We spend a lot of our time doing things with other people. But we spend even more of our time thinking about other people. “Why did my brother sound so weird on the phone when he told me about his new girlfriend?” “Is my new friend from psych class a real friend, or does she just want somebody to help her with her homework?” “How embarrassing that I forgot my professor’s name—I wonder what she thinks of me?” Thoughts like these can swirl through your head, hour after hour, day after day.

Social psychologists call these thoughts social cognition. Social cognition refers to people’s beliefs, opinions, and feelings about the individuals and groups with whom they interact socially. Three main topics in the study of social cognition are attributions, attitudes, and stereotypes.

In everyday life, we often ask why: Why aren’t my grades higher? Why are people saying no when I ask them out? Your answers to such questions are called attributions. An attribution is a belief about the causes of a social behavior.

Attributions are important in two ways. First, they affect motivation and emotion. For example, students who attribute school difficulties to a lack of ability rather than to a lack of effort become less motivated (Weiner, 1985). Teenagers who attribute personal problems to causes that cannot be changed, rather than to changeable ones, tend to become depressed (Hankin, Abramson, & Siler, 2001). Second, accurate attributions—correct diagnoses of the causes of events—are a first step in altering behavior you want to change. If you want to improve your grades or get people to start saying “yes” when you ask them out, you first want to figure out the cause of the current problems.

525

PERSONAL VERSUS SITUATIONAL CAUSES. When we try to figure out the causes of someone’s behavior, two types of causes stand out: personal and situational. If a friend says, “I am struggling in calculus; I might not pass,” the cause could be:

What determines whether we make personal or situational attributions? According to an attribution theory developed by Harold Kelley (1967; Kelley & Michela, 1980), the key factor is the degree to which potential causes covary with the behavior to be explained—that is, whether the potential causes co-occur with the behavior. Consider two cases:

THE FUNDAMENTAL ATTRIBUTION ERROR. Thus far, attributions seem simple, so you might expect people to be very good at attributional reasoning. But it turns out that people commonly commit an error—the fundamental attribution error (Ross, 1977). The fundamental attribution error is a pattern of thinking in which people underestimate the causal influence of situational factors on people’s behavior and overestimate the causal influence of personal factors. Recall, as an example, the case of Kitty Genovese. If you first thought that the cause of people’s failure to help was the callous personality of city dwellers, you committed the fundamental attribution error.

The fundamental attribution error is evident in a study in which participants made attributions about the causes—personal or situational—of the content of a political essay (Jones & Harris, 1967). They read an essay expressing either favorable or unfavorable opinions about the Cuban revolutionary leader Fidel Castro. In different experimental conditions, participants learned that the essay writer had either (1) a choice of what to write (pro- or anti-Castro) or (2) no choice, that is, the writer had to compose either a pro- or an anti-Castro essay. After reading the essay, participants judged the writer’s true opinion of Castro; that is, they made an attribution about whether a personal factor (the writer’s political opinion) determined the essay’s content.

Imagine yourself in this experiment. In the choice condition, the task is easy: The writer chose the essay content, so it must match his true opinion of Castro. In the no-choice condition, it also seems easy; because the writer had no choice about the content, you have no idea of his true opinion. You might expect that, in the no-choice condition, participants’ thoughts about the essay writer would be the same whether the essay was pro- or anti-Castro.

Now look at the results (Figure 12.4). In the choice conditions, when the essay said good (or bad) things about Castro, participants of course judged that the writer’s true opinion about Castro was positive (or negative). The surprising result occurred in the no-choice conditions. Here, people committed the fundamental attribution error: They judged that the writer’s personal opinion was the cause of the essay’s content, even though a situational factor, the instruction given to the writer, was its true cause.

526

In another fundamental attribution error study, participants played a quiz game in which pairs of participants were randomly assigned to the role of questioner or answerer (Ross, Amabile, & Steinmetz, 1977). The questioner role was relatively easy: devising quiz questions on a topic of one’s own choosing. The answerer role was difficult: trying to answer the questions devised by the questioner. After playing the game, both the questioner and answerer rated their own “general knowledge” and that of the other participant. A third person, who simply watched the quiz, also rated their general knowledge.

On logical grounds, there is no reason to think one participant was more knowledgeable, in general, than the other. Even if the answerer had trouble answering questions correctly, this doesn’t mean the questioner was smarter. Participants were assigned to these roles randomly. If questioners had been answerers instead, maybe they wouldn’t have looked so smart. Yet when people made their ratings after the quiz game, the winner was the fundamental attribution error: Contestants and observers of the game made personal attributions, judging that the questioner was more knowledgeable than the answerer.

