Chapter 1: Introducing Social Psychology PDF

Chapter 1 Introducing Social Psychology Wolfgang Stroebe, Miles Hewstone and Klaus Jonas Source: © Action Plus. Used under licence from Getty Images. KEY TERMS attribution theories authoritarian personality autokinetic effect balance...

Chapter 1 Introducing Social Psychology Wolfgang Stroebe, Miles Hewstone and Klaus Jonas Source: © Action Plus. Used under licence from Getty Images. KEY TERMS attribution theories authoritarian personality autokinetic effect balance theory Bennington study consistency theories covariation theory crisis in social psychology demand characteristics European Association of Social Psychology  experiment experimenter expectancy effects field experiment field theory laboratory experiment methodological individualism minimal group paradigm priming realistic conflict theory rebound effect scapegoat theory social facilitation social loafing CHAPTER OUTLINE INTRODUCTION: SOME CLASSIC STUDIES How do social psychologists go about addressing research questions? WHAT IS SOCIAL PSYCHOLOGY? How do social psychologists define their discipline? HOW DOES SOCIAL PSYCHOLOGY DIFFER FROM OTHER DISCIPLINES? What differentiates social psychology from related disciplines such as personality psychology and sociology? A BRIEF HISTORY OF SOCIAL PSYCHOLOGY The beginning The early years The years of expansion The crisis years Overcoming the crisis SOCIAL PSYCHOLOGY IN EUROPE SOCIAL PSYCHOLOGY TODAY ROUTE MAP OF THE CHAPTER Most textbooks introduce social psychology with examples of everyday experiences of social behaviour or even with a formal definition. We thought that a better way of familiarizing you with our discipline was to present some examples of classic studies. These should give you an impression of the research questions social psychologists address and of the methods they use to tackle these questions. Only then do we present a formal definition of social psychology and discuss the differences between social psychology and related areas. The second part of the chapter is devoted to the history of social psychology, which we trace from its starting years around 1900 until today. As our American colleagues like to point out, much of this history took place in the US. However, as we Europeans like to point out, this development was strongly influenced by European researchers, even before the establishment of social psychology in Europe during the last four decades.  INTRODUCTION: SOME CLASSIC STUDIES How do social psychologists go about addressing research questions? In 1954, Muzafer Sherif (see the Leader in the Field, Muzafer Sherif, in Chapter 14), who was then professor of social psychology at the University of Oklahoma (US), conducted one of a series of classic studies with 11–12 year-old boys, who had been sent to a remote summer camp at Robbers Cave State Park, Oklahoma. None of the boys knew each other before the study. They were divided into two groups, who stayed in cabins far apart from each other and did not know of each other’s existence. For one week, each of the groups enjoyed the typical summer camp life, engaging in fun activities like camping out, transporting canoes over rough terrain to the water and playing various games. They had a great time. It is therefore not surprising that at the end of the week, group members had grown very fond of one another and the groups had developed strong group identities. Each chose a name for itself (the ‘Rattlers’ and the ‘Eagles’), which they proudly displayed on shirts and flags. At the end of the week, each of the groups was told that there was another group in the vicinity. As though acceding to the boys’ requests, the staff arranged tournaments of games (e.g. touch football, baseball, tug of war) between the groups. The winning team would receive a cup, and members of the winning team would each be given a new penknife. The tournament started in the spirit of good sportsmanship, but as it progressed, hostilities between the groups began to develop. ‘Soon members of each group began to call their rivals “stinkers”, “sneaks” and “cheats”... Near the end of this stage, the members of each group found the other group and its members so distasteful that they expressed strong preferences to have no further contact with them at all’ (Sherif, 1967, p. 82). What was the point of all of this? What can tales about boys in a summer camp tell us about real life? The answer is a great deal. These Robber’s cave studies actually mark a turning point in the study of prejudice (i.e. dislike for members of an outgroup), because they challenged the then dominant view of prejudice as either an outflow of a prejudiced personality disposition (authoritarian personality; see Chapter 14) or as the result of displaced frustration (scapegoat theory). There was no indication that these boys had prejudiced personalities or needed scapegoats to displace their aggression. And yet, they developed strong dislikes for the members of the other group (the ‘stinkers’ and ‘sneaks’), because they were competing with them for some valued good which only one of the two groups could attain. Sherif interpreted these findings as support for his realistic conflict theory, which assumed that intergroup hostility and intergroup prejudice are usually the result of a conflict of interest between groups over valued commodities or opportunities. Goals were the central concept in Sherif’s theory: he argued that when two groups were competing for the same goal, which only one could achieve, there would be intergroup hostility. ---------- scapegoat theory the theory that prejudice is due to aggression displaced towards members of an outgroup (scapegoats), because the group (or circumstances) that was the source of frustration is not within reach. ---------- ---------- realistic conflict theory a theory developed by Sherif that holds that conflict and competition between groups over valued resources can create intergroup hostility and prejudice. ---------- Not surprising, you might say. After all, this is the reason why football supporters beat each other up every so often before and after games between their clubs. And yet, this is not the full story. Nearly two decades later, Henri Tajfel (see Leader in the Field, Henri Tajfel, in Chapter 14), then professor of social psychology at Bristol University (UK), and colleagues conducted a series of studies that called into question the assumption that competitive goals are a necessary condition for the development of intergroup hostility (Tajfel, Billig, Bundy, & Flament, 1971). Participants in these studies were 14- to 15-year-old schoolboys, who all knew each other well and came to the psychology laboratory in groups of eight to participate in an experiment (see Chapter 2) on visual perception. Their task was to estimate the number of dots that were flashed onto a screen. After completion of this task, they were told that they would also participate in a second experiment and, for the ease of coding, would be divided on the basis of the dot estimates they had just made. Half the boys were then (randomly) assigned to the ‘under- estimators’ group, the other half to the ‘over-estimators’ group. (In later studies, boys were often divided on the basis of their alleged preference for paintings by Klee or Kandinsky, an equally irrelevant criterion for boys of that age.) The boys then had to assign rewards to other individuals in real money. They did not know the identity of the other individuals, but only their code numbers and their group membership. This experimental procedure became known as minimal group paradigm. These groups were minimal, because they were created using arbitrary criteria, involved no interaction between members of the two groups, and group members had no knowledge of who belonged to the group. And yet Tajfel could show that members of these groups displayed intergroup discrimination. When asked to divide money between a member of their own group and a member of the other group, most boys gave consistently more money to members of their own group than to members of the other group (see Chapter 14). These studies were again quite innovative, because they showed that intergroup conflict was not an essential cause of intergroup discrimination (or at least ingroup favouritism). Apparently, the mere fact of division into groups was sufficient to trigger discriminatory behaviour. ---------- minimal group paradigm a set of experimental procedures designed to create groups based on essentially arbitrary criteria (with no interaction within or between them, and with no knowledge of who else belongs to each group) whose members show intergroup discrimination. ---------- You might now believe that you have some idea of what social psychology is all about and how social psychologists conduct their research. You might also think that the approach of Sherif was more in line with what you had expected, but that the studies by Tajfel, despite their artificiality, led to some interesting results. However, you will be somewhat premature in your confidence. A clearer and more appropriate picture of the field of social psychology will emerge after considering some additional studies, described below. In 1994, Neil Macrae (then at Cardiff University) and colleagues studied people’s ability to suppress their prejudicial thoughts (Macrae, Bodenhausen, Milne, & Jetten, 1994). After all, there is a great deal of evidence that people acquire their prejudices quite early and may not be able to get rid of them later in life, even if these prejudicial thoughts have become inconsistent with their egalitarian values (Wilson, Lindsey, & Schooler, 2000). Thus, if people cannot forget their prejudicial thoughts, it would be good if, at least, they could inhibit them and prevent them from affecting their actions. As the studies by Macrae et al. (1994; see Chapter 4) show, this may be more difficult than one would think. Participants in these studies were students. When they arrived at the laboratory, they were told that they were to participate in an investigation of people’s ability to construct life event details from visual information. They were then presented with a colour photograph of a skinhead and were asked to write a short essay about a typical day in the life of a skinhead (Figure 1.1). Skinheads were chosen here not only because there is widespread prejudice against them, but also because, unlike prejudice towards other minority groups, expressing prejudice towards skinheads is not (yet) politically incorrect. Half of the participants were asked to suppress their prejudice against skinheads in writing this essay. They were told to try to write their essay without being influenced by their stereotypes about skinheads – that is, the beliefs they might have about the characteristics of skinheads in general. The other half (i.e. the control group) were not given this instruction. FIGURE 1.1 How easy is it for people to suppress their prejudice towards skinheads? Source: © BROOKS WALKER. Used under licence from Getty Images.  