Person in a World of People part 7 Chapter 13 @ Social Perception and Attitudes Chapter 14 # Social Influences on Behavior We humans are social beings through and through. We are motivated to understand others; we are concerned about what others think of us; and our understanding of ourselves is strongly affected by our perception of what others think of us. This two-chapter unit is on social psychology— the attempt to understand human thought and behavior in relation to the social contexts in which they occur. The first chapter is about the mental processes involved in understanding others, ourselves, and the social world in general. The second is about some of the ways in which the presence or activities of other people, real or imagined, influence our behavior. \\ 495 jcial Perception and Attitudes lans are intensely social animals. We are designed, by evolution, to depend ie another for even our most basic needs. We are not good, for example, at obig food alone; we need the help of others and the knowledge that is shared by fers of a human community. Throughout our evolutionary history, to be m out of the tribe was tantamount to death. We are also thinking animals— Sbg social animals. Most of what we think about is other people, ourselves, our feiships with other people, and the social conventions and norms that are es-Ifaspects of life in any human society. Fhis is the first of a two-chapter sequence on social psychology, the subfield of lology that deals most explicitly with how people are influenced by each other, {chapter focuses on person perception, the processes by which people perceiveV inderstand each other and themselves, and on attitudes, the evaluative beliefs/ feople have about their social world and the entities within it. The next chap-;es on the effects of those perceptions and beliefs on the person's emotions ions. four themes run through this chapter. The first is that biases in social percep-jand attitudes can lead people to make judgments that are objectively untrue lfoir. The second theme is that social perceptions and attitudes serve life-hm% functions for the individual. The third is that culture plays a powerful \ shaping our social perceptions and attitudes; some social-psychological ámena that occur reliably in North America and western Europe do not , serving their behavior (Ambady & others, 1995; Funder, 1995). Yet, as Heider himself pointed out, the accuracy of our judgments c I 'uhu sometimes suffers from certain consistent mistakes, or biases. These bias« ■■ cu most often when we are not using our full mental resources, or have mil) ' i.i I« information with which to reason, or have unconscious motives for reaching pra ticular conclusions. Such biases interest social psychologists for two re-at>oi l'ir^í they provide clues about the mental processes that contribute to accurate a- i .1' * inaccurate perceptions and judgments. In this regard, social psychologists' ml ' in biases is analogous to perceptual psychologists' interest in visual illusion aIi1 (as discussed in Chapter 8) provide clues to understanding normal, accura usu perception. Second, an understanding of bias can promote social justice. B> i-Liiti-fying and teaching others to recognize psychological tendencies that conti n i (. prejudice and unfair treatment of people, social psychologists can help imp * liu social world. 2. How does the process of attribution contribute to person perception? 3. According to the logic outlined by Kelley, when should an attribution be made to the person and when should it be made to the situation? \\ tvt u nib Forming Impressions About People from Their Behavior Actions are directly observable and thoughts are not. Therefore, our j' Lin i about the personalities of people we encounter must be based primarily oh ■ I * observe of their actions. As "naive psychologists" we intuitively, in our evf ■ periences, form impressions of people's personalities on the basis of the For example, if a new acquaintance smiles at you, you do not simply re j feet that she smiled; rather, you interpret the smile in terms of its meanin that interpretation to infer something about her personality. Depending oi text and any prior information you have about her, you might decide that ■ ■«■ *iiiiki represents friendliness, or smugness, or guile. What you carry away fro ■■ ■! ■ iu-j counter is not so much a memory that the person smiled as a memory th ' ■ «*Js| friendly, smug, or deceitful. That memory is added to your growing imp i > i ul§ her and may affect your future interactions with her. 3 Any such judgment about another person is, in essence, a claim ab< ■'■ iw^ tion. It is an implicit claim that the person's behavior is caused in pat- i< 41101 more or less permanent characteristic of the person, such as friendliness ■ 1 '■ ■ i&' fulness. In normal English usage, any claim about causation is called an / './ Tŕ In the study of person perception, an attribution is any claim about the ca.- ut someone's behavior. As Heider (1958) pointed out, a major problem in ji ■■ n$ someone's personality on the basis of his or her action is that of determining ti ik- gree to which the action truly represents something unique and lasting abc ■ t>iC person or, instead, represents a normal human response to a particular situai1"1!^ set of circumstances. The Logic of Attributing Behavior to the Person or the Situation If you see a man running and screaming and then see that a tiger is chasing liii ■ might logically attribute his fear to the situation rather than to any special asj 1u his personality; almost anyone would be afraid of a loose and charging tir. 11 build a useful picture of a person on the basis of the person's actions, you nv ■!■. cide which actions imply something unique about the person and which , ■ ■> would be expected of anyone under similar circumstances. Heider noted ttu i'n behavior is clearly appropriate to the environmental situation, people conuno.n .' tribute the behavior to the situation rather than to the behaving person's pcrso.i ''n In line with Heider's general ideas about attributions, Harold Kelley''. >i 1973) developed a logical model for judging whether a particular action should \ attributed to some characteristic of the acting person or to something about ti immediate environment. The essence of the model is that the perceiver ennside three questions in making an attribution: (1) Does this person regularly behaved way in this situation? (2) Do many other people regularly behave this \\iur in tl Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 499 Attributional problem: We are caught in a traffic jam and Susan, the driver, is expressing anger. Does her anger tell us something useful about her personality? Logic of solution: Compare the observed behavior with the person's typical behavior in this situation (consistency), with other people's typical behavior in this situation (consensus), and with the person's behavior in other situations (distinctiveness). Questions: (I) Does Susan regularly get angry'in traffic jams? No I t Yes No basis^for^ttributing Ker anger to either J Attribution: her personality or lie v situation; This may beafluke. . „ (2) Do many other ' people get angry ' in traffic jams? Yes Situational attribution: Traffic jams'make people angry,- No [ (T) Does; Susan get ang ry 111 111 1111 ulili! till! IIIIIII-- Yes Personality attribution general: S111 in is easily asm 11 (Be careful around hej ) No Personalit* attribution particular: Su*> 111 can't toler.11« traffýxjam*- I \*im«I riding with Ik 1 ) XL....on? (3) Does this person behave this way in many other situations? For a fully hrai ked-out illustration of the model, linking the answers to these questions to the lattribntion that would logically follow, examine Figure 13.1. There is nothing surprising in this model. It is simply a statement of the logic ftbat you or I or anyone else—with sufficient motivation and information—would ■'se in deciding whether or not an observed bit of behavior tells us something inter-sting about the person. It states explicitly the logic that leads us to conclude that a man's repeated fearful reaction to a gentle poodle tells us more about the man than oes his fearful reaction to a loose and raging tiger. Not surprisingly, a number of research studies have shown that when people re asked to explain the cause of a particular behavior and are given sufficient infor- ■ íation to answer the three questions, they usually do make attributions that accord . ith the model just described (McArthur, 1972). But often people lack the infor- I íation, the time, or the motivation to make a logical attribution. In that case they iay take shortcuts in their reasoning, which may result in certain consistent errors, r biases. The Person Bias in Attributions II his original writings about attribution, Heider (1958) noted that people tend to ve too much weight to personality and not enough to the environmental situation ■hen they make attributions about others' actions. Concerning the example in sgure 13.1, they tend to ignore the traffic jam and attribute Susan's anger too h iavily to her personality. Subsequently, other researchers confirmed the existence ~ůás person bias in attribution. For instance, in one experiment male college stu-■1 ;nts listened to a student who they were told was assigned to read a political statement written by someone else (Gilbert & Jones, 1986). Even when the assignment was made by the observers themselves, so they could be sure that the reader had not chosen it himself, observers tended to rate the reader as politically liberal when the statement he read was liberal and as politically conservative when the statement was conservative. Although there was no logical reason to assume that the statement had anything to do with the reader's own political beliefs, the students made that attribution. Some of the most socially relevant examples of the person bias have to do with the effects of a person's social role on others' perceptions of the person. When we observe a police officer, nurse, teacher, or student carrying out his or her duties, we tend—in accord with the person bias—to attribute the action to the individual's personality and to ignore the constraints that the role places on how the person can figure 13.1 The logic behind an attribution According to Kelley, our decision to attribute an observed behavior to internal causes (the person) or external causes (the situation) depends, ideally, on our answers to three questions. Although Kelley did not specify a sequence for asking the questions, this flowchart depicts the most efficient sequence and the attribution that would follow each combination of yes and no answers. In Kelley's original model the three questions shown here were described, respectively, as issues of consistency, consensus, and distinctiveness. 4. What evidence supports the existence of a person bias in attributions, and why is the bias often called the "fundamental attribution error"? 500 Part 7 The Person in a World of People A victim of bias? Leonard Nimoy called his autobiography I Am Not Spock. He has apparently often encountered the fundamental attribution error. 5. What conditions seem to promote a person bias or a situation bias, and how did an experiment demonstrate the effects of these conditions? act. We might develop quite different impressions of the same person if we saw hi or her in out-of-role situations. In an experiment demonstrating this effect of rol< Ronald Humphrey (1985) set up a simulated corporate office and randomly i signed some volunteer subjects to the role of manager and others to that of clei The managers were given interesting tasks and responsibilities, and the clerks we given routine, boring tasks. At the end of the study, the subjects rated various í pects of the personalities of all subjects, including themselves. Compared wi those in the clerk role, those in the manager role were judged by others more pc itively; they were rated higher in leadership, intelligence, assertiveness, support« ness, and likelihood of future success. In keeping with the person bias, the subja apparently ignored the fact that the role assignment, which they knew was randos had allowed one group to manifest characteristics that the other group could m The bias did not hold when the subjects rated themselves, but it did hold when th rated others who had been assigned to the same role as themselves. By the mid-1970s so much evidence appeared to support the person bias ti / Lee Ross (1977) called it the fimdamental attribution error, a label designed to si nify the pervasiveness and strength of the bias and to suggest that it underlies mi other social-psychological phenomena. That label is still in use despite growings idence that the bias may not be quite as fundamental as Ross and others thought Conditions That Promote a Person Bias or a Situation Bias The studies that supported the pervasiveness of the person bias may themsi have been biased. Volunteers for psychological studies tacitly agree to coopéi with the researcher. In experiments on attributions, they may believe tha-is to make some sort of attribution about the performer's personality o '■ -i" even with insufficient information (Leyens & others, 1996). In real life i --ii people might not make judgments about a person's characteristics on tli i b such scanty information as the person's reciting of an assigned statement -i i u of an assigned role. When volunteers are invited to explain samples of behavior in what . t o they choose, they most often give explanations that cannot be classed as i I ■ p» son attributions or situation attributions (Malle & others, 2000). In one li -to (Lewis, 1995), many of the attributions were stated in terms of the goals ' i . tion, such as "She is typing diligently at the computer in order to fin i paper that is due tomorrow." A goal lies in both the person and the envii The environment sets the challenge (the paper is due), and the persoi meet the challenge. In real life people are not just interested in judging o I sonalities; they are also interested in the challenges that are set by various i iU0 and in the ways that people go about meeting those challenges. Other research suggests that even in the laboratory the person bia L,iO very much on the conditions of the study. It occurs most frequently wb i J t subjects' task or goal is clearly to assess the personality of the target indi 11 :l a. (b) the subjects are provided with little opportunity or motivation to h Üi controlled reasoning powers to bear on the problem, so they rely primal »vl tomatic mental processes. In one such experiment, female college stu ills o served a silent videotape in which a young woman being interviewed wa< I' linií very nervously (Krull, 1993). The subjects were told that the interview to] i nag or might not be highly anxiety-provoking (the possible topics ranged &■< n ft« travel to sexual fantasies) but were not told the topic of the interview th< my« manipulate the subjects' goal, some were told that their task would beto j "itfi anxious the woman was in her everyday life and others were told that Huff | would be to judge the degree to which the interview topic was anxiety-p mm To manipulate the opportunity for controlled thought, some subjects in e ■ h two goal groups were kept cognitively busy with another task (rehear digit number) as they watched the video and others were permitted t< full attention to the video. Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 501 \ller watching the video, all subjects were asked to judge (a) the degree to ■ li die interviewee was an anxious person in her everyday life and (b) the degree v!i ch the topic of the interview was one that would provoke anxiety in most i! . Only those subjects who were kept cognitively busy showed evidence of bi-I i-Msoning. The others attributed the woman's anxiety about equally to her j» ality and to the situation. Of the cognitively busy subjects, only those who ' en asked in advance to evaluate the person manifested the typical person V I'hcy judged the woman to be a very anxious person and judged the interview | ■■ 10 be only moderately anxiety-provoking. Most interesting, the cognitively S "ibjccts who had been asked to judge the interview topic manifested an oppo-, "it 11 is—it situation bias rather than a person bias. They rated the interview topic \ liiiiily .mxiety-producing and the woman as only a moderately anxious person. 1' i tlie basis of this and other evidence, the researchers proposed a two-stage •I I of the process of making attributions (Krull & Erickson, 1995, elaborating íodel proposed earlier by Gilbert, 1989). The first stage is rapid and auto-iii- md typically leads to a judgment that is biased in accordance with the reason - goal—a person bias if the goal is to judge the person and a situation bias if _• al is to judge the situation. The second stage is slower, is controlled, and i* ii t occur if the person is mentally busy with another task or is not motivated to Ian - mental resources to the task. At this stage the person corrects the automatic tir I don by taking into account the entity (situation or person) that is not the di-i irget of the attributional goal. The model is illustrated, with a new example, mi 'ire 13.2. £ Annburionai problem: Joe is laughing hysterically while watching a television comedy, f Hint can we conclude? Tvpcal solutions: «...... ill. ftu s«n 'tiffM-rsini \iinnn iii< ill I miil.iiľ F'mschi itinfiiuKm. (Jot. IniL'hs« i»lv ) Ueyisiufi »t uiiihiili'iu (ľhiiinn In lun tuiuii «oiuliIv * pi i hips 1«. ÜOLSIl11 lll_!l si» l isili ) ijdu. mim nifiii imm is ih ii I \ Sllll H H HI lltlllllltlltll (h's i tut timm ( York every article published in i1 Journal, a Chinese-language newspaper published in New York, (.once m _ ľ specific mass murders that took place in 1991. The researchers found tha >o Ji cles in the Times focused most heavily on personality characteristics of t m ltd ers—their traits, attitudes, character flaws, mental disorders, and so on. ii lift the articles in the Chinese newspaper focused most heavily on the lite s im» the murderers—their living conditions, their social relationships, and h1 'nst tions that might have provoked their actions. The Actor-Observer Discrepancy in Attributions Many studies suggest that the person bias is weaker, and the smia'i mi luv ^/stronger, when people make attributions about their own behavior thai lund *i make attributions about someone else's. This difference is referred to i ■ drf \ observer discrepancy (Nisbext tk others, 1973). The person who perfori ■ m. (the actor) commonly attributes the action to the situation—"I am wL .1 ly cause it is a beautiful day," or "I read those political statements because in-.* to read them." In contrast, another person (the observer) who sees d\e sa i i uo likely to attribute it to the actor's internal characteristics—"She is whistln - n >í she is a cheerful person," or "He read those statements because he i i 'hue liberal." What causes the actor-observer discrepancy? According to one I mihi people know from experience that their own behavior changes from o *>m«tj to another, but they do not have as much evidence that the same is tin ■l <>tlC For example, you may assume that your psychology professor's calm d nm the classroom is indicative of his or her behavior everywhere and thus ati i n" l personality, but this may be only because you haven't seen your profess i tin in traffic court, or on the softball diamond. Consistent with tins " >*\j across-situations" hypothesis, people usually judge the behavior of ■! r c friends as more flexible—more determined by the situation and less b; n v ii personality traits—than the behavior of people whom they biov * (Prentice, 1990; Sande & others, 1988). Another hypothesis holds that the actor-observer discrepancy sti» it» /basic characteristic of visual perception: Our eyes point outward, awai ■■< ui" selves. When we watch someone else perform an action, our eyes focus imo not the environment to which the actor is responding, so we tend to se u e* caused by the person rather than the situation. But when we perform an • u in see the surrounding environment, not ourselves, so we tend to attril'iMi u properties to the situation. Consistent with this "visual-orientation" '■ luth one experimenter found that the actor-observer discrepancy was revcrst I actor and the observer watched videotaped replays of the action from * sual orientations (Storms, 1973). When people watched themselves or they attributed relatively more of their behavior to their own traits ,in(' situation. When people watched a videotape of another person's peifon Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 503 1 made dirterem * interpret ma set up for dien, becauSe people^ W ^ £ of others'interpretations, our ,rior impression of us. ^ce, and, hke m ^ d people (the pr iments have ^^ *f**££ (the witches and ogre ) are ug^y P intelligent^ m and the bad people Cth, ^ commonly judged as Eagly ^ U that physically ^^A^ less attractive people (Uio , and Lent, sociable, and moral than ers were given repor 8. ľ " - """ 0 n,ppxístíng schema bias d0 first impressions often res,st change? rXnstatpacp.e.a.eabou.a person? 504 Part 7 The Person in a World of People 10. How might a sex difference in facial features contribute to perceived psychological differences between women and men? FIGURE 13.4 Who would deceive you? Adults whose faces are babyish (left) are commonly seen as more naive, honest, helpless, kind, and warm than are mature-faced adults (right). The characteristics of a baby face include a round head, large forehead, large eyes, short nose, and small chin. attractive (Dion, 1972). In yet another study, which analyzed actual couri judges regularly gave longer prison sentences to unattractive persons than tractive persons convicted of comparable crimes (Stewart, 1985). The Baby-Face Bias Another pervasive bias, although less well known, concerns a person's fecial rity. Some people, regardless of their age, have facial features resembling th baby—a round rather than elongated head, a forehead protruding forware than sloping back, large eyes, and a small jawbone (see Figure 13.4). In ,i s experiments conducted in both the United States and Korea, baby-faced were perceived as more naive, honest, helpless, kind, and warm than m.itui adults of the same age and sex, even though the perceivers could tell thai ih faced persons were not really younger (McArthur & Berry, 1987; Zebrowitz ers, 1993). Leslie Zebrowitz and Susan McDonald (1991) found that the ba'-bias, like the attractiveness bias, can influence the outcome of actual small court cases. Baby-faced defendants were much more frequently found mm cases involving intentional wrongdoing than were mature-faced defendaľ they were neither more nor less frequently found innocent in cases involvin.' gence (such as performing a contracted job incompetently). Apparently, j udit hard to think of baby-faced persons as deliberately causing harm bul do ]>■ it hard to think of them as incompetent or forgetful. Zebrowitz (1996) has also found evidence that differences in facial iiutu tween men and women may contribute to differences in how the two sexes ceived. Women, on average, are more baby-faced than men, and women ai on average, judged as kinder, more naive, more emotional, and less sociall; nant than men. In an experiment, Zebrowitz and her colleagues presented ■ students with schematic drawings of men's and women's faces in which tacif I rity was varied by altering the size of the eyes and the length of the jaw (Fi ľ & Zebrowitz, 1992). When the typical differences between men's and \ ■ faces were present, students judged the man as more dominant and less wa ■ the woman. But when the faces were equivalent on the maturity dimensii , students judged the two as equal in dominance and warmth. Zebrowitz did i gest that facial features are the sole determinant of the different perception1 have of women and men under more natural conditions, but she did Migg such features may contribute to the difference. / The ethologist Konrad Lorenz (1943, 1971) suggested long ago that i Ifacial features act as sign stimuli (a concept discussed in Chapter 3) to elicit niis-i -Ig I 'KT of ■:..b \A >ini- - .li- .-! x • ' ■« l-ili ■■■it jii1 ■nid i k-' i r- ■"■, 1.1 m* ■I L.0 111 .liS- < ii-iii>; ' n \i$ ) ..- 'iej ■i -u j- . ■ -|.le' Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 505 ■n ■ e response of compassion and caring. He also noted that the same features ! I as to perceive some animal species (such as bunnies and pandas) as particularly mi innocent, and needing care, regardless of the animals' actual behaviors. /■ ■ owitz's work suggests that we generalize this response not just to babies and i ii als but also to adult humans whose faces resemble those of babies. For another -«i pie, see Figure 13.5. !'rom an evolutionary perspective, it is noteworthy that human adults of both are much more baby-faced than the adults of our closest primate relatives, lypical adult human face is more like that of an infant chimpanzee than like tat of an adult chimpanzee. This difference is generally attributed to the expanded [cranial cavity that came with enlargement of the brain in humans. But I wonder if $hat is the whole explanation. In the course of human evolution individuals who í had babyish faces may have been treated more benignly than those who had more 'mature feces, and perhaps this helped promote our species' evolution toward baby-éfacedness. This is speculation, but it is supported by evidence that baby-faced chil-iren and adolescents are less often physically abused than are their age mates who lave more mature faces (McCabe, 1984). As further speculation, perhaps the pro-ective effect was of greater value for girls and women than for boys and men—for •easons having to do with the general sex difference in strength and aggressiveness—leading to sex difference in baby-facedness observed today. (Can you imag-ne any ways to test this theory? Send me an e-mail if you can.) figure 13.5 The evolution of innocence in Mickey Mouse Mickey Mouse began life, in the 1928 cartoon Steamboat Willie, as a mischievous, teasing character who delighted in cranking a goat's tail. In response to social pressure from citizens concerned about Mickey's moral effects on children, the Walt Disney Company made him increasingly innocent over the years. As part of this change, Disney artists made him look more innocent and cute by augmenting his juvenile features. According to measurements made by Stephen Jay Gould, over a 50-year period Mickey's eye size increased from 27 to 42 percent of his head length, his head length increased from 43 to 48 percent of his body length, and the apparent vault of his forehead increased markedly as a result of the gradual migration of his front ear toward the back of his head. (Based on S.J. Gould, 1980, pp. 96-97.) Effects of Stereotypes We all carry in our heads Schemas not just for individual persons but also for whole groups of people. You may have Schemas