Investigating the Generation and Spread of Numerical Misinformation: A Combined Eye Movement Monitoring and Social Transmission Approach

Abstract Numerical facts play a prominent role in public discourse, but individuals often provide incorrect estimates of policy-relevant numerical quantities (e.g., the number of immigrants in the country). Across two studies, we examined the role of schemas in the creation of numerical misinformation, and how misinformation can spread via person-to-person communication. In our first study, we combined eye movement monitoring and behavioral methods to examine how schemas distorted what people remembered about policy-relevant numerical information. Then, in a second study, we examined the consequences of these memory distortions via the social transmission of numerical information, using the serial reproduction paradigm. We found that individuals misremembered numerical information in a manner consistent with their schemas, and that person-to-person transmission can exacerbate these memory errors. Our studies highlight the mechanisms supporting the generation and spread of numerical misinformation and demonstrate the utility of a multi-method approach in the study of misinformation.