Download Self and Social Identity: Understanding Personal and Collective Categorizations and more Exams Social Psychology in PDF only on Docsity!
PERSONAL AND SOCIA L I DENTI T Y: SELF AND SOCIAL CONTEXT
John C_._ Turner, Penelope J_._ Oakes, S. Alexander Haslam and Craig McGarty Department of Psychology Australian National University
Paper presented to the Conference on "The Self and the Collective" Department of Psychology, Princeton University, Princeton, NJ, 7-10 May 1992
A revised version of this paper will appear in Personality and Social Psychology Bulletin Special Issue on The Self and the Collective
Professor J. C. Turner Department of Psychology GPO Box 4, ANU Canberra, ACT 2601 Australia
Tel: 06 249 3094 Fax: 06 249 0499 Email: JCT655@CSCGPO.ANU.EDU.AU 30 April 1992
Abstract
Social identity and self-categorization theories provide a distinctive perspective on the relationship between the self and the collective. They assume that individuals can and do act as both individual persons and social groups and that, since both individuals and social groups exist objectively, both personal and social categorical self-categorizations provide valid representations of self in differing social contexts. As social psychological theories of collective behaviour, they take for granted that they cannot provide a complete explanation of the concrete social realities of collective life. They def i ne their task as providing an analysis of the psychological processes that interact with and make possible the distinctive "group facts" of social life. From the early 1970s, beginning with Tajfel's research on social categorization and intergroup discrimination, social identity theory has explored the links between the self- evaluative aspects of social'identity and intergroup conflict. Self-categorization theory, emerging from social identity research in the late 1970s, made a basic distinction between personal and social identity as differing levels of inclusiveness in self-categorization and sought to show how the emergent, higher-order properties of group processes could be explained in terms of a functional shift in self-perception from personal to social identity. It suggested that the basic capacity of people to engage in collective behaviour (group formation, social influence, social stereotyping etc.) is related to the essential character of the self-process. The collective does not merely impinge on , influence or modify the psychologically real individual as a set of external social forces, but is as much an authentic expression of the self as is the individual behaviour we describe in terms of personality or individual differences.
Latterly, our research has sought to elucidate how social relationships and the social context lead to variation in the social categorizadon of self and others. T he guiding theme is rh:;t self- categorizing is inherently variable, fluid and context-dependent, since self-categories are social- comparative and always reladve to a frame of reference. This notion has major implications for accepted ways of thinking about the relationship between the self and the social context. In one sense the self may be regarded as the psychological vehicle by which the collective shapes the cognitive functioning of the individual in terms of his or her changing social relationships and group memberships. The present paper will describe some of this more recent research and discuss these implications.
depersonalized. That is, individuals tend to define and see themselves less as differing individual persons and more as the interchangeable representatives of some shared social category membership. For example, when an individual man tends to categorize himself as a man in contrast to women, then he (really "we") tends to accentuate subjectively his similarities to other men (and reduce his idiosyncratic personal differences from other men) and enhance perceptually his stereotypical differences from women (Hogg and Turner, 1987). His self changes in level and content and his self-perception and behaviour becomes depersonalized. Depersonalization of the self is the subjective stereotyping of the self in terms of the relevant social categorization.
Much if not most of our research over the last decade or more has been directed at explaining and showing how depersonalization produces a qualitative transformation of individual behaviour from the interpersonal to the intergroup level, and how this transformation produces the higher-order, emergent processes of group behaviour (in the areas of group formation and cohesiveness, social cooperation and competition, and social influence, especially group polarization and minority influence; see Turner, 1991; Turner and Oakes, 1986, 1989; Turner et al., 1987). We tried to show that the hypothesis of the social categorical or collective self was necessary to provide a satisfactory and heuristic explanation of the major group phenomena and that traditional theories assuming a dominant role for personal self- interest had reached the end of their useful life.
