Do We Trust Our Gut? The Causal Relationship between Implicit Group Attitudes and Behavior

Considerable evidence suggests that implicit attitudes co-vary with behavior (Greenwald, Poehlman, Uhlmann \u26 Banaji, 2009). Within the domain of stereotyping and prejudice, in particular, implicit group attitudes have been shown to correlate with behavior towards individual group members. Notably, little experimental evidence demonstrates that implicit group attitudes cause behavior towards individual group members. In five experiments, I created (Experiments 1, 3, 4, \u26 5) or manipulated (Experiment 2) implicit attitudes, and measured these attitudes as well as behavior towards individual group members. Although an evaluative conditioning procedure reliably affected implicit attitudes, it did not have any impact on behavior by itself (Experiments 1 \u26 2). The addition of a narrative vignette to the manipulation increased condition differences in implicit attitudes (Experiment 3) and impacted behavior (Experiments 4 \u26 5). However, multiple mediation analysis revealed conflicting evidence regarding the roles of implicit and explicit attitudes in affecting behavior. In Experiment 4, implicit but not explicit attitudes mediated condition difference in behavior; in Experiment 5, explicit but not implicit attitudes mediated condition differences in behavior. This suggests that any causal relationship between implicit group attitudes and individual level behavior may be smaller and more tenuous than previously assumed