Nonunique invulnerability: Singular versus distributional probabilities and unrealistic optimism in comparative risk judgments

Five studies showed that people assess vulnerability to future controllable negative life events differently depending on whether it concerns: (a) generalized targets (e.g., the average peer), or (b) concrete and familiar targets (self or nonself). In the former case, a distributional framework is applied which focuses on available statistical information such as groups' base-rates. In the latter case, people apply a singular judgmental perspective and focus on available individuating information such as personal dispositions and preventive actions. This perspective would typically lead to an optimistic bias in favor of concrete and familiar rather than generalized targets. Participants in Study 1 evaluated both self's and a spontaneously selected (non-intimate) peer's risk prospects as lower than those of their average peer in controllable but not uncontrollable future events or controllable and uncontrollable everyday negative events. Study 2 replicated the main findings in direct comparative judgments. In Study 3, verbal accounts and a measure of interest in the base-rate supported the singular-distributional analysis. In Studies 4 and 5, an optimistic bias was found in favor of a randomly assigned member of their small group to whom participants were yoked. These findings suggest that in addition to the classic ''illusion of unique invulnerability'' (e.g., Perloff & Fetzer, 1986), whereby participants judge themselves to be less vulnerable than the average group-member, there is an ''illusion of nonunique invulnerability,'' whereby participants judge also their singular group-members to be less vulnerable than the average group-member. Finally we discuss conditions in which a pessimistic rather than the standard optimistic bias might occur in comparative risk comparisons. (C) 1996 Academic Press, Inc.