Where does the distinction between personal and social identity come from? Why is it that sometimes we define ourselves as social groups and at other times as separate persons? What determines this variation in the level of self-categorization? We do not thin k that there is anything odd, special or aberrant about the collective self. We suppose that variation in self categorization is normal and ordinary and that the collective self arises as part of this normal variation and as a result of the general processes that govern it. Following Bruner. Tajfel and Rosch, we explain variation in self-categorization as a function of an interaction between the relative accessibility of a particular self-category (or "perceiver readiness", the readiness of a perceiver to use a particular categorization) and the fit between category specifications and the stimulus reality to be represented (the match between category meaning and reality).
Relative accessibility reflects a person's past experience, present expectations and current motives, values, goals and needs. Generally, it reflects the active selectivity of the perceiver in being ready to use categories which are relevant, useful and likely to be confirmed by the evidence of reality.
Fit has two aspects (Oakes, 1987; Oakes et al., 1991): comparative fit and normative fit. Comparative fit is defined by the principle of meta-contrast (Turner, 1985), which states that a collection of stimuli is more likely to be categorized as an entity (a higher-order unit) to the
degree that the a v erage differences perceived between them are less than the average differences perceived between them and the remaining stimuli which comprise the frame of reference. Stated in this form, the principle defines fit in terms of the emergence of a focal category against a contrasting background. It can also be used to define fit for the salience of a dichotomous classification: for example, any collection of people will tend to be categorized into distinct groups to the degree that average intragroup differences are less than average intergroup differences w ithin the relevant comparative context.
Normative fit refers to the content aspect of the match between category specifications and the instances being represented. For example, to categorize a group of people as Catholics as opposed to Protestants, they must not only differ (in attitudes, actions, etc.) from Protestants more than from each other (comparative fit), but must also do so in the right direction on specific content dimensions of comparison. Their similarities and differences must be consistent with our normative beliefs about the substantive social meaning of the social category.
We assume that the interaction between perceiver readiness and fit is a general process at work in categorization, not merely one that applies to social and self-categorization. In our work, however, we have been primarily concerned with the role of fit in determining the salience of social categorizations of self. Three studies that illustrate what we mean by the meta-contrast principle in relation to social categorization are described in Oakes et al. (1991) and Hogg and Turner (1987).
Oakes et al (1991) show in two studies how the social categorization of people into males and females and Arts and Science university students is determined by comparative and normati v e f it It is where men and women or .Arts and Science students disagree between social groups but agree within social groups in a direction consistent with observers’ stereotypes of their beliefs that they are socially categorized as such, but not otherwise. T here must be a correlation (Tajfel, 1969) between the intragroup and intergroup differences observed between interactors and the relevant social categorization before they will be perceived in terms of their shared social identity as men or women. Arts or Science students. Thus, collective conflict, in particular, is a powerful determinant of social categorization. Hogg and Turner (1987) show how males and females define and stereotype themselves more as males or females when they make intergroup rather than intragroup comparisons. Intergroup comparisons focus on the perceived differences between men and women compared to intra-sex differences, whereas intragroup comparisons focus on the larger perceived differences between personal self and other men relative to intra-personal differences (for men) and between personal self and other women relative to intra-personal differences (for women).
(both theoretically and empirically) the classic judgemental differences between extremists and moderates in these terms (Sherif and Hovland, 1961). T he comparative differences between own and others' positions which exist for extremists predict that they will tend to categorize people into us and them more sharply and polarize in their judgements more than moderates (seeing the world as more black and white than grey, and more black than white). We have also been able to reverse the classic pattern under certain conditions - getting extremists to assimilate more than moderates and moderates to polarize and contrast more than extremists - by varying their relationship to the majority of others within the context as prescribed by the meta-contrast principle.
The meta-contrast principle of comparative fit can predict the effects of extent of comparative context, intergroup versus intragroup comparisons, minority versus majority status, and relative extremity of own and others' position in increasing the level of inclusiveness of self-categorization so that others categorized as different and contrasted away from self at a lower level are re-categorized as self and assimilated towards self at the more inclusive level. Fundamentally, it is where intergroup differences tend to be perceived as larger than inttagroup differences that we tend to categorize self as "we" instead of "1" and see the included other(s) as similar instead of different.
