Social Psychology, Chapter 12 – Flashcards

Beliefs that certain attributes are characteristic of members of particular groups. Stereotyping involves thinking about person not as an individual, but as member of group, and bringing to bear what you know about the group onto your expectations about the individual. Most concern with stereotyping has focused on those thought to be the most questionable and those most likely to give rise to the most pernicious forms of prejudice and discrimination.

A negative attitude or affective response toward a certain group and its individual members. Negative attitudes received most attention, but it's possible to be positively prejudiced toward particular group. Prejudice involves prejudging others because they belong to a specific category.

Unfair treatment of members of a particular group based on their membership in that group. Treatment based not on their character or abilities but on their membership in group.

Roughly speaking, these three statement refer to the belief, attitudinal, and behavioral components, respectively, of negative intergroup relations. They don't often go together, and one person can discriminate without prejudice, it's also possible to prejudiced and yet not discriminate.

Changes with forms of discrimination, sociallly acceptable of expressing prejudices and stereotypes has created conflict in many people between what they really think and feel and what they think they should think and feel. It also created conflict between competing beliefs and values or between competing abstract beliefs and good level reactions. Old fashioned racism has largely disappeared in the US but has been supplanted by subtler, more modern counterpart, for similar accounts of homophobia and sexism, respectively. Prejudice directed at other racial groups that exists alongside rejection of explicitly racist beliefs. Beliefs fuel modern racism whether they are true or not. Many Americans hold strong egalitatian values that lead them to reject prejudice and discrimination. And whether they act in a prejudiced or discriminatory manner depends on the details of the situation.

Participants were in position to aid a white or black individual in need of medical assistance. Participants who think they are the only one who could help came to aid the black victim more often. But when they thought that other people were present and their inaction could be justified on nonracial grounds, they helped the black victim much less often than the white victim. Prejudice or discrimination is masked and the individual remains comfortably unaware of being racist.

White participants evaluated black and white applicants to college. Participants who scored high on the attitudes toward Blacks Scale and participants who scored low rated white and black applicants the same when the applicants excelled on all pertinent dimensions or were below par on all dimensions. When the applicants excelled on certain dimensions and were below average on others, the rating of prejudiced and unprejudiced participants diverged: the prejudiced participants rated the black applicants less favorably than did the unprejudiced participants. The desire to appear unprejudiced is sometimes sufficiently strong. Bias directed at the ingroup. White participants who read about black applicants who were strong on some dimensions and weak on others rated them more favorable than they rated comparable white applicants.

Stereotype does not necessarily be negative or harmful. Many "isms" are ambivalent, containing both negative and positive features.

Peter Glick and Susan Fiske, 2001. Interviewed 15000 men and women in 19 nations and found that benevolent sexism often coexist with hostile sexism. They argued that such partly positive stereotypes aren't necessarily benign. When we idealize only certain members of outgroups, those who meet our positive expectations, we are likely to disparage those who don't fulfill that positive stereotype. Those who hold ambivalent attitudes tend to act positively toward members of outgroups only if they fultill their idealized image of what such people should be like. Those who deviate tend to be treated with hostility.

A technique for revealing non-conscious prejudices, even those parts who advocate universal equality and high regard for all groups. Series of words and/or pictures are presented on computer screen, and respondent is told to press key with left hand if the picture or word conforms to one rule and press another key with the right hand if it conforms to another rule. Result: responders would react faster when they were to press one key for members of particular group and words stereotypically associated with group than when they were to press the same key for members of that group and words that contradict the stereotype associated with that group.

Participants asked to press one key for both positive words and either photos or names of people in one group, and another key for both negative words and people in another group. People prejudiced against old people should be faster to press the appropriate key when the same key is used for old faces and negative words and slower when the same key is used for old faces and positive words. Nonconscious prejudice toward old people would be captured by the difference between the average time takes to respond to young faces, negative words. Result: both young and older individuals show pronounced prejudice in favor of the young over the old, and about two-thirds of white respondents show strong or moderate prejudice for white over black. About half of all black responses on the IAT are predictive of behavior that is more significant than pressing computer keys.

Participants in brain-imaging machine were shown pictures of black and white faces. Participants earlier IAT responses were significantly correlated with heightened neutral activity in the amygdala(brain region related to emotional learning and evaluation) in response to the black faces and their scores were not correlated with difference in neural activity, suggesting that IAT assessed an important component of their attitudes that participants were unable or unwilling to articulate. Participants first interacted with a white experimenter, took the IAT, and then interacted with a black experimenter. Result: their IAT scores, were predictive of the discrepancy between how much they spoke to the white versus the black experimenter, how often they smiled at the white versus the black experimenter, and the number of speech errors and hesitations they exhibited when interacting with the white versus the black experimenter.