In summary, when considering the causes of others’ behavior, people often overestimate the influence of stable personal qualities (e.g., attitudes, personality, intelligence) and under-estimate the impact of situational forces (but see This Just In).

THIS JUST IN

Attributions, as you have seen, involve thinking (about causes of behavior). In a second aspect of social cognition, thoughts combine with feelings to form social attitudes.

528

A quick question: If you’re the average American, what feature of our society did you encounter about 40,000 times a year while growing up (Comstock & Paik, 1991)?

If there’s a television nearby while you’re reading this, you’ve got a big hint: It’s TV commercials. Businesses spend billions of dollars a year on the tens of thousands of advertisements that you see on TV—plus the Internet, radio, billboards, and so on. What exactly are they trying to accomplish with these ads?

In the long run, they’re trying to get you to buy their products. Yet, many ads don’t actually say, “Buy our products!” They’re more subtle. Advertisers present their product alongside images of sex, power, prestige, or sheer fun. They know that people tend to like sex, power, prestige, and sheer fun, and hope that some of this liking will “rub off” on their product. Advertisers thus try to shape your thoughts and feelings; they want you to think more positive thoughts about Coke than Pepsi (or iPhones than Samsung Galaxys, etc.).

In a word, ads are designed to change your attitudes. An attitude is a combined thought and feeling that is directed toward some person, object, or idea. “I don’t like opera,” “I’m against the death penalty,” and “I think all the Kardashians are terrific” are examples of attitudes.

Different types of factors can cause attitude change. Let’s look at three: (1) persuasive messages and their influence on thinking processes; (2) cognitive dissonance that occurs when people act in a manner inconsistent with their attitudes, and (3) mere exposure, that is, attitude change that occurs merely as a result of repeatedly being exposed to an object.

PERSUASIVE MESSAGES AND ATTITUDE CHANGE. Suppose a political ad tells you that a smiling, well-groomed candidate has “the best experience for the job.” How—that is, through what psychological processes—might the ad affect your attitudes? A theory developed by Shelly Chaiken and her colleagues (Chaiken, Liberman, & Eagly, 1989) identifies two types of thinking processes (or “information processes”) that come into play: systematic and heuristic.

Any given persuasive message may have different effects, depending on whether people are thinking systematically or heuristically. This is evident from research that experimentally manipulated two factors: (1) argument strength (participants heard arguments that were either strong or weak), and (2) information-processing style (in one of two conditions, participants were distracted and given little time to think about the arguments, thereby forcing them to think heuristically rather than systematically). Stronger arguments produced more attitude change—but only when participants could think systematically (Mackie & Worth, 1989).

This finding explains why advertisers often do not even bother to present strong, carefully worded arguments in TV commercials. Much of the time, people are so distracted that they wouldn’t pay enough attention for the arguments to make any difference.

Systematic and heuristic information processes are two psychological routes through which persuasive messages can change attitudes. Another way of changing attitudes is cognitive dissonance.

COGNITIVE DISSONANCE AND ATTITUDE CHANGE. Cognitive dissonance is a negative psychological state that occurs when people recognize that two (or more) of their ideas or actions do not fit together sensibly. Here’s an example. Suppose you are an environmentalist, but to make some extra money, you take a part-time job for a company that is drilling for oil in a wildlife preserve. Taking the job conflicts with—in other words, is “dissonant” with—your environmentalist attitudes. When you recognize this, you may feel bad about the conflict. The bad feeling is known as cognitive dissonance.

529

Cognitive dissonance is psychologically uncomfortable. People thus are motivated to reduce it. One way to reduce dissonance is to change your attitudes. If you convince yourself that “maybe oil drilling in wildlife preserves isn’t all that bad,” your behavior (taking the job) and attitude (“isn’t all that bad”) are not as dissonant anymore. Cognitive dissonance theory predicts that when people engage in behavior that conflicts with their attitudes, the attitudes will change.

Researchers tested this theory by targeting people’s attitudes about a boring task (Festinger & Carlsmith, 1959). In the first step of the experiment, participants spent an hour working on a task that was extraordinarily dull and monotonous (participants turned pegs in a pegboard). Their attitudes toward the task were negative because it was so boring. Afterwards, participants were told that the experiment was over—but in reality, it had barely begun. There were three more research steps:

530

After all this, who do you think would have more positive attitudes about the experiment? You might guess it was the people who were paid $20 rather than $1; people like getting money. But cognitive dissonance theory predicts the opposite result. When paid $20, people experience little cognitive dissonance because they know why they said that the boring task was interesting: to earn a lot of money. But when paid only $1, the low income does not justify the action of lying to a fellow student. In this condition, participants experienced dissonance between two conflicting ideas: (1) “the task was boring” and (2) “I said it was interesting (for a mere $1).” Just as the theory predicted, this cognitive dissonance produced attitude change (Figure 12.6). In the $1, but not the $20, condition, people’s attitudes toward the task changed, becoming positive. People reduced cognitive dissonance by changing their attitudes.