for men, women, Asians, African-\mericans, Californians, Catholics, and college professors. Such Schemas are called stereotypes. The first person to use the term stereotype in this way was the journalist 1 Valter Lippmann (1922), who defined it as "the picture in the head" that a person. ' nay have of a particular group or category of people. Some stereotypes may accu-tately portray the distinguishing characteristics of a group, others may exaggerate those characteristics, and still others may be total fabrications. Later in this chapter and the next, you will encounter some ideas about how stereotypes are acquired and the psychological and social functions they may serve. For now, however, our concern is with the effects of stereotypes on perceptions of individuals. Whether or not a stereotype accurately portrays the average member of a particular group, it can bias our assessment of any individual member who differs (as everyone does) from the average. The Problem of Identifying Stereotypes At one time, psychologists in the United States had no difficulty identifying people's stereotypes of various ethnic groups. All they had to do was ask. Typically they asked respondents to check off, on a list of traits, those that best characterized the group in question. In one such study in the early 1930s, Princeton University students described black people as superstitious (84 percent checked this trait) and lazy 506 Part 7 The Person in a World of People 11. What is the distinction among public, private, and implicit stereotypes, and how do psychologists identify implicit stereotypes? Overcoming a stereotype We tend to stereotype people who look different from us or from what we conceive as the norm. This woman, Barbara Tiemann, has a genetic condition called Turner's syndrome and is a past president of the Turner's Syndrome Society. Because women with this condition look different, many people mistakenly assume that they are intellectually slow. The society serves as a support group to help members deal with this kind of stereotyping. 12. How have researchers shown that stereotypes can lead to prejudice and discrimination even in the absence of conscious prejudice? What different roles do automatic and controlled mental processes play in reactions to stereotyped individuals? (75 percent), Jews as.shrewd (79 percent) and mercenary (49 percent), and C as säentificatty minded (84 percent) and industrious (65 percent) (Katz & Ural; A decade later, after World War II broke out, the prominent stereotypes c and Jews were little changed, but that of Germans changed to include the t rogant and cruel (Seago, 1947). Today it is not so easy for psychologists to assess stereotypes. People in ture, particularly college students, are sensitized to the harmful effects ol types and are reluctant to admit holding them, especially negative one socially oppressed groups. Some social psychologists today distinguish three levels of stereotypes: public, private, and implicit (Dovidio & oilier; -The public level is what we say to others about the group. The private levc we consciously believe but generally do not say to others. The implicit lev set of learned mental associations that can guide our judgments and actions our awareness, whether or not the associations coincide with our consciou: Much recent research on stereotypes has centered on the implicit level. As you may recall from Chapter 9, cognitive psychologists often use a called priming to learn how knowledge is organized in people's minds. The behind this method is that any concept presented to a person activates (pi the person's mind the entire set of concepts that are closely associated v concept. Priming the mind with one concept makes the related concepts it ily retrievable from long-term memory into working memory. For cxan word apple, presented as a prime, enables a typical person to respond more ■ i to yes-or-no questions about such related concepts as red, round, pie, and * person whose task is to press one button for yes or a different button for n> spond a few milliseconds faster to the question "Is red a color?" after sc word apple as a prime than after seeing banana as a prime. Social psychologists have adopted the priming method to identifj peo plicit stereotypes. So that subjects do not consciously counteract the cttcc primes and suppress their stereotypes, the primes are either presented as i distractors or flashed so quickly that they are registered unconscious! y bu consciously perceived (a procedure discussed in Chapter 8). The priming s experiments dealing with stereotypes of black people and white people an. cases words—such as BLACK or WHITE—and in other cases pictures of white faces. Such studies reveal that implicit stereotypes are very much alive, c\e sons who don't subscribe to the stereotypes in their explicit statements ■ tionnaires (Dovidio & others, 1996; Kawakami & Dovidio, 2001). Prii concept of a black person typically leads white subjects to respond more q questions about such concepts as lazy, hostile, musical, and athletic; and prii concept of a white person leads them to respond more quickly to such coi conventional, materialistic, ambitious, and intelligent. Although positive and traits appear in both stereotypes, the experiments reveal that the implici types white students have of blacks are significantly more negative than tl have of whites (Dovidio & others, 1986; Fazio & others, 1995). Converse! study black Americans manifested significantly more negative stereotypes ■ than of blacks (Fazio & others, 1995). Effects of Implicit Stereotypes on Actions and Judgments Implicit stereotypes can promote prejudicial actions and attributions to stereotyped group, even in people who are not consciously prejudiced study, John Dovidio and his colleagues (1997) found that white college nonverbal signs of discomfort (rate of eye blinks and failure to make eye when interviewed by black interviewers correlated significantly and quite with their implicit negative stereotypes of black people, which had been using a priming method. In contrast, the same study revealed no covrel tween the measures of discomfort and the students' explicit views of lilac which had been assessed with a questionnaire. Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 507 In an earlier study, Patricia Devine (1989) worked with white college students as subjects. Half were primed to activate their stereotypes of black men. Then all subjects heard a story about a man named Donald who engaged in actions that might or might not be attributed to hostility. The story did not mention Donald's race Devine assumed that subjects in the stereotype-primed condition would have in their unconscious mind the image of a black man as they heard the story, and she predicted that they would interpret Donald's actions in terms of that stereotype. Consistent with Devine's prediction, those in the stereotype-primed condition rated Donald higher in hostility and unfriendliness than did those in the unprimed condition. This was as true of subjects who manifested the least prejudice on a questionnaire designed to assess explicit prejudice toward blacks as it was of those • who manifested the most prejudice by that measure. Other studies have shown that college students more often make biased attributions based on stereotypes when they are tired or mentally preoccupied than when they are refreshed and able to bring their full conscious attention to the attri-butional task (Bodenhausen & others, 1999; Gilbert & Hixon, 1991). Stereotypes apparently provide automatic shortcuts to judgment in situations where we lack the information, time, mental resources, or motivation to evaluate logically the facts of the individual case. Overcoming prejudice, therefore, is like resisting any well-learned habit. Devine and Margo Monteith (1993, 1999) have found that people who consciously dispute the culture's stereotypes feel discomfort or guilt when they ■ind themselves reacting automatically in stereotype-consistent ways. They suggest "hat this discomfort can lead people to work deliberately at countering their auto-natic reactions and eventually overcoming them. Person Perceptions as Self-Fulfilling Prophecies Our beliefs and expectations—whether they are initially true or false—can to some legree create reality by influencing the behavior of ourselves and others. Such erects are sometimes called self-fulfilling prophecies (Merton, 1948). In George ■iernard Shaw's play Pygmalion (upon which the musical My Fair Lady was based), he impoverished cockney flower girl Eliza Doolittle becomes a fine lady in large ,>art because of her response to the expectations of others. Professor Higgins assumes that she is capable of learning to talk and act like a fine lady, and Colonel dickering assumes that she is truly noble at heart. Their actions toward Eliza lead iier to respond in ways that transform their assumptions into realities. A number of experiments have affirmed the "Pygmalion effect." When persohwľ V is led to believe that person Z has a particular characteristic, A!s behavior toward J í can lead Z to manifest that characteristic. In one such experiment, Robert Rosenthal and Lenore Jacobson (1968) led elementary school teachers to believe hat certain students would show a spurt in intellectual growth during the next few nonths, as indicated by a special test that all students had taken. In reality, the students labeled "spurters" had been selected not on the basis of a test score but at ran-lom. Yet when all the students were tested 8 months later, the selected students howed significantly greater gains in IQ and academic performance than did their lassmates. Subsequent research on this Pygmalion in the classroom effect indicates ■hat it occurs through differences in the ways that teachers behave toward students fhom they believe will excel compared with the other students. Teachers are /armer toward those students, give them more time to answer difficult questions, _ive them more challenging work, and notice and reinforce more often their self-oitiated efforts (Cooper & Good, 1983; Rosenthal, 1994). In short, either con-ciously or unconsciously, they establish for those students a better learning nvironment, one in which the teachers' expectations of the students become self-i drilling prophecies. Other researchers have documented self-fulfilling effects of cultural stereotypes. J. Michael Palardy (1969) identified, with a survey, a group of first-grade teachers who believed that boys learn to read more slowly than do girls and another V 13. What is some evidence that beliefs about a person or a group can affect that person or group in such a way as to become a self-fulfilling prophecy? 508 Part 7 The Person in a World of People group who believed that boys and girls learn to read at equal rates. Then, 5 months later, he examined the reading achievement scores of students who had subsequently entered the classrooms of the two groups of teachers. The results conformed with the teachers' expectations. The boys scored lower in reading than die girls in classrooms where the teacher believed in the sex difference but not in classrooms where the teacher did not believe in the sex difference. Other research suggests that parents' beliefs about biologically based sex differences in math, sports, and social relationships can influence the development of their sons and daughters in the expected directions {Jacobs & Eccles, 1992). The perceptions and misperceptions of others can affect not just the behavior1 of the target person but also that person's self-concept, the topic to which wc now turn. SECTION SUMMARY As intuitive psychologists, people form impressions of others' personalities by ( •* ing others' behaviors. In line with Kelley's model, distinctive behaviors (those tha ' •' most from the way that a typical person would behave in similar circumstanct w most informative of personality. Nondistinctive behaviors are more reasonably lmm uted to the situation than to anything unique about the person. Often, however, V do attribute nondistinctive behaviors to personality. This person bias, or "runda. I ' attribution error," occurs in experiments in which the subjects' goal is to asse » personality of someone performing an action. When the goal is to assess the sit . ■ i that provokes the action, the opposite bias—the situation bias—occurs, Both oi , biases are especially strong when subjects are kept mentally occupied so the • ' judgments are based on automatic rather than controlled thought processes. Other studies indicate that the person bias is more characteristic of pei •> Western cultures than in Eastern cultures. Moreover, the person bias is weaki ní; the situation bias is stronger, when people make attributions about their own be • ** than when they make attributions about someone else's. This so-called actor-ot it* discrepancy might derive from the more extensive knowledge that people h . uf. themselves than of others, or it might derive from the fact that people's eyes • incused on the person when they watch another perform an action and on the sit *. .n when they themselves perform an action. Preexisting beliefs about a person can influence the way in which that pe 'is, behavior is judged. Physical appearance is one source of such beliefs. For ex< baby-faced people are viewed as more naive, innocent, and incompetent tfa ■ tá mature-faced people, and their behavior is judged in that light First impressio also affected by cultural stereotypes concerning race, gender, age, and