Let's Gamble: How a Poor Visualization Can Elicit Risky Behavior
This work addresses the problem of suboptimal visualization choices influencing risky decisions, which is relevant for designers and users in data communication, though it is incremental in building on existing risk perception research.
The study investigated how different visualization designs affect risk perception and decision-making in a gambling context, finding that icon arrays promoted economically sound behavior while area-proportioned triangles and circles increased gambling likelihood.
Data visualizations are standard tools for assessing and communicating risks. However, it is not always clear which designs are optimal or how encoding choices might influence risk perception and decision-making. In this paper, we report the findings of a large-scale gambling game that immersed participants in an environment where their actions impacted their bonuses. Participants chose to either enter a lottery or receive guaranteed monetary gains based on five common visualization designs. By measuring risk perception and observing decision-making, we showed that icon arrays tended to elicit economically sound behavior. We also found that people were more likely to gamble when presented area proportioned triangle and circle designs. Using our results, we model risk perception and discuss how our findings can improve visualization selection.