MERE EXPOSURE AND ATTITUDE CHANGE. The first two attitude change techniques we have discussed take effort. To change your attitudes with persuasive messages, companies have to create advertisements. To change people’s attitudes through cognitive dissonance, you have to get them to perform behaviors that contradict those attitudes. A third method of attitude change takes no effort at all. In the mere exposure effect, people’s attitudes toward an object become more positive simply as a result of being exposed to the object repeatedly (Zajonc, 2001).

The mere exposure effect is surprising. You might expect people to be bored by things they see repeatedly, and to like them less. Yet mere exposure reliably makes attitudes more positive (Zajonc, 1980, 2001). In research demonstrating this, participants see pictures of objects. Some are displayed only once or twice, whereas others are shown repeatedly (e.g., 25 times). Participants then indicate how much they like each object. Inevitably, objects seen repeatedly are liked more.

531

TRY THIS!

Mere exposure should seem familiar to you from this chapter’s Try This! activity. There you saw exactly the sort of research procedures that provide scientific evidence of the influence of mere exposure on attitudes.

Three aspects of the mere exposure effect are particularly remarkable:

Mere exposure helps to explain phenomena that otherwise would be puzzling. Consider what happens when people are asked which of two photos they like more: a normal picture of themselves or a left–right reversed picture (created by “flipping” the normal image horizontally). A large majority prefer the reversed photo (Mita, Dermer, & Knight, 1977). Why? Mere exposure! People usually see themselves by looking in a mirror. Mirrors reverse images from left to right. (If you look in the mirror and touch your right ear, you appear to be seeing someone who is touching his or her left ear.) Because people have been exposed to a reversed image of themselves repeatedly, they like that image more.

FROM ATTITUDES TO BEHAVIOR. You’ve seen how attitudes can change. A second question is whether attitudes predict behavior.

How could people’s attitudes not predict their behavior? If you have more positive attitudes toward Pepsi than Coke, you buy Pepsi. If you like candidate Williams more than Washington, you vote for Williams. But, reflecting on what you’ve learned in this chapter, you will realize that people’s attitudes often do not predict their behavior. “Teachers” in the Milgram study did not express negative attitudes toward “learners,” yet they administered seemingly lethal shocks to them. It is unlikely that bystanders in the Darley and Latané study had negative attitudes about the person having a seizure, yet many failed to help.

So when do attitudes predict behavior? Under three circumstances (Ajzen & Fishbein, 1977):

Again, the first question is general, and the second one is specific.

Attitudes most strongly predict behavior when the attitude and the behavior being measured are equally specific (Ajzen & Fishbein, 1977). In our example, people’s attitudes about music in general might predict their overall music purchasing, and their attitudes about a particular band might predict their purchasing behavior toward that band. But when attitudes and behavior are at different levels—for example, attitudes toward music in general, and behavior toward one particular band—the behavior and attitudes may be unrelated. Lots of factors other than your attitudes toward music in general will determine your behavior toward a particular band.

533

In summary, attitudes do predict behavior, but not always. They predict best when situational factors are weak, people think about their attitudes, and measures of attitudes and behavior are equally specific.

Not all beliefs are equally important to the life of society. Some, such as our views of sports teams, soft drinks, or celebrities, play only a minor role. Others affect the fabric of social life. One of these is stereotypic beliefs.

A stereotype is a simplified set of beliefs about the characteristics of members of a group. For instance, a person who believes that everyone of British ancestry is intelligent but aloof, or that all people of Italian ancestry are friendly and emotional, would be said to hold stereotypes about those groups.

Stereotyping has been a part of human life throughout history. One after another, different groups have been the target of negative stereotypes, with members of other groups seeing them as, say, cheap, lazy, aggressive, unintelligent, or unscrupulous. Sometimes stereotypes contain positive content, too; for example, in U.S. society there is a stereotype that Asian Americans are particularly skilled at mathematics (Cheryan & Bodenhausen, 2000). (Though it contains positive content, this stereotype can have social costs for Asian Americans; Wong & Halgin, 2006.)