- Which specific self-category tends to become salient at any given level? Self-categories must match the relations between self and others in terms of normative fit, content. Oakes et al. (1991) showed that the Arts/Science social categorization was only used to define a target person as a typical Arts student where her attitudes were in line with subjects' stereotype of what Arts students thought on the issue. We know what particular social categories are like or are supposed to be like and we employ social categorizations that are consistent with our background knowledge and theories tcf Murphy and Medin, ly«5j.
But it is not only that categorizations must match in terms of the specific content dimensions of comparison; it is also true that the meaning of the salient social categorization will vary to reflect die content of the diagnostic differences between groups in specific contexts. The content of categories is selectively varied to match what is being represented in terms of our background theories and knowledge. Directional differences are observed on specific content dimensions, and we selectively construct a meaningfully matching category. It is not a fixed category content being applied, the category content is being selectively defined by what it best represents. There is no one-to-one correspondence between the long-term knowledge we have about what different kinds of people are like and the actual social category which is constructed to represent them in a given setting.
Examples are provided by our studies of stereotyping and outgroup homogeneity. The "Iraq" study (Haslam et al., 1992b) showed that the content of the stereotype of Americans
(what Americans are like) varied significantly from the beginning to the end of the Gulf War and w ith the groups comprising the frame of reference. For example, compared to the Soviet Union at this time, Americans were seen as aggressive; compared to Iraq, they were seen as less aggressive. At another time, compared to another group, being American meant something else. In the homogeneity study (Haslam et al„ 1992a), Australians judged on their own (by Australians) are seen as happy-go-lucky, straightforward and sportsmanlike; judged in the context of Americans (where ingroup homogeneity increases), they are seen as even more sportsmanlike but less happy-go-lucky, and a new trait also makes its appearance, pleasure-loving_._
Comparative and normative fit are inseparable. A fitting category must optimize meta contrast on the right dimensions and the differences which optimize such contrast selectively define category content. Self-categories are selected and constructed to represent meaningfully the observed relations between self and others on specific content dimensions in terms of people's background knowledge and theories and also vary in content as a function of these represented relations. The content of a self-category is not a fixed set of attributes applied in an all or none manner, but is shaped selectively by the context of its application (this was also true in Oakes et al., 1991).
3_._ The internal (graded) structure of the relative prototypicality of members of self categories varies with the context within which the category is defined. Categories are not defined by a fixed prototype (nor a fixed set of exemplars), but vary in the relative prototypicality of their members as a function of context. This follows from and can be easily demonstrated by means of the meta-contrast principle (Turner and Oakes, 1986, 1989; McGarty et al., 1992b). What the typical "psychologist" is like will vary with the ingroup and outgroup members compared in any sett i ng_._ Change tne outgroup ana tne most prototypical psychologist (defined in terms of the meta-contrast between his or her average difference from outgroup members divided by his or her average difference from ingroup members) will change. We have applied this analysis to the explanation of group polarization (the tendency of group discussion to extremitize members' attitudes, beliefs, etc. in the direction to which they were already tending) - in fact the analysis was developed precisely to make sense of this kind of group process (Turner, 1991; Turner et a l, 1987). It can be shown that as a group becomes more extreme in an intergroup context, its more extreme members will gain in relative prototypicality over more moderate members. Thus the member who best defines what the group has in common (its consensual, normative position) will become more extreme than the mean position of members as the group tends to differ in one direction from other groups in the context. Group polarization (which we define as conformity to a polarized ingroup norm) is a product of asymmetrical intergroup comparisons. It occurs because people are conforming to what they have in common as m gm up im. atm m tsz. it* other groups (ie., an emergent category
perceiver, definitions of the individual in terms of his or her (and others') contextual properties. The meaning and form of the self-category derives from the relationship of the perceiver to the social context. The perceiver gains identity from being placed in context.
