language comprehension relies on the support from working memory (WM), but it is known that WM capacity is limited. To combat this limitation, comprehenders form structured representations (chunks) comprising multiple words at various levels of complexity. How the WM encoding, maintenance and retrieval processes support structures of varying complexity remains poorly understood. This dissertation project examines how linguistic representations are formed and stored in WM by identifying the costs and benefits associated with supporting complex representations in WM. While there has been previous work on complexity, this project aims to provide a more nuanced perspective by examining different aspects of linguistic complexity. By contributing to the body of work on the interplay between WM and language comprehension, the project also has potential clinical implications for language-related disorders such as specific language impairment and various forms of aphasia.The project investigates the relatively unexplored interaction between representational complexity and the three working memory processes (encoding, maintenance, and retrieval). The first set of studies investigates whether forming complex representations with more features can facilitate or hinder one or more of the three core WM processes. The second set of studies further examines how the facilitation or interference effects are modulated by the similarity between elements within a chunked representation. Finally, the third set of studies examines how the formation of a hierarchically chunked structure affects the retrieval of its individual components. To provide a multidimensional view of the interaction between WM processes and linguistic complexity, the project employs both behavioral (self-paced reading) and electrophysiological (EEG) methodologies.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.