Summary

The general assumption is that "people categorize others in order to render the social world a meaningful and predictable place in which we can act efficaciously." (p. 59) We place people - including ourselves - in categories as a way to handle complexity (both Malcolm Gladwell's Blink and Gigerenzer's work on heuristics may be related to this aspect of cateogorization).

Members of a social category bear to varying degrees resemblance to a category prototype, an abstraction of properties. Depersonalization is the process by which individuals are not perceived in terms of their unique characteristics and contexts, but are perceived in terms of their match to the relevant category prototype. Polarization in groups can also be explained in terms of the category prototype - "polarization emerges where people identify with a group that has a polarized prototype" (p. 64).

Given a category prototype for a group, then there will be "fringe, marginal, or peripheral members" (p. 66) who are perceived to match this prototype weakly. Such members will be perceived as deviants within the groups. Deviancy may be positive or negative, where the former occurs through being over-achievers. Norms of the group may be embodied in the category prototype. A group leader who more closely matches the prototype may be more influential within the group, and also less likely to be harmful to the group. Leaders may exercise power in harmful ways when the leader is no longer prototypical to the followers, while the leader also feels less in-group identification with the group.

Questions

1. Is a stereotype a type of category prototype?
2. The author uses the terms depersonalized social attraction and personal attraction - are they equivalent to common identity and common bond as used in the other readings this week?

Connections

The explanation of group polarization with respect to category prototypes has a connection to Sunstein's (2001) study of polarization on the Internet. The content of a web site may be an expression of the category prototype. So polarization may emerge with some web site readers who identify strongly with the category prototype. The extent to which this happens is unclear. 

Meta-contrast is the ratio of between group differences and within group differences.

Prototype of a social category is determined with this ratio, and helps explain why the average of the group beliefs is not the prototype.

Some implications of meta-contrast in group identity formation:
- Groups pick a prototype that maximizes this ratio; polarization happens b/c people move towards the prototype
- Decisions about who is in and who is out are based on this prototype
- Leader of group tends to be close to this prototype

After reading this paper I was left wondering if it makes sense to divide categorization into social versus non-social categorization. I think the part that bothers me the most is the assumption that all social categorization involves the self, while non-social categorization does not. I can think of instances of social categorization that do not appear to involve the self in a meaningful way (e.g. trying to determing membership of two outgroups who are rivals with each other). Also, there may be instances of self-relevant categorization of objects. For example, people who have collections (say, of stamps) probably have a sense of what genres of stamps they prefer, and preferences for objects seem to implicate the self (e.g. the "endowment effect"). Similarly, when purchasing clothes people have to determine which articles of clothing "are me." Again this seems to be a level of choice that implicates the self.

One of the reasons why groups research has lost its central place within psychology is that it is difficult to find situations that are best explained at the group level rather than the individual level. This chapter covers a few (conformity, risky shift) and there are a few others out there (e.g. some research on leadership) but these effects just don't seem to cohere together.

The one effect that was new to me that I read about in this article was the "black sheep effect". Basically, the finding is that people reject non-prototypical ingroup members more strongly than non-prototypical outgroup members. This is interesting because it reverses typical effects of group membership.