534

The presence of stereotypic beliefs in a society fosters prejudice, which is a negative attitude toward individuals based on their group membership. Disparaging thoughts and feelings about individuals in a group are the hallmark of prejudice.

Stereotyping and prejudice, in turn, contribute to discrimination, or the unjust treatment of people based on their membership in a particular group. The history of the United States, a nation founded on the notion that all people are created equal, is laced with examples of stereotyping, prejudice, and discrimination. For example, members of the group “women” could not vote until the year 1920. Members of the group “Black” could not be sure of exercising this basic right until the passage of the national Voting Rights Act in 1965.

SOCIAL-COGNITIVE PROCESSES IN STEREOTYPING AND PREJUDICE. Most people are aware of society’s stereotypes—it’s hard to miss them—but do not want to make prejudiced statements. Can they avoid it? Research suggests that it is nearly impossible to stop stereotypic thoughts from coming to mind, but it is possible to stop yourself from saying them in public. To understand this, one has to distinguish between two types of thinking: automatic and controlled (Devine, 1989; also see Chapter 6). Automatic thinking processes occur rapidly—so fast that you cannot control them. Controlled thinking processes require effort and can be influenced by the person doing the thinking (Kahneman, 2011).

In a study of automatic and controlled thinking (Devine, 1989), a researcher first identified two groups of participants: people (1) high and (2) low in racial prejudice toward African Americans. She then gave both groups two tasks:

On the first task, the groups did not differ. The automatic thinking measure revealed that stereotypic thoughts can pop into anyone’s mind. But on the second task, they did differ; low-prejudice people wrote down fewer negative thoughts about African Americans. The findings show that although the minds of people low in racial prejudice did contain society’s negative stereotypes about African Americans, these individuals could exert the mental effort needed to prevent themselves from expressing those thoughts (Devine, 1989).

The lesson is that, if you live in a society with stereotypes, it is almost impossible not to learn them. However, you can counteract prejudice by reminding yourself, through controlled thinking processes, that stereotypes are simplistic, inaccurate portrayals of your fellow citizens.

STEREOTYPE THREAT. The research on stereotyping we just discussed focuses on people who hold stereotypes. But what about those who are the targets of stereotypes?

Research pioneered by Claude Steele (whom you met in the opening pages of this book) shows that stereotypes can harm people’s intellectual performance. This occurs through stereotype threat, which is a negative emotional reaction people have when they realize that they might confirm a negative stereotype about their group (Steele, 1997). When threatened by a stereotype, people perform less well on tests of intellectual ability.

535

Stereotype threat occurs when three factors coincide:

When these occur together, people face a psychological threat: They might confirm the negative stereotype. Even if they know the stereotype to be false, they still may worry about performing in a way that confirms it. These worries can interfere with people’s performance, causing them to achieve at levels below their true ability.

Steele tested this idea in a study of stereotype threat among Black college students taking an intellectually challenging test. White and Black students completed the test in two different experimental conditions. In one, they simply took the test. In the other, they first completed a demographic questionnaire asking them to indicate their racial background. This simple manipulation affected test scores. When participants indicated their race—which activated stereotype threat among Black students—Black students obtained lower test scores, and White students outperformed them to a large degree. When students did not indicate their race, the difference between Black and White students’ scores was substantially reduced (Figure 12.7).

Racial and ethnic stereotypes are not the only ones that can create stereotype threat. Another is gender stereotypes. Women are subjected to a stereotype portraying them as inferior to men in math. Research shows how this stereotype can affect performance. In one experimental condition, male and female participants were told that, on a set of math problems they were to attempt, men’s and women’s test scores usually differed (information consistent with the stereotype). In this condition, men outperformed women. However, in another experimental condition, participants were told that men’s and women’s scores usually did not differ on the math problems they would be given—an instruction that reduced stereotype threat. Here, women and men performed equally well (Steele, 1997). Women therefore had the ability to perform as well as men, but did not do so when subjected to stereotypes.

A meta-analysis of more than 100 stereotype threat studies confirms Steele’s original finding, that stereotype threat can impair intellectual performance (Nguyen & Ryan, 2008). When researchers study the flow of conscious thinking during task performance (see Chapter 9), they find that stereotype threat causes people’s thoughts to “wander,” that is, it causes people to become distracted and to begin thinking about topics other than the task at hand. These mental distractions impair performance (Mrazek et al., 2011).

