Philosophers have done significant work on concepts of ‘race’ and ‘racism’, on the ethics of a spectrum of race-conscious policies proposed to address a history of discrimination, and on identity and the experience of race (Outlaw 1996; Boxhill 2001; Bernasconi 2001a; Goldberg 1990; Appiah and Gutmann 1996). That work consists primarily of conceptual and normative analyses of prejudice and of the social policies designed to address it. Philosophers have also considered internal questions of racism within the canonical history of Western Philosophy, in for example Aristotle, Kant, and Hegel (Popkin 1980; Bernasconi 2001b). What has been lacking, however, is sustained philosophical analysis regarding issues raised in the extensive social psychological literature: questions regarding the nature and formation of prejudice, questions regarding the social dynamics of prejudice, and questions regarding prospects for prejudice reduction. Here, explanation is central. If we cannot accurately explain how prejudice occurs and how it can be reduced, how are we to construct adequate public policy? The lack of philosophical attention in this area is thus particularly conspicuous and unfortunate. As a first step toward remedying this situation—and with an eye toward public policy— we apply spatialized game theory and multi-agent computational modeling as philosophical tools: (1) for assessing the primary social psychological hypothesis regarding prejudice reduction, and (2) for pursuing a deeper understanding of the basic mechanisms of prejudice reduction. Social modeling in general has a philosophical pedigree that extends at least back to Hobbes and Locke. The particular techniques of social simulation employed here are relatively new, however, and raise important questions for the philosophy of science. For that reason we proceed reflexively, commenting throughout on both the promise of simulational techniques for social psychology and public policy and their inherent limitations.