Synthesis produced by the Fondation Descartes of the following research paper:
Lyons, B. A., Montgomery, J. M., Guess, A. M., Nyhan, B., & Reifler, J. (2021). Overconfidence in news judgments is associated with false news susceptibility. Proceedings of the National Academy of Sciences, 118(23).
The Dunning-Kruger effect is a cognitive bias that describes a tendency to overestimate one’s knowledge of a given subject. It is possible that this phenomenon is playing a central role in the diffusion of fake news online. Indeed, overconfidence in one’s capacities for discernment could lead to a decrease in overall vigilance. Overlooking our difficulties in discerning fake news means that we have no incentive to increase our vigilance online.
This study by Benjamin Lyons and his colleagues supports this theory. On average, in this study, 70% of participants surveyed overestimated their ability to distinguish between reliable information and fake news. To demonstrate this, the authors asked a large panel representative of the U.S. population to determine the degree of accuracy of a series of Facebook posts, half of which were fake news. Then, participants were asked to estimate their own performance compared to other participants. The authors were thus able to determine that about 70% of the participants wrongly believed that they were better than others at assessing the accuracy of Facebook content. Participants, on average, overestimated their actual test score by 20%. Similarly, the more they overestimated their capacities for discernment, the less participants were able to distinguish between true and false information. This result confirms the existence of a Dunning-Kruger effect in the detection of false information online.
The authors then recorded participants’ Internet activity to determine whether overestimating their ability to assess the accuracy of information was associated with visiting websites known to publish fake news. The results support this hypothesis: overconfidence in one’s capacities for discernment is indeed associated with a greater likelihood of visiting websites that spread fake news. Individuals who overestimate their capacities for discernment the most are 11% more likely to have visited an unreliable website than people who overestimate it the least. Lastly, this overestimation is associated with a greater willingness to spread the fake news presented in the study.
This study by Benjamin Lyons and his colleagues is the first to establish a statistical relationship between the Dunning-Kruger effect and the spread of fake news. Now, it is necessary to better characterize the causal relationship between these two phenomena.