A procedure used to increase the accessibility of a concept or schema (e.g. a stereotype). Prejudices that individuals might not know they have, or that they may wish to dent, by using number of priming procedures. When giving a word previously, one was asked to recognize two new words, one is related with the previous word and one does not. Individuals recognize the related words quicker than the other one. An implicit measure of prejudice can thus be derived by comparing a person's average reaction time to positive and negative words preceded by faces of members of the target category. People are not lying when they deny such prejudices and they do not have conscious access to many of their attitudes and beliefs.

The economic view of prejudice and discrimination makes the same claim about groups. Groups develop prejudices about one another and discriminate against one another when they compete for material resources. Religious groups, racial groups, and cultural groups all have the capacity, like Cain, to protect and promote their own interests by lashing out at those they perceive to be threatening them.

A theory that group conflict, prejudice, and discrimination are likely to arise over competition between groups for limited resources. It acknowledges that groups sometimes confront real conflict over what are essentially economic issues. Prejudice and discrimination often arise from competition over limited resources. and should increase under conditions of economic difficulty. They should be strongest among groups that stand to lose the most from another group's economic advance. eg. People in the working class in the United States exhibited the most antiblack prejudice in the wake of the civil rights movement.

Glorifying one's own group while vilifying other groups. People in the outgroup are often thought of in stereotyped ways and are treated in a manner normally forbidden by one's moral code. At the same time, loyalty to the ingroup intensifies. Consider the 9/11 attack, telling white students that the attacks were directed at all Americans, regardless of race and class, served to reduce prejudice toward African-Americans

Twenty-two fifth-grade boys were taken to Robbers Cave State Park. They had signed up for 1.5 weeks summer camp experience and also unknown, a study of intergroup relations. 22 boys were average in nearly every respect and none of them knew each other beforehand, were divided into two groups of 11 and taken to separate areas of the park. 1. Neither differences in background, nor differences in appearance, not prior histories of conflict are necessary for intergroup hostility to develop but groups into competition for goals. 2. Competition against outsiders often increase group cohesion. 3. Intergroup conflict can be diminished with superordinate goals that keeps everyone's eyes on the prize and away from meddlesome subgroup distinction.

The first phase of the experiment, the two groups Eagles and Rattlers, independently engaged in activities designed to foster group unity and contented themselves with common camp activities. Neither group even knew of the other's existence. Consistent hierarchical structure also emerged within each group, with "effective initiators"—the boys who made suggestions that the others accepted. In the second phase, the Eagles and Rattlers were brought together for a tournament and each member of the winning team would receive a medal and a highly coveted pocket knife while members of the losing team would get nothing. The tournament was designed to encourage each group to see the other as an impediment to the fulfillment of its own goals. Result: From the very first competitive encounter, and with steadily increasing frequency throughout the tournament, the two groups hurled insults at each other. They also differ markedly from referring to members of their own group, primarily self-glorifying and congratulatory comments about fellow group members. members. The expression of intergroup hostility, moreover, was not limited to words. issued. Internal dynamics of the two groups changed as they became locked into this competitive struggle. Boys who were either athletically gifted or who advocated a more aggressive stance toward the outgroup tended to gain in popularity.

Investigators conducted numbers of more tightly controlled assessments of the degree to which boys tended to look favorably on members of their own group while derogating members of the other group. Investigators scattered a large quantity of beans around a field and asked the two groups to pick up as many as they could in a 1-minute period. The group collecting the most would receive $5. Image of each boy's collection of beans was briefly projected on a wall, and everyone was asked to estimate the number of beans that the boy had collected. In reality, the same quantity of beans (35) was always shown, but this was impossible to discern because of the brief duration of the projection and the large quantity of beans shown. Result: boys' estimates revealed clear ingroup favoritism: each group estimated that boys who were members of their group had collected more beans than boys who were members of the other group.

Goals that transcend the interests of one individual group and that can be achieved more readily by two or more groups working together. Two groups were simply brought together in various noncompetitive settings to ascertain whether their hostility would dissipate which does not work. Investigator then contrived to confront boys with number of crises that could only be resolved through cooperative effort of both groups. Result: relations between the two groups quickly showed the effects of these superordinate goals. When returned back home, boys insisted that everyone return on the same bus rather than on the separate buses by which they had arrived. When the bus pulled over at a roadside diner, the group that had won $5 in the bean collection contest decided to spend their money on malted milks for everyone, Eagles included. The hostility produced by five days of competition was erased by the joint pursuit of common goals, resulting in a happy ending.