othei w categorizing people. Experiments involving the' technique of priming have re< that even people who are not consciously prejudiced carry the culture's racial í types in their heads and that those stereotypes can bias judgments about a pe act/ons. Preexisting beliefs can sometimes result in self-fulfilling prophecies. In one ' iment, teachers were falsely informed that certain of their students had perform ceptionally well on a test of academic potential. The teachers subsequently be' differently toward those students than toward others, in ways that led those sti to achieve more than they otherwise would have. PERCEIVING AND EVALUATING THE SELF Self-awareness is often described as one of the hallmarks of our species, though' difficult to judge the degree to which members of other species may be awr themselves as entities. At about 15 months of age, human infants stop treating" image in a mirror as if it were another child and begin to treat it as a reflccti themselves. If a researcher surreptitiously places a bright red spot of rouge o~ child's nose before placing the child in front of the mirror, the 15-monlli-c sponds by touching his or her own nose to feel or rub off the rouge; a younger Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 509 i _ contrast, touches the mirror or tries to look behind it to find the -nosed child (Lewis & Brooks-Gunn, 1979). The only other animals ides ourselves who have passed the rouge test of self-recognition are the other apes—chimpanzees, bonobos, orangutans, and at least one gorilla and one gibbon (Parker & others, 1994; Ujhelyi & others, '2000). Other animals, including all species of monkeys tested so far, ■continue to treat the mirror image as another animal—a creature to threaten and try to chase away—no matter what their age or how much experience they have had with mirrors. Research with chimpanzees suggests that for them social interaction is crucial for self-recognition. Chimpanzees raised in isolation om others of their kind did not learn to make self-directed responses fp their mirror images, whereas those raised with other chimps did |Gallup & others, 1971). Many psychologists and sociologists have argued that .the f-concept, for humans as well as chimps, is fundamentally a soäal.product. lb be-e aware of yourself, you must first become aware of others of your species and 'Wen become aware, perhaps from how others treat you, that you are one of them. In humans, self-awareness includes awareness not just of the physical self, reflected ä in mirror images, but also of one's own personality and character^reflected psychcf ||gieally in the reactions of other people. It's me By pointing to her own nose as she looks at herself in the mirror, this child demonstrates her understanding that the mirror image is indeed of her. Seeing Ourselves Through the Eyes of Others Many years ago the sociologist Charles Cooley(1902/1964) coined the term looking-glass self to describe what he considered to be a very large aspect of each person's self-concept. The "looking glass" to which he referred is not an actual mirror but other jnepple who react to us. He suggested that we all naturally infer and imagine what Iphers think of us from their reactions, and we use those inferences and images to |pid our own self-concepts. As Eliza Doolittle said to Colonel Pickering in Pygmalim: "You see, really and truly,. .. the difference between a lady and a flower girl is not how she behaves but how she's treated. I shall always be a flower girl to Professor Higgins, because he always treats me as a flower girl, and always will; but I know I can be a lady to you, because you always treat me as a lady, and always will." From a functionalist perspective, it makes sense that our self-concepts should be founded largely on what others think of us. Our self-concepts allow us to predict how others will respond to us so that we can fit into society in ways consistent with others' expectations. A flower girl who thought she was a fine lady but could convince nobody else of it would be in for a hard time. Effects of Others' Appraisals on Self-Understanding The concept of the looking-glass self may help explain the effects that teachers' beliefs have on their pupils' behavior, discussed in the previous section. Children treated as if they have a particular quality may incorporate that quality into their self-concepts and therefore express it more fully in their actions. Studies in which children were asked to describe themselves have confirmed that their self-descriptions change in accordance with evaluations by their teachers and peers (Cole, 1991; Jussim, 1991). Moreover, in experiments where children were told explicitly that they were a certain kind of person, they responded by behaving in accordance with the attribute they were told they had. In one such experiment, some children were told in the course of classroom activity that they were neat and tidy {attribution condition); others were told that they should be neat and tidy {persuasion condition); and still others were given no special treatment {control condition). The result was that those in the attribution condition showed significantly greater gains in neatness, as measured by the absence of littering, than did those in either of the other conditions (Miller & others, 1975). Similarly, children who were told that they were good at math showed greater improvements in math scores than did those who were told that they should try to 14. What evidence in contemporary psychology supports, and delimits, Cooley's concept of the looking-glass self? 510 Part 7 The Person in a World of People I 100 n 60- a 40- 20- Pretest figure 13.6 Effect of attribution compared with persuasion (a) Fifth-graders who were repeatedly told that they were neat and tidy (attribution condition) showed greater gain in use of the wastebasket than did those in the other conditions, (b) Second-graders who were repeatedly told that they were good at math (attribution condition) showed greater improvement in math scores than did those in the other conditions. In each case, the students were tested three times: once right before (pretest), once immediately after (immediate posttest), and once a few weeks after (delayed posttest) the experimental conditions were in effect. (Adapted from Miller & others, 1975.) 15. What might lead us to develop multiple self-concepts, and why might they be useful? 19-1 18- 17- 16- 15- 14- Immediate Delayed Pretest Immediate Delayed posttest posttest posttest posttest (a) — "*™ Attribution condition "■**" Persuasion condition ^^— Control condition (b) become good at math (see Figure 13.6). In these experiments the change in ■ '• *-ior presumably occurred because of a direct effect of the appraisals on the ch s-lr m* self-concepts, which they then strove to live up to. Of course, people's self-concepts are not always as moldable as the expe -i i '. just cited might suggest. The effects are strongest with young children ai ■ ■■". characteristics for which people do not already have firm self-beliefs. Ado" i \ and adults often respond to such appraisals in ways that seem designed to ■ ■ -i i t what they perceive to be another person's misperception of them (Swann, 1 ,,r Pygmalion in reverse. In one experiment, adults who perceived themselves ,■ ■ i"-inant became all the more dominant in their behavior if their conversation ■ ■■ ■ ■ *r initially thought they were submissive, and those who perceived themselves missive became all the more submissive if their partner initially thought they we dominant (Swann & Hill, 1982). As another example, Zebrowitz and her colleagii (1998) have found that baby-faced teenage boys and young men often behave1 ways that seem to be designed to counteract the baby-face stereotype. For instani in World War II and the Korean War, baby-faced soldiers undertook more dan; ous missions and won more military awards, on average, than did mature-faced sô] diers, apparently to counteract others' expectations that they lacked couraj (Collins & Zebrowitz, 1995). Social Roles and Multiple Selves You have read of evidence that the raw material for self-understanding lies in o: social interactions with others. But any given person may have quite different of interactions with different groups. As William James (1890/1950) put it I ago, "Many a youth who is demure enough before his parents and teat ■ ■ and swaggers like a pirate among his 'tough' young friends." From su ■ -I tions^James argued'that each person has not just one self-concept but in iih, ii| corresponding to his or her relationship with a different person or s< i ' n up] Psychologists who emphasize the relationship between self-concepts and sc have expanded on this idea. Each of us plays a number of different roles in i < and we have a somewhat different concept of ourselves associated with eacb I u father, son, neighbor, and college professor to different people. When I think cl i in each role, different sets of traits and abilities come to mind. My understan vi£* myself in each role is mediated partly by the larger society's stereotype of i thers, sons, neighbors, and college professors are like and partly by the specific pectations of the individuals to whom I am those things. Research has shown that people's self-descriptions vary depending on whii of their social roles has been mentally activated. Such work has led to webli Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 511 Hard-working Authoritative lodels of the self-concept, in which some self-perceived raits are attached to specific roles (the nodes in the web) řhiíe others are attached to several or all roles, tying the ^^^^^ odes together (Hoelter, 1985; Rosenberg, 1988). For ex- Employer mple, a person might see herself or himself as authoritative , I. the role of employer, submissive in the role of daughter or on, companionable in the role of wife or husband, and caring rt all these roles. For this person, the trait of being caring ies the roles together and is a source of consistency in the elf-concept (see Figure 13.7). Self-report studies suggest JKt,such consistent traits lead to a general sense of the self Éfecuts across roles, and people feel most comfortable in lose roles where the role-specific self most closely matches the general self Roberts & Donahue, 1994). You might expect that having multiple self-concepts, each associated with a dif-;rent role, would be psychologically stressful, but research suggests that the oppo-ite is more often true (Dance & Kuiper, 1987; Linville, 1985, 1987). Thesense of Ping multiple roles and a wide variety of traits seems to protect a person from derision when one role is lost or diminished in importance—as might happen in a ivorce, or when children grow up and leave home, or when a job is lost. Moreover, Éserise of having many roles and traits to draw, on apparently adds to a person's fidence in his or her ability to handle new situations (Sande & others, 1988). Companionable Witty- Caring' Husband ■ Romantic Son Dutiful Submissive figure 13.7 The multiple nature of the self-concept The self-concept can be represented as a web with different nodes for one's different social roles. Some self-perceived traits may be tied to just one role (such as husband or employer), and others may be tied to several or all of one's roles. comparing and Contrasting Ourselves to Others Ithough our self-concepts are to some degree sketched out for us by others' per-eptions and the roles we occupy in society, we do not passively accept those ketches. Rather, we use the sketches as models, actively selecting from among Igraand modifying, them to build our own self-concepts. As part ofthat process, ře,compare ourselves to others. In perception everything is relative to some frame of reference, and in self-lerception the frame of reference is other people. To see oneself as short, consci-ntious, or good at math is to see oneself as those things compared with other people. i^pVopess of comparing ourselves with others to, identify our unique character-licsJand evaluate our abilities is called social comparison. A direct consequence of ocial comparison is that the self-concept varies depending on the reference fäup, the group against whom the comparison is made. If the reference group gainst which I evaluated my height was made up of professional basketball play-:rSj I would see myself as short, but if it was made up of jockeys, I would see my-elf as tall. In one series of studies that illustrates the role of the reference group, chil-Iren's self-descriptions were found to focus on traits that most distinguished them rom others in their group (McGuire & McGuire, 1988). Thus, children in racially íomogeneous classrooms rarely mentioned their race, but those in racially mixed :lassrooms quite commonly did, especially if their race was in the minority, children who were unusually tall or short compared with others in their group nore frequently mentioned height, and children with opposite-gender siblings nore frequently mentioned their gender than did other children (see Figure 13.8). figure 13.8 Evidence that children define themselves in terms of differences from their reference group As shown here, children were more likely to mention their gender when describing themselves if their gender was in the minority in their household than they were if it was in the majority. (Adapted from McGuire & McGuire, 1988.) 