RESUMO
Previous research indicates that rejection by a group causes aggressive responses. However, in these previous studies, rejected participants were led to believe that they were liked and accepted before the rejection; likely, this rejection was highly unanticipated. Sociometer theory (Leary et al., 1995) proposes the existence of a psychological mechanism (a "sociometer") that enables individuals to detect potential rejection via others' reactions; a properly working sociometer affords a person predictive control over an interaction. We hypothesized the lack of predictive control inherent in previous rejection studies was a critical contributor to participants' aggressive responses; predictive control should lead to decreased aggression. To test this, we manipulated predictive control by varying confederate behavior toward participants before a rejection manipulation. Results indicate that unpredictable rejection undermined participants' belief that they could predict other's behavior (i.e., led to the perception of a broken sociometer) and led to higher levels of aggression.