After the participants had finished the first essay, they were given a photo of another skinhead and asked to write another essay. This time, however, they were not given any instructions about suppressing stereotypes. Both essays were then rated by independent raters, who did not know whether a given essay had been written by a participant from either the experimental or the control group and who evaluated the extent to which writers expressed stereotypes about skinheads. With regard to the first essay, results were not very surprising. As one would expect of ‘good’ (i.e. obedient) participants, individuals who had been instructed to suppress their stereotypes in their first essay did so quite successfully. Their essays were much less stereotypic than the essays of the control group. However, the analysis of their second essays provided a striking finding: there was a rebound effect (see Chapter 4). The second essay of these ‘suppressors’ was more stereotypic than that of the control group. Thus, when people no longer tried to suppress their stereotypes, they showed a higher level of stereotypical thinking than if they had never tried to suppress their thoughts in the first place. Although these are fascinating results, Macrae and his colleagues were not satisfied with merely showing a rebound effect of stereotype suppression on thinking (see Chapter 4): they also wanted to know whether attempts to suppress one’s stereotype would affect people’s behaviour. They therefore conducted a second study. The first part of this study was identical to that of their first experiment. However, after having written an essay under either stereotype suppression or no-suppression instructions, participants were told that they would now go next door to meet the person depicted in the photograph (i.e. the skinhead). When they entered the room next door, there was a row of chairs standing next to each other, but no skinhead. However, on the first chair there was a denim jacket and bag. The experimenter told the participant that the other person must just have gone to the toilet and would return shortly and that the participant should sit down on one of the chairs in the meantime. The measure of interest in this case was the seating position, that is, how far the participant would choose to sit away from the skinhead he or she was supposed to meet. We would all acknowledge that the distance we keep from someone is an indication of our liking for that person (Macrae et al., 1994). And in line with the findings of the previous study, participants who had (successfully) suppressed their stereotype on writing the essay now chose a chair that was significantly further away from the skinhead than did individuals in the control group. Thus, the rebound effect of stereotype suppression affected not only thoughts, but also behaviour (but for some constraints on the general effect, see Monteith, Sherman, & Devine, 1998). As surprising as these findings were, the impact of stereotypes on behaviour was still restricted to the way the individual behaved towards a member of the group towards whom the stereotype was held. As we will see in the next experiment, the impact of stereotypes can be even more pervasive. This study was conducted by John Bargh (see Leader in the Field, John Bargh, in Chapter 4) and his colleagues (Bargh, Chen, & Burrows, 1996) at New York University (US). In the first part of this experiment, participants had to complete a ‘scrambled sentence test’ in which they had to form sentences from scrambled sets of words. For participants in the experimental group, these sentences contained words that were part of the (American) stereotype of the elderly, such as ‘Florida’, ‘Bingo’ and ‘grey’. This procedure is known as ‘priming’ (see Chapter 4), because these words will bring the elderly stereotype to participants’ minds (i.e. make it more accessible), including characteristics of elderly people that were not even mentioned in the priming procedure. One such characteristic that is typically attributed to the elderly, but which was not mentioned in the priming procedure, is that elderly people move rather slowly. The researchers assumed that participants who were primed with the stereotype of the elderly would also think of ‘moving slowly’ as another salient characteristic of the elderly. It was further assumed that this thought would affect the participants’ own behaviour. The researchers predicted that participants primed with the elderly stereotype would move more slowly than participants in the control condition who had been exposed to neutral primes. The experimenters then measured the time it took participants to walk from the experimental room to the nearest lift. In line with the hypothesis, participants who were primed with the elderly stereotype took significantly longer to reach the lift than did participants who had been primed with neutral words. It appears that thinking of the concept ‘slow’ influenced behaviour, and that consciousness did not play any part in this process, because participants were aware neither that they had been primed nor that they had been led to walk more slowly (see Research Close- Up 4.1 in Chapter 4). We hope that reading about these studies has stimulated your interest in social psychology. If it has, you can read more about the first two studies in Chapter 14 (Prejudice and Intergroup Relations). The last two studies are discussed in Chapter 4 (Social Cognition). Given that the research we have discussed so far is quite varied in its research questions, scope and methods, we now turn to a more general discussion of the nature of social psychology.  WHAT IS SOCIAL PSYCHOLOGY? How do social psychologists define their discipline? When social psychologists are called upon to define their discipline, they usually refer to the definition given by Gordon Allport (1954a) (see Leader in the Field, Gordon Allport, in Chapter 14) in his classic chapter on the history of social psychology, published in the second edition of the Handbook of Social Psychology: ‘Social psychology is the attempt to understand and explain how the thoughts, feelings, and behaviours of individuals are influenced by the actual, imagined, or implied presence of other human beings’ (p. 5). With ‘imagined presence’ Allport referred to the influence of reference persons (e.g. our parents) whose expectations might influence our behaviour. With the ‘implied presence’ he acknowledged the fact that much of our behaviour is shaped by social roles and cultural norms. This is quite a good definition, which can accommodate the studies that we have described earlier. One characteristic of social psychology, which Allport implied but did not mention specifically, is the use of scientific methods. The scientific method of choice used in the studies we have just described was the experiment. We will discuss this method only briefly, because you will learn more about the experimental method in the chapter on methods (Chapter 2). Experiments are a method in which the researcher deliberately introduces some change into a setting to examine the consequences of that change. The typical procedure used in experiments is that conditions in which a change has been introduced (i.e. an independent variable manipulated) are compared to conditions in which this has not been the case, the so-called control group. By randomly assigning participants to either experimental or control group, the researcher can be reasonably certain that any difference between the two groups was due to the manipulation of the independent variable. Thus, Macrae and colleagues asked half their participants to suppress their stereotype of skinheads, and compared their thoughts and behaviour to those of a control group of individuals who had not been asked to suppress their stereotype. Bargh and colleagues compared the walking speed of participants who had been primed with the elderly stereotype with that of (control) participants who had not been primed. The study by Sherif is somewhat deficient in this respect, because he did not really have a proper control group. He compared the impact of the introduction of intergroup competition on group members’ behaviour over time. The control conditions in the Tajfel experiment are difficult to explain without a more detailed description of the study. You may remember that Tajfel and colleagues assessed how the boys would divide money between a member of their own group and a member of the other group. As a control for ingroup bias, they simply reversed the alleged group membership of the two individuals between whom the money had to be divided. Another methodological difference between the study by Sherif and those of the other researchers is that Sherif’s study was a field experiment rather than a laboratory experiment: he used a natural setting (summer camp) to test his hypotheses. The other studies were all laboratory experiments which used settings that were specially created by the experimenter. For example, Macrae and colleagues led their participants to believe that they were in a study of people’s ability to construct life event details from visual information. This is also an example of a darker aspect of social psychology, namely, that we often have to use deception to test our predictions. But if the participants in the study by Macrae and colleagues (1994) had known the real purpose of the study, this would have influenced their thoughts and behaviour and the results of such a study would have been meaningless. (We therefore often disregard the data of participants who guess the purpose of our experiments.) Field and laboratory experiments are not the only scientific methods used by social psychologists to test their hypotheses. You can read about other methods in Chapter 2 (Research Methods in Social Psychology). ---------- field experiment a true randomized experiment conducted in a natural setting. ----------  ---------- laboratory experiment a study, conducted in the laboratory, in which the researcher deliberately introduces some change into a setting, while holding all other factors constant, to examine the consequences of that change. ---------- Obviously, the use of scientific methods is not a characteristic that allows one to distinguish social psychology from other social sciences, as by definition all social sciences use methods they consider scientific, and for many of them, experiments are the method of choice. A more distinctive characteristic introduced by Allport is the fact that social psychology is concerned with social influence, and that it studies the impact of others on individuals’ thoughts, feelings and behaviours. All the studies we described earlier tried to understand and explain how the thoughts, feelings and behaviours of their participants were influenced by the presence of other human beings. In the case of the study by Sherif, these human beings were mainly the members of the other group with whom the boys competed, although the members of their own groups also influenced the behaviour of these boys. In contrast to the Sherif study, where the others were actually present, the presence of others was imagined rather than real in the Tajfel study (recall that Allport’s careful definition allowed for the impact of the imagined presence of others). Finally, in the studies by Macrae and by Bargh and colleagues, it was not really the presence of others that influenced participants’ thoughts or behaviour, but the suppression or activation of their beliefs about other groups. The studies by Macrae and Bargh are also good examples of an aspect of social psychological research that is less clearly emphasized in Allport’s definition, namely, the fact that we are interested not only in the impact others have on our thoughts, feelings and behaviour, but also in the cognitive processes by which our thoughts, emotions and goals guide our understanding of the world around us and our actions. You can read more about this in Chapter 4 (Social Cognition). A final characteristic of social psychology emphasized in Allport’s definition is that social psychologists study the impact that the implied or actual presence of others has on the thoughts, feelings and behaviours of individuals. Thus, even when we study social groups, we examine the impact groups have on the individual group members. For example, in the classic study of conformity with group majorities, Asch (1956) examined the impact of the majority opinion on the judgements of individual participants (see Chapter 8). Similarly, Tajfel and colleagues (1971) studied the impact of the mere categorization of others into ingroup and outgroup on the way individuals distributed money between them. This emphasis on the individual is actually a very important point which had already been made by the elder brother of Gordon Allport, Floyd Allport, in his classic textbook of social psychology: ‘There is no psychology of groups which is not essentially and entirely a psychology of individuals. Social psychology must not be placed in contradistinction to the psychology of the individual; it is a part of the psychology of the individual, whose behaviour it studies in relation to that sector of his environment comprised by his fellows’ (F. Allport, 1924, p. 4). The emphasis on the individual does not deny the importance of the social context as a determinant of individual behaviour, but it rejects the existence of a group consciousness or a collective mind as separate from the minds of the individuals who comprise the group.  