16. What is some evidence that people construct a self-concept by comparing themselves with a reference group and that a change in reference group can alter self-esteem? c 20- bß ß u f "v / -s / a» 15- \ / VBoys /biris Q-'C" \ / s* d X * 10- \S* o> v §a /^ \ u ^^^ \ c *"fc>^ 2 5 - ^**™**,™~*~™^~=*« c "* o a. ír, tí U n u •O -3 U -1 1 1 i Female Equal Male o majority numbers majority Household gender composition 512 Part 7 The Person in a World of People fu "Of course you're going to be depressed if you keep comparing yourself with successful people." Effect of the Reference Group on Self-EvaluaUon The evaluative aspect of social comparison can be chargi I Aldi We are pleased with ourselves when we feel that we n-i ■ >vl reference group and distressed when we don't. A chj'ij uf group, therefore, can dramatically affect our sell-csteen \1 iiiy college students who earned high grades in high,scl."'l when their marks are only average or less compared vh'p tin new reference group of college classmates. Researcher-1 w academically able students at nonselective schools typ-* ill", hat academic self-concepts than do equally able students í 'ii 'lily schools (Marsh, 1991; Marsh & others, 2000), a pli- iimne called the big-fish-in-little-pond effect. William James > ^"t1/! fleeting on extreme instances of selective comparison urns have the paradox of the man shamed to death because n ond pugilist or second oarsman in the world. Thai he h whole population of the globe minus one is nothing; hi self to beat that one and as long as he doesn't do "I counts." In a follow-up of James's century-old idea, Victori Mi iIvl colleagues (1995) analyzed the televised broadcasts of ■■■ ľ''J2 Olympics for the amounts of joy and agony expresscc I bronze medalists after each event. The main finding ^ medalists (the second-place finishers) showed less joy "n than did the bronze medalists (the third-place finisher ■ > defeated. This seemingly paradoxical finding makes sense if \v i groups were implicitly making different comparisons. The sihci i most come in first, so the prominent comparison to them—afLei before—was likely that of themselves to the gold medalists, and i they were losers. In contrast, the bronze medalists had barely group that received a medal at all, so the prominent comparison . likely that of themselves to the nonmedalists, and in that they we s niily r and therefore don't see that others perform better than they do. As Charles rwm (1871/1965) put it long ago, "Ignorance more frequently begets confidence n the reverse." Iturat Dependence of the Self-Serving Bias The better-than-average and self-serving effects just described are so common in North America that some psychologists take their absence as a sign of psychological depression. Depressed people, according to some reports (Peterson & Scligman, 1984), evaluate themselves more accurately than do nondepressed people: If depressed individuals are average at some task, they see themselves as average; if they cause an accident, they blame themselves. As you will discover in I 1 100 n 70- 50-40- Perceived ability Perceived test score Actual test score Bottom 2nd 3rd Top quartile quartile quartile quartile figure 13.9 Perceived and measured logical reasoning ability In an experiment, students took a test of logical reasoning and immediately afterward estimated, on a percentile scale, how well they thought they did on the test and how skilled they were in "general logical reasoning" compared to their classmates. As shown here, those with the lowest scores greatly overestimated their performance and skill. Similar findings occurred in other experiments having to do with sense of humor and grammatical ability. (Kruger & Dunning, 1999, p. 1125.) 514 Part 7 The Person in a World of People 18. Whai evidence suggests that the self-enhancing biases observed in Western culture may not characterize people everywhere? Chapter 15, some personality theorists believe that certain illusions are normal healthy. But what is true in the West is not necessarily true everywhere. Sometim what Western psychologists report to be human nature is not human nature Western nature. Hazel Markus and Shinobu Kitayama (1991) asked universitys dents in the United States and Japan to estimate what percentage of their cl mates had higher intellectual abilities than their own. The average response for Americans was 30 percent, consistent with the better-than-average phenomena but the average for the Japanese was 50 percent, which of course is what d age would have to be for both groups if they were estimating accurately. In another study, Michael Bond and Tak-Sing Cheung (1983) asked ur ■ students in the United States, Hong Kong, and Japan to describe themselv open-ended way by completing 20 statements, each of which began with th 1 am. They then analyzed the statements for evaluative content and found ■ ■ ratio of positive to negative self-statements was nearly 2 to 1 for the Amerii dents, 1 to 1 for the Hong Kong students, and 1 to 2 for the Japanese stud ■ other words, the Japanese students showed a self-effacing bias that was as st the Americans' self-enhancing bias. In Japan—according to sociologists ai chologists—the ideal person is not someone who thinks highly of himself oi I but someone who is aware of his or her deficiencies and is working haul I come them (Heine & others, 1999). 19. What value might lie in our flexible ability to think of ourselves in terms of both personal and social identities? Seeing Ourselves and Others as One: Social Identity You have been reading of evidence that the self-concept is social in thai oil involved in its construction: We see ourselves reflected in others' reaction ■ and we understand ourselves by comparing our properties with those of oi It B the self-concept is social in another sense as well. Others are not just involv« i construction; they are also part of its contents. We describe and think oř oi i not just in terms of our individual characteristics—"I am short,. . . advcntm somewhat shy'—but also in terms of the groups to which we belong and uitl we identify—"I am a French Canadian, . . . Roman Catholic, . . . wanbci University Marching Band." Self-descriptions that pertain to the per^n i- > s.|i individual are referred to as personal identity, and those that pert "u inihi categories or groups to which the person belongs are referred to a \miah& (Tajfel, 1972). Adult self-concepts are relatively consistent from situation to siin.UK i they are not rigid (Oakes & others, 1994). We think of ourselves differently i ■! ferent times, in ways that help us meet the ever-changing challenges of soc ' I Sometimes, for some purposes, we find it most useful to think of our uniqat i n erties and motives; other times, for other purposes, we find it most useful t( of ourselves as interchangeable components of a larger unit, the group, i >ni i tion as a social species entailed a continuous balance between the need i»,«." and assert ourselves as individuals and the need to cooperate wmVotn i\ may have selected for a capacity to hold both personal and social uko (Guisinger & Blatt, 1994). In evolutionary history the groups with which -erated included some that may have been lifelong, such as the family and triL others that were more ephemeral, such as a hunting party organized to track1 a particular antelope. Today the relatively permanent groups with which wé rify may include our family, ethnic group, religious affiliation, and occupa colleagues. The temporary groups include the various teams and coalitkr which we affiliate for particular ends, for periods ranging from minut * ■■ When we see some interest in common, we can be remarkably adept ni '• m. ii our differences and thinking of ourselves and our group-mates as one (a i ■ lj sued more fully in Chapter 14). Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 515 elauonships of Social Identity to Self-Esteem I- dings about ourselves depend not just on our personal achievements but also ■i 11 achievements of the groups with which we identify, even when we ourselves pi ■• ľ de or no role in those achievements. Social psychologists have found, for ex- that sports fans' feelings about themselves rise and fell as "their" team wins '■I' ■ es (Hirt & others, 1992). Similarly, people feel good about themselves when i own, university, or place of employment achieves high rank or praise. i- some situations, the very same event—high achievement by other members n' ■■ i group—can temporarily raise or lower our self-esteem, depending on um iľ t our social identity or personal identity is most active. When our social 11«" ŕ predominates, our group-mates are part of us and we experience their suc- ours. When our personal identity predominates, our group-mates are the ■ ■■ ce group against which we measure our own accomplishments, so their suc- i ly diminish our view of ourselves. Social psychologists have demonstrated „ J these effects by priming people to think in terms of either their social or eir personal identities as they hear of high accomplishments by others in their 0" i Brewer & Weber, 1994). You read previously that students at highly selec- p. i tools think worse of themselves as scholars than do equally able students at ictive schools because of the difference in their relative standing in their ref- encc group. A subsequent study indicated that this is true for those who think rimai íly in terms of their personal identities but not for those who think primarily terms of their social identities (McFarland & Buehler, 1995). When the social identity is foremost, self-feelings are elevated, not diminished, by evidence of -oup-mates' excellent performances. Graceful winners of individual achievement wards know this intuitively, and to promote good feelings, they activate their group-mates' social identities by describing the award as belonging properly to the hole group. Other studies reveal that the better-than-average phenomenon and the self- erving attributional bias apply at least as much to our judgments about our groups as to those about ourselves as individuals (Rubin & Hewstone, 1998). In fact, the ""ncept of social identity first became prominent in social psychology when Henri jfel (1972, 1982) used it to explain people's strong bias in favor of their own oups over other groups in all sorts of judgments. He argued that we exaggerate c viimes of our own groups to build up the part of our self-esteem that derives )in our social identities. Tajfel and others showed that the bias is so strong that we ply Jt even when we have no realistic basis at all for assuming that our group dif- ■s horn another. In one laboratory experiment, people who knew they had been -igncd to one of two groups by a purely random process—a coin toss—neverthe- ■s rated their own group more positively than they did the other group (Locksley 20. How does the distinction between social identity and personal identity help explain (a) the two opposing effects that our group-mates' excellent performance can have on our self-esteem and (b) extensions of the self-serving bias to attributions about our group-mates? 'i-m KV ijfASirt TO *f«w4iíSr&=__ ojea.it ir fM«t. 516 Part 7 The Person in a World of People 21. How does Triandis characterize individualist and coUectivist cultures, and what differences have been found between the two in people's self-descriptions? Walking a line between two cultures Masako Owada was born in Japan, attended a public high school in Massachusetts, graduated from Harvard in 1985, distinguished herself as a brilliant executive in Japanese foreign trade, and then, in 1993, after much hesitation, accepted the marriage proposal of Japanese Crown Prince Naruhito. Her new role required that she give up much of her Western independence and refrain from expressing her own ideas direcdy and publicly. & others, 1980). You don't need a degree in psychology to know il ii llife posing baseball teams see the same plays differently, in ways th-i il .w leave the game believing that theirs was the better team, regai- ■ ■>■• (if Strike three is attributed by one group to the pitcher's sparkling 11- I'.'U other to the umpire's unacknowledged need for eyeglasses. ■tiuh I --LI Cross-Cultural Differences in the Balance of Social identity and Personal Identity Although both personal and social identities exist among peoplt * two aspects of the self are differentially strengthened or weaken- ■ tures. Harry Triandis (1995), one of the pioneers of cross-i if i chology, distinguishes between individualist cultures, wjii Ii siiu identities, and coUectivist cultures, which strengthen social iaéniiiiLs In ' tures predominate in western Europe, North America, and Aus , a sophical and political traditions-emphasize personal freedom, *■■ lt-d and mdividualcompetition. CoUectivist cultures predominate in '• ■ iul Africa and Latin America, where philosophical and political ti In ,the inherent connectedness and interdependence of people wiilim y the family, workplace, village, and nation. Whereas people in ii ' ■ ui tend to define their lives in terms of self-fulfillment, those in colle ■■»-= tend to define theirs in terms of fulfilling their duties to, and promot .r of, the groups of which they are members. Each type of culture has its benefits and costs from the pcrspeci .