IMPLICIT MEASURES OF STEREOTYPES AND PREJUDICE. In the study of stereotypes and prejudice, a big scientific challenge involves measurement. Suppose one tried to measure racial stereotypes by asking people, “Do you think that [people of a given ethnic group] are inferior to your group?” You can see the problem: Even people who believe their group is superior may not want to admit that they think this. People might not answer honestly. What you need is some strategy to measure stereotypic beliefs that does not rely on explicit self-reports. Fortunately, psychologists have created one: implicit measures.

An implicit measure is one in which participants cannot tell what a psychologist is measuring. The test does not contain explicit statements that would give participants clues about its purpose, and when taking the test, participants do not directly report on their personal qualities. Nonetheless, the test measures people’s beliefs. To see how implicit tests work, let’s look in detail at one: the implicit association test, or IAT (Greenwald, McGhee, & Schwarz, 1998).

536

In an IAT that measures racial stereotypes, participants are shown two types of stimuli: (1) names that are typically European American (Heather) or African American (Ebony), and (2) words that are either emotionally positive (“happy”) or negative (“evil”). When a name or word appears, participants must press, as quickly as possible, one of two keys on a computer keyboard. On some experimental trials, participants are asked to press the keys according to the following rule:

On other trials, they press the keys according to another rule:

When people’s personal thoughts match the rules of the task, they perform the task more quickly. If, for example, people have mostly positive thoughts about European Americans and negative thoughts about African Americans, they would perform the task more quickly when a negative word is paired with an African American name than when a negative word is paired with a European American name. The measure of stereotyping compares the speed with which people perform the task under the two conditions.

Research shows that European Americans perform the task more quickly under the first condition than the second. Even among European Americans who disavow racial stereotypes when asked explicitly about race, the IAT reveals an implicit preference for European American names (Greenwald et al., 1998). In the years since the IAT’s development, many studies have found that implicit measures reveal stereotypic beliefs and attitudes that are not identified by traditional questionnaires (Petty, Fazio, & Briñol, 2009). Implicit measures of attitudes open a window into the mind, revealing mental content that previously was hidden.

PREJUDICE: ORIGINS AND REDUCTION. How did prejudice originate? And, whatever its origins, how can it be reduced? Let’s conclude our coverage of stereotyping and prejudice by addressing these questions.

When searching for origins of prejudice, there are two factors to consider: (1) psychological processes and (2) historical factors, specifically, the history of the people toward whom prejudice is directed.

Gordon Allport (1954) discussed three psychological processes that contribute to prejudice: generalization, hostility, and scapegoating:

An analysis of general psychological processes, such as Allport’s, is valuable yet limited. It implies that individual cases of prejudice can be understood as examples of the three psychological processes (hostility, generalization, and scapegoating). Yet some cases of prejudice may require not only a psychological analysis, but also a historical one. Consider the case of African Americans.

The psychologists Stanley Gaines and Edward Reed (1995) have argued that a full understanding of prejudice and the African American experience requires that one confront the unique history of this group. The African American experience involves not only generalization, hostility, and scapegoating, but also “the exploitation of a people [who were] stripped forcibly of home, property, and, to some extent, culture and family” (p. 99) in the slave trade.

Gaines and Reed draw on the insights of the great African American scholar W. B. Du Bois, who identified a unique aspect of African American psychological experience. Du Bois explained that African Americans experience a “divided social world” (Gaines & Reed, 1995, p. 99). On the one hand, they are Americans—citizens of the nation. On the other, they find it hard to identify with their nation because, historically, it treated African Americans as property. Even after slavery was abolished, mainstream society subjected African Americans to prejudice and discrimination. If you are African American, they explain, you are “both an American and not an American” (Gaines & Reed, 1995, p. 97)—a citizen, yet aware that your nation has treated your group shamefully. Others similarly explain that the African American experience includes a unique “tension between the individual’s ‘blackness’ and the broader white society” (Sellers et al., 1998, p. 21).

Whatever its roots, can prejudice be reduced? A large body of research (Paluck & Green, 2009) suggests that it can—through intergroup contact. In intergroup contact interventions, psychologists arrange for members of different groups to meet and spend time together. For instance, students from different racial and ethnic groups may spend time together in the same classroom, or employees from different groups may work together on the job. An analysis of results from more than 500 intergroup contact studies yields results that are encouraging for those hoping to reduce prejudice (Pettigrew & Tropp, 2006). Groups that spend more time together are found to express less prejudice toward one another. (This finding should remind you of Sherif’s “Rattlers” and “Eagles” study, discussed earlier in this chapter.)

The cases of stereotyping and prejudice we have reviewed occurred in one region of the world, the United States. Let’s now widen our world view by considering questions of culture.

538