Rainfall data usually comprise large sets of complex information, with data points spread across both time and space. To be understood, such environmentally significant data must be integrated—but distribution shapes can bias this integration process. Distribution biases can lead to differing levels of concern with rainfall, and in turn, alter the perceived importance of climate adaptation measures. In three experiments, participants reported greater concern about rainfall and reported stronger climate adaptation intentions when rainfall distributions were negatively skewed compared to positively skewed, even though both distributions contained identical total rainfall. This cognitive bias from skewed distributions persisted in both retrospective judgments of entire stimulus sets, and in evaluations of additional instances of rainfall, and it depended on foreknowledge of stimulus context (i.e., the shape and endpoints of the distribution). The results demonstrate how cognitive bias can influence the formation of environmental attitudes.