Effects of fact-checking warning labels and social endorsement cues on climate change fake news credibility and engagement on social media

Online fake news can have noxious consequences. Social media platforms are experimenting with different interventions to curb fake news’ spread, often employing them simultaneously. However, research investigating the interaction of these interventions is limited. Here, we use the Heuristic-Systematic Model of Information Processing (HSM) as a theoretical framework to jointly test two interventions against fake news that are implemented at scale by social media platforms: (1) adding warning labels from fact checkers to initiate systematic processing and (2) removing social endorsement cues (e.g., engagement counts) to reduce the influence of this heuristic cue. Moreover, we accounted for dispositions previously found to affect a person’s response to fake news through motivated reasoning or cognitive style. An online experiment in Germany (N = 571) confirmed that warning labels reduced the perceived credibility of a fake news post exaggerating the consequences of climate change. Warning labels also lowered the (self-reported) likelihood to amplify fake news. Removing social endorsement cues did not have an effect. In line with research on motivated reasoning, left-leaning individuals perceived the climate fake news to be more credible and reported a higher likelihood to amplify it. Supporting research on cognitive style, participants with lower educational levels and a less analytic thinking style also reported a higher likelihood of amplification. Elaboration likelihood was associated only with age, involvement, and political leaning, but not affected by warning labels. Our findings contribute to the mounting evidence for the effectiveness of warning labels while questioning their relevance for systematic processing.

在线虚假新闻可能产生有害后果。社交媒体平台正在尝试不同的干预措施以遏制虚假新闻的传播，通常这些措施是同时实施的。然而，对这些干预措施之间相互作用的调查研究却相对有限。本研究以信息处理的启发式-系统模型（HSM）为理论框架，联合检验了社交媒体平台大规模实施的两种针对虚假新闻的干预措施：一是添加事实核查者的警告标签以启动系统性处理，二是移除社会认可线索（例如，参与度计数）以减少此启发式线索的影响。此外，我们还考虑了先前发现会影响个人对虚假新闻反应的动机推理或认知风格。德国的一项在线实验（N = 571）证实，警告标签降低了夸大气候变化后果的虚假新闻帖子的感知可信度。警告标签还降低了（自我报告的）放大虚假新闻的可能性。移除社会认可线索没有产生效果。与动机推理的研究一致，左翼倾向的个人认为气候虚假新闻更具可信度，并报告了更高的放大其的可能性。与认知风格的研究相吻合，教育水平较低且分析性思维风格较差的参与者也报告了更高的放大可能性。详尽性只与年龄、参与度和政治倾向相关联，但不受警告标签的影响。我们的发现为警告标签的有效性提供了越来越多的证据，同时也对其在系统性处理中的相关性提出了质疑。