No abstract

This article reviews literature relevant to the out-group homogeneity effect. The review assesses whether the effect emerges in both natural-and minimal-group contexts. Data relevant to the out-group homogeneity effect are examined for 3 types of dependent measures. Whereas strong support for the effect is obtained across all measures in natural-group settings, no consistent effect is observed in minimal-group settings. Some theories (need-based motives, salience of self, and generalized homogeneity beliefs) predict the occurrence of the effect in both natural and minimal groups, whereas others (group-specific homogeneity beliefs and information encoding and retrieval) predict the occurrence of the effect only for natural groups. The question of whether conditions exist under which the out-group homogeneity effect can be produced in the minimalgroup setting is addressed.Group membership dramatically affects social perception and categorization. Pioneering research by Tajfel (1969; see also Tajfel, 1982b) on the accentuation principles showed that group membership leads to accentuation of intergroup differences and accentuation of intragroup similarities. One important offspring of this research reflects the asymmetrical accentuation of intergroup differences in favor of the own group. That is, people tend to be more favorable, in terms of affective reactions and resource allocation, to members of their own group (the in-group) than to members of other groups (the out-group)-a phenomenon termed ethnocentrism or in-group favoritism (Brewer, 1979;Tajfel & Turner, 1986). The second offspring of Tajfel's (1969) pioneering research reflects the asymmetrical accentuation of intragroup similarities in favor of out-group homogeneity. That is, people judge members of out-groups as more similar to one another than they do members of ingroups. Males may perceive females as more similar to one another than they perceive males and vice versa. Pro-lifers may judge pro-choicers to be more similar to one another than they judge pro-lifers to be and vice versa. Academicians may perceive business people to be more similar to one another than they perceive fellow academicians and vice versa. This phenomenon is known as the out-group homogeneity effect and is the focus of the present article.The out-group homogeneity effect can be considered a special case of stereotyping. In the early phases of stereotyping research, stereotypes were conceptualized in terms of the prototypic trait dimensions of a given social group (Gilbert, 1951; Karlins, Coffman, & Walters, 1969; Katz & Braly, 1933). Stereotypes were viewed as the perceived central tendency of the We thank Katherine Gannon, William von Hippel, and especially

Issues regarding the fairness of lineups used for criminal identification are discussed in the context of a distinction between nominal size and functional size. Nominal size (the number of persons in the lineup) is less important for determining the fairness of a lineup than is functional size (the number of lineup members resembling the criminal). Functional size decreases to the extent that the nonsuspect members of the lineup are easily ruled out as not being suspected by the police. The extent to which the identification of the suspect can be considered an independently derived piece of incriminating evidence is positively related to functional size. Empirical estimates of functional size can be obtained through pictures of the corporal lineup from which mock witnesses make guesses of whom they believe the police suspect. A distinction is made between a functional size approach and hypothesis testing approaches. Uses of functional size notions in the court, by police, and in research are discussed.

Judgments about others are often based on memory for information about the persons being judged. Three studies are reported that use decision time to determine what information subjects selectively recall when they make memorybased person judgments. Each study employed a sequential judgment paradigm in which a subject first made an impression judgment about a person on one dimension while stimulus information was continuously available. Immediately thereafter, the subject made a second judgment about the same person on a different dimension without the stimulus information being available. It was concluded that subjects' memory-based judgments were based on memory for their first impression judgments combined with a selective memory search for negative stimulus information.

No abstract

Male and female subjects (n-65) witnessed a staged theft in which either an expensive object (high seriousness) or an inexpensive object (low seriousness) was stolen, and subjects either had prior knowledge of the object's value or learned of its value only after the theft. When witnesses had prior knowledge of the object's value, accurate identification of the thief was more likely when the theft was of high rather than of low seriousness. When knowledge of the crime's seriousness was gained after the theft, seriousness did not affect identification accuracy. These results suggest that the effect of perceived seriousness on accuracy is mediated by processes that operate during rather than after the viewing interval, processes such as selective attention and encoding. The present study also found that certainty of choice in the identification task was unrelated to accuracy of choice. Portions of this paper were presented at the meeting of the American Psychological Association, Washington, D.C., September 1976. The authors would like to thank Tonnie Hoyle and Steven Kunkler for helping to conduct the experiment, and Michael H. Baumgardner for comments on an earlier draft of the manuscript.

People have a more differentiated cognitive representation of in-groups than of out-groups. This has led to the prediction that memory should be better for in-group information than for out-group information. However, past research has provided equivocal support for that prediction. This article advances a differential processing hypothesis that offers a solution to this paradox. The hypothesis suggests that whereas in-group information is organized by person categories, outgroup information is organized through attribute categories. In-group membership alters the categorical basis of memory for person information, but these categories are not necessarily superior to the attribute categories that are used to organize out-group information. That is, both person and attribute categories elicit equal amounts of recall for the in-group and the out-group. Three experiments are reported that support the differential processing hypothesis.People respond to out-group members differently than they do to in-group members. There is an abundance of research on how in-group versus out-group membership influences a wide range of social responses, including resource allocation, ethnocentrism, attribution of stereotypical characteristics, perception of group homogeneity, and social influence. Reviews of much of this literature can be found in the articles by Abrams and Hogg (1990), Diehl (1990), Park, Judd, andRyan (1991), Messick and Mackie (1989), Ostrom and Sedikides (1992), Spears and Manstead (1990), and van Knippenberg and Ellemers(1990).One approach to understanding such phenomena is through investigating whether the cognitive representations developed for in-group information differ from those formed for outgroup information. To the extent that judgments, decisions, feelings, and actions toward others are mediated by the nature of the cognitive structures drawn on, an analysis of those representations is a precondition to theory development. This article explores differences between the structures used in representing in-group versus out-group information. Previous Conceptions In-Group Versus Out-Group RepresentationThere have been several attempts to specify the different structures that are used to store and process in-group and out-

In 1979, Latane, Williams, and Harkins observed that individuals working together put out less effort than when they work alone, an effect these researchers termed social loafing. Subsequent research (Williams, Harkins, & Latane, 1981) has suggested that this effect arises, at least in part, because when participants work with others on these tasks their individual outputs are lost in the crowd, and, thus, they can receive neither credit nor blame for their performances. However, it is unlikely that participants found the tasks used in these experiments (e.g., clapping, shouting) to be involving. The possibility that personal involvement could moderate the social loafing effect was tested in a 2 (high/ low involvement) X 2 (high/low identifiability) factorial design across three replications. Replicating previous loafing research, we found that under conditions of low involvement, participants whose outputs were identifiable worked harder than those whose outputs were pooled. However, when the task was personally involving, the loafing effect was eliminated. Participants whose outputs were pooled worked as hard as those participants whose individual outputs could be identified.In 1979, Latane, Williams and Harkins rekindled interest in a phenomenon they termed social loafing. Social loafing refers to the finding that participants put out less effort when working together than when working alone. This effect has been demonstrated on tasks that require physical effort (shouting: Latane et al.