Many analysis cite the integration of blacks, hispanics, and native Americans, and whites in the military. To engaged cooperative, interdependent action to accomplish shared goals, group or category to which a person belongs recedes in importance, and what he or she can contribute to the joint effort becomes more prominent. Intergration effort on college campuses may be less successful than it has been in the military.

Intergroup hostility, can develop even in the absence of competition.In Robbers Cave experiment, signs of increased ingroup solidarity when the two groups first learned of each others' existence. Existence of another group make boys take their group membership much more seriously. When both groups learned the existence of the other group, they want to run them off and challenge them. Existence of group boundaries among any collection of individuals are sufficient to initiate group discrimination.

An experimental paradigm in which researchers create groups based on arbitrary and seemingly meaningless criteria and then examine how the members of these "minimal groups" are inclined to behave toward one another. Participants first perform a rather trivial task and are then divided into two groups, ostensibly on the basis of their responses. 1. Participants estimate the number of dots projected on a screen. Some participants are told they belong to a group of "overestimators," and others belong to a group of "underestimators." In reality, the participants are randomly assigned to the groups, and they only learn about the group to which they themselves belong--the minimum group team. 2. Participants are taken individually to different cubicles and asked to assign points redeemable for money to successive pairs of their fellow participants. They do not know the identity of those to whom they are awarding points; they only know their "code number" and group membership. Some of the options participants can choose provide slightly more for the member of the outgroup; some maximize what the ingroup member can receive but still result in more points for the members of the outgroup; and some maximize the relative ingroup advantage over the outgroup but don't provide much in the way of absolute reward for members of the ingroup. Majority of participants are interested more in maximizing the relative gain for members of their ingroup than in maximizing the absolute gain for their ingroup. The participants do not know who the ingroup and outgroup members are; the choices are never for themselves; and, of course, the ostensible basis for establishing the two groups is utterly trivial. Yet, they still exhibit a tendency to favor their minimal ingroup! Moreover, they are willing to do so at a cost to the ingroup, which gets less than it would if the focus were on absolute gain rather than "beating" the other group, explains why we slip into thinking in terms of "us" versus "them". The us/them distinction may be one of the basic cuts we make in dividing up and organizing the world. Still, the ingroup favoritism observed in the minimal group situation cannot be the product of cognition alone. People's cognitive processes might lead them to make the us/them distinction, but cognitive processes alone cannot lead to one group being favored over the other.

A theory that a person's self-concept and self-esteem not only derive from personal identity and accomplishments, but from the status and accomplishments of the various groups to which the person belongs. Because our self-esteem is based in part on the status of the various groups to which we belong, we might be tempted to do what we can to boost the status and fortunes of these groups and their members. By giving advantage to fellow members of an ingroup, we boost the group's standing and thereby potentially elevate our own selfesteem. Thus, feeling better about the group leads us to feel better about ourselves. Assessed participants' self-esteem after they have had an opportunity to exhibit ingroup favoritism in the minimal group situation. Those who had been allowed to engage in intergroup discrimination had higher self-esteem than those who had not been given the opportunity to discriminate. Other research has shown that people who take particularly strong pride in their group affiliations are more prone to ingroup favoritism when placed in a minimal group situation. And people who are highly identified with a particular group react to criticism of the group as if it were criticism of the self.

Social identity theory also receives support from the everyday observation that people go to great lengths to announce their affiliation with a certain group when that group is doing well. The tendency to take pride in the accomplishments of those with whom we are in some way associated (even if it is only weakly), as when fans identify with a winning team. He investigated the tendency by recording how often students wore their school sweatshirts and T-shirts to class after their football team had just won or lost a game. As expected, students wore the school colors significantly more often following victory than after defeat. Cialdini and his colleagues also tabulated students' use of first-person ("We won") and third-person ("They lost") pronouns following victory and defeat. The inclusive "we" was used significantly more often after a win, and the more restrictive "they" was used more often after a loss. As social identity theory predicts, the triumphs and failings of the groups with which we affiliate impact our self esteem—even when the group is simply a favorite sports team.

Half of the participants had their self-esteem threatened by being told they hand just performed poorly on an intelligence test, the other half were told they had done well. The participants then watched a videotaped interview of a job applicant. The content of the videotape made it clear to half of the (non-Jewish) participants that the applicant was Jewish, but not to the other half. When later asked to rate the job applicant, those who thought she was Jewish rated her more negatively than did those who were not told she was Jewish, but only if they had earlier been told they had performed poorly on the intelligence test. In addition, the participants who had had their self-esteem threatened and had "taken it out" on the Jewish applicant experienced an increase in their self-esteem from the beginning of the experiment to the end. Stereotyping and derogating members of outgroups, it appears, serve to bolster self-esteem.

