Despite early promise, scholarship has shown little empirical evidence of learning from the news on social media. At the same time, scholars have documented the problem of information ‘snacking’ and information quality on these platforms. These parallel trends in the literature challenge long-held assumptions about the pro-social effects of news consumption and political participation. We argue that reliance on social media for news does not contribute to people’s real level of political knowledge (objective knowledge), but instead only influences people’s impression of being informed (subjective knowledge). Subjective knowledge is just as important for driving political participation, a potentially troubling trend given the nature of news consumption on social media. We test this expectation with panel survey data from the 2018 U.S. midterm elections. Two path model specifications (fixed effects and autoregressive) support our theoretical model. Implications for the study of the ‘dark side’ of social media and democracy are discussed.

中文翻译：

---

重新思考社交媒体上的良性循环假设：主观与客观知识与政治参与

尽管早有希望，但学术研究几乎没有显示出从社交媒体新闻中学习的经验证据。同时，学者们也记录了这些平台上信息“偷吃”和信息质量的问题。文献中的这些平行趋势挑战了长期以来关于新闻消费和政治参与的亲社会影响的假设。我们认为，依赖社交媒体获取新闻并不能提高人们的真实政治知识水平（客观知识），而只会影响人们被告知的印象（主观知识）。主观知识对于推动政治参与同样重要，鉴于社交媒体上新闻消费的性质，这是一个潜在的令人不安的趋势。我们使用 2018 年美国的面板调查数据来检验这一预期 中期选举。两种路径模型规范（固定效应和自回归）支持我们的理论模型。讨论了研究社交媒体和民主的“黑暗面”的意义。