HOW DOES SOCIAL PSYCHOLOGY DIFFER FROM OTHER DISCIPLINES? What differentiates social psychology from related disciplines such as personality psychology and sociology? In addition to using examples of studies as well as a definition to illuminate the nature of social psychology, it might be helpful to contrast social psychological research to that of research in related disciplines. As in the previous section, we will use the example of an experimental study to clarify these differences. This study was conducted at a small elite college in the United States and was announced as an experiment on perception. The experimental sessions were held in a small classroom and eight participants attended each of the sessions. The participants, who were seated in two rows of four, were presented with sets of four lines of different length, a standard line and three comparison lines. Their task consisted of the comparison of the standard line with the three other lines, one of which was equal to the standard line. The comparison lines were numbered from 1 to 3, and the participants stated their judgements by calling out one of the numbers (see Figure 1.2). FIGURE 1.2 Example of the stimulus pattern used in the conformity studies of Asch (1955). Source: Original data from Asch (1955), Scientific American, 193, 31–35.  Obviously, this was a simple experiment in visual discrimination in which the experimenter probably wanted to find out how accurately participants could differentiate between lines of different lengths and where the threshold lay at which people would begin to make mistakes. However, there is one feature of the experiment which does not fit with standard procedures in perception experiments – namely that participants judged these lines in groups. This would not have been a problem had the experimenter ensured that judgements were written down, to exclude the possibility that participants would be aware of each others’ judgements. But in the present experiment, participants were asked to call out their judgements to the experimenter. This appears to be a serious methodological fault. Any determination of a difference threshold based on such data would be flawed, because participants might have been influenced by the earlier judgements that they overheard. Let us assume that the first participant calling out his judgements committed an error. The second participant, who might normally have given a correct response, might now have become uncertain and given the same erroneous response as the first participant. In this way, an experiment on perception might, in fact, have become a study of social influence. Since we are concerned here with social psychology, it will not come as a surprise that the experimenter, a professor of social psychology at Swarthmore College, Pennsylvania, was not really interested in measuring perceptual thresholds, but in the extent to which individuals would be influenced by a discrepant majority judgement. In fact, only one of the eight participants in each session was a ‘naïve’ participant; all the others were confederates of the experimenter and were instructed to give unanimous but wrong judgements on 12 out of the 18 trials. These judgements were so easy that participants who judged the stimuli in individual sessions made practically no incorrect judgements. And yet, when participants were exposed to the incorrect judgements of a unanimous majority, 36.8 per cent of their judgements were incorrect (Asch, 1955). With this experimental setting, Asch created a situation that is familiar to most of us from everyday life. We have probably all had the experience of members of our group disagreeing with us on some issue, then having to decide whether we should go with the group or stick to our own position at the risk of becoming disliked or of looking foolish. Naturally, we do not usually disagree about the length of lines, but about some issue of greater importance, and often the disagreeing majority is not unanimous. However, the setting that was developed by Asch would allow us to manipulate all these variables, and most of them have indeed been investigated in subsequent research (for a review, see Allen, 1966). Our decision to stick to our guns or go with the group will depend very much on how confident we are of the correctness of our own opinion, on how important a correct decision is for us and for the group, and on how well we know the other group members. We are probably also more willing to conform to a majority if we are confronted with a majority that is unanimous rather than divided. If we return to Gordon Allport’s definition of social psychology, it is easy to see that the Asch experiment fits all of the characteristics: Asch used a laboratory experiment to study the social influence which a (false) majority judgement would have on the thoughts and behaviours (i.e. judgements) of individuals. The Asch experiment also allows us to demonstrate the difference between social psychology and ‘asocial’ general psychology. If Asch had been interested in studying perceptual thresholds, he would have varied the difference in the lengths of his standard and comparison stimuli systematically, to assess the extent to which such variations affected perceptual judgements. The (perceptual) judgements would have remained the same, but they would now be investigated in relation to variations in the physical aspects of the stimuli, while keeping the social context constant. In contrast, Asch kept the physical stimulus constellation relatively constant and was interested in the effect that varying the social context (i.e. majority size and unanimity) had on perceptual judgements. The Asch situation is also useful for demonstrating the difference between social and personality psychology. As a social psychologist, Asch was interested in the impact that characteristics of the social situation had on the thoughts and behaviours of his participants. Does the rate of conformity increase if we increase the number of majority members who give erroneous judgements? Does the conformity rate decrease if participants are allowed to give their judgements anonymously? Asch’s approach is typical of social psychological research, which usually manipulates important aspects of the social context in order to assess the impact these changes have on the thoughts, feelings and behaviour of the target person. Personality psychologists, on the other hand, might be less interested in the impact of the social context on behaviour and, instead, ask themselves why some participants are influenced by the erroneous judgements of the majority while others remain unaffected. Thus, the personality psychologist would be interested in the personality traits that are responsible for the fact that different individuals act differently in what is essentially the same social situation. The personality psychologist might test whether intelligent individuals are less likely than unintelligent ones to conform to majorities, or whether conformity is more prevalent among authoritarian rather than non- authoritarian personalities (see the discussion of the authoritarian personality in Chapter 14; Adorno, Frenkel-Brunswik, Levinson, & Sanford, 1950). Personality psychologists would not, however, only address the question of individual differences as determinants of conformity; they would also want to know how these individual differences came about. Is it possible to relate differences in authoritarianism to differences in the way parents brought up their children, and what aspects of a person’s upbringing determine his or her self-esteem? Thus, one could try to separate the disciplines of social and personality psychology as follows: individual behaviour is determined by three factors: (1) the biological constitution of individuals, (2) their acquired traits, and (3) the social and physical context. Whereas personality psychologists are mainly interested in studying how particular traits are acquired and how these traits influence the individual’s behaviour, social psychologists study the impact of the social situation on individual behaviour. Unfortunately, such a distinction would oversimplify the differences between social and personality psychology (for more details, see Krahé, 1992) because one of the central concepts of social psychology, namely social attitudes, is defined by many social psychologists (e.g. Eagly & Chaiken, 1993) as a tendency (i.e. individual disposition) to evaluate an attitude object positively or negatively (see Chapter 6). Even though social psychologists are mainly interested in studying how attitudes change in response to social influence attempts (see Chapters 7 and 8), they also use attitudes to predict individual behaviour (see Chapter 6). Furthermore, within social psychology, researchers have often been interested in studying individual difference variables, such as the degree to which individuals are prone to prejudice and susceptible to Fascist ideologies (‘authoritarianism’; Adorno et al., 1950; see Chapter 14), or the degree to which individuals are oriented to situational cues or reactions of others (‘self-monitoring’; Snyder, 1974). Since there is a great deal of agreement that individual behaviour is influenced by personality traits (see Chapter 9 on aggression) as well as the social context, the two fields of personality psychology and social psychology are, in fact, difficult to separate. It is therefore not surprising that the leading social psychological journal is the Journal of Personality and Social Psychology, and that most American social psychologists are members of the Society of Personality and Social Psychology. However, there are subtle differences in focus. Social psychologists are typically interested in personality variables as moderators. They look for the extent to which the impact of an independent variable on a dependent variable is qualified by, or depends on, the level of an individual’s score on a personality measure. For example, there is a higher correlation between attitudes and behaviour for ‘low’ than for ‘high’ self-monitors (Snyder & Kendzierski, 1982). Many of the chapters in this volume refer to such personality influences on social behaviour. Social psychologists also tend to emphasize that the impact of personality variables on social behaviour is weaker in ‘strong’ compared to ‘weak’ social situations (Mischel, 1977). Thus, social psychologists emphasize the power of strong social situations to relegate personality influences to the background. This occurs, for example, in experiments investigating helping in emergencies (Latané & Darley, 1976; see Chapter 10) and obeying an authority figure’s orders to behave in immoral ways (Milgram, 1974; see Chapter 8). After the difficulties we experienced in distinguishing social psychology from personality psychology, distinguishing it from neighbouring social sciences such as sociology might seem easy. It would appear that sociology differs from social psychology both in the issues it studies and in the level of analysis at which it addresses these issues. Unfortunately, things are again not that simple. First, there is quite a bit of overlap between the issues studied by social psychologists