Participants were either praised or criticized by a white or black doctor. Sinclair and Kunda predicted that the participants would be motivated to cling to the praise they received but to challenge the criticism—and that they would use the race of their evaluator to help them do so. In particular, they thought that individuals who received praise from a black doctor would tend to think of him more as a doctor (a prestigious occupation) than as a black man, whereas those who were criticized by a black doctor would tend to think of him more as a black man than as a doctor. To test their predictions, Sinclair and Kunda had their participants perform a lexical decision task right after receiving their feedback from the doctor. That is, the researchers flashed a series of words and nonwords on a computer screen and asked the participants to indicate, as fast as they could, whether each string of letters was a word. Some of the words were associated with the medical profession (for example, "hospital," "prescription") and some were associated with common stereotypes of blacks (for example, "rap," " jazz"). Result: Participants were particularly fast at recognizing words associated with the black stereotype when they had been criticized by the black doctor, and particularly slow to recognize those words when praised by the black doctor (see Figure 12.3A). When he criticized them, in other words, participants saw him as a black man, but not when he praised them

The theory that frustration leads to aggression. People are particularly likely to vilify outgroups under conditions that foster frustration and anger. If the source of frustration is the very group to which prejudice and discrimination are directed, frustration-aggression theory is both an economic and a motivational account. But sometimes the source of frustration is not the targeted group. In these cases, the motivation is not economic competition, and the two accounts diverge. By itself, the link between frustration and aggression cannot explain the origins of prejudice and discrimination because frustration leads to generalized aggression. Often we cannot lash out at the true source of our frustration without getting into further difficulty, so we displace our aggression onto safer target. Frustration-aggression theory predicts that hardship will generate malevolence directed at minority groups, who, by virtue of being outnumbered and in weaker position, constitute particularly safe and vulnerable targets. eg. Antisemitism. Jews have been welcomed and accepted into numerous societies that, when times go tough, suddenly targeted Jews as scapegoats and directed their anger at the Jewish community.

Relationship between price of cotton and number of lynchings of blacks in the South between 1882 and 1930. It was assumed that times were good and frustrations low when the price was high, and times were tough and frustrations high when the price was low. Result: Strong negative correlation between price of cotton in given year and number of lynchings that took place that year. For the relationship between cotton price and lynchings of white is also negative correlation. The relationship is stronger for blacks than for whites also consistent with idea that frustration leads to aggression that tends to be displaced toward relatively powerless groups.

The strength of the motivational perspective is that it builds on two undeniably important elements of the human condition. First, people readily draw the us/them distinction, and the various groups to which they belong are intimately connected to the motive to enhance self-esteem. Second, people tend to react to frustration with aggression, and often direct their aggression at the "safest" and least powerful targets.

From the cognitive perspective, stereotyping is inevitable. It stems from the ubiquity and necessity of categorization. All categorizing works to simplifies the task of taking in and processing the incredible volume of stimuli that confronts us. Stereotypes provide us with those simpler models that allow us to deal with great blooming, buzzing confusion of reality. According to cognitive perspective, stereotypes are a natural result of way our brains are wired to store and process information.

According to the cognitive perspective, stereotypes are useful cognitive categories that allow us to process information efficiently. If so, we should be particularly inclined to use them when we are overloaded, tired, or mentally taxed in some way—that is, when we are in need of a shortcut. eg. students were shown to be more likely to invoke stereotypes when tested at low point of their circadian rhythm. "Morning people" were more likely to invoke a common stereotype and conclude, for example, that a person charged with cheating on an exam was guilty if he was an athlete—but only when they were tested at night. "Night people" were more likely to conclude that a person charged with dealing drugs was guilty if he was black—but only when they were tested in the morning. Thus, people are most likely to fall back on mindless stereotypes when they lack mental energy.

If the use of stereotypes conserves intellectual energy, then encoding information in terms of relevant stereotypes should furnish extra cognitive resources that can be applied to other tasks. Resources not used on one task can be applied to another. Students were asked to perform two tasks simultaneously. One required them to form an impression of a (hypothetical) person described by a number of trait terms presented on a computer screen (for example, "rebellious," "dangerous," "aggressive"). The other task involved monitoring a tape-recorded lecture on the economy and geography of Indonesia. For half of the students, the presentation of the trait terms was accompanied by an applicable stereotype (for example, skinhead); for the remaining students, the trait terms were presented alone. The key questions were whether the applicable stereotype would facilitate the students' later recall of the trait terms they had seen and, more important, whether it would also release extra cognitive resources that could be devoted to the lecture on Indonesia. Students were given a brief quiz on the contents of the lecture. As the experimenters anticipated, the use of stereotypes eased the students' burden in the first task and thereby facilitated their performance on the second.