* ual psychology. Individualist cultures may foster personal freedom a- I ■ a cost of loneliness and insecurity, and coUectivist cultures may Ibstei i ■ longing and security at a cost of reduced individual initiative n i CoUectivist cultures typically have less conflict within groups but mou tween groups than is the case for individualist cultures (Triandis, 19""i In where group identities are strong, the distinction between "wc" i greater than in cultures where group identities are weak. Consistent with Triandis's view are numerous studies indicating ľ Eastern cultures such as Japan, Korea, China, and India describe thci" entry than do people in Western cultures (Markus & Kitayama, 19"'l former describe themselves more often in terms of their social group nul less often in terms of consistent personality traits that cut across th i roles. Asked to describe themselves, they are more likely to make su 1 as "I am a student at University X" or "I am the eldest daughter in j .' m less likely to make such statements as "I am easygoing" or "I am am n: i are also more likely to attribute their achievements to their group i nil themselves as individuals (Chen & others, 1998). When they do il ■ ~i' selves with personality traits, they typically delimit the traits to p contexts—"I am easygoing with my friends," "I am ambitious at w 1989). William James's idea that people have multiple self-concepts i multiple roles seems to be even more true of people in coUectivist (■ those in individualist cultures. .in] i ires ^SECr/ON^SaMMARY^ Our self-concepts are social constructs. We acquire them in part from oV> actions to us and beliefs about us. For that reason we may have mu/t/p/* each related to a different social role and a different group with which a also construct aspects of our self-concepts by comparing ourselves to • see ourseives as good or bad at a given task depending on the abilil '•( ence group with which we compare ourselves. People in Western cultures tend to have inflated views of themseli*-non that may be explained in part by biased feedback from others, by f definitions of success, by the self-serving attributional bias, and by tht Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 517 npetent to judge their own incompetence. Some studies indicate that such self- |t/on does not occur in Asian cultures, perhaps due to cultural traditions that pro- • e communal, less individualistic outlook. • 'y culture, people describe themselves partly in ways that emphasize their bn> *Urf (1992) surveyed thousands of primary and secondary school teacher; in 2(11 tries around the globe. He chose teachers because they are literate (a -I >> lüj questionnaires) and because he assumed that they, as conveyors tsf v ■■■! ■• u>^ dren and adolescents, would hold values representative of the culture :■■ íi-y* teacher filled out a form rating each of 56 values in terms of its imp- ■! unn, j guiding principle" of his or her life. Schwartz then analyzed the rcspoi .* I» the degree to which each value correlated positively or negatively wii i h i value in the list. (Two values correlate positively if people who rate on ľuli q( tend to rate the other in the same direction. Iwo values correlate negí i »if{ pie tend to rate them in opposite directions.) The final result of this n iří the wheel-like structure of values portrayed in Figure 13.10. 1 he v | mented into 10 value categories, each representing a cluster of values 11 rcorfl strongly and positively with each other. The categories are arranged »■> one is flanked by the categories that correlate most positively with it ai Hum site those that correlate most negatively with it. The figure tells you, ■ that self-direction correlates most strongly in a positive direction with . and stimulation (which flank self-direction in the wheel) and most strom I tive direction with security, conformity, tradition, and power (which lie |> direction). This general pattern of correlations occurred within each o" ■■ ľ m Schwartz studied, suggesting that the pattern is universal. To explain the universality of the value wheel, Schwartz (1992,19'-that the value types derive from basic human social needs and moliv logically to one another as similars or opposites, as depicted by the wh the pattern for yourself, make your way around the wheel and consii i 'io value could promote a person's survival and how it either supports or c values in the wheel. For example, a decision to act in accordance with ■ l'-i! figure 13.10 The value wheel The universal pattern of values proposed by Schwartz is based on the degree to which values support or oppose one another. Each value category occupies a wheel segment adjacent to the categories that correlate most positively with it and opposite those that correlate most negatively with it. Conformity and tradition occupy the same segment because they each correlate about equally strongly with benevolence and security. The value categories are further grouped into four clusters that define two dimensions of the wheel: self-enhancement versus self-transcendence and openness to change versus conservation. Hedonism is marked off with dashed rather than solid lines because it straddles the openness-to-change and self-enhancement clusters. (Based on Figure 1.1 in Schwartz, 1996, p. 5.) Openness to change —— Sell-direction (heetlom, independence, curiosity, creativity) ---------VII ii ubít l„j „ L _ (equality, justice, tolerance, preservation of nature) Stimulation (excitement, novelty, challenge) Hedonism \ (pleasure, enjoyment * -flife) Sili-iiilnnHöK-ii Achievement (competence, ambition, success) Power (status, authority, wealth) Benevolencí (helpfulness., responsibility, friendship, love) Conformity (politeness, ^obedience, honoi of elders) 1> h i (Ju i iih , a 11 ■■■(.] n'ijui» Taiwan, gave the highest rankings to values that fall under security and con-rmity (conservation), whereas those in individualist cultures, such as New aland, placed relatively more weight on self-direction and stimulation (openness change). But contrary to what you might predict, collectivist and individualist hires did not differ reliably on the dimension of self-enhancement versus self-cendence. This finding is consistent with other studies suggesting that people collectivist cultures value helping others as a duty arising from tradition and so-'ál norms, whereas people in individualist cultures value helping others as a free, rsonal choice (Markus & Kitayama, 1994; Moghaddam & others, 1993). 'alues as Predictors of Actions bu certainly can't predict everything people will do by knowing their stated values, utyou can make some general predictions that are reliable on a statistical basis. In idies of Israeli citizens, Schwartz (1996) found that people's relative ratings of the tlue types shown in Figure 13.10 correlated significantly with measurable aspects " their behavior. People who rated the openness-to-change values highest most ■.ten supported liberal political candidates, and those who rated the conservation i dues highest most often supported conservative candidates. In a laboratory game I'tring people's tendency to cooperate against their tendency to compete, people *ho rated the self-transcendence values highest chose most often to cooperate with ■ ihers, and people who rated the self-enhancement values highest chose most often ■ »compete. In research that preceded Schwartz's, Milton Rokeach, one of the pioneers in i'-e study of values, found similar correlations between people's relative ranking of irnculär values and their real-world behaviors. For example, he found that white ■ illege students who ranked equality especially high were more likely than other * hite college students to make eye contact when speaking with a black person, to in a political group supporting equal rights, and to participate in a rally for equal i $hts (Rokeach, 1980). 24. What is some evidence that people's values can be used to predict their behavior? Attitudes as Social Norms \ dues and other attitudes are properties not just of the individual person but also ■■I the social groups to which the person belongs. It is not hard to think of many asons why an interacting group of individuals would have attitudes in common. ■' 'ople living in the same conditions and communicating with one another are ex- tsed to similar information, are subject to the same persuasive messages, may have ■me together partly because of preexisting similar attitudes, and may have modi- d their attitudes to gain greater acceptance in the group. The best-known study of the effects of a social group on attitude change was . dated in 1934, in the midst of the Great Depression, by Theodore Newcomb 1 ?43) at Bennington College in Vermont, which then was a small women's college. nnmgton at that time had a politically liberal faculty but drew most of its stu-: nts from wealthy, politically conservative families. Most first-year students ired their parents' conservative views, but with each year at the college they be-|tame more liberal. In the 1936 presidential election, for example, 62 percent of the 25. How did a long-term study at Bennington College and follow-up studies of Bennington graduates illustrate the role of social forces in attitude change? The Person in a World of People first-year students, 43 percent of the sophomores, and only 15 percent niors and seniors favored Alf Landon, the conservative Republican, over Democrat, Franklin Roosevelt. By the time the first-year students beca and seniors, they too had become politically liberal. Certainly the economic crisis, the Depression, played a role in d change that Newcomb observed, but more directly influential w ere the pressed by the dominant members of the college. People who occupio prestigious positions—the faculty, older students, and leaders of various camp. ganizations—were politically liberal, and new students could gain social accep' by shirting their expressed attitudes in that direction. In interviews, many stií said that at first they expressed liberal attitudes at least partly to m< i ■ i i ' gain prestige, but over time the attitudes became part of their pri - iH public ways of thinking. The relatively few students who rem.iine ■ i rv throughout their 4 years said that they felt socially isolated and no i m y the college community. In follow-ups, 25 years after the initial sind ■ m I years after that, the Bennington graduates whom Newcomb had fir;....... located and interviewed again (Alwin & others, 1991; Newcomb & '■ i I The follow-up studies showed that most of the graduates reiajni.' ' ■ ' views throughout their lives. Their self-reports suggested that ihev ■ i ■,1 eral at least partly because they continued to associate primarily with ■ views were like their own. Other researchers have focused on regional differences and gen i.i -i I ferences in attitudes (Cohen, 1996; Duncan & Agronick, 1995; Schnn ■"» 1989). Such work suggests that lifelong attitudes tend to jell dnrin hood. Young adults who experience similar events and a shared soci:' to help them interpret those events tend to acquire and maintain sin ■■ Attitudes as Rationalizations to Attain Cognitive Consisten A century ago, Sigmund Freud began developing his contro\ers ■ ■ human beings are fundamentally irrational. What pass for reason' ■ ■■ Freud, are most often rationalizations designed to calm our anxieties n ' !■■ self-esteem. You will read more about Freud's view in Chapter 15./ i ■■ . ■ ate view, to be pursued now, is that we are rational but the machi ner; ' ■ r so is by no means perfect. The same mental machinery that producer i _■ un duce pseudo-logic. In the 1950s, Leon Festinger (1957) proposed what he called tht u »Kim sonance theory^ which ever since has been one of social psycholog)- n '« ideas. According to the theory, we have, built into the workings <>■ mechanism that creates an uncomfortable feeling of dissonance, or h< 1 when we become aware of some inconsistency among the various au ■' and items of knowledge that constitute our mental store. Just as tht ' hunger motivates us to seek food, the discomfort of cognitive disson i 3JsjoseeVwaysto_ resolve contradictions or inconsistencies among mi ■' Such a mechanism colHa^veirhave~evoived to serve adaptive fimci i i r i logic. Inconsistencies imply that we are mistaken about something ■■ 1 '■ can lead to danger. Suppose you have a favorable attitude about si ■ '■ iii you learn that overexposure to the sun's ultraviolet rays is the leadin cancer. The discrepancy between your preexisting attitude and yoi i * edge may create a state of cognitive dissonance every time you think ' ■■ the sun. To resolve the dissonance, you might change your attitii bathing from positive to negative or you might bring in a tin "Sunbathing is relatively safe, in moderation, if I use a sunscreen loti As with all of our psychological machinery, our dissonance-rei nism does not always function adaptively. Just as our hunger can ■ things that aren't good for us, our dissonance-reduction drive can li Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 521 Educe dissonance in illogical and maladaptive ways. Those are the effects that jjcularly intrigued Festmger and many subsequent social psychologists. hing Dissonant Information ice heard a person cut off a political discussion with the words, "I'm sorry, but I use to listen to something I disagree with." People don't usually come right out ftay that, but have you noticed how often they seem to behave that way? Given ioicc of books or articles to read, lectures to attend, or documentaries to watch, iple generally choose those that they believe will support their existing views, t observation is consistent with the cognitive dissonance theory. One way to [id dissonance is to avoid situations in which we might discover facts or ideas i run counter to our current views. If we avoid listening to or reading about the fence that ultraviolet rays can cause skin cancer, we can blithely continue to jay sunbathing. People certainly don't always avoid dissonant information, but a iiderable body of research indicates that they very often do (Frey, 1986; Jonas & |bs, 2001). iPaiiJ Sweeney and Kathy Gruber (1984) conducted a study during the 1973 ate Watergate hearings that documented this phenomenon. (The hearings unci illegal activities associated with then-president Richard Nixon's reelection [ i i against George McGovern.) By interviewing the same voters before, iL, u d after the hearings, Sweeney and Gruber discovered that (a) Nixon sup-tt i- i. jided news about the hearings (but not other political news) and were as ohli. ■« ipportive of Nixon after the hearings as they were before; (b) McGovern [»IM i eagerly sought out information about the hearings and were as strongly ■) Nixon afterward as they were before; and (c) previously undecided vot->l'i I i'ioderate attention to the hearings and were the only group whose attitude :d Nixon was significantly influenced (in a negative direction) by the hearings. insistent with the dissonance theory, all but the undecideds approached the igs in a way that seemed designed to protect or strengthen, rather than chal-their previous view. jrm/ng Up an Attitude to Be Consistent with an Action : make most of our choices in life with less-than-absomte certainty. We vote for I candidate not knowing for sure if he or she is best, buy one car even though some §f the evidence favors another, or choose to major in psychology even though some |dier fields have their attractions. After we have irrevocably made one choice or he other—after we have cast our ballot, made our down payment, or registered for pur courses and the deadline for schedule changes is past—any lingering doubts be discordant with our knowledge of what we have done-, so, according to : the cognitive dissonance theory, ■we s\vou\dbe motivated to set them aside. 