Integrating Linguistic and Social Meaning in Imprecision Resolution

每当我们说话时，我们不仅描述事物是怎样的，而且还揭示了我们是谁的信息——我们的出身、身份和个性。该项目旨在探索对说话者身份的推断如何影响我们对他们所说内容的理解，以及我们应该在多大程度上相信他们所说的话。它通过关注人们如何根据说话者的典型特征来解释句子的意思来解决这个问题——例如，他们是否看起来或听起来像一个“冲浪老兄”；一个“书呆子”;“山谷女孩”；法学院的学生；或者其他一些社会公认的刻板印象。调查这个问题不仅会让我们更好地理解我们在日常生活中如何交换信息，而且最重要的是，对句子的解释如何导致人类对某些社会群体的成员给予更多的信任，而不是其他人，反映和加强对某些人的偏见和偏好，而排斥其他人。因此，这项研究将揭示人类交流反映和重塑社会和文化的隐含的、往往是无意识的机制；并揭示这些过程可能对人们生活造成的潜在有害和歧视性后果。本项目采用实验方法，探讨理解者如何整合说话人身份的社会信息来计算句子所传达的信息。它通过（非）精确的视角来研究这个问题，即当说话者用一种偏离字面意思的方式来使用词语时所观察到的现象——比如当他们把312美元的实际价格描述为300美元时。特别令人感兴趣的是数字的解释如何受到特定说话者所体现的社会刻板印象的影响。先前的研究表明，说话人精确地和不精确地使用数字与不同的个性和身份特征有关：例如，一个把价格描述为“312美元”的人往往被认为比一个把价格描述为300美元的人更善于表达、更好学、更迂腐。这个结果激发了一个假设，相反，对话者会推断说话者的社会身份，以决定如何准确地解释他们所说的话：也就是说，一个“书呆子、能言善语”的说话者说出的一句话“价格是300美元”，比一个“冷静、悠闲”的说话者说出的同样一句话更有可能被理解为确切的300美元。我们进一步询问，与“冷静”的说话者相比，“书呆子”说话者对同一个句子的不精确使用是否更容易被相信是正确的。通过将人类交流的社会、认知和描述基础结合起来，本研究在不同的语言研究方法之间建立了新的桥梁，并加深了语言学和心理学的理论和传统的整合。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

Whenever we talk, we not only describe how things are, but also reveal information about who we are — our origin, identity, and personality. The proposed project aims at exploring how reasoning about a speaker's identity affects how we interpret what they are saying, and to what extent we should believe what they say. It pursues this question by focusing on how people interpret the meaning of a sentence depending on stereotypical features of a speaker — for example, on whether they look like, or sound like, a "Surfer Dude"; a "Nerd"; a "Valley Girl"; a Law School student; or some other socially recognizable stereotype. Investigating this issue will lead us to better understand not only how we exchange information in everyday life, but also, and most importantly, how the interpretation of sentences leads humans to ascribe more trust to members of some social groups as opposed to others, reflecting and reinforcing biases and preferences towards some people to the exclusion of others. This research will thus shed light on the implicit, and often unconscious, mechanisms whereby human communication reflects and reshapes society and culture; and to unveil potentially harmful and discriminatory consequences that these processes might have for people's lives. Drawing on experimental methods, this project explores how comprehenders integrate social information about speaker identity to compute the message conveyed by a sentence. It investigates this question through the lens of (im)precision, the phenomenon observed when speakers use words in a way that deviates from their literal meaning — such as when they describe an actual price of $312 as $300. Of particular interest is the question of how the interpretation of numerals is affected by the social stereotype embodied by a particular speaker. Previous work has shown that speakers using numerals precisely vs. imprecisely are associated with different sets of personality and identity traits: for example, someone describing the price as "$312" tends to be perceived as more articulate, studious, and pedantic than someone describing it, more liberally, as $300. This result motivates the hypothesis that, conversely, interlocutors reason about the social identity of a speaker to determine how precisely to interpret what they say: that is, that a sentence such as "The price is $300" uttered by a "nerdy, articulate" speaker is more likely to be interpreted as meaning exactly $300 than the same sentence uttered by a "chill, laid-back" one. We furthermore ask whether imprecise uses of the same sentence are more likely to be believed to be true when said by a "nerdy" speaker as opposed to a "chill" one. By bringing together the social, cognitive and descriptive underpinnings of human communication, this research builds new bridges between different approaches to the study of language, and deepens the integration of theories and traditions across linguistics and psychology.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.