Gained in efficiency is paid by occasional inaccuracy and error. Invoking the stereotype may save time and effort, but it can lead to mistaken impressions and unfair judgments about individuals. Inaddition, biased information processing can help explain why even stereotypes completely lacking in validity nevertheless develop and endure. If we suspect—because of what we've been told, or the implications of a joke we heard, or a hard-to-interpret performance difference—that a particular group of people might differ from the mainstream in some way, it is shockingly easy to construe pertinent information in such a way that our suspicion is confirmed, solidified, and elaborated. The cognitive perspective on stereotyping does more than point out the obvious fact that stereotypes can distort our perceptions of others.

Merely dividing a continuous distribution into two groups leads people to see less variability within each group and more variability between the two. In early experiment, participants were shown a series of lines, with adjacent lines in the series varying from one another in length by a constant amount. When the series was split in half to create two groups, the participants tended to underestimate the differences between adjacent lines within each group and to overestimate the difference between the adjacent lines that formed the intergroup border. In more social tests of this idea, participants are divided into two "minimal" groups. They then fill out an attitude questionnaire twice—once to record their own attitudes and once to record how they think another ingroup or outgroup member might respond. Participants consistently assume that their beliefs are more similar to those of another ingroup member than to those of an outgroup member—even when group membership is arbitrary. What is remarkable, and potentially troubling, is that people make such assumptions even when the groups are formed arbitrarily or when they are formed on the basis of a dimension (for example, skin color) that may have no bearing on the particular attitude or behavior under consideration.

The tendency to assume that within-group similarity is much stronger for outgroups than for ingroups. Princeton and Rutgers students were showed a videotape of other students making a decision, such as whether to listen to rock or classical music or whether to wait alone or with other participants during a break in an experiment. Half of the Princeton and Rutgers students were told that the students shown on the tape were from Princeton; half were told they were from Rutgers. After watching the tape, the participants estimated the percentage of students at the same university who would make the same choices as those they had seen on the tape. The results indicated that the participants assumed more similarity among outgroup members than among ingroup members. Princeton students who thought they had witnessed the behavior of a Rutgers student were willing to generalize that behavior to other Rutgers students. In contrast, Princeton students who thought they had witnessed the behavior of a Princeton student were less willing to generalize. The opposite was true for Rutgers students. People see more variability of habit and opinion among members of the ingroup than they do among members of the outgroup. We typically have much more contact with fellow members of an ingroup than with members of an outgroup, giving us greater opportunity to encounter evidence of divergent opinions and discrepant habits among ingroup members.

As we have just seen, people are more likely to assume that an individual action is typical of a group if the group is not their own. invite it. In general, people are more likely to extrapolate from behaviors they already suspect may be typical of an individual's fellow group members. Stereotypes can therefore be self-reinforcing. Actions that are consistent with an existing stereotype are noticed, deemed significant, and remembered, whereas those at variance with the stereotype may be ignored, dismissed, or quickly forgotten. Stereotypes also influence how the details of events are interpreted. White participants watched a videotape of a heated discussion between two men and were asked, periodically, to code the behavior they were watching into one of several categories (for example, "gives information," "playing around," "aggressive behavior"). At one point in the video, one of the individuals shoved the other. For half the participants, it was a black man doing the shoving; for the other half, it was a white man. The race of the person made a difference in how the action was seen. When perpetrated by a white man, the incident tended to be coded as more benign (as "playing around," for example). When perpetrated by a black man, it was coded as a more serious action (as "aggressive behavior," for example). The influence of stereotypes is likely to be even greater when the episode is presented to people secondhand and is therefore more amenable to differential construal. Participants listened to a play-by-play account of a college basketball game and were told to focus on the exploits of one player in particular, Mark Flick. Half the participants saw a photo of Mark that made it clear he was African-American, and half saw a photo that made it clear he was white. When participants rated Mark's performance during the game, their assessments reflected commonly held stereotypes about black and white basketball players. Those who thought Mark was African-American rated him as more athletic and as having played better; those who thought he was white rated him as having exhibited greater hustle and as having played a more savvy game. Studies such as these make it clear that people do not evaluate information evenhandedly. Instead, information that is consistent with group stereotypes typically has more impact than information that is inconsistent with it.

