Impure or Just Weird? Scenario Sampling Bias Raises Questions About the Foundation of Morality

Moral psychologists have used scenarios of abuse and murder to operationalize harm and chicken-masturbation and dog-eating to operationalize impurity. These scenarios reveal different patterns of moral judgment across harm and purity, ostensibly supporting distinct moral mechanisms, modules, or “foundations.” However, these different patterns may stem not from differences in moral content per se but instead from biased sampling that confounds content with weirdness and severity. Supporting this hypothesis, frequently used impurity scenarios are weirder and less severe than both harm scenarios (Study 1) and participant-generated impurity scenarios (Study 2). Weirdness and severity—not content—also appear to drive differences between act and character evaluations (Study 3). Also problematic for modular accounts are extremely high correlations between harm and impurity ( rs > .86), and findings that harm scenarios assess impurity better than researcher-devised impurity scenarios. Overall, patterns of moral judgment previously ascribed to distinct moral mechanisms may reflect domain-general moral cognition.