The aim of this experiment is to examine whether communicating the results of social psychological research improves out-group stereotypes and diminishes in-group bias. The experimental material consisted of 2 communications: one described Hamilton and Gifford's (1976) experiment on illusory correlation (Experiment 1); the other described Sherif's (1966) studies on summer camps. The results of the present experiment show that knowledge of Sherif's findings had no effect on evaluations, whereas an awareness of the experiment on illusory correlation produced a boomerang effect, accentuating, rather than diminishing, in-group bias. A second experiment revealed that the persuasive power of a scientific message on stereotypes depends on whether in-groups and out-groups are cogni-tively present in the message acquisition phase.