Acting in a way that tends to produce the very behavior we expected in the first place, as when we act toward members of certain groups in ways that encourage the very behavior we expect from them. eg. Teacher who thinks that members of a particular group lack intellectual ability may fail to offer them adequate instructions, increasing the chances that they will indeed fall behind their classmates. Princeton students interviewed both black and white men pretending to be job applicants . The interviews were monitored, and it was discovered that the students (the white interviewers) unwittingly treated black and white applicants differently. When the applicant was black, the interviewer tended to sit farther away, to hem and haw throughout the session, and to terminate the proceedings earlier than when the applicant was white. The second phase of the experiment showed just how difficult it had been for the black applicants. Interviewers were trained to treat new applicants, all of whom were white, the way that either the white or the black applicants had been treated earlier. These interviews were tape-recorded and later rated by independent judges. Those applicants who had been interviewed in the way the black applicants had been interviewed earlier were evaluated more negatively than those who had been interviewed the way the white applicants had been interviewed. In other words, the white interviewers' negative stereotypes of blacks were confirmed by placing black applicants at a disadvantage.

An erroneous belief about a connection between events, characteristics, or categories that are not in fact related. Distinctive events capture attention. We attend more closely to distinctive events, we are also likely to remember them better, with the result that they may become overrepresented in our memory. Minority groups are distinctive to most members of the majority, and so minority group members stand out. This means that negative behavior on the part of members of minority groups is doubly distinctive and doubly memorable. And because negative behavior by the majority or positive behavior by the minority is not as memorable, negative actions by the minority are likely to seem more common than they really are. Minority groups are therefore often thought to be responsible for more problematic behavior than they actually engage in.

The pairing of two distinctive events that stand out even more because they co-occur. Participants were shown a series of 39 slides, each of which described a positive or negative action initiated by a member of "group A" or "group B." The groups were completely mythical, and so any judgments made about them could not be the result of any preexisting knowledge or experience on the part of the participants. Two-thirds of the actions were attributed to group A, making A the majority group. Most of the actions attributed to each group were positive, and this was equally true of both groups. There was thus no correlation between group membership and the likelihood of positive or negative behavior. After viewing the entire series of slides, the participants were shown just the behaviors they had seen earlier and were asked to indicate the group membership of the person who had performed each one. They were also asked to rate the members of the two groups on a variety of trait scales. Both measures indicated that the participants had formed a distinctiveness-based illusory correlation. They overestimated how often a negative behavior was performed by a member of group B (the smaller group), and they underestimated how often such a behavior was performed by a member of group A (the larger group). As a result, they also rated members of the larger group more favorably. Hamilton and Gifford showed that an illusory correlation was also obtained when it was positive behaviors that were less common. Under these circumstances, participants overestimated how often a positive behavior was associated with the smaller group. The first is that their results are particularly impressive because they were obtained in a barren context that excluded elements that might encourage illusory correlations in everyday life. Their analysis predicts that people should be prone to develop illusory correlations between any two variables that are jointly distinctive. But that does not happen. They overpredicts and where their analysis falls short, then, is in specifying which jointly distinctive pairings are likely to form the core of commonly held stereotypes, and which are not.

How people respond to stereotype disconfirmation varies with factors such as how emotionally involved they are in the stereotype, whether they hold the stereotype in isolation or belong to a group that preaches it, and so on. One thing is clear, however: people do not give up their stereotypes easily. As numerous studies have demonstrated, people evaluate disconfirming evidence in a variety of ways that have the effect of dampening its impact. The first thing to note is that no stereotype contains an expectation of perfectly invariant behavior. Groups thought to be dishonest, lazy, or carefree are thought to be dishonest, lazy, or carefree on average, or more dishonest, lazy, or carefree than other groups. It is not expected that all of their members behave in those ways all the time.

People remain unmoved by apparent disconfirmations of their stereotypes because anyone who acts at variance with the stereotype is simply walled off into a category of "exceptions." Explaining away exceptions to a given stereotype by creating a subcategory of the stereotyped group that can be expected to differ from the group as a whole. Sexists who believe that women are passive and dependent and should stay home to raise children are likely to subcategorize assertive, independent women who choose not to have children as "militant" or "strident" feminists, thereby leaving their stereotype of women largely intact. This echoes the more general truth that evidence that supports a stereotype is treated differently from evidence that refutes it. Supportive evidence tends to be accepted at face value, whereas contradictory evidence is often critically analyzed and discounted. One way we do this is by attributing behavior consistent with a stereotype to the dispositions of the people involved and attributing inconsistent behavior to external causes. An anti-Semite who believes that Jews are "cheap" is likely to dismiss a Jew's acts of philanthropy as due to a desire for social acceptance, but to attribute any pursuit of self-interest as a reflection of some "true" Jewish character. Thus, episodes consistent with a stereotype reinforce its perceived validity; those that are inconsistent with it are deemed insignificant. Another way that we differentially process supportive and contradictory information is by varying how abstractly we encode the actions of people from different groups. The more concrete the description, the less it says about the individual involved. If people's evaluations are guided by their preexisting stereotype, we might expect them to describe actions that are consistent with stereotype in broad terms. Stereotypes may insulate themselves from disconfirmation by influencing the level at which relevant actions are encoded.

