In 5 studies, the authors investigate the impact of self-activation on the occurrence and direction of social comparison effects. They show that self-evaluative comparison effects are more likely to occur when self-related cognitions are made cognitively accessible. Contrast occurs when personal self-construals ("I") are accessible, whereas assimilation occurs when social self-construals ("we") are activated. These effects of self-construal activation are similar to the impact of self-unrelated information processing styles that are often associated with personal and social self-accessibility (i.e., differentiation and integration mind-sets). However, whereas self-construal activation elicits self-serving social comparisons, activation of self-unrelated processing styles results in non-self-serving social comparison effects. Implications of these results for understanding the cognitive processes underlying social comparison effects are discussed.

No abstract

How does power affect behavior? We posit that this depends on the type of power. We distinguish between social power (power over other people) and personal power (freedom from other people) and argue that these two types of power have opposite associations with independence and interdependence. We propose that when the distinction between independence and interdependence is relevant, social power and personal power will have opposite effects; however, they will have parallel effects when the distinction is irrelevant. In two studies (an experimental study and a large field study), we demonstrate this by showing that social power and personal power have opposite effects on stereotyping, but parallel effects on behavioral approach.

The authors investigated how a collective self-construal orientation in combination with positive social comparisons "turns off" the negative effects of stereotype threat. Specifically, Experiment 1 demonstrated that stereotype threat led to increased accessibility of participants' collective self ("we"). Experiment 2 showed that this feeling of "we-ness" in the stereotype threat condition centered on the participants' stereotyped group membership and not on other important social groups (e.g., students). Experiment 3 indicated that in threat situations, when participants' collective self is accessible, positive social comparison information led to improved math test performance and less concern, whereas in nonthreat situations, when the collective self is less accessible, positive comparison information led to worse test performance and more concern. Our final experiment revealed that under stereotype threat, only those comparison targets who are competent in the relevant domain (math), rather than in domains unrelated to math (athletics), enhanced participants' math test performance.

The authors hypothesize that social comparisons can have automatic influences on self-perceptions. This was tested by determining whether subliminal exposure to comparison information influences implicit and explicit self-evaluation. Study 1 showed that subliminal exposure to social comparison information increased the accessibility of the self. Study 2 revealed that subliminal exposure to social comparison information resulted in a contrast effect on explicit self-evaluation. Study 3 showed that subliminal exposure to social comparison information affects self-evaluations more easily than it affects mood or evaluations of other people. Studies 4 and 5 replicated these self-evaluation effects and extended them to implicit measures. Study 6 showed that automatic comparisons are responsive to a person's perceptual needs, such that they only occur when people are uncertain about themselves. Implications for theories of social cognition, judgment, and comparison are discussed.