CAREER: Large-Scale Examination of Problematic Online Behaviors and Their Regulators

该项目将通过确定在不同社交媒体平台上规范跨党派敌意和虚假信息的制约因素，帮助提高在线对话和信息的质量。这项研究将政治传播和社会经济理论的优势与计算和实验方法的严谨方法结合起来，将确定：(1)哪些监管机构或策略适合或最有效地打击网上虚假信息和跨党派敌意；(2)它们的优势在不同行为和社交媒体平台上有何不同；以及(3)这些监管机构如何互动，有时会相互破坏或相互支持。学者们曾认为，在线社交媒体平台将带来民主讨论和辩论的新时代。然而，学者和用户现在最关心的是这些平台的阴暗面--无礼、跨党派敌意和虚假信息等问题在网上都很常见。虽然已经努力解决这些问题--例如使用道德劝说来遏制不文明行为和媒体素养来遏制错误信息--但到目前为止，这些办法缺乏一个统一的理论框架，无法系统地探索解决办法的空间。这项研究将开发一个框架，将线上和线下规范行为的三种模式联系起来：(1)规范通过社区的制裁或规则进行约束。(2)市场通过价格进行约束。(3)架构-在线空间中构建的环境或代码-通过其施加的结构负担进行约束。这些模式对虚假信息和跨党派敌意的影响将通过开发广泛的方法方法来审查，这些方法横跨机器学习、网络科学和因果推理等领域。首先，该项目将贡献丰富的数据集和可扩展的机器学习和网络科学方法，用于在线识别跨党派的敌意和虚假信息。其次，该项目将把法律和政治沟通学术的理论优势与计算机和信息科学的计算优势结合起来，以打击在线问题行为。它将通过自然和随机实验来调查不同监管模式的有效性，并使用结构方程建模来确定它们之间的相互依赖关系。第三，它将通过检查不同平台上监管机构的行为和效率来确定策略的普适性。最后，大多数在线解决问题行为的方法都将个人作为分析单位。然而，结构监管者对社区采取行动。该项目将通过对个人和社区进行比较分析，克服通常破坏个人层面研究的问题。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 project will help improve the quality of conversations and information online by identifying the constraints that regulate cross-partisan animosity and disinformation across different social media platforms. Combining the strengths of political communication and socioeconomic theories with the methodological rigor of computational and experimental approaches, this research will identify: (1) which regulators, or strategies, are suitable or most effective for combating disinformation and cross-partisan animosity online; (2) how their strengths vary across behaviors and social media platforms; and (3) how these regulators interact, at times undermining or supporting each other. Scholars once thought that online social media platforms would bring in a new era of democratic discussion and debate. However, scholars and users alike are now mostly concerned about the dark side of these platforms - problems such as incivility, cross-partisan animosity, and disinformation are all commonplace online. While there have been efforts to combat these problems - such as the use of moral suasion to curb incivility and media literacy to curb misinformation - the approaches thus far lack a unified theoretical framework that allows for a systematic exploration of the solution space. This research will develop a framework connecting three of the modalities that regulate behavior online and offline: (1) Norms constrain through the sanctions or rules of a community. (2) Market constrains through price. (3) Architecture - built environment or code in online space - constrains through the structural burdens it imposes. The impact of these modalities on disinformation and cross-partisan animosity will be examined by developing a broad range of methodological approaches, spanning fields such as machine learning, network science, and causal inference. First, the project will contribute rich datasets and scalable machine learning and network science approaches for identifying cross-partisan animosity and disinformation online. Second, this project will bring together the theoretical strengths of legal and political communication scholarship and the computational strengths of computer and information sciences to combat problematic behaviors online. It will investigate the efficacy of different modalities of regulation through natural and randomized experiments and identify their interdependencies using structural equation modeling. Third, it will determine the generalizability of strategies by examining the behavior and efficacy of regulators across different platforms. Finally, most approaches that address problematic behaviors online treat individuals as the unit of analysis. However, structural regulators act upon communities. This project will overcome issues that generally undermine research at the individual level by performing comparative analyses across not just individuals but also communities.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.