- Self-categories are veridical in that their variation is systematically related to changes in social reality. The variability of self-categorization is not a sign of the tine identity of the person being distorted by external circumstances. On the contrary, variability is necessary if self-perception is to be veridical, accurate and useful, if it is to get right the changing contextual properties of people. Identity varies in order to represent the perceiver's changing relationship to reality.
We can speculate that self-categories are the variable products of an active, constructive process of veridical social judgement in which the perceiver is defined in terms of his or her changing relationships to others within the frame of reference, presumably to enable the individual to regulate him- or herself in relation to an ever-changing social reality. They are selective in reflecting perceiver readiness as well as matching stimulus characteristics and embody long-term knowledge, beliefs and theories about category meanings. They are veridical in the sense that they are selected and constructed to match reality.
This point about veridicality is important because there is a long tradition in social psychology which equates the social identity level of self-categorization with social prejudice and cognitive distortion. Research on social stereotyping tends to take for granted - and has done so for 70 years with only 2 or 3 exceptions (Asch, Sherif, and maybe Vinacke) - that stereotyping is over-simplification, over-generalization, seeing people as more similar within groups and more different between groups than they really are, cognitive error in the service of of social, motivational and cognitive needs. Social stereotyping , however, from a cognitive point of view, is only (as G. W. Aliport and Tajfel pointed out) the perceptual accentuation of intra-category similarities and inter-category differences as a result of the salience of ingroup- outgroup categorizations of self and others. It is the perceptual effect of the social identity level of self-categorization. If the traditional perspective were accepted, then self-categorization theory would imply that group formation and group processes were founded on a cognitive distortion of self, perceiving self as more similar to ingroup others than we really are and more different from outgroup others than we really are, since the theory explains group processes as based on the depersonalization of self. We find this is an unpalatable and implausible inference.
But is n't it so? Surely when, solely as a function of context, as in the Haslam and Turner (in press) studies, another person who differs from own position by some constant amount is categorized as similar and assimilated in one extended context and as different and contrasted in another restricted context, this must be distortion? Own position has not changed, other's
position has not changed and yet the relevant including or excluding self-category does change, resulting in differential perception of the target other. This example points out the error of interpretation normally made. The "distortion" only appears as such if we begin with the assumption that self-categories reflect enduring, fixed, absolute properties of the perceived, bul this is mythical - all perception is relative; all perception reflects categorization, and categorization we have argued is intrinsically comparative and relational. When one changes the context of comparison, then by definition the relationship of the self and target individual to the context must change. The social contextual, relational, comparative properties of the self and other must therefore change, and it is this relative reality that self-categories embody, the relationship of the people to the context. It is not tine then that social and self-categorization vary independently of reality - which would indeed be distortion. In our research we never find that self-categories vary without some corresponding, lawful change in the self, other or context.
From the perspective of self-categorization theory stereotyping is not cognitive distortion or error. We take it for granted that social groups and collective relationships exist as much as do individual personalities and individual differences. There is a distinctive aspect of reality which social categorizations of self and others fit and that is the reality of the group. We act as social groups in some contexts and as individual persons in others. We can and do do both. The fit hypothesis that governs the salience of self-categorizations applies at both the personal and group level. Moreover, that principle (of meta-contrast) is rational, reasonable and directed at representing the reality of our social relationships in a sensible way. It is no more a distortion to see people in terms of their social identity than in terms of their personal identity - both are products of the same general categorization processes. It is not true that individual differences are real but that social similarities are fictions. It is a purely ideological assumption to say that one level of self-categorization is inherently more real than another (it is the individualistic thesis of F. H. Allport. 1924; see Asch, 1952. for a refutation).