and those that interest sociologists. Thus, social groups and group norms are topics that are of equal interest to sociologists and social psychologists (see Chapter 12). The sociologist George Homans wrote one of the classic monographs on social groups (Homans, 1950) and the sociologists Hechter and Opp (2001) edited a volume that summarizes the important work of sociologists in the area of social norms. Although there are sociological approaches that, influenced by the work of Talcott Parsons and Emile Durkheim, emphasize that sociological facts should not be explained through psychological processes (Vanberg, 1975), most sociologists would no longer accept this position. In fact, sociologists have made major contributions to the development of individualistic social psychological theories. Thus, the sociologists Homans (1961) and Blau (1964) have written monographs on exchange theory, a theory that has become central in social psychology through the classic Social Psychology of Groups written by the social psychologists Thibaut and Kelley (1959), but is now more frequently referred to as interdependence theory. The central tenet of this theory is that individuals interact with those others who provide the greatest rewards for the least costs (see Chapter 11). Thus, most sociologists agree with social psychologists in espousing what has been called methodological individualism, namely the assumption that even collective behaviour is essentially behaviour of the individuals who form the collective and therefore has to be explained in terms of rewards and costs of this behaviour to the individual (e.g. Klandermans, 1997). ----------  methodological individualism the assumption that collective action must be explained by showing how it results from individual decisions and behaviour; collective behaviour is seen as essentially behaviour of the individuals who form the collective. ---------- Even though there is a great deal of overlap between sociology and social psychology, there are also major differences in the way these areas approach social behaviour. Sociologists are more likely to trace social behaviour upwards to structural variables such as norms, roles or social class, whereas social psychologists will trace it downwards to the individual’s goals, motives and cognitions. For example, both sociologists and social psychologists are interested in aggression and violence. Social psychologists have studied the cognitive and affective processes through which anger can, given the right contextual cues, explode in aggressive behaviour – that is, behaviour performed with the express intention of hurting another person (Figure 1.3; Chapter 9). Sociologists, on the other hand, have been more interested in why levels of aggression are higher in some societies or groups than in others. Why is the murder rate in the US so much higher than in Canada, even though guns are widely available in both countries? Since a possible difference could be the type of guns that are available in the two countries, with hunting rifles being more prevalent in Canada and hand guns or assault weapons more frequently held in the United States, the potential answer might lie in the aggressive images that will be activated by different types of weapons, leading us back to individual psychological processes. Thus, even though sociologists are more likely to link individual behaviour to social structural variables, while social psychologists are more likely to study individual processes, a combination of the two approaches might often provide a fuller explanation than either discipline can offer on its own. FIGURE 1.3 How is the way social psychologists study aggression and violence different from the approach of sociologists? Source: © Hurricane. Used under licence from Shutterstock.  A BRIEF HISTORY OF SOCIAL PSYCHOLOGY The beginning Who conducted the first experiment and who wrote the first textbook? Authors who write about the history of a scientific discipline usually like to report dates that mark the origin of that discipline. Often these are the years in which the first textbooks or handbooks bearing the name of the discipline were published. In social psychology, 1908 is usually noted as the year when the first two textbooks of social psychology were published, one by a sociologist (Ross, 1908), the other by a psychologist (McDougall, 1908). However, since both texts cover very little material that we would consider social psychological these days, 1908 may not be the best choice for the birth year of social psychology. One could also argue that using the date of the first textbook to mark the beginning of a discipline is questionable anyway, because it would be difficult to write a textbook about a discipline that did not already exist. There must first be relevant theorizing and research available with which to fill the pages of a textbook. It is probably for this reason that another date has become quite prominent in chapters on the history of social psychology, namely the date of (presumably) the first social psychological experiment, a study published in 1898 by Norman Triplett. Triplett appeared to have been a fan of bicycle races. He was interested in the phenomenon whereby racing cyclists go faster when racing with others or when being paced than when riding alone, racing against the clock (Figure 1.4). Illustrating the research tool of ‘archival analysis’ (see Chapter 2), Triplett (1898) used records of the average speed of cyclists under these different conditions; he could indeed demonstrate that cyclists ride faster in competition or with pacers than when riding alone. However, as Triplett recognized, the shortcoming of this kind of quasi-experimental evidence is that different racers participate in different kinds of races and select those in which they do particularly well. The differences in speed between cyclists racing against the clock or racing in competition could therefore have been due to self-selection. To rule out this explanation, Triplett (1898) conducted an experiment in which schoolchildren performed a simple task (turning a fishing reel) either alone or in competition with another participant (see Research Close-Up 1.1). The experiment is usually cited as demonstrating the effects of what later became known as social facilitation, the phenomenon whereby the performance of simple tasks is facilitated by the presence of an audience or of others working on the same task (see Chapter 8). However, if you read the Research Close-Up, you will realize that his data do not warrant such a strong conclusion. FIGURE 1.4 It is a well known phenomenon that racing cyclists go faster when racing with others or when being paced than when riding alone, racing against the clock. Source: © AISPIX. Used under licence from Shutterstock. ---------- social facilitation an improvement in the performance of well-learned/easy tasks and a worsening of performance of poorly-learned/difficult tasks due to the presence of members of the same species. ----------  Although the study by Triplett (1898) had the elegance and clarity that became the hallmark of experimentation in social psychology, its historical significance has been challenged by scholars who doubted whether it really was the first social psychological experiment. For example, Haines and Vaughan (1979) have argued that there were other experiments before 1898 deserving to be called social psychological, such as studies on suggestibility by Binet and Henri (1894; see Stroebe, 2012). But social psychological experiments may have been performed even earlier by the French agricultural engineer Max Ringelmann, who between 1882 and 1887 conducted investigations into the maximum performance of workers pulling a load under different conditions (Kravitz & Martin, 1986). Although the comparison of individual and group performance was of only secondary interest to Ringelmann, he found the first evidence of productivity loss in groups, a phenomenon that was later named social loafing (see Chapter 13). Ringelmann found that eight men who pull at a rope together achieve only about 50 per cent of the pulling power that could be expected on the basis of their pulling measured individually. However, since Ringelmann only published this research in 1913, Triplett predates him, certainly as far as publication is concerned. RESEARCH CLOSE-UP 1.1 TRIPLETT’S CLASSIC STUDY OF SOCIAL FACILITATION EFFECTS Triplett, N. (1898). The dynamogenic factors in pacemaking and competition. American Journal of Psychology, 9, 507–533. Introduction Triplett suggested several theoretical explanations for the superiority of cyclists who are paced or ride in a competition over those who ride alone. For example, there could be aerodynamic or psychological advantages for a racer who was paced or racing in a competition. However, Triplett was most interested in what he called ‘dynamogenic factors‘, namely that ‘the presence of another rider is a stimulus to the racer in arousing the competitive instinct; that another can thus be the means of releasing or freeing nervous energy from him that he cannot of himself release’ (p. 516). His concept of nervous energy appears to be similar to the modern concept of arousal. Method Participants Forty schoolchildren (14 boys, 26 girls) from 8 to 17 years participated in the study. Design and procedure The apparatus consisted of two fishing reels fixed to a wooden frame and placed sufficiently far apart so that two people could turn them side-by-side. A cord was looped from each of the fishing reels through corresponding wheels fixed to the opposite end of the frame, creating two closed circuits (like the circuit formed by the chain on a bicycle). A small white flag was sewn to each of the cords. By winding the fishing reel the flag traversed the course from the fishing  reel to the wheel at the opposite end and back again. A trial consisted of four circuits of the course and the time taken for this performance was measured by stopwatch. After an initial practice session, each participant performed six trials. Triplett created a somewhat complicated but quite ingenious design, with the sequence of alone and competition trials differing for Groups A and B in a systematic way (see Table 1.1). Table 1.1 Sequence of alone and competition trials for Groups A and B in Triplett’s design. The rationale behind these two sequences was that it eliminated practice as well as fatigue effects. Both effects are related to the number of trials performed by a participant (i.e. trial order). One would expect that during the first few trials performance improves due to practice. In later trials fatigue might set in and begin to slow down participants. Triplett controlled for both of these effects from trial 2 onwards by having the task performed under both conditions during each of the trials. Results Triplett categorized his participants into three groups on the basis of their performance during the various trials: those who were stimulated positively by competition and performed faster (N = 20), those who were ‘overstimulated’ and performed more slowly (N = 10), and those who were not affected (N = 10). From visual inspection of his data, Triplett concluded that his findings showed ‘that the bodily presence of another contestant participating simultaneously in the race serves to liberate latent energy not ordinarily available’ (p. 533). This conclusion formed the basis of the myth reported in most textbooks that his study provided the first evidence for social facilitation. These days conclusions based on visual inspection of data are no longer accepted and journal editors require that researchers conduct (often extensive) statistical analysis of their results, and test how likely the results are to be due to chance. When Strube (2005) reanalysed Triplett’s data more than a century after the publication of the original  study using modern statistics, he found very little