Each participant was shown a number of sketches depicting a member of their own contrada or of the rival contrada engaged in an action. The contrada membership of the person depicted was established simply by having the color of the protagonist's shirt match that of one contrada or another. Some of the sketches portrayed desirable actions (for example, helping someone), and some portrayed undesirable actions (for example, littering). After inspecting each sketch, the participants were asked to describe what it depicted, and their responses were scored for level of abstraction. Actions consistent with a participant's preexisting orientation (that is, positive actions by a member of one's own contrada; negative actions by a member of the rival contrada) were described at a more abstract level than actions at variance with a participant's preexisting orientation (that is, negative actions by a member of one's own contrada; positive actions by a member of the rival contrada). This asymmetry feeds the tendency to perceive the ingroup in a favorable light.

Some of the cognitive processes that give rise to stereotyping and prejudice are rather deliberate, elaborate and mindful. Other cognitive processes give rise to stereotyping and prejudice rapidly and automatically, without much conscious attention and elaboration. This is likely to be the case for distinctiveness based illusory correlations and the outgroup homogeneity effect. Our reactions to different groups of people are to surprising degree guided by quick and automatic mental processes that we can override, but not eliminate. Devine sought to demonstrate that what separates prejudiced and nonprejudiced people is not their knowledge of derogatory stereotypes, but whether or not they resist the stereotypes. Stereotypes can be triggered even if we don't want them to be.

Processes that occur outside of our awareness, without conscious control

Processes that occur with conscious direction and deliberate thought

Groups of high and low prejudiced participants on the basis of their scores on the Modern Racism Scale. She present each participant a set of words, one at a time, to briefly that they could not be consciously identified. Some participants were shown neutral words and some were shown related to blacks. Participants were then presented with written description of individual who acted in ambiguously hostile manner, target individuals refused to pay his rent until his apartment was repaired. Result: He was primed more hostile and more negative by participants who had earlier been primed by words designed to activate their stereotypes of blacks. This was equally true of prejudiced and nonprejudiced participants. The stimulus words unconsciously activated their stereotypes, the nonprejudiced participants were unable to suppress the automatic processing of stereotypical information.

Participants list characteristics of black Americans. The two groups differed substantially in output f consciously controlled procedure: prejudiced participants listed many more pejorative characteristics stereotypically associated with blacks then did nonprejudiced participants. Even though both prejudced and non prejudiced individuals have stored their minds the same nagetive stereotypes of black Americans, prejudiced individuals believe them and are willing to voice these beliefs, whereas nonprejudiced individuals reject them. Nevertheless, even among nonprejudiced individuals, there is often a rift between the beliefs and sentiments elicited by automatic processes and those elicited by more controlled processes. This was shown in an interesting way by a study that examined the areas of the brain that were activated when white participants were shown pictures of black faces and white faces. The key manipulation in this study was the amount of time participants were exposed to the black and white faces. When shown the faces for only 30 milliseconds, the participants exhibited greater activation in the amygdala (which registers emotional response) after exposure to black faces than after exposure to white faces. Furthermore, the amount of amygdala activation was related to participants' implicit prejudice as measured by the IAT. But when the faces were shown for 525 milliseconds, there was no difference in amygdala activation when exposed to black versus white faces, suggesting that these participants—all of whom had expressed a strong desire to avoid prejudice—initially had an automatic response to black versus white faces that they then tried to control. Indeed, at 525 milliseconds, black faces caused more activity in the prefrontal cortex—an area of the brain associated with cognitive and behavioral regulation—than did white faces.

Payne (2001) had participants decide as quickly as possible whether an object depicted in a photo was a handgun or a hand tool (for example, pliers). Each photograph was immediately preceded by a picture of either an African-American or a white face. Payne found that the (white) participants were faster to identify a weapon as a weapon when it was preceded by an African-American face and faster to identify a hand tool as a hand tool when it was preceded by a white face. This is the facilitation caused by a stereotypical association between handguns and African-Americans that exerts its effect even among nonprejudiced individuals. Judd and his colleagues found that African-American faces facilitated the recognition of both positive and negative stereotypical items (handguns and sports equipment), but not the nonstereotypical items (insects and fruits), regardless of whether they were positive or negative.