We argue (Oakes and Turner, 1990; Oakes et al., in preparation) that social categorizations of self and others fit aspects of reality as do other self-categorizations, that (1) there is a group reality for them to fit, that (2) ingroup-outgroup categorizations become salient appropriately, as a function of a principle directed at the veridical perception of both personal and collective relationships, and (3) that the perceptual accentuation effects of ingroup-outgroup categorization reflect the rational selectivity of perception in which it is more appropriate see people in some contexts at the level of social category identity than at the level of personal identity. What is important is that we vary the category identity of self and others to reflect their comparative relationships with others. Moreover we do not impose a fixed mental image, but construct the ingroup-outgroup category meaningfully to explain, describe and justify intergroup relations (Tajfel, 1981). Never to see people as social categories would be getting it
w ays of perceiving, thinking and doing which we assume to be appropriate in terms of the demands of objective reality. The issues and research are complex and cannot be easily summarized, but a central point which emerges is that physical reality testing (which is really direct individual perceptual, cognitive and behavioural testing) and social reality testing (consensual validation, seeking the agreement of ingroup others) are not alternatives, as Festinger ( 1950) suggested, but are interdependent and mutually complementary phases or aspects of achieving valid social cognition. Individual perception and cognition rests upon socially validated knowledge, theories, methods and categories just as the informational power of consensual validation depends upon the recipients' assumption that individuals have independently tested the views they are exchanging. Social reality testing is a basic extension of human cognitive activity and competence. It is not true that human information processing is individual, private, asocial and non-normative - this is a superficially convenient but misleading fiction of modem psychology. The facts of social influence as studied by social psychologists for 70 or more years are witness to how individual cognitive activity is qualitatively transformed into an emergent group process by the functioning of the collective self. The knowledge, theories, understandings embodied in the meanings of categories, schemata, cognitive structures (including self-categories and self-schemata) are collectively produced and validated. Categories are not only cognitive structures, they are also implicit social norms (McGarty and Turner, in press; McGarty et al., in press).
We suggest that the way we think about people and things, the way we categorize them, and the meaning of the categories we employ vary with judgements from others that we accept or reject and that acceptance or rejection, agreement or disagreement, is a function of context- dependent self-categorization. The same information may be accepted or rejected by the same kind of people in different social contexts depending on how they categorize self and others at the time (Wetherell, 1987). We have shown this in group polarization studies where groups may polarize in opposite directions in different intergroup contexts, agreeing, for example, to be "risky" in the context of a cautious outgroup and "cautious" in the context of a risky outgroup. The group norm, which specifies the correct judgement, will vary, becoming more or less extreme, as a function of the relationship of the ingroup to the social context (McGarty et al., 1992b).
Similarly, in minority conversion studies, where subjects may be influenced by a minority to extend their private category "green" to include degrees of "blue" (as in Moscovici and Lage, 1976), David and Turner (1992) have shown in a series of experiments that this kind of indirect influence from a minority to change one's private categories depends on the minority being categorized as ingroup rather than outgroup, which in turn varies with whether the social context is intergroup or intragroup. Minorities, we argue, tend to have indirect (private, long term, general) influence rather than direct (public, temporary, limited) influence, because they
are categorized a t a subordinate level as different from self in the immediate intragroup context of disagreement, but as similar to self at the higher group level in the w ider context of societal outgroups. In the party room we reject the factional minority's views as wrong, but outside, later, facing our common enemies, we begin to embrace them.
What we call private acceptance or conversion or informational influence can easily be called private category change. We have known since Sherif and Hovland and have confirmed it within the present framework (Haslam and Turner, 1992; McGarty and Turner, in press) t hat categorization-accentuation effects, reflecting the functioning of a cognitive process of categorizing (Tajfel and Wilkes, 1963), vary with own position and self-categorization. The group polarization and minority influence studies demonstrate that own position and ingroup- outgroup categorization vaiy with the context within which self is defined. Categorical judgements are norms implicit in social identi t y and therefore are subject to the same variability as self-categorization. Who one is directly affects the normative categories employed to represent the world.
- Putting these points together, a different view of self emerges from that curren t ly dominant (Turner, 1988, 1989). The self is not a fixed mental structure but an active process of social judgement producing variable self-categories which represent the individual in terms of his or her social contextual properties. We should not think of self-perception or self definition as the activation of pre-formed, already stored self-concepts (whose meaning is defined prior to their activation), but as a flexible, constructive process of judgement and meaningful inference in which varying self-categories are created to fit the perceiver's relationship to social reality. The varing self-concepts reflect the varying outcomes of this dynamic process.
