CAREER: Information Misperceptions in the Internet Era

学者们观察到，网络新闻和政治言论有可能促使人们相信虚假或误导性的事实说法，他们经常将此归因于互联网的独特特征，比如没有看门人，可以自由地筛选令人不快的证据，以及保护新闻消费者不受令人不快的事实影响的个性化系统。然而，这些机制在很大程度上是投机性的，与现有数据不太相符。这个项目朝着对互联网时代政治误解的理论基础和经验检验的理解迈进。假设既关注与使用在线新闻媒体相关的广泛影响，也关注这些影响发生的机制。具体地说，一系列媒体效应假说解决了(1)媒体曝光决策中的党派偏见的流行和后果，(2)将社交媒体用作政治新闻来源，以及(3)自动(通常是看不见的)个性化技术塑造在线新闻消费者看到的内容。第二种更微妙的预测是关于这些技术如何使个人更有可能接受真实的、不准确的政治信息。这些假设将所讨论的技术的属性与关于加工流利性、影响和有偏见的同化的既定理论工作联系起来。为了验证假设，这项研究将对美国有代表性的样本进行多波调查，并进行一系列对照实验，以评估理论上提出的具体机制。第一年将进行三波调查，第五年将进行两波调查，与美国总统选举相对应。衡量受访者在前几波中对各种在线政治新闻来源和服务的使用情况，以及评估受访者在后几波中存储的政治知识和误解，将为使用这些互联网技术的后果提供明确的证据。第一个调查还将包括一项嵌入式实验，以测试加工流利性在现场的影响。在其间的几年中，将进行一系列相互关联的实验。第二年将重点测试元认知经验，如处理流利性，对接受虚假信息和事实更正的影响。在第三年，重点将放在情感上，测试对政治主张的情感反应对个人评估的影响。第四年的实验将集中在有偏见的同化，因为它告知可信度的影响，以及机构信任和个人判断对参与者信念的相对重要性。这些数据集将与学生和其他学者共享，进一步增强研究努力的影响。一门新的本科课程，旨在帮助学生成为更知情的政治信息消费者，将基于这项研究的理论和实证工作。通过公共宣传，向记者和广大公众以及科学场所的出版物传播有关政治误解的来源和补救的重要经验教训，将提高人们对评估政治信息的有效战略的认识。

Scholars have observed that online news and political talk have the potential to promote belief in false or misleading factual claims, frequently attributing this to distinctive characteristics of the Internet such as the absence of gatekeepers, the freedom to screen out disagreeable evidence, and personalization systems that shield news consumers from uncomfortable truths. However, these mechanisms are largely speculative, and do not align well with existing data. This project moves toward a theoretically grounded and empirically tested understanding of political misperceptions in the Internet era. Hypotheses focus both on broad effects associated with use of online news media, and on the mechanisms through which these effects occur. Specifically, a series of media-effects hypotheses address the prevalence and consequences (1) of partisan bias in media exposure decisions, (2) of the use of social media as a source of political news, and (3) of automated (and often invisible) personalization technologies that shape what online news consumers see. A second more nuanced set of predictions concerns how these technologies could lead individuals to be more likely to accept as true inaccurate political information. These hypotheses link the attributes of the technologies in question with established theoretical work on processing fluency, affect, and biased assimilation. To test the hypotheses, the study will pair multi-wave surveys of representative samples of Americans with a series of controlled experiments designed to evaluate the specific mechanisms theorized. A three-wave survey will be conducted in year one, and a two-wave survey in year five, corresponding with U.S. Presidential elections. Measuring respondents' use of various online political news sources and services in early waves, and assessing respondents' store of political knowledge and misperceptions in later waves will provide clear evidence concerning the consequences of using these Internet technologies. The first survey will also include an embedded experiment in order to test the influence of processing fluency in the field. A series of interrelated experiments will be conducted in the intervening years. Year two will focus on testing the influence of metacognitive experiences, such as processing fluency, on the acceptance of false information and factual corrections. In year three the emphasis will be on affect, testing the influence of emotional responses to political claims on individuals' assessments. Year four experiments will focus on biased assimilation as it informs credibility effects, and on the relative importance of institutional trust and individual judgment on participants' beliefs. The datasets will be shared with students and other scholars, further enhancing the impact of the research endeavor. A new undergraduate course, intended to help students become more informed consumers of political information, will be based on the theoretical and empirical work coming out of this research. Public outreach, communicating key lessons concerning the sources and remedies of political misperceptions to journalists and the broader public, as well as publications in scientific venues, will raise awareness of effective strategies for assessing political information.