A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
The minimal group paradigm demonstrates that people show ingroup favoritism and discrimination based on entirely arbitrary and meaningless group assignments. Even when group membership is random, offers no material benefit, and provides no interaction history, people favor their own group. This finding reveals how minimal group identity is sufficient to trigger intergroup bias, suggesting social identity is a fundamental human motivation.
From social identity theory, you know that people derive part of their self-concept from the groups they belong to, and that they are motivated to maintain a positive social identity. The minimal group paradigm (Tajfel et al., 1971) was designed to identify the *minimum conditions* necessary to produce intergroup discrimination — and the result was deeply counterintuitive: almost no conditions at all. Participants were sorted into groups based on trivial or entirely arbitrary criteria (preference for Klee vs. Kandinsky paintings, or even a coin flip), then asked to distribute points between anonymous ingroup and outgroup members. Despite having no history with their group, no face-to-face interaction, no group name that carried meaning, and no material benefit from the allocation, participants systematically favored ingroup members.
The most striking finding wasn't just that people gave more to the ingroup — it was *how* they gave. Tajfel and colleagues showed that participants often preferred strategies that maximized the relative difference between ingroup and outgroup allocations over strategies that maximized the *absolute* amount their ingroup received. Given a choice between giving 7 to ingroup / 1 to outgroup versus 12 to ingroup / 11 to outgroup, many participants chose the former — a smaller absolute gain for their group, but a larger *relative* advantage. This suggests the motivation isn't purely economic or rational; it's about maintaining a positive distinction between "us" and "them," even at a cost to one's own group.
Why does mere categorization produce this effect? Social identity theory offers an account grounded in self-enhancement motivation: if my group's value is part of my self-concept, then making my group look better (relative to other groups) is a form of self-enhancement. The categorization alone is enough to activate this dynamic — no conflict, no competition for resources, no historical grievance required. The implication is stark: intergroup bias doesn't need realistic conflict (Sherif's robbers cave) or ideological content (racism, nationalism) as a precondition. Category membership alone, however arbitrary, triggers psychological differentiation between "us" and "them" and motivates behavior that advantages the ingroup.
The paradigm doesn't imply that all intergroup bias is equally arbitrary or that context doesn't matter. Real-world discrimination involves history, power, cultural meaning, and material stakes that the laboratory strips away. What the minimal group paradigm reveals is the *psychological floor*: the baseline bias that can be triggered by categorization alone, before any of those complicating factors are added. This sets a troubling baseline — it suggests that any social categorization (race, gender, team, neighborhood, shirt color) can activate ingroup favoritism as a default, and that overcoming bias requires active counterforces, not just the absence of explicit hostility.
Understanding the minimal group paradigm also illuminates real-world phenomena that otherwise seem puzzling. Sports team fandom, national identity, corporate culture, school spirit — all of these produce strong ingroup favoritism despite the fact that, rationally, "which soccer team I support" should not affect how I treat strangers. The minimal group findings suggest that these are not aberrations or failures of rationality but expressions of a very basic psychological dynamic: categorization activates social identity, social identity motivates positive distinctiveness, and positive distinctiveness generates favoritism toward members of one's own category. Managing this dynamic — rather than pretending it doesn't exist — is the practical starting point for reducing intergroup bias.