INTERPERSONAL RELATIONS AND GROUP PROCESSES Group Conversion Versus Group Expansion as Modes of Change in Majority and Minority Positions: All Losses Hurt But Only Some Gains Gratify Radmila Prislin and P. Niels Christensen San Diego State University This study examined reactions to minority and majority positions that were either stable or reversed through group conversion that transformed opponents (supporters) of the minority (majority) into supporters (opponents) or through group expansion that brought new supporters (opponents) for the minority (majority) into the group. Minorities who became majorities through group expansion, com- pared with those who changed through group conversion, perceived their supporters and the overall group as significantly more similar to the self, and had significantly higher expectations for future positive interactions within the group. Perception of similarity with the supporters mediated the effect of the experimental conditions on perception of the overall group–self similarity. Implications of changes through conversion and expansion for the functioning of social groups are discussed. The dynamic aspects of social life, evident in changes occurring within and between groups over time, are conspicuously absent from the contemporary group research (Levine & Moreland, 1985, 1998). The paucity of empirical research is even more surprising given that at least two broad models of group dynamics over time, developmental and cyclical, have been proposed. Developmental models conceptualize change as a distinct and predictable, though not necessarily linear and irreversible, series of phases in the life of a group (Arrow, McGrath, & Berdahl, 2000; Tuckman & Jensen, 1977; Wheelan & McKeage, 1993; Worchel, Coutant- Sassic, & Grossman, 1991) and its members (Levine & Moreland, 1985; Moreland & Levine, 1988; Ziller, 1965, 1977). Cyclical models also assume the predictability of change; however, change is conceptualized as a cyclical recurrence in the perpetual battle for preferred positions within a group or a larger social system. The underlying idea of these models is that parties involved in the cycle may reverse their positions, but the group or the larger system within which the cycle perpetuates remains the same. This idea, echoing Machiavelli’s (1525/1988) view of history as a series of cyclical ups and downs, is incorporated in several sociological models (e.g., Harper, 1998; Prewit & Stone, 1973) and a rare sociopsychological model (Taylor & McKirnan, 1984) of social change. In spite of its intuitive appeal, the conceptualization of change as a cyclical reversal in positions is mechanical and of limited heu- ristic value. Cyclical models invite questions about determinants of change but not questions about its outcomes, which presumably always involve a return to the beginning. This likely is a reason for the dearth of literature on consequences of change, which stands in sharp contrast to the long tradition of research on many anteced- ents of change, including persuasion (e.g., Eagly & Chaiken, 1993; Petty & Wegener, 1998), minority influence (e.g., Moscovici, 1976; Wood, Lundgren, Ouellete, Busceme, & Blackstone, 1994), and conformity (e.g., Cialdini & Trost, 1998). The present analysis assumes that change is more than a me- chanical reversal of positions. A recently proposed gain–loss asymmetry model of change in majority and minority position within a group (Prislin, Limbert, & Bauer, 2000) postulates that change in these positions alters reality not only for the involved factions but also for the entire group. The group is assumed to be fragile immediately following the change because general loss aversion (Kahneman & Tversky, 1979) causes negative reactions to a loss of the majority position to be stronger than positive reactions to a comparable gain of the majority position. That is, the hypothesized intense and instantaneous detachment from the group in response to changes away from the majority position, which cannot be counterbalanced by less intense and gradual attachment to the group in response to the corresponding changes toward the majority position, should leave the group vulnerable to disruptions. Initial support for this asymmetry model was obtained from two studies in which groups of 1 participant and 3 confederates indi- vidually expressed their opinions about important social issues. The participant either was supported by 2 of the 3 confederates to enjoy an initial majority position or was opposed by all 3 confed- We thank Judy Price for her helpful comments on an earlier version of this article. Correspondence concerning this article should be addressed to either Radmila Prislin or P. Niels Christensen, Department of Psychology, San Diego State University, 5500 Campanile Drive, San Diego, California 92182-4611. E-mail: rprislin@sunstroke.sdsu.edu or niels@sunstroke.sdsu .edu Journal of Personality and Social Psychology, 2002, Vol. 83, No. 5, 1095–1102 Copyright 2002 by the American Psychological Association, Inc. 0022-3514/02/$5.00 DOI: 10.1037//0022-3514.83.5.1095 1095 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.