Date created: | Last Updated:

Category: Project

Description: Visualization supports exploratory data analysis, but charts can also mislead people into drawing unwarranted conclusions from spurious patterns. We evaluate interventions to prevent false discovery from visualized data. We investigate whether operationalizing analyst beliefs as part of the visual analytic process can improve inference quality. In two experiments, we exposed participants to both noisy and 'true' scatterplots, and assessed their ability to infer data-generating models that underlie those samples. Participants who underwent prior belief elicitation made 21% more correct inferences along with 12% fewer false discoveries. This benefit was observed across a variety of sample characteristics, suggesting broad utility to the intervention. However, additional interventions to highlight counterevidence and sample uncertainty did not provide a significant advantage. These findings suggest that lightweight, belief-driven interactions can yield a reliable, if moderate, reduction in false discovery. Our work also suggests additional research avenues to further improve visual inference and reduce confirmation bias.