support for the original conclusion. Only one of several analyses resulted in a significant effect. This effect disappeared, however, when Strube eliminated the data of two left-handed participants, for whom the task would have been more difficult, as they were instructed to turn the wheel with their right hand. As you will learn in Chapter 8, the presence of others facilitates performance only for easy tasks, but inhibits performance with difficult tasks. Elimination of the left- handed participants should therefore have strengthened rather than weakened the social facilitation effects. Strube concluded that ‘the analysis of Triplett’s data... indicate barely a statistical hint of the social facilitation of performance to which his experiment has been credited’ (2005, p. 280). This is quite different from the conclusions usually drawn in textbooks of social psychology. Discussion If significance tests had already been available and required by journal editors in 1898, Triplett’s study might never have been published and historians of social psychology would have had to look elsewhere for the first experiment (Stroebe, 2012; Strube, 2005). It is nevertheless puzzling that Triplett failed to find clear evidence of social facilitation. After all, turning a fishing reel is a simple mechanical task and competition should clearly have facilitated performance. It is interesting to note that these early experiments were studies of an applied nature in areas which later became known as sports psychology and psychology of work performance. There were other studies of this nature available in other applied areas (e.g. Mayer, 1903; Moede, 1920) and it needed somebody to recognize that the study of the impact of the social context on performance was really a discipline by itself, namely, social psychology. It may then be justifiable to choose the date of the first textbook or handbook about a discipline as its ‘origin’ insofar as a discipline is characterized not only by its content, but also by its disciplinary identity. Thus, it is not sufficient that research that is vaguely social psychological has been conducted in the area of sports psychology or even agriculture. There needs to be somebody who pulls all this research together and declares the emergence of a new area (in which, incidentally, this applied research then becomes fundamental). In our view, this was first achieved by Floyd Allport (1924), who in his textbook made several major contributions towards defining the field of social psychology (see Leader in the Field, Floyd Henry Allport). He declared the study of social behaviour as the subject of social psychology. He defined social behaviour as ‘behavior in which the responses either serve as social stimuli or are evoked by social stimuli’ (p. 148). As mentioned above, he postulated that social psychology ‘is part of the psychology of the individual, whose behavior it studies in relation to that sector of the environment comprised by his fellows’ (p. 4). He had noted earlier, in the same volume, ‘For... only within the individual can we find the behavioral mechanisms and the consciousness which are fundamental in the interactions between individuals’ (p. vi). Another contribution, which may be less embraced today, was his emphasis on the experimental method. Although the experimental method is still one of the major research tools of social psychologists, other research methods have become equally accepted these days. However, in Allport’s time, the emphasis on experiments was probably essential for establishing the scientific respectability of social psychology. It would also have helped to distinguish it further from sociology, a discipline that still prefers surveys and field studies to conducting experiments. It is interesting, though, that with the exception of his chapter on the ‘Response to social stimulation in groups’, Allport (1924) did not review a great deal of experimental evidence of a social psychological nature.  LEADER IN THE FIELD Floyd Henry Allport (1890–1978), the elder brother of Gordon Allport (see Leader in the Field, Chapter 14), received both his undergraduate degree (1914) and his PhD (1919) from Harvard University. His dissertation was based on his studies on social facilitation, a research area that had been suggested to him by Hugo Münsterberg, then professor of psychology at Harvard. In 1922 Allport obtained a position as Associate Professor of Social Psychology at the University of North Carolina at Chapel Hill. It was here that he began writing his Social Psychology, which was widely praised and adopted as a text. This book and his studies on the impact of the group on individual cognitive performance are his major contribution to social psychology. He retired from Syracuse University in 1956. Allport’s conception of social psychology derived from his research on social facilitation (see Chapter 8). His experimental paradigm, which had the defining characteristics that he attributed to all social psychology research, had been developed ‘in Germany by August Mayer, Meumann, Moede and others... prior to 1915’; Allport, 1919, p. 304). This strong German influence on his dissertation research is no coincidence, given that his dissertation supervisor was Hugo Münsterberg, a German psychologist who was then head of the psychology department at Harvard. Allport (1924, p. vii) acknowledges Münsterberg’s influence in the preface to his book, where he writes ‘For the origins of my interest in social psychology I am indebted to the memory of Hugo Münsterberg. It was he who suggested the setting for my first experiment and who foresaw many of the possibilities which have been developed in this book.’  The early years What were the key contributions to social psychology during the first half of the twentieth century? It would be an exaggeration to claim that the publication of Floyd Allport’s textbook stimulated an exponential growth in social psychological research immediately. In fact, not that many milestones are to be reported for the period before World War II. A rather doubtful one is the publication of the first Handbook of Social Psychology by Carl Murchison (1935). We call it doubtful because this handbook covers many topics which nobody would consider social psychological today, such as ‘Population behavior of bacteria’ or the ‘Social history of the yellow man’. There are really only three chapters included in this volume that are truly social psychological in the sense that they would still be included in social psychology handbooks today: the chapter by Gordon Allport on attitudes, that by Dashiell on ‘Experimental studies of the influence of social situation on the behaviour of individual human adults’, and Cox Miles’s chapter on gender differences. But there are also other chapters that could be part of a social psychology curriculum, even though they are not included in most modern handbooks. For example, there is a chapter on language by Esper, on age in human society by Miles, on material culture by Wissler, and on the physical environment by Shelford. Three other significant events during this early period were the publication by Thurstone (1928) of a paper with the provocative title ‘Attitudes can be measured’, The Psychology of Social Norms by Sherif in 1936, and Newcomb’s (1943) Personality and Social Change, a study of attitude formation in the student community of Bennington College (see Leader in the Field, Theodore Newcomb). Thurstone’s article was remarkable because he described the first psychometrically sound method for the measurement of attitudes. Sherif’s study became a classic, because he devised an experimental paradigm that allowed him to study the development of group norms in a laboratory situation (see Chapter 8). Participants in his study were repeatedly exposed to a stationary light source in a darkened room. Sherif made use of the fact that participants perceive this light source as moving (autokinetic effect) and that, if asked to judge the movement over repeated trials, they establish relatively stable individual norms. By putting individuals who had developed widely differing individual estimates into a group situation, Sherif could demonstrate that individuals in groups develop a joint and stable group norm, which they then maintain even when they continue to make their estimates again in individual situations. ---------- autokinetic effect perceptual illusion, whereby, in the absence of reference points, a stationary light appears to move. ----------  LEADER IN THE FIELD Theodore Newcomb (1903–1984) received his undergraduate degree from Oberlin College in 1924 and entered Union Theological Seminary in New York intending to become a Christian Missionary (Converse, 1994). However, more attracted by the psychology courses taught at Columbia University across the road, he switched to psychology and received his PhD from Columbia in 1929. He joined Bennington College in 1934, a newly founded women’s college that drew its students from the politically conservative ‘upper crust’ of Vermont’s society, but had a famously liberal atmosphere. His Bennington study of the change in attitudes that these young women underwent during their studies became a classic. Not only was the longitudinal design innovative at that time, but the study captured the interplay between individual and group processes and thus supported one of the central assumptions of social psychology. After a stint of wartime research, he became director of a joint doctoral program of the departments of sociology and social psychology at the University of Michigan, where he stayed for the remainder of his career. Intrigued by the work of Fritz Heider (see Leader in the Field, Chapter 3), Newcomb developed his own interpersonal version of balance theory. Finally, Newcomb’s Bennington study became a classic, because it is an ingenious longitudinal field study of social influence on a college campus. It maps out the way in which the political attitudes of students, all women who came from conservative homes, changed over time towards the liberal attitudes that were predominant on this college campus. Thus, it illustrates how individual beliefs and attitudes can be shaped by the group context, and thus supports one of the basic assumptions of social psychology. The study is particularly interesting because these students were followed up for 50 years, allowing researchers to demonstrate the stability of their attitude change over a lifetime (Alwin, Cohen, & Newcomb, 1991).  ---------- Bennington study a longitudinal field study of social influence showing how political attitudes of initially conservative female students changed over time towards the liberal attitudes that were predominant on this college campus. ----------  The years of expansion How did Adolf Hitler inadvertently further the development of social psychology in the US? Somewhat tongue in cheek, Cartwright once wrote that the one person who most furthered the development of social psychology in North America was Adolf Hitler (Cartwright, 1979; Figure 1.5). This observation is correct insofar as Hitler’s actions had an important impact on the development of social psychology in the US. World War II greatly stimulated interest in social psychological research. The Information and Education Branch of the US Army initiated surveys and experiments to assess the impact of army propaganda films on the morale of their soldiers. One social psychologist who became heavily involved in this work was Carl Hovland (see Leader in the Field, Carl Iver Hovland). Originally a learning theorist, Hovland became fascinated by the experimental study of the determinants of attitude change. The work he directed during his army years on experiments in mass communication was eventually published as one of the volumes of the American Soldier series under the editorship of the sociologist Stouffer (Hovland, Lumsdaine, & Sheffield, 1949). FIGURE 1.5 How did Hitler’s actions affect the development of social psychology? Source: © Dariush M. Used under licence from Shutterstock.  