Shared Representations and Asymmetric Social Influence Processes in Small Groups

Abstract

There are two basic theoretical assumptions that guide the interpretations of the various findings discussed in this chapter. The first serves as the cornerstone of Davis’ (1973) social decision scheme theory and has played an important role in small-group research since the early work of Lorge and Solomon (1955) and Taylor (1954). Social decision scheme theory starts with the assumption that small-group interaction can be seen as a combinatorial process, where task ele­ ments (be they ideas, task responses or preferences, resources, etc.) must be combined in such a way as to allow a group to reach consensus on a particular task (Davis, 1982). The theory then goes on to represent such processes in terms of a social decision scheme matrix, the elements of which represent conditional probabilities linking distinguishable preference distributions among the group members to particular group-level outcomes. (See Davis, 1973, 1982, or Stasser, Kerr, & Davis, 1989, for more complete descriptions of the theory.)Combinatorial processes in ad hoc, task-oriented groups (i.e., groups not constrained by rules of order or leader veto powers, etc.) can often be seen as mainly a function of social influence (either normative or informational-often both), where members with opposing preferences try to convince each other of the merits of their preferred alternatives. A common theme in social influence research has been the power of large versus small factions (though for coun­ terexamples, see Moscovici, 1976; Nemeth, 1986). Many recent models of social influence, such as social impact theory (Latane, 1981; Latane & Wolfe, 1981), the other-total ratio (Mullen, 1983), and the social influence model (Tanford & Pen­rod, 1984) all use faction size as the central component. Research specifically focused on influence in small groups has demonstrated the power of larger versus smaller factions (e.g., Tindale, Davis, Vollrath, Nagao, & Hinsz, 1990), and majority/plurality and related faction-size models have often been found to provide excellent fits to empirical data (e.g., Davis, 1982; Hastie, Penrod, & Pennington, 1983; Tindale & Davis, 1983, 1985). Recent theories of self-categori­ zation in groups (Turner & Oakes, 1989), and self-attention as it relates to groups (Mullen, 1987) also provide compelling reasons why members of larger factions within groups, as compared to members of smaller faction, would be less easily influenced and more likely to define themselves as members of the faction. In addition, in societies based on democracy, majority-wins rules are often man­ dated by law, probably leading to general perceptions that majorities are most fair, right most of the time, and so forth. Thus, for many small decision-making groups, a majority or faction-size model of social influence in groups should provide a good baseline prediction.