Knowing what to think by knowing who you are: Self‐categorization and the nature of norm formation, conformity and group polarization*

TLDR

Social influence theories differ between interpersonal dependence (normative/informational) and group membership (self‑categorization) frameworks. The study argues that both uncertainty‑reducing comparisons and normative compliance rely on perceiving the influencer as an in‑group member. The authors tested the theories with three paradigms: Sherif’s autokinetic effect, Asch’s conformity, and a group‑polarization discussion. Across the experiments, making group membership salient reduced norm formation by out‑group confederates, limited normative pressure from out‑group observers, increased uncertainty from in‑group confederates, and inhibited convergence in polarized discussions, underscoring self‑categorization as a key determinant of social influence.

Abstract

We contrast two theoretical approaches to social influence, one stressing interpersonal dependence, conceptualized as normative and informational influence (Deutsch & Gerard, 1955), and the other stressing group membership, conceptualized as self‐categorization and referent informational influence (Turner, Hogg, Oakes, Reicher & Wetherell, 1987). We argue that both social comparisons to reduce uncertainty and the existence of normative pressure to comply depend on perceiving the source of influence as belonging to one's own category. This study tested these two approaches using three influence paradigms. First we demonstrate that, in Sherif's (1936) autokinetic effect paradigm, the impact of confederates on the formation of a norm decreases as their membership of a different category is made more salient to subjects. Second, in the Asch (1956) conformity paradigm, surveillance effectively exerts normative pressure if done by an in‐group but not by an out‐group. In‐group influence decreases and out‐group influence increases when subjects respond privately. Self‐report data indicate that in‐group confederates create more subjective uncertainty than out‐group confederates and public responding seems to increase cohesiveness with in‐group — but decrease it with out‐group — sources of influence. In our third experiment we use the group polarization paradigm (e.g. Burnstein & Vinokur, 1973) to demonstrate that, when categorical differences between two subgroups within a discussion group are made salient, convergence of opinion between the subgroups is inhibited. Taken together the experiments show that self‐categorization can be a crucial determining factor in social influence.