View article

Many people judge that it is permissible to harm one person in order to save many in some circumstances but not in others: it matters how the harm comes about. Researchers have used trolley problems to investigate this phenomenon, eliciting moral judgments or behavioral predictions about hypothetical scenarios where five people can be saved at the cost of harming one other person. We operationalized trolley problems in the laboratory, with economic incentives and real-life consequences, allowing us to observe not only judgments but actual decisions. We varied whether the five were saved by clicking a switch that diverted the harm to the one or by dragging the one in front of the harm. We found differences in moral judgments between the two tasks, but no differences in behavior. The judgments of actors and observers also differed, with observers judging it more right to act. Our results suggest that the …