One of the most troublesome issues of social categorization is the tendency for humans to automatically categorize individuals along certain lines, such as race. A new study has looked at ways to reduce racial bias through identification with a new group.
Jay Van Bavel and William Cunningham created a series of experiments where white participants were arbitrarily assigned to one of two mixed race groups. They were not given any reason for the assignment, creating what is typically known as a “minimal group” situation. They were then shown pictures of members from both groups. In a final task, participants were shown faces from both of the groups and given an automatic measure of evaluation: faces from both teams were attached with a positive or negative word, and participants had to identify whether that word was positive or negative. Faster responses indicate stronger indirect associations between the face and the concept (positive or negative).
Their results indicate that participants did not demonstrate a racial bias towards members of their own group (the ingroup) or white individuals in the other group (the outgroup); these individuals were all evaluated relatively positively. However, outgroup black individuals were still subject to the traditional racial bias. When compared to a control group, this effect was not driven by stigmatizing the outgroup black individuals, but by increasing favoritism for ingroup black members.
Categorization by race, a bias built upon years of exposure through media and social interactions, was overridden merely by recategorization with a novel group, one with minimal context and low meaning. While this would seem to suggest that it would be useful for reducing prejudice, this benefit comes at the cost of increasing ingroup favoritism. For example, assigning mixed race work groups might reduce racial bias within the group, but it might also decrease one’s willingness to cooperate with other work groups, desiring that credit and benefits be given to one’s ingroup and not others, even if they are more diserving. As the experimenters note, although reducing racial biases is desirable, doing so at the cost of creating additional biases is not optimal.
It may be the case that group membership can override racial biases in evaluation, but do not reflect changes in attitudes, only the salient categorization at that moment. In this experimental situation, participants were likely focused on the experimental manipulation of group, appropriately understanding that if group assignment was manipulated, it must be important. This may have attenuated the effects other categories had, such as race. More generally speaking, these experiments provide evidence that categorization is a dynamic construct, changing to reflect environmental demands. Accordingly, if novel group membership is no longer salient, it is likely that general attitudes (such as attitudes towards race) should resurface.
These experiments found that classification with a group–even one minimally defined–provides a number of benefits to group members, including reduction in longstanding racial biases. However, what is it about being a part of a group that drives these benefits? Why are group members seen in a positive light? The next experiment we review examines some of these assumptions.