Wre doubt whether the idea of self as a relatively fixed mental structure is meaningful or necessary. If self-categories are contextual definitions of the individual, how can they be stored prior to their use? How can they be stored as pre-formed givens independent of the context in which they are used? Social contexts are infinitely variable, as are our relationships to them, yet we are never at a loss for an appropriate self-definition. If a stored set of self- concepts is adjusted in some way for new contexts, then theoretically what is needed is an explanation of how the adjustment occurs, a principle of the genera t ion of the concepts used, and once we have one (as in the fit hypothesis, for example), it is not clear tiiat the notion of prior-concepts-waiting-to-be-activated plays any further heuristic or explanatory role.
The concept of self as relatively fixed, bounded mental structure does not seem necessary since we can assume that all and any cognitive resources, including long-term knowledge, theories etc., are recruited, used, and deployed as and when necessary to create the needed self-category. Rather than a distinction between the activated self and the stored, inactive self.
other about an object is an agreement and that therefore the object is to be defined in terms of this "similarity" of attribu t es. Or, the opposite, that the judgement of a contrasting outgroup other is a disagreement and defines a "difference", to be resolved by some other kind of categorizing.
To conclude, the readiness to create particular categories, the background knowledge used to give them form and meaning and the tendency to consttue certain stimulus properties as sharing identi t y or not on the basis of agreement or disagreement, all v ary with the context- dependent self. Perception ref l ects the cognitive structures created to represent the stimulus world. These structures are actively created through processes anchored in the varying self and the variable self expresses the relationship of the individual to the social context. The collective self is therefore a mechanism for the social determination of cognition, for translating variation in one's "social place" into relevant "cognitive choice". Cognition not only reflects the stimulus array to be represented but also always the social context within which it takes place. All cognitive activity has, depends upon, a collective dimension via the collective self. The fact of self makes the study of cognition necessarily social psychology. All cognition is social cognition, from the perspective of the mechanisms of cognition, whether or not it is people or objects that are perceived.
References Abrams, D_._ and Hogg, M. A. (1990) Social identification, se l f-categorization and social ifluence. In W. Stroebe and M. Hewstone (ed.s), European Revie w of Social Psychology. Vol. 1, 195-228, Chichester, UK: Wiley. Abrams, D„ W etherell, M. S., Cochrane, S., Hogg, M. A. and Turner, J. C. (1990) Knowing what to think by knowing who you are: A social identity approach to norm formation, conformity and group polarization. British Journal of Social Psychology. 29, 97-
Allport, F_._ H. (1924) Social psychology. New York: Houghton, Mifflin. Asch. S. E. ('1952') Social psychology. New Jersey: Prentice-Hall. David, B. and Turner, J. C. (1992) Studies in self-categorization and minority conversion. Paper presented in Symposium on "Minority Influence" at the Joint Meeting of the EAESP/SESP, Leuven/Louvain -la Neuve, Belgium, July. Doise, W. (1969) Intergroup relations and polarization of individual and collective judgments. Journal of Personality and Social Psychology. 12, 136-143. Festinger, L_._ (1950) Informal social communication. Psychological Review_._ 57, 271-82. Gaertner, S. L., Mann, J., Murrell, A. and Dovidio, J. F. (1989) Reducing intergroup bias: The benefits of recategorization. Journal of Personality and Social Psychology. 57, 239-249. Haslam, S. A. and Turner, J. C. (1992) Context-dependent variation in social stereotyping 2: The relationship between frame of reference, self-categorization and accentuation. European Journal of Social Psychology. 22, 251-277. Haslam, S. A. and Turner, J. C. (1992) Context-dependent variation in social stereotyping 3: The social reality of extremism as a basis for accentuation. ANU. Haslam, S. A., Oakes, P. J., McGarty, C. and Turner, J. C. (1992a) Studies of ingroup homogeneity. ANU. Haslam, S. A., Turner, J. C., Oakes, P. J., McGarty, C. and Hayes, B. K. (1992b) Context- dependent variation in social stereotyping 1: The effects of intergroup relations as mediated by social change and frame of reference. European Journal of Social Psychology. 22, 3-20. Hogg, M. A. and Turner, J. C. (1987) Intergroup behaviour, self-stereotyping and the salience of social categories. British Journal of Social Psychology. 26, 325-340. Hogg, M. A., Turner, J. C. and Davidson, B. (1990) Polarized norms and social frames of reference: A test of the self-categorization theory of group polarization Basic and Applied Social Psychology. 11, 77-100. McGarty, C., Haslam, S. A., Oakes, P. J. and Turner, J. C. (1992a) The effects of relative group size on category entitativity. ANU. McGarty, C. and Turner, J. C. (in press) The effects of categorization on social judgement. British Journal of Social Psychology.
