Borkenau Peter, Mauer Nadine, Riemann Rainer, Spinath Frank M, Angleitner Alois
Department of Psychology, Martin-Luther Universitat, Halle, Germany.
J Pers Soc Psychol. 2004 Apr;86(4):599-614. doi: 10.1037/0022-3514.86.4.599.
Self-reports, peer reports, intelligence tests, and ratings of personality and intelligence from 15 videotaped episodes were collected for 600 participants. The average cross-situational consistency of trait impressions across the 15 episodes was .43. Shared stereotypes related to gender and age were mostly accurate and contributed little to agreement among judges. Agreement was limited mainly by nonshared meaning systems and by nonoverlapping information. Personality inferences from thin slices of behavior were significantly associated with reports by knowledgeable informants. This association became stronger when more episodes were included, but gains in prediction were low beyond 6 episodes. Inferences of intelligence from thin slices of behavior strongly predicted intelligence test scores. A particularly strong single predictor was how persons read short sentences.
我们收集了600名参与者的自我报告、同伴报告、智力测试以及来自15个录像片段的人格和智力评分。15个片段中特质印象的平均跨情境一致性为0.43。与性别和年龄相关的共同刻板印象大多是准确的，对评委之间的一致性贡献不大。一致性主要受非共享意义系统和非重叠信息的限制。从行为薄片中得出的人格推断与知识渊博的 informant 的报告显著相关。当纳入更多片段时，这种关联会变得更强，但超过6个片段后预测增益较低。从行为薄片中得出的智力推断强烈预测了智力测试分数。一个特别强的单一预测因素是人们阅读短句的方式。