Participants were instructed to "shoot" if the target individual was holding a gun and to press a different response key if he was not. Participants were instructed to respond as quickly as possible, which guaranteed there would be occasional mistakes. The pattern of mistakes is shown in Figure 12.8, and it is clear that participants treated African-American and white targets differently. They made both types of mistakes—shooting an unarmed target and not shooting an armed target—equally often when the target individual was white. But for African-American targets, they were much more likely to make the mistake of shooting if the target was unarmed than failing to shoot if the target was armed.

We all tend to stereotype and that we all have the capacity to harbor troubling prejudices—prejudices we are often unaware we have.

Members of stereotyped or stigmatized groups are also typically aware of the stereotypes that others hold about them, and this awareness also can have negative effects on them. Members of stigmatized groups cannot tell whether many of their experiences have the same origins as those of everyone else or whether they are the result of prejudice.

In one study, African-American and white students received flattering or unflattering feedback from a white student in an adjacent room. Half the participants were led to believe that this other student could see them through a one-way mirror, and half were led to believe they could not be seen (because a blind covered the mirror). Whether or not they could be seen had no effect on how white students reacted to the feedback. But it did affect how black students reacted. When black students thought the other person could not see them—and therefore didn't know their race—their self-esteem went down from the unflattering feedback and was boosted by the positive feedback. When they thought the other person could see them, their self-esteem was not injured by the bad news, nor was it enhanced by the good news. Result: members of stigmatized groups quite literally live in a less certain world, not knowing whether to attribute positive feedback to their own skill or to others' condescension and not knowing whether to attribute negative feedback to their own error or to others' prejudice.

Performance of members of stigmatized groups can be impaired by stereotype threat—the fear that they will confirm the stereotypes that others have regarding some salient group of which they are a member. mathematics. In one condition, Participants were told there was no gender difference on a particular test they were about to take. Other participants were told that there was a gender difference in favor of men. Men and women performed equivalently when told there was no gender difference on the test, but women performed worse than men when they were told that there was a gender difference. The manipulation of stereotype threat for it to have an effect. Michael Inzlicht and Talia Ben-Zeev (2000) had university women take a math test either in the company of two other women or in the company of two men. Those who took the test with other women got 70 percent of the problems right on average. Those who took the test with men got 55 percent right on average.

Black and white Stanford University students were given a difficult verbal test taken from the Graduate Record Exam. Half of the students were led to believe that the test was capable of measuring their intellectual ability, and half were told that the investigators were in the early stages of trying to develop the test and that nothing could be learned about intellectual ability from their scores. This manipulation had no effect on the performance of white students. African-American students did as well as white students when they thought it was the test that was being tested, but they performed much worse than white students when they thought their intellectual ability was being tested. Moreover, a blatant manipulation was not required to produce a significant effect on the performance of African-Americans. It was enough simply to have them indicate their race at the top of the page to cause their performance to be worse than in a control condition in which they did not indicate their race

Stereotype threat leads to increased arousal, which can directly interfere with performance on complex tasks and serve as a source of distraction that interferes with concentration on the task at hand. Furthermore, knowing that one's group is "suspect" in the eyes of others tends to elicit negative thinking, which can both directly undermine performance and lead individuals to "play it safe" by being more obsessed with avoiding failure than reaching for success.

One factor that is both cause and consequence of characteristics developments is the increased day-to-day interaction between members of different groups. When people interact frequently, it becomes easier to see one another more as individuals and less as representatives of particular groups. Some types of contact are more helpful than simple contact between cross sections of different groups.

Contact between different groups is likely to be more positive and more productive if certain conditions are met. First, the different groups need to have equal status. If one group feels superior and the other resentful, it is unlikely that harmonious, productive interactions will be the norm. Second, as we saw in the Robbers Cave study, productive intergroup interactions are facilitated if the different groups have a shared goal that requires their cooperative interaction. Third, broader social norms should be supportive of intergroup contact. Finally, the contact should encourage one-on-one interactions between members of the different groups. Doing so puts each person's identity as an individual in the foreground and downplays a person's group membership. Studies with thousands of students in over 25 different nations, found that when most of these conditions are met, contact between members of different groups does indeed tend to be effective in reducing prejudice. Even if increased intergroup contact were entirely effective in eliminating prejudice, even if the level of intergroup hostility were reset to zero, the ideas discussed in this chapter make it clear that new prejudices and animosities might soon arise.