The impact of cognitive biases on the believability of fake news
通过调查社交媒体用户，识别出五种最影响假新闻可信度的认知偏见（从众、框架、过度自信、确认和锚定），并提出减少这些偏见的缓解方法。
Modern technologies, especially social networks, contribute to the rapid evolution and spread of fake news. Although the creation of fake news is a serious issue, it is the believability of fake news and subsequent actions that produce negative outcomes that can be harmful to individuals and society. Prior research has focused primarily on the role of confirmation bias in explaining the believability of fake news, but other biases are likely. In this research, we use theories of truth and a taxonomy of 10 cognitive biases to conduct an exploratory, qualitative survey of social media users. Five cognitive biases (herd, framing, overconfidence, confirmation, and anchoring) emerge as the most influential. We then propose a Cognitive Bias Mitigation Model of methods that could reduce the believability of fake news. The mitigation methods are grouped according to three themes as they relate to the five biases.