Workshop: Exploring the (un)expected--The role of informativity in language production and comprehension (Special Session at the 2015 CUNY Conference on Human Sentence Processing)

研讨会：探索（意外）预期——信息性在语言生成和理解中的作用（2015 年纽约市立大学人类句子处理会议特别会议）

• 批准号: 1423514
• 负责人: Elsi Kaiser
• 金额: $ 4.29万
• 依托单位: University of Southern California
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2014
• 资助国家: 美国
• 起止时间: 2014-09-01 至 2015-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1423514&HistoricalAwards=false
• 关键词: Workshop; Exploring; un; expected; role

Language provides us with a powerful means of sharing information with other people. Within a sentence, the informativity of expressions varies. For example, some parts of the sentence may be relatively uninformative because they are highly predictable based on what has already been said, or because they involve information that the hearer already knows. In contrast, other parts of a sentence are more informative because they provide new information and mention things previously not known by the listener. When we speak, we continually have to make choices (largely unconsciously) about how to order linguistic elements of differing levels of informativity within an utterance and how to realize them - for example, when to say something like "My neighbor Pat did not break the window!" or when to just say "He didn't!." These issues are also relevant when understanding language: When we listen to someone else or when we read text, we are faced with the task of extracting the intended meaning. A growing body of research suggests that informativity plays a key role in language. However, many questions about the role of informativity in language processing remain open, and different research traditions have approached these questions in different ways. Significant advances in addressing these open questions would be facilitated by bringing together researchers from these various traditions and providing a forum for exposing future generations of researchers to a variety of perspectives on the question of informativity in language processing.In this project, the investigators will organize a special conference session on these issues, to be held in March 2015 at the University of Southern California, in conjunction with the 28th CUNY Conference on Human Sentence Processing. CUNY conferences are attended by a rich cross-section of the academic community, from undergraduates through senior scholars. From its inception, this conference series has been committed to attracting graduate students, post-doctoral fellows, and other new researchers, encouraging them to attend and also to participate in the proceedings. Holding the special session in conjunction with CUNY will thus ensure a wide audience. The special session is designed to evaluate how informativity impacts language production and comprehension, in order to contribute to our understanding of how questions of informativity can inform theories of language processing. The investigators have invited six distinguished scholars with special expertise to give spoken presentations at this special session. Presentations from the broader community of researchers working on these issues will also be solicited. This special session aims to draw together diverse strains of research; this can inform the work currently being done within each subfield and can potentially result in new collaborative ideas. Issues related to informativity provide a fertile ground for tools and insights from computer science and computational linguistics, in addition to psychology and theoretical linguistics, thereby opening the door to new cross-fertilization between these fields. The special session will also contribute in important ways to the training of graduate students and post-doctoral fellows. Educating and inspiring young scientists has a powerful impact on the future of scientific research.

语言为我们提供了与他人分享信息的强大手段。在一个句子中，表达的信息量是不同的。例如，句子的某些部分可能相对没有信息，因为它们是基于已经说过的高度可预测的，或者因为它们涉及听者已经知道的信息。相比之下，句子的其他部分信息量更大，因为它们提供了新的信息，并提到了听者以前不知道的事情。当我们说话时，我们不断地（在很大程度上是无意识地）选择如何在话语中排列不同信息量水平的语言元素，以及如何实现它们--例如，什么时候说“我的邻居帕特没有打破窗户！或者什么时候说“他没有！“这些问题在理解语言时也是相关的：当我们听别人说话或阅读文本时，我们面临着提取预期含义的任务。越来越多的研究表明，信息性在语言中起着关键作用。然而，关于信息性在语言处理中的作用的许多问题仍然是开放的，不同的研究传统以不同的方式处理这些问题。通过将这些不同传统的研究人员聚集在一起，并为未来的研究人员提供一个论坛，让他们从各种角度了解语言处理中的信息性问题，将有助于在解决这些开放性问题方面取得重大进展。在本项目中，研究人员将于2015年3月在南加州大学组织一次关于这些问题的特别会议，第28届纽约市立大学人类句子处理会议（28 th CUNY Conference on Human Sentence Processing）纽约市立大学的会议由学术界的丰富的横截面参加，从本科生到高级学者。从一开始，这个系列会议就致力于吸引研究生，博士后研究员和其他新的研究人员，鼓励他们参加并参与会议。因此，与纽约市立大学同时举行特别会议将确保有广泛的听众。特别会议旨在评估信息性如何影响语言产生和理解，以促进我们对信息性问题如何为语言处理理论提供信息的理解。调查人员邀请了六位具有专门知识的杰出学者在这次特别会议上作口头报告。还将请从事这些问题研究的广大研究人员介绍情况。这次特别会议旨在汇集各种研究;这可以为每个子领域目前正在进行的工作提供信息，并可能产生新的合作想法。与信息性相关的问题为计算机科学和计算语言学以及心理学和理论语言学的工具和见解提供了肥沃的土壤，从而为这些领域之间的新交叉施肥打开了大门。特别会议还将对研究生和博士后研究员的培训作出重要贡献。教育和激励年轻科学家对科学研究的未来具有强大的影响。