People often make inferences with incomplete information. Previous research has led to a mixed picture of how people treat missing information. To explain these results, the authors follow the Brunswikian perspective on human inference and hypothesize that the mechanism's accuracy for treating missing information depends on how it is distributed in a certain environment. The hypothesis is supported by the results of a simulation study, which also shows that the mechanism for treating missing information has a much stronger impact on overall accuracy than the most accurate inference strategies considered. The conclusion is that how people react to missing information could be an adaptive response to specific environments. (Contains 2 tables, 3 figures and 1 footnote.)