Which thoughts can kill a boxer? naive theories about cognitive and emotional antecedents of suicide.

摘要

OBJECTIVES: We investigated naive theories regarding the association among beliefs, emotions and behaviours to test Rational Emotive Behaviour Therapy's (REBT) assumption that rational cognitions and adaptive emotions lead to functional behaviours whereas irrational cognitions and maladaptive emotions trigger dysfunctional reactions. Design: We applied an experimental between-subjects design. METHODS: Participants read newspaper articles about the defeat of a boxer. In one condition, the authentic article informed participants that he committed suicide and in the other, a fictitious article about the same defeat described the athlete as successfully continuing his career. Different question formats were employed to assess participants' assumptions about the stimulus person's defeat-related cognitions and emotions: rating scales, sentence completion and free responses. RESULTS: Participants assumed significantly more irrational beliefs (e.g. I absolutely have to win) on the side of the boxer in the suicide scenario than in the non-suicide version. This finding was obtained by directive and non-directive assessment methods. Additionally, participants expected the suicidal stimulus person to be experiencing maladaptive emotions (e.g. depression, guilt) whereas a successful resumption of his career lead to expectations of adaptive affects (e.g. sadness, concern). Ratings of the functionality revealed that sadness, fear, annoyance and concern were expected to be more functional than depression, anxiety, rage and guilt. CONCLUSIONS: The results show that naive psychological theories about the antecedents of dysfunctional behaviour are in accordance with theoretical assumptions of REBT: Irrational beliefs are viewed to be connected with maladaptive emotions and to result in dysfunctional behaviour, and adaptive emotions are thought to be of higher functional value than their maladaptive counterparts. The use of different question formats and a between-subject design excluded that results are due to methodological artifacts or contrast effects.