The predictive power of exponential numeracy

Abstract

Many fundamental relationships, from the spread of disease to climate change, as well as common scales, such as the Richter, pH, and Decibel scales, are characterized by exponential relationships. Across six pre-registered studies (N = 3522) and three supplemental studies (N = 1079), we introduce and test a measure of exponential numeracy. We demonstrate that individuals with greater exponential numeracy are better able to recognize, understand, and predict exponential trends. In doing so, we make several important contributions. First, we show that exponential numeracy helps people predict consequential phenomena, such as the spread of COVID-19 and sales of a popular product after launch. Second, compared to people low in exponential numeracy, people high in exponential numeracy are less influenced by anchoring when forecasting exponential phenomena. Third, we link exponential numeracy with the ability to understand common data presentations. We demonstrate that different presentations of exponential data predictably influence forecast accuracy, but that exponential numeracy mitigates these effects. Our findings have important implications for communicating exponential trends and advance our understanding of how individuals make decisions involving exponential relationships.