Moderators of automatic gender stereotyping: Motivated expectancies, cognitive constraints, and the environment

Three experiments were conducted with undergraduates at a private university to provide evidence for the automatic operation of gender stereotypes, and to provide an assessment of theoretically specified conditions under which automatic stereotyping may be moderated. An established semantic priming procedure was utilized to reveal a particular form of automatic stereotyping that emerges when situational and cognitive constraints prevent perceivers from readily controlling their response. All three experiments demonstrated automatic gender stereotyping in the form of faster responses to target names when preceded by stereotypic than counter-stereotypic attributes, under conditions conventionally accepted to reveal an automatic process (e.g., 200-350 ms. stimulus onset asynchrony, SOA). More importantly, this form of automatic stereotyping was shown to be moderated by motivated expectancies, cognitive resources, and perceivers' environment. Automatic stereotyping was significantly reduced by subjects motivated to use a counter-stereotype expectancy, compared to baseline conditions and to subjects motivated to carry out the opposite, stereotype expectancy strategy. As predicted, cognitive constraints played a critical role in the success of a motivated strategy. Subjects were able to completely reverse automatic stereotyping under low cognitive constraints (2000 ms. SOA) and they were able to reduce it significantly from baseline under relatively high cognitive constraints (350 ms. SOA). However, subjects ability to counter automatic stereotyping was seriously impaired under more severe cognitive constraints (250 ms. SOA). Finally, automatic stereotyping was reduced by the presence of counter-stereotypic events in subjects' environment. That is, after exposure to counter-stereotypic sentences, significantly less automatic stereotyping was obtained compared to a baseline condition. Importantly, this effect was shown to generalize to new category exemplars and to occur without subjects' intent or awareness that their response had been altered. The results of these experiments support recent proposals that stereotyping has both automatic and controlled components, and that it is contextualized by perceivers' environment. This research program highlights the complexity with which those factors may interact to produce a range of outcomes that may have important implications for the perpetuation versus reduction of prejudice and discrimination.