After the war, Hovland returned to his academic career and founded the Yale Communication and Attitude Change programme. This programme attracted young researchers from a variety of universities and generated a stream of collaborative studies that defined attitude change research for decades to come (see Chapter 7). The programme resulted in the publication of four highly influential volumes on studies of the determinants of persuasion and attitude change. In the first of these volumes, Hovland, Janis, and Kelley (1953) explored the impact of communicator variables (e.g. prestige, credibility and expertise), communication variables (e.g. fear appeals) and context variables (e.g. salience of reference groups). Although the theoretical perspective of the programme was eclectic, Hovland himself was most comfortable with the view that attitude change was a special form of human learning (Jones, 1998).  LEADER IN THE FIELD Carl Iver Hovland (1912–1961) received his bachelor’s and master’s degrees from Northwestern University in 1932 and 1933. He then moved to Yale to work for his PhD under the prominent learning theorist Clark Hull. After finishing his dissertation in 1936, Hovland was invited to join the Yale faculty, of which he remained a member for the rest of his life. Hovland never abandoned his interest in learning theory. Even when he became fascinated by persuasion and attitude change during his wartime leave from Yale in the period from 1942 to 1945, he used learning theory principles as a theoretical perspective. His wartime research was published (with Lumsdaine and Sheffield) in 1949 in Experiments in Mass Communication. After returning to Yale, Hovland established the Yale Communication and Attitude Change programme, which he directed until his premature death in 1961. The research conducted there by Hovland and 30 students and co-workers over a 15-year period established the field of attitude change research as we know it today (Shepard, 1998). A second action of the Hitler regime that advanced the development of social psychology in the US was the forced emigration of Jewish (e.g. Koffka, Lewin, Wertheimer) and even some non-Jewish (e.g. Köhler) academics from Germany. The most important of these émigrés for social psychology was undoubtedly Kurt Lewin, considered by many to be the most charismatic psychologist of his generation (Marrow, 1969). Lewin left the Berlin Psychological Institute in 1933 for the Department of Home Economics at Cornell University, to move in 1935 to the Iowa Child Research Station. In 1945 he established the Research Center for Group Dynamics at the Massachusetts Institute of Technology, which, after his premature death in 1947 at the age of 57 years, was moved to the University of Michigan (see Leader in the Field, Kurt Lewin).  LEADER IN THE FIELD Kurt Lewin (1890–1947) studied psychology and philosophy in Berlin. After fulfilling the formal requirements for a PhD in 1914 (a degree he received only in 1916), he volunteered for the army and spent the next four years fighting World War I (Marrow, 1969). He then returned to the University of Berlin to join the Gestalt psychologists Köhler and Wertheimer at the Institute of Psychology, where he stayed until his (permanent) move to the United States in 1933. The time at Berlin University was probably Lewin’s most productive period. He attracted an international group of students, developed his field theory, which argued that behaviour is a function of both the person and the environment, and supervised a series of classic studies, mainly conducted by his students as part of their dissertation. These studies addressed fundamental issues of the psychology of motivation. Lewin’s interest in social psychology developed only after his move to the United States (Marrow, 1969). In the US he first worked at Cornell University, then moved to the University of Iowa. During his 10 years at the University of Iowa (1935– 1945) Lewin conducted some classic experimental studies in social psychology, such as the experiment on the impact of authoritarian and democratic leadership styles on group atmosphere and performance (Lewin, Lippitt, & White, 1939), which later stimulated research on participative leadership to overcome resistance to change (Coch & French, 1948). Lewin became more and more interested in social processes, and in 1944 he moved to the Massachusetts Institute of Technology, where he founded the Research Center for Group Dynamics. It is difficult to understand today how and why Lewin became such a key figure in social psychology. As is the case today, the impact of a researcher in those days was mainly determined by three factors: (1) a great number of publications in top journals; (2) the development of a theory, which stimulated a great deal of research; or (3) training of a stream of outstanding graduate students, who would later continue the work. Lewin did not score all that well on the first two criteria. Although Lewin’s students were highly productive, he himself published only a few empirical studies in social psychology, the most well known being the study of autocratic and democratic leadership (Lewin et al., 1939), which initiated interest in the impact of leadership styles on group atmosphere and performance (see Chapter 13). His field theory provided a framework for looking at the forces (e.g. positive and negative valences) that influence the individual in a social situation. However, it did not lend itself easily to the derivation of testable hypotheses. Even his own empirical work was only very loosely related to that theory. And as Morton Deutsch (1968), one of Lewin’s most eminent students, concluded two decades after Lewin’s death: ‘It cannot be said that field theory as a specific psychological theory has much current vitality. Nor can it be said that Lewin’s specific theoretical constructs... are central to research now being carried out in social psychology’ (p. 478). ---------- field theory a framework adopted by Kurt Lewin which represented the individual as an element in a larger system of social forces. ---------- So how could Lewin become so influential? As Deutsch (1968) explains, Lewin’s ‘impact is reflected instead in his general orientation to psychology, which has left its impression on his colleagues and students.’ (Deutsch, 1968, p. 478). Lewin believed that psychological events must be explained in psychological terms and that central processes in the ‘life space’ or psychological field of the individual such as cognition, motivation and goals are the proper focus of investigation. This theoretical perspective offered an exciting alternative to the behaviouristic theories that dominated psychology at that time. Furthermore, Lewin’s approach to social psychology had two characteristics which were novel at the time. For him, a problem was only worth studying if addressing it would make a difference with regard to actual problems in the world (Festinger, 1980). Second, and more importantly, he insisted on studying such problems experimentally and on creating in the laboratory powerful situations that made a big difference (Festinger, 1980). Lewin instilled these ideas in his graduate students, and his impact on social psychology was mainly due to these graduate students, who, as a group, were highly influential during the second half of the twentieth century. LEADER IN THE FIELD Leon Festinger (1919–1990) completed his undergraduate studies at City College in New York, and his graduate research at the University of Iowa with Kurt Lewin (see Leader in the Field, Chapter 1). After receiving his PhD in 1942, and a stint of wartime research, he rejoined Lewin and the newly formed Center for Group Dynamics at the Massachusetts Institute of Technology in 1945. In 1948, he moved with the Center to the University of Michigan, from there to the University of Minnesota in 1951, on to Stanford in 1955, and finally, in 1968, to the New School for Social Research in New York, where he stayed until his retirement (Schachter, 1994). During his period at MIT, Festinger, Schachter, and Back (1950) conducted their classic study of friendship patterns and residential proximity (see Chapter 11). Festinger (1950) published his first theoretical paper in social psychology on informal social communication and the process, via social comparison, of establishing the correctness of one’s beliefs. These ideas were later elaborated in his paper on social comparison processes (Festinger, 1954). Soon afterwards, Festinger (1957) published the work for which he is best known, his theory of cognitive dissonance (see Chapter 7). It marked the end of his interest in social psychology, which shifted, first to the visual system and perception, then to archaeology and the history of religion. All these individuals shaped the field of experimental social psychology in the post-war period, but the most illustrious among them was undoubtedly Leon Festinger, whose theory of cognitive dissonance dictated the research agenda in social psychology during the 1960s and 1970s (Festinger, 1957; see also Chapter 7; Leader in the Field, Leon Festinger). The theory of social comparison processes, which he had developed earlier (Festinger, 1954), had less of an immediate impact, but is still influential today (see e.g. Chapters 5, 8, 10 and 12). Another important émigré was the Austrian, Fritz Heider (see Leader in the Field, Fritz Heider, in Chapter 3), although in this case Hitler cannot be blamed for his emigration. Heider came to the US in 1930 to work with Kurt Koffka, who was then at Smith College in Northampton, Massachusetts. He had initially planned to stay for only one year, but decided to remain when he fell in love with Grace Moore, whom he later married. He moved to the University of Kansas in 1947, where he remained until his retirement. His impact on the field is intriguing, because he was not a prolific writer, attracted few graduate students, and published no experimental research in social psychology. Yet he stimulated two of the theoretical traditions which dominated social psychology during the second half of the last century, namely consistency theories and attribution theories. ---------- consistency theories a group of theories (see balance theory, cognitive dissonance theory) proposing that people prefer congruence or consistency among their various cognitions, especially among their beliefs, values, and attitudes. ---------- ---------- attribution theories a group of theories about how individuals manage to infer the ‘causes’ underlying the behaviour of others, or even their own behaviour. ---------- With his paper on balance theory in 1946, Heider developed the notion central to consistency theories that inconsistency between our attitudes and beliefs creates tension in our cognitive system and a tendency to establish consistency. Although only a limited amount of research has been conducted to test Heider’s balance theory, the theory stimulated the development of other consistency theories, most importantly the theory of cognitive dissonance.  ---------- balance theory a cognitive consistency theory that assumes that individuals strive to maintain consistency or balance (objects or persons perceived as belonging together are evaluated similarly) in their social perceptions. ---------- With his paper on phenomenal causality, published in 1944, and his monograph The Psychology of Interpersonal Relations, published in 1958, Heider initiated another important theoretical perspective, namely attribution theory (see Chapter 3). Attribution theory is a social psychological theory about how individuals manage to infer the ‘causes’ underlying the behaviour of others, or even their own behaviour. In trying to interpret behaviour, we will typically attempt to disentangle the contribution of internal causes (e.g. personality traits, motivation) from external causes (e.g. situational factors). For example, if a mother learns that her son has received a poor grade in his first maths test, she will wonder whether this poor result is due to lack of ability, lack of motivation or to an overly zealous maths teacher who gave too tough a test. Deciding between these alternatives will be important for her, because it will suggest different strategies to prevent this situation from happening again. The impact of attribution theory in stimulating a great deal of research in the 1960s and 1970s is intriguing, because neither Heider’s (1958) monograph nor his 1944 article was written in a way that would make it accessible or appealing to the average researcher in North America. There was also very little research to back up Heider’s ideas. It is generally accepted that attribution theory became influential because three major figures in the field of social psychology – Edward Jones, Harold Kelley and Bernard Weiner – adopted it and translated it into a language that was more accessible to social psychologists and yielded clear, testable hypotheses (Jones & Davis, 1965; Kelley, 1967; Weiner, 1986). Probably most influential was Kelley’s (1967) covariation theory (see Chapter 3). This model was appealing because Kelley argued that, in inferring causes of behaviour, our inference process would be analogous to conducting an analysis of variance, a statistical procedure highly familiar to social psychologists. Other influential adaptations of attribution theory were Jones and Davis’s (1965) correspondent inference theory and Weiner’s (1986) application of attribution theory to achievement motivation and emotion. A final way in which Hitler influenced the development of social psychology is by stimulating interest in topics such as obedience and authoritarianism. Why did the German people accept such an authoritarian regime, and why did so many of them carry out commands they must have perceived as immoral even at the time? These questions stimulated some of the most influential research in social psychology: researchers studied the authoritarian personality (Adorno et al., 1950), the determinants of conformity (Asch, 1955) and obedience (Milgram, 1963). Lewin’s interest in the effects of authoritarian and democratic leadership styles can be seen as an attempt to demonstrate the superiority of the democratic style, an attempt that was only partly effective because in his research autocratically led groups outperformed the democratic groups with regard to quantity of production, although democratic leadership produced more creative groups whose performance did not deteriorate so dramatically when the leader was removed (White & Lippitt, 1968).  The crisis years How and why did the crisis in social psychology develop? So far the history of social psychology appears to have been one of unmitigated success. Stimulated by World War II, social psychological research expanded enormously and there was soon no single psychology department at a top university that did not also have a strong social psychology unit. But just when social psychology was on the up and up, a crisis in social psychology developed that led to years of infighting about the right course one should follow. This crisis was probably initiated by two critical papers published in 1967 and 1973. ---------- crisis in social psychology a short-lived crisis of confidence among social psychologists during the 1960s, when social psychologists questioned the values, the methods and the scientific status of their discipline. ---------- The first of these two papers was a paper by Kenneth Ring entitled ‘Experimental social psychology: Some sober questions about some frivolous values’, published in the highly respected Journal of Experimental Social Psychology. In this paper, Ring contrasted the vision of Kurt Lewin of a social psychology that would contribute to the solution of important social problems with what he called the ‘fun and games’ attitude of the social psychology of his days. He argued that: ‘Experimental social psychology today seems dominated by values that suggest the following slogan: ‘Social psychology ought to be and is a lot of fun... Clever experimentation on exotic topics with zany manipulations seems to be the guaranteed formula for success... One sometimes gets the impression that an ever-growing coterie of social psychologists is playing (largely for another’s benefit) a game of “can you top this?”’ (pp. 116–17). Although Ring did not refer to any specific examples of this fun and games approach, his criticism was probably directed at some of the work conducted in tests of dissonance theory. Ring, although a respected researcher, was not a very central figure in the social psychology of his days. Therefore, the paper stimulated some discussion but did not really have a serious impact on the field. However, in 1973, one of the golden boys of experimental social psychology, Kenneth Gergen, published an article entitled ‘Social psychology as history’ in the top journal of our discipline, the Journal of Personality and Social Psychology. As the title already suggests, Gergen’s article was not an attack on the values directing social psychological research. Much more seriously, he questioned its scientific value. His two most important arguments were (1) that knowledge of social psychological principles could change our behaviour in ways which would negate these principles, and (2) that since the basic motives assumed by many of our theories are unlikely to be genetically determined, they might be affected by cultural change. As an example of the first principle, Gergen argued that once groups were aware of their tendency to make extreme decisions (i.e. group polarization; see Chapter 8), they might consciously counteract this tendency in their decision-making. As an example of the second principle, Gergen used social comparison and dissonance theory. Social comparison theory assumes that people have a desire to evaluate themselves accurately and do this by comparing themselves to others. Gergen argued that one could easily imagine societies in which such a desire would not exist. Similarly, dissonance theory assumes a need for consistency, which not everybody might share. Gergen saw these problems as the main reason why, as he claimed, social psychological research often failed to be replicable, and hence did not result in a body of cumulative knowledge. Most researchers these days would accept these arguments without questioning the scientific status of social psychology. With regard to Gergen’s first point, we would argue that it would be difficult, even for a trained social psychologist, to keep in mind all situations in which our behaviour might be affected by others, to recognize all the relevant cues signalling such situations, and then to counteract the situational pressures. With regard to his second point, the jury is still out. There is increasing evidence that repeating the same social psychology study in different parts of the world often leads to rather different results (see Chapter 15). However, such variation does not necessarily challenge the assumption that there are universal social processes. After all, cultural differences might merely qualify such universal processes. For example, it is hard to imagine societies in which people do not engage in social comparison, because the evaluation of one’s own abilities through social comparison is highly functional and essential for effective action. However, we do know that there are individual differences in the need for social comparison (Gibbons & Buunk, 1999), as there are in individual need for consistency (Cialdini, Trost, & Newsom, 1995). Since there are substantial individual differences within cultures, differences between cultures would not necessarily imply that the theories of social comparison or cognitive consistency should not apply to other cultures. Gergen’s (1973) critique would probably have been less effective had it not come at a time when the collective self-esteem of social psychologists had been undermined by other developments. For one, there was an attack on the usefulness of a concept that Allport (1935) had hailed as the most central concept of social psychology. In a review of studies that empirically assessed the value of social attitudes in predicting behaviour, the sociologist Alan Wicker (1969) drew the following conclusion: ‘Taken as a whole, these studies suggest that it is considerably more likely that attitudes will be unrelated or only slightly related to overt behavior than that attitudes will be closely related to actions.’ (p. 65). This conclusion was highly damaging, since social psychologists were interested in attitudes mainly because they expected them to predict behaviour. Since attitude change in most studies is assessed through an individual’s self-rated position on some attitude dimension, the news that such ratings might be unrelated to behaviour was devastating. A second development with a negative impact on the collective self-esteem of the scientific community of social psychologists was the publication of a series of papers that were highly critical of the experimental method (see Chapter 2). Thus, Martin Orne (1962) had suggested that most experimental situations contained demand characteristics, which would help research participants to guess the hypothesis to be tested in a given study. Since participants typically tried to be ‘good subjects’, Orne argued, they would then do their best to support these hypotheses.1 Even more damaging was the suggestion of Robert Rosenthal (Rosenthal & Fode, 1963) that the expectations of the experimenter might influence the behaviour of research participants, even without their knowledge (experimenter expectancy effects). The impact of these expectations on the behaviour of research participants could, for example, be mediated by experimenters’ reacting positively to responses that supported their hypotheses and negatively to responses that were inconsistent with expectations. ---------- demand characteristics cues that are perceived as telling participants how they are expected to behave or respond in a research setting; that is, cues that ‘demand’ a certain sort of response. ---------- ---------- experimenter expectancy effects effects produced unintentionally by the experimenter that increase the likelihood that participants will confirm the experimenter’s hypothesis. ---------- The reaction to these critical voices was the organization of numerous conferences in which the crisis was discussed, sometimes in rather heated language. Although these conferences resulted in a number of crisis books (e.g. Strickland, Aboud, & Gergen, 1976), they failed to bridge the theoretical and methodological chasm that separated the critics from mainstream social psychology. The critics finally founded their own social psychological schools, such as social constructionism in the United States (e.g. Gergen, 1999) and discourse analysis in the United Kingdom (e.g. Potter & Wetherell, 1987), which developed their own methodologies in an attempt to address these problems.  