26. How does the cognitive dissonance theory explain people's attraction to some information and avoidance of other information? 27. How does the cognitive dissonance theory explain why people are more confident about a choice just after they have made it than just before? Dissonance abolished This couple may have agonized long and hard before signing on the dotted line, but once they signed, they set their doubts aside and focused on the positive qualities of their new home. According to the cognitive dissonance theory, people are generally more confident about the correctness of their choices after those choices are made than before. 522 Part 7 The Person in a World of People 28. How does trie cognitive dissonance theory explain why people who behave in a manner contrary to their attitude are likely to change their attitude? A number of studies have shown that people do tend to set tri' ii iluti after making a decision. Even in the absence of new information,1 myl? become more confident of their choice after acting on it than they1 * t it 1 example, in one study, bettors at a horse race were more confident i nl would win if they were asked immediately after they had placed inir they were asked immediately before (Knox & Inkster, 1968). In an -'inr ers who were leaving the polling place spoke more positively abo': ilitir-candidate than did those who were entering (Frenkel & Doob, 19j '■' ■L uid i In./ ■nu 1. J Hill ■ i íphjr ■Ute . ..f 1 '.UlCui "ito1 lllihli ■ iiiytft ■ in nic »ill ,* i-1. A ■ M'lWC , i d. i- Ins Changing an Attitude to Justify an Action Sometimes people behave in ways that run counter to their attiti faced with the dissonant cognitions, "I believe this, but I did that." their deed, but they can relieve dissonance by modifying—maybe < their attitude. More than 200 years ago, the great inventor, states! of practical psychology Benjamin Franklin recognized this pheno* it to his advantage. Franklin (1818/1949) describes in his autobi" changed the attitude of a political opponent who was trying to bl ment to a high office: I did not like the opposition of this new member, who was a gentleir and education with talents that were likely to give him in time great i I did not, however, aim at gaining his favour by paying any servile r< but after some time took this other method. Having heard that he had a certain very scarce and curious book, I wrote a note to him express ■ of perusing that book and requesting he do me the favour of lending few days. He sent it immediately; and I returned it in about a week note expressing strongly my sense of the favour. When we next met he spoke to me (which he had never done before), and with great d ever afterwards manifested a readiness to serve me on all occasions, came great friends, and our friendship continued to his death. This. stance of the truth of an old maxim I had learned, which says, "He done you a kindness will be more ready to do you another than he wl self have obliged." According to the cognitive dissonance theory, what might h . iia the former opponent's mind to change his attitude toward Franlil n: lie Franklin's request to borrow the book, and for reasons of which in may been fully aware, such as the simple habit of courtesy, he did not I ■ i ii it J once he sent the book to Franklin, he was thrown into a state oi m-iul' nance. One thought, / do not like Ben Franklin, was discordant wit' iin»thc just lent Franklin a very valuable book. The second of these could no< ■ Uini that was objective fact, so dissonance could best be relieved by chán j l^ rlw Franklin isn't really a bad sort. At least I know he's honest. If he weren't • '■».*,/ wouldn't have lent him that valuable book. Such thinking reduced or i *.i nance and set the stage for new, friendlier behaviors toward Franl h i "ii Notice that, according to this analysis, the man changed hi' ■■■ilude Franklin because he saw the decision to lend the book as his own c i no good reason why he should have made that choice if he didn'i Franklin had paid him or threatened him to get him to lend the 1 ■■ of it would not have created dissonance with the belief that he disli would say, "I lent the book to Franklin only because he paid me" ■ >\ cause he threatened me." Since either of these would have been si I ľ k iií j tion, no dissonance would have resulted and no attitude change '■ \ without the change, the person cannot justify his or her already compku. Many dozens of experiments have demonstrated this effect and havefhJ|i tify the conditions required for its occurrence (Harmon-Jones & I...I-1 'I- lilii I Irr Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 523 onditions That Optimize the Insufficient-Justification Effect )ne requirement for the msurncient-justification effect to occur is that there be no uvious, high incentive for performing the counterattitudinal action. In an early demonstration of this, Leon Festinger and James Carlsmith (1959) gave college stu-ents a boring task (loading spools into trays and turning pegs in a pegboard) and en offered to "hire" them to tell another student that the task was exciting and en-tle. Some students were offered $1 for their role in recruiting the other student, d others were offered $20 (a princely sum at a time when the minimum wage in e United States was $1 an hour). The result was that those in the $1 condition |hanged their attitude toward the task and later recalled it as truly enjoyable, whereas äiose in the $20 condition continued to recall it as boring. Presumably, students in áe$l condition could not justify their lie to the other student on the basis of the little they were promised, so they convinced themselves that they were not lying. Those in the $20 condition, in contrast, could justify their lie: "I said the task was enjoyable when it was actually boring, but who wouldn't tell such a small lie for $20?" Another essential condition for the insufficient-justification effect is that subX ^ jects must perceive their action as stemming from their own free choice, y* ' Otherwise, they could justify the action—and relieve dissonance—simply by say-/ ing, "I was forced to do it." In one experiment demonstrating the requirement of free choice, students were asked to write essays expressing support for a bill in the state legislature that most students personally opposed (Linder & others, 1967). Students in the. free-choice condition were told clearly that they didn't have to write the essays, but they were encouraged to do so and none refused. Students in the no-^ choice condition were simply told to write the essays, and all complied. After writing the essays, all students were asked to describe their personal attitude toward the \ bill. Only those in the free-choice condition showed a significant shift in the direction of fevoring the bill; those in the no-choice condition remained as opposed to it I as did those who had not written essays at all. Using essentially the same procedure, other researchers found that students in the free-choice condition not only changed their attitude more than did those in the no-choice condition but also manifested more psychological discomfort and physiological arousal as they wrote their essays, a finding consistent with the view that they were experiencing greater dissonance (Elkin & Leippe, 1986; Elliot & Devine, 1994). 29. How have researchers identified three conditions that increase the likelihood that the insufficient-justification effect will occur? By OffifiMtms linr 5H0ULD I HIRE YOU AS r\r CONSULTANT? BNT ITSTRANOE THAT YOU HAVE THIS DEAD END JOB WHEN YOU'RE TUICE A5 5HARTA5 YOUR BOSS? I'LL USE/TT SPECIAL PROCESS OF COGNITIVE DISSONANCE TO IHPROVE EMPLOYEE HORALE. THE HOURS ARE L0NG> THE FAY IS HEDIOCRE, NOBODY RESPECTS YOUR CONTRIBUTIONS, AND YET YOU FREELY CHOOSE TO IU0RK HERE IT'S ACKURDI N0> WAIT... THERE Í1U5T BE A REA50N ... I MUST UJORK MERE BECAUSE I LOVE THE UJ0K.K.J ^... S-1 WHEN PEOPIE ARE IN AN ABSURD SITUATION. THEIR rUND5 RATIONALIZE IT ĎY INVENTING A COI1F0RTABLE ILLUSION. 524 Part 7 The Person in a World of People 1. Research also indicates that the insufficient-justification eflect is st when the action to be justified would, from the viewpoint of the original a be expected to cause harm to others or to oneself. In this case, the thought lying the attitude change might be, "I would not deliberately do something ful; therefore I must believe that what I did is helpful." In one expcrimi example, students who wrote essays counter to their initial attitude showed j attitude change if they were led to believe that their essays could influence than if they were led to believe that their essays could not influence policy & Cooper, 1989). Other experiments, however, have shown that some clef dissonance-induced discomfort and attitude change can occur even \» ■ ul'i know that their jx)unterattitudinal statements will be immediately cl n I ■! nobody will know what they had written (Harmon-Jones, 2000). Taking all this research into account, modern-day Ben Franklins > 'i ■ change someone's attitude by inducing the person to behave in a way il ii <«ů diets the old attitude are most likely to succeed if they (a) minimize i. uln ^incentive for the behavior, (b) maximize the appearance of free ch" • in ^y f choose a behavior that would seem harmful if viewed from the persp i '.i o \pld attitude. Using Attitudes to "Justify" Injustice Perhaps we develop unrealistic attitudes not only to explain our oiyn ľ'li inexplicable actions but also to make sense of the chaos and injustice v>- '" the world around us. Melvin Lerner (1980) summarized evidence thí' t our culture, people tend to believe life is fair, a tendency he labeled th /» a í bias. In line with Freud's general way of thinking, Lerner suggested th;ii *ť ^ life is fair because to believe otherwise would induce more anxiety dian vti-erate. Unconsciously we may reason: "If life is not fair, then no i ■■ i h have, or how worthy I am, something terrible could happen to m I ml idea, so life must be fair." The just-world bias may motivate us i » il take precautions that indeed do promote our survival and well-fa i ' uiď reason it may be adaptive. However, it may promote truly unfair ui'i^s toward people who surfer from mistreatment or misfortune. lb maintain the illusion that life is fair and predictable, we n ii ■ I ^u planations of others' misfortunes to make it seem that people desí > «hat to them, a phenomenon called blaming the victim (Lerner & Gold ■ 1971). If millions of Jews were killed in the Holocaust, they must i i>l J thing to bring it on or failed to do something to prevenf it. I f bia- \int fer from poverty and discrimination, it must somehow be ihcir &'■!■ If suffer from discrimination, harassment, and AIDS, homosexualit, i"»* my neighbor contracts Lyme disease, she was foolish to have go i \ .< woods. Numerous studies have shown that victims of rape, robbc i. 't. cidents, illnesses, poverty, and social injustice often suíícr doubl i. "ii misfortune itself and again from the subtle or not-so-subtle hl m » ' from others for "causing" or "allowing" the misfortune to ha lj«.a Miller, 1978; Maes, 1998). In one laboratory experiment, collegi ^fld ated another student's character less favorably if they believe ceive a series of painful electric shocks as part of the experi they believed she wasn't (Lerner & Simmons, 1966). Attitudes as Guides to Action As you have seen, attitudes can help us define ourselves, ad ľ t loa and rationalize our actions or beliefs. But common sense tells ■ - ihu function of attitudes is to guide our behavior effectively. Pr. -> n ibl those objects, people, events, and activities about which we li n* p and avoid those about which we have negative attitudes. Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 525 Social psychologists first became interested in attitudes because they conceived of them as mental guides of behavior (Allport, 1935). But early research on the topic: was most remarkable for its failure to find reliable relationships between attitudes and actions. In one classic study, for instance, students in a college course filled out a questionnaire aimed at assessing their attitudes toward cheating, and later in the semester they were asked to grade their own true-false tests (Corey, 1937). The tests had already been graded secretly by the instruc-toi. so cheating could be detected. The result was that a great deal of cheating occurred and no correlation at all was found between the attitude measure and actual cheating. A strong correlation was found, however, between cheating and the student's true score on the test: The lower the true store, the more likely the student was to try to raise it by cheating. Subsequent studies, however, have revealed many examples of reliable coi relations between attitudes and behavior. Among these are Schwartz's study correlating people's expressed values with their subsequent choice to cooperate or compete and Rokeach's study correlating privileged people's ranking of equality as a value with their subsequent actions toward members of a discriminated-against group. Today essentially all social psychologists agree that attitudes do play a role in people's behavioral choices and the question is, "When, or under what conditions, do they play that role?" 30. How did an early study demonstrate attitude-behavior inconsistency? Attitudes Must Be Retrieved from Memory to Affect Behavior Attitudes, like any other cognitions, are stored in long-term memory and can influence a person's behavioral choice only if recalled into working memory at the time the choice is made. Behavior occurs in a continuous flow, and we rarely stop to think about each of our relevant attitudes before we act. In Corey's experiment on cheating, the students who cheated may have been immediately overwhelmed by their poor performance on the test, which reminded them strongly of their negative attitudes toward failing a course but did not remind them of their negative attitudes toward cheating. Perhaps with more time to think about it, or with more immediate inducement to think about it, fewer would have cheated and a correlation would have been found between their anticheating attitudes and their behavior. This line of reasoning suggests that cues reminding a person of his or her attitudes at the time of action would increase the attitude-behavior correlation. Indeed, experiments have shown that if people are presented with a task that requires them to think about their attitude on an issue shortly before they must act on it, the correlation between the attitude and the behavior increases markedly (Aronson, 1992; Snyder & Swann, 1976). Other experiments have shown that the presence of a mirror can promote attitude-behavior consistency (Wicklund & Frey, 1980). Apparently their physical reflection reminds people of all aspects of themselves, including their central attitudes, or values. In one experiment, for example, trick-or-treaters were told to help themselves to a specific amount of candy, under conditions in which they believed nobody could see them. They took extra candy more often if no mirror was present than they did if a mirror was present behind the candy bowl (Beaman & others, 1979). Russell Fazio (1986, 1990) has argued that the strongest attitude-behavior correlation occurs when the attitude is acquired through direct experience with its object, because then the object automatically reminds the person of the attitude. If you have a negative attitude toward bacon because on one or more occasions you ate it and got sick, then the sight and smell of bacon will automatically elicit your negative attitude and you won't eat it. However, if you have a negative attitude toward bacon only because you read that it is high in nitrates and nitrates are bad for you, your attitude will not be elicited automatically by the sight and smell of bacon. In that case you will need some other cue, extrinsic to the bacon, to remind you of the attitude, or you will need to rely on a habit of checking your set of food-related attitudes Irbu promised yourself you'd do something, someday. S=ľ:i Promise kept. Yes, III hdp needy children ihrough sponsorship. mih ííkI J k« of tovc. Thai' ii be a íia™ the (Jtífdfcn íponawr lake much *orfd.JiBrj24 aSi you need Sponsorship b a £irai wjy io iho*v your J íůfKím and hdp bring poucivc, listing change J ro itisadwuttiigd chüdrun. Your comribiirjcuu É willbccornbiixd wiihihoKofnjihťripoiiwn L to crcai tí: about the decision-making process, IcekAjzen (1985,1991) proposal -ilm lw a theory of planned behavior. Planned behavior is defined as thai w ■■ i .o1 from one's conscious intention to behave in a particular way. Accordin.' tlii ory, in developing a behavioral intention, people take into account t! three types of cognitions: (a) their attitude, defined as their personal i ■ have that way or not; (b) the subjective norm, defined as their belief abon *li t ers who are important at the moment would think about the action; un! ■«.) perceived control^ defined as their sense of their own ability or inability i ■ ' the action (see Figure 13.11). Suppose you have a strongly favorable an ■ V becoming an astronaut but fail to act on it because others in your ■ 1 would think you strange for pursuing that line of work or because yo' I n you have the ability for it or could raise the funds needed for training. \ » r ill in that case, is controlled by the subjective norm or by your perceived i nf trol, despite your attitude. figure 13.11 A theory of planned behavior According to Ajzen's theory, the decision to behave in a certain way is a product of three categories of cognitions. (Adapted fromAjzen, 1987.) Perceived behavioral control Research based on Ajzen's model has shown that, depending on factors, any of the three inputs to behavioral intention may account lb variability among people in their behavior. In a study of dieting and wt ■■ instance, perceived control (confidence in the ability to stay on a diet) predictor of weight loss than was either attitude (desire to lose weight ■ ■ jective norm (belief about whether others thought one should lose w« (Schifter & Ajzen, 1985). In a study of decisions about how much tii studying, some students were most influenced by the subjective norm ( what other students believe is an appropriate amount of study time) wli were most influenced by their own attitudes (Miller & Grush, 1986). \ ingly, a number of studies have shown that the subjective norm contrib behavioral intention among people who identify strongly with the reft than among those who don't identify strongly (Terry & others, 1999, ?'• in Míľ1 ■n ■»: . ,i i, Nit the mr "i n ■so" Attitudes as Products of Information Processing In the previous discussions you have already encountered some ide* I m origins of attitudes. To some degree we inherit attitudes from our ;■ ■■ m'|I environment. To some degree we manufacture attitudes to create i lu-m consistency. To the degree that we use attitudes for utilitarian fi guide our behavior toward or away from particular objects and even sense that we would also construct attitudes from actual informatioi us about those objects and events. In some cases this construction pie Chapter 13 SOCIAL PERCEPTION AND ATTITUDES 527 tic, engaging no or little conscious thought, and in other cases the process is ply thoughtful. ííucfes Through Classical Conditioning: No Thought ~ical conditioning, a basic learning process discussed in Chapter 4, can be üght of as an automatic attitude-forming system. A new stimulus (the condi-. stimulus) is paired with a stimulus that already elicits a particular reaction ^unconditioned stimulus), and, as a result, the new stimulus comes to elicit, on "e presentations, a reaction similar to that elicited by the original stimulus, the language of the present chapter, we can say that Pavlov's dog entered .experiment with a preexisting positive attitude toward meat powder. When lov preceded the meat powder on several occasions with the sound of a bell, dog acquired a positive attitude toward that sound. The dog now salivated and ď its tail when the bell rang, and if given a chance, it would have learned to le bell itself. The neural mechanisms underlying classical conditioning evolved, presumably, use in evolutionary history such conditioning produced adaptive reactions more n than maladaptive ones. Conditioning leads us to approach objects and events ave been linked in our experience to pleasant, life-promoting occurrences, and eads us to avoid objects and events that have been linked to unpleasant, life-"atening occurrences. In today's world, however, where many of the stimulus we experience are the creations of advertisers and others who want to manipu-iis, classical conditioning can have maladaptive consequences. All those ads in h cigarettes, beer, and expensive gas-guzzling cars are paired with beautiful je, happy scenes, and lovely music are designed to exploit our most thoughtless rude-forming system, classical conditioning. The advertisers want us to salivate, our tails, and run out and buy their products. irendy the technique works; if it didn't, they wouldn't tinue spending money on such ads. Laboratory experi-ts have shown that it is easy to condition positive atti-s, in university students, toward a fictitious brand of tithwash, or other product, by pairing the brand with sant, though irrelevant, scenes (Grossman & Till, 1998). sJDirect evidence that classical conditioning can create 'tudes in the absence of conscious thought comes from eriments in which the unconditioned stimulus is pre-ted too rapidly for conscious detection. In one such ex-jiment, subjects viewed slides of a woman engaged in "Ous mundane activities, such as shopping. Each slide preceded by a quickly flashed scene (9 milliseconds' du--on) designed to induce either a positive or a negative otional reaction. Among the positive scenes were kit's, a bridal couple, and a child with a Mickey Mouse doll. the negative scenes were a werewolf, a bucket of akes, and a depiction of open-heart surgery. After all the Her royal pleasure John B. Watson, the founder of behaviorism (discussed in Chapters 1 and 4), became an advertising specialist after leaving academia. This ad, created by Watson, helped sell a skin cream by associating it with the wealth and beauty of the queen of Spain. According to the fine print, the queen uses Pond's vanishing and cold creams and "has expressed her loyal pleasure in them." 33. How do advertisers use classical conditioning to influence our attitudes, and how did one experiment demonstrate that classical conditioning can influence attitudes without requiring conscious thought? HMVr CTORIA EUG ENiA QueenofSpmn - >jr Spain, Li rh« tEnbhlimctitDc"queenly beauty, Sire it bnrjnJJjuj;hitr of Qntcti VicEiXd» meče iff K:o£ Er that than for other purposes. SECTION SUMMARY iftftudes can serve value-expressive, sociai-adjustive, defensive, and utilitarian functions. The value-expressive function is served especially by central attitudes, or values, ihat are part of one's self-concept. Schwartz found that values tend to cluster into 10 ■ategories that can be arranged in a wheel in which similar values lie adjacent to one mother and contradictory values lie opposite one another. This value structure appears to be universal, but the relative importance of each value category varies from 'ulture to culture. The sociai-adjustive function of attitudes was demonstrated in a classic study by tewcomb at Bennington College. New students changed their initially conservative attitudes to match the liberal attitudes of the older students and professors as a means or gaining social acceptance. The defensive function of attitudes is illustrated by studies of cognitive disso-lance. To avoid the discomfort that arises from awareness of inconsistency in their Defers and actions, people will (a) avoid information that contradicts their present attitudes and (b) alter their attitudes to match their actions. People are especially likely o alter an attitude to match an action when they have no easy alternative means of 'Xplaining why they did what they did (the insufficient-justification effect). Another example of the defensive use of attitudes derives from the just-world bias: To convince hemselves that the world is fair (and they are safe), people develop negative, blaming itt/tucfes toward those who suffer. Attitudes serve a utilitarian function to the degree that they guide behavior in use-ul ways. Some attitudes, such as those derived from classical conditioning, automatically come into play in response to the object of the attitude, but attitudes that are acquired intellectually must be recalled and thought about if they are to influence behavior. Anything that reminds a person of his or her attitudes—such as the presence of a mirror—tends to increase the consistency between attitudes and actions. According to the theory of planned behavior, the intention to behave in a particular way is influenced not just by one's own attitude but also by the subjective norm (one's sense of what others would think of one for behaving that way) and perceived control (one's confidence in one's own capacity to behave that way).