Dixon et al. suggest that the psychological literature on intergroup relations should shift from theorizing “prejudice reduction” to “social change.” A focus on social change exposes the importance of psychological theories involving collective phenomena like social norms and institutions. Individuals' attitudes and emotions may follow, rather than cause, changes in social norms and institutional arrangements.

A view of inequality as a relationship between the advantaged and the disadvantaged has gained considerable currency in psychological research. However, the implications of this view for theories and interventions designed to reduce inequality remain largely unexplored. Drawing on the literature on close relationships, we identify several key features that a relational theory of social change should include.

Different groups, because they are perceived to pose different threats, elicit different prejudices. Collective action by disadvantaged groups can amplify the perception of specific threats, with predictable and potentially counterproductive consequences. It is important to carefully consider the threat-based psychology of prejudice(s) before implementing any strategy intended to promote positive social change.

Individuals differ in their support for social change. We argue that examinations of inequality and change would benefit from consideration of underlying personality processes. New data suggest that Right-Wing Authoritarianism and Social Dominance Orientation, indicators of support for inequality, may be motivated by biologically driven personality processes, particularly those related to positive-approach motivation.

A theoretical framework is required that explains why and how cross-group contact reduces collective action and how the demobilizing effects can be counteracted. We propose that at least two mechanisms are involved: an affective process whereby the positive affect created offsets negative emotions and action tendencies, and a more strategic process whereby individual advancement comes to seem like a possibility.

We disagree with Dixon et al. by maintaining that prejudice is primarily rooted in aversive reactions toward out-group members. However, these reactions are not indicative of negative attributes, such as trait bigotry, but rather normative homophily for peers with similar perceived attributes. Cognitive biases such as stereotype threat perpetuate perceptions of inequipotential and subsequent discrimination, irrespective of individuals' personality characteristics.

The focus on negative attitudes toward other groups has led to a dichotomy between the prejudice reduction and the collective action approach. To solve the resulting problems identified by Dixon et al., we suggest analyzing the psychological processes underlying the construction of relationships (and their alternatives) between own and other groups.

We also critique the myopic focus on prejudice reduction, but we do not support the call for a reconceptualization of prejudice. Redefining key psychological constructs is unproductive. Also, we point to interpersonal dynamics in cross-group interaction as a key mechanism in the prejudice reduction/collective action paradox and point to solutions involving intrapersonal/interpersonal processes, as well as broader structural intergroup relations.

This response clarifies, qualifies, and develops our critique of the limits of intergroup liking as a means of challenging intergroup inequality. It does not dispute that dominant groups may espouse negative attitudes towards subordinate groups. Nor does it dispute that prejudice reduction can be an effective way of tackling resulting forms of intergroup hostility. What it does dispute is the assumption that getting dominant group members and subordinate group members to like each other more is the best way of improving intergroup relations that are characterized by relatively stable, institutionally embedded, relations of inequality. In other words, the main target of our critique is the model of change that underlies prejudice reduction interventions and the mainstream concept of “prejudice” on which they are based.