Why don't bystanders help? Because of ambiguity?

Two experiments investigated the effects of ambiguity of an emergency situation on helping behavior. In Experiment I, 70 male undergraduates waiting either alone, with a stranger, or with a friend heard a maintenance man fall and cry out in agony. Half of the two-person groups were naive; the other half included a confederate who was instructed to react as passively as possible. In contrast to the finding of Latane and Rodin's lady in distress study, all subjects intervened. In Experiment II, subjects either alone, with one other, or with four others were exposed to either a nonambiguous or an ambiguous emergency situation. At least one subject in all of the nonambiguous conditions responded to the needs of the victim, whereas approximately 30% of the subjects in the ambiguous condition included a helper. Subjects in the latter twoand five-person groups were less likely to help and intervened slower than was expected on the basis of the alone subjects' performances. These results were attributed to the degree of ambiguity and seriousness of consequences employed in the emergency situation,