McGarty, C., Turner, J. C., Hogg, M. A., David, B. and Wetherell, M. S. (1992b) Group polarization as conformity to the prototypical group member. British Journal of Social Psychology. 31, 1-20.
Turner, J. C. (1981) The experimental social psychology of intergroup behaviour. In J. C_._ Turner and H. Giles ( eds.), Intergroup behaviour. Oxford: Basil Blackwell and Chicago: University of Chicago Press, Ch. 3, pp. 66-101. Turner, J_._ C. (1985) Social categorization and the self-concept: a social cognitive theory of group behaviour. In E. J. Lawler ( ed.~). Advances in Group Processes. Vol. 2. Greenwich, Connecticut: JAI Press, pp. 77-122. Turner, J. C_._ (1988) The self, social identity and social cognition. In symposium on "Social cognition: Critical perspectives from social and developmental psychology", 24th International Congress of Psychology, Sydney, 28 August-2 September. Turner, J. C. ( 1989) Is the self a cognitive structure? The Henri Tajfel Memorial Lecture, University of Bristol, UK, 25 January. Turner, J. C. (1991) Social influence. Milton Keynes, England: Open University Press and Pacific Grove, Calif.: Brooks/Cole. Turner, J. C., Hogg, M. A., Oakes, P_._ J., Reicher, S. D. and Wetherell, M. S. (1987) Rediscovering the social group: A self-categorization theory. Oxford and New York: Basil Blackwell. Turner, J. C. and Oakes, P. J. (1986) The significance of the social identity concept for social psychology with reference to individualism, interactionism and social influence. British Journal of Social Psychology. 25. 237-252. Special issue on the Individual/Society Interface. Turner, J. C. and Oakes, P. J. (1989) Self-categorization theory and social influence. In P. B. Paulus (ed.), The psychology of group influence. Hillsdale, New Jersey: Erlbaum, 2nd edition, pp. 233-275.
Turner, J. C., Oakes, P. J. and Haslam, S. A. (1991) Politics. Prejudice and Myth in the Study of Stereotypes. Paper presented at the 4th International "Nag's Head" Conference on Stereotypes and Intergroup Relations, Highland Beach, Florida, USA, June 16-21.
Turner, J. C., W etherell, M. S. and Hogg, M. A. (1989) Referent informational influence and group polarization. British Journal of Social Psychology_._ 28. 135-147.
van Knippenberg, A. and Wilke, H. (1988) Social categorization and attitude change. European Journal of Social Psychology. 18, 395-406.
Wetherell, M. S. (1987) Social identity and group polarization. In Turner, J. C., Hogg, M. A., Oakes, P. J., Reicher, S. D. and Wetherell, M. S., Rediscovering the social group: A self-categorization theory. Oxford and New York: Basil Blackwell.
Wilder, D. A. and Thompson, J. E. (1988) Assimilation and contrast effects in the judgements of groups. Journal of Personality and Social Psychology. 54, 62-73.