Overcoming the crisis How was the crisis overcome? In mainstream social psychology a number of developments were initiated, which over the years helped to alleviate some of the problems highlighted by these critics: Social psychologists began to demonstrate their ability to contribute to the solution of real-life problems by developing several applied areas, which contributed to resolving important societal problems. To mention only one such area, health psychology is an application of social psychology. One of the major research areas in health psychology is aimed at changing health-impairing behaviour patterns in our society (e.g. smoking, eating too much, drinking too much alcohol, practising unsafe sex). Social psychologists have helped to understand the reasons why people engage in these behaviours as well as to develop interventions aimed at changing them (Stroebe, 2011). The following chapters in this volume give many additional examples of how social psychology can be, and has been, applied to real social issues. The impression that social psychological research did not result in cumulative knowledge may have been the result of improper strategies of reviewing, a problem that was mostly resolved with the development of meta-analytic procedures (see Chapter 2). When reviewing research areas, researchers often erroneously concluded that support for a theory was missing or inconsistent, because few studies supported the theory by yielding significant results, whereas the majority of studies failed to find significant results. In the meantime, we have realized as a discipline (as has science, in general) that the failure to find significant results may simply have been due to conducting a study with an insufficiently large number of participants. If the effects we were looking for were small, this might have resulted in insignificant findings, even though the differences between conditions might all have been in the predicted direction. Since then, meta-analytic procedures have been developed which allow us to integrate statistically the results of independent studies of a given phenomenon, with a view to establishing whether the findings exhibit a pattern of relationships that is reliable across studies (Cooper & Hedges, 1994). We now know that attitudes are predictive of behaviour but that this relationship is often obscured in studies which employ inappropriate procedures in measuring the two components (see Chapter 6). As Ajzen and Fishbein (1977) demonstrated in their classic review, attitudes are related to behaviour if both components are assessed with measures that are both reliable and compatible. To be reliable, measures have to consist of multiple items rather than a single item. To be compatible, attitude and behaviour have to be assessed at the same level of specificity. Thus, if we want to predict whether people are likely to engage in physical exercise to improve their health, we should not measure their attitude towards their health, but their attitude towards engaging in physical exercise. The latter attitude is likely to be highly correlated with an aggregate measure of a variety of exercise behaviours (such as jogging, walking, going to the gym). If one wanted to predict specific exercise behaviour, such as whether an individual is likely to jog, one should measure his or her attitude towards jogging rather than towards physical exercise in general. Finally, social psychologists have tried to design their experimental manipulations in ways that would minimize the threat of demand characteristics and experimenter expectancy effects. Furthermore, the fact that many research participants do not even meet experimenters any more (because experiments are often run on the computer by computer programs) should certainly rule out experimenter expectancy effects. The depressing fact that most experiments do not work out the way they were expected to by the experimenter who designed them also appears to suggest that demand characteristics and experimenter expectancy effects cannot be all that powerful. Festinger (1980) most aptly expressed these feelings when he wrote: ‘I’ve always wondered why, if these spurious experimenter effects were so strong, so many of my own experiments did not show the expected results.’ (p. 252.)  SOCIAL PSYCHOLOGY IN EUROPE How did social psychology develop in Europe? Until the end of World War II, the development of social psychology as a discipline was restricted to the US. However, even before the influx of academic refugees in the 1930s, there had been a great deal of European influence on this development. For example, as mentioned earlier, F. Allport’s (1924) work on social facilitation had been stimulated by one of his academic teachers at Harvard, the German Hugo Münsterberg, who in turn was familiar with similar work that had been done in Germany by Moede (1920). The experimental work of Bartlett (1932) in Britain on remembering can be viewed as a major precursor of contemporary research on social cognition. And finally, the theorizing underlying Sherif’s (1936) studies of norm development is heavily influenced by Gestalt psychology. However, even though there were individuals in Europe who conducted research that could be considered social psychological, there was no unitary social psychology. This situation continued into the 1960s, even though social psychology groups had been established at a number of European universities. But while there was social psychology in Europe, there was no European social psychology: there was no European collaboration, and most European researchers had not met each other, nor were they even aware of each other’s work. Obviously, a European network was not necessary for the development of a strong social psychology in some of the European countries where effective social psychology research groups already existed (e.g. Belgium, Britain, the Netherlands and Germany). However, in some other countries it would probably have taken many decades for social psychology to develop. Furthermore, since most of the European researchers met each other, if at all, only at conferences held in the US, without the foundation of a European association European social psychology would probably have remained a minor appendix of North American social psychology rather than developing its own theoretical perspective. Thus, the foundation of the European Association of Experimental Social Psychology was critical.  Given the dominance of North American social psychology at that time, even in Europe, it is no coincidence that it was again an American, John Lanzetta, who set things in motion in 1963. During a sabbatical year in London, Lanzetta, then Professor of Social Psychology at the University of Delaware, visited various social psychology groups in Europe. He was struck by the fact that many of these colleagues, though well informed about US social psychology, were not really aware of what was going on in the social psychology departments of neighbouring European countries. He decided to change this and raised funds for a first European Conference on Experimental Social Psychology, held in Sorrento, Italy, in 1963 (Nuttin, 1990; Moscovici & Marková, 2006). One of the main initiatives which emerged from this and two follow-up conferences was the foundation of the European Association of Experimental Social Psychology (EAESP) in 1966. The European Association engaged in a number of regular activities, which had great impact on the development of social psychology in Europe. These included: ---------- European Association of Social Psychology an association formed by European researchers in 1966 to further social psychology in Europe (originally named the European Association of Experimental Social Psychology). ---------- Summer schools for advanced students, taught by outstanding researchers. Publication of the European Journal of Social Psychology from 1970, which included most of the early research thought of (then at least) as typically ‘European’ (e.g. studies of intergroup relations or minority influence). Other key European publications were, first, the European Monographs series and, later, the European Review of Social Psychology. The first edition of the textbook you are reading now was published in 1988, in part to counteract the tendency of American textbooks to under-report the work of European social psychologists. The regular organization of conferences, including plenary meetings of the whole membership, and special East–West meetings (the latter were  particularly effective forums at a time when travel and currency restrictions made it extremely difficult for social psychologists from Eastern and Western Europe to meet). Membership in the European Association of Social Psychology (EASP), as it is known now after dropping ‘experimental’ from its name in 2008, has grown at a phenomenal rate, from less than 100 in 1970 to more than 1,000 members in 2010. During this period, scientific development in social psychology also changed from being a one-sided enterprise, with American ideas being adopted in Europe, to a mutual development, with European ideas being taken up enthusiastically in the United States and ever-increasing collaboration leading to scientific growth. It is now accepted practice for prominent North American journals (Journal of Personality and Social Psychology, Journal of Experimental Social Psychology and Personality and Social Psychology Bulletin) to have at least one European editor, and likewise for the European Journal of Social Psychology to have non- European editors. Probably the two most important examples of European ideas influencing social psychology in the United States are research on intergroup behaviour and on minority influence. Although Henri Tajfel was not the first to conduct experimental research on intergroup behaviour (that honour goes to Sherif), he developed the paradigm (the minimal group paradigm) that turned intergroup behaviour into a major research area (see Chapter 14). The minimal group paradigm offered an easy and very economical procedure for the study of intergroup behaviour, but Tajfel and Turner (1979, 1986) developed from it a theoretical framework that could account for these findings, social identity theory (see Chapter 14). The second theoretical innovation that was started in Europe and then accepted in the United States is research on minority influence. Social influence research in North America focused exclusively on conformity, that is, on explaining how majorities influence minorities. It was Moscovici who first pointed out that this type of theorizing could hardly explain social or religious innovations, where powerless minorities influenced powerful majorities (e.g. women’s rights, Christianity). After Moscovici and his colleagues in Paris (e.g. Moscovici, Lage, & Naffrechoux, 1969) had published a number of studies demonstrating minority influence, and again with the development of a theory that could account for these effects, research on minority influence became a major research area both in the US and in Europe (Moscovici, 1976; see Chapter 8).  SOCIAL PSYCHOLOGY TODAY What new theoretical perspectives have emerged during the last few decades? In the 1980s most of the researchers who had contributed to modern social psychology, and who, as often as not, had come from the research centres directed by either Lewin or Hovland, were still alive and active (Cartwright, 1979). In the meantime, not only have many of these pioneers retired or died, but so also have most of the students whom they, in turn, had trained. The field has grown at an exponential rate. There are now chairs in social psychology at practically all major universities in the United States, in Europe, and in Australasia, and social psychologists number in the thousands rather than a few hundreds. Social psychology has also become an essential part of the psychology curriculum in these countries. Not surprisingly, social psychology has also changed over these decades. Major scientific perspectives, such as consistency theory or attribution theory, have faded and new perspectives, such as social cognition, evolutionary social psychology and social neuroscience, have emerged. Jones (1998) colourfully described these changing trends in research as ‘band wagons and sinking ships’ (p. 54). Social cognition research is an application of principles of cognitive psychology to the area of social psychology (see Devine, Hamilton, & Ostrom, 1994). Unlike other psychological disciplines, social psychology has always placed a strong emphasis on how individuals internally represent their environment. Many of