Social learning vs. outgroup bias: Minimal group effects in emotion categorization

Happy faces are categorized faster and more accurately than faces with negative expressions. This happy face advantage is moderated by social category cues, such as sex and race, in evaluative congruent ways. The happy advantage is larger for the more positively evaluated faces. With a minimal group paradigm, we examined whether recent and artificially acquired group membership will be sufficient to elicit a comparable evaluative bias in an emotion categorization task. To test this, group membership was created with a bogus personality test, where participants were assigned different personality types, indicated by the colour green or orange. Caucasian male faces presented on a green or orange background, indicating their personality type, were then categorized as happy or angry as quickly and accurately as possible. Consistent with the suggestion that emotion recognition can be affected by arbitrarily assigned social categories, a larger happy face advantage was predicted for the ingroup compared to the outgroup faces. A minimal group effect was evident in the error rates, where fewer errors were made categorizing happy than angry ingroup faces. There was no effect of group membership on categorization times or explicit ratings of the faces. The results suggest that artificially created groups can elicit an evaluative bias in emotion perception and provide some support for the evaluative congruence account as an explanation for the effects of social category cues on emotion categorization.