Introducing Social Psychology

There once was a man whose second wife was a vain and self­ish woman. This woman’s two daughters were similarly vain and selfish. The man’s own daughter, however, was meek and unselfish. This

sweet, kind daughter, whom we all know as Cinderella, learned early on

that she should do as she was told, accept ill treatment and insults, and

avoid doing anything to upstage her stepsisters and their mother.

But then, thanks to her fairy godmother, Cinderella was able to

escape her situation for an evening and attend a grand ball, where she

attracted the attention of a handsome prince. When the love-struck

prince later encountered Cinderella back in her degrading home, he

failed to recognize her.

Implausible? The folktale demands that we accept the power of

the situation. In the presence of her oppressive stepmother, Cinder­

ella was humble and unattractive. At the ball, Cinderella felt more

beautiful—and walked and talked and smiled as if she were. In one

situation, she cowered. In the other, she charmed.

The French philosopher-novelist Jean-Paul Sartre (1946) would

have had no problem accepting the Cinderella premise. We humans

are “first of all beings in a situation,” he wrote. “We cannot be distin­

guished from our situations, for they form us and decide our possibili­

ties” (pp. 59-60, paraphrased).

What is social psychology?

What are social psychology’s big ideas?

How do human values influence social psychology?

I knew it all along: Is social psychology simply common sense?

Research methods: How do we do social psychology?

Postscript: Why I wrote this book

4 Chapter 1

social psychology The scientific study of how people think about, influence, and relate to one another.

Throughout this book, sources for information are cited parenthetically. The complete source is provided in the reference section that begins on page R-1.

FIGURE :: 1.1 Social Psychology Is .. .

WHAT IS SOCIAL PSYCHOLOGY? I Define social psychology and explain what it does.

Social psychology is a science that studies the influences of our situations, with spe­ cial attention to how we view and affect one another. More precisely, it is the scientific study of how people think about, influence, and relate to one another (Figure 1.1).

Social psychology lies at psychology’s boundary with sociology. Compared with sociology (the study of people in groups and societies), social psychology focuses more on individuals and does more experimentation. Compared with personality psychology, social psychology focuses less on individuals’ differences and more on hovf individuals, in general, view and affect one another.

Social psychology is still a young science. The first social psychology experi­ ments were reported barely more than a century ago, and the first social psychol­ ogy texts did not appear until approximately 1900 (Smith, 2005). Not until the 1930s did social psychology assume its current form. Not until World War II did it begin to emerge as the vibrant field it is today. And not until the 1970s and beyond did social psychology enjoy accelerating growth in Asia—first in India, then in Hong Kong and Japan, and, recently, in China and Taiwan (Haslam & Kashima, 2010).

Social psychology studies our thinking, influences, and relationships by asking questions that have intrigued us all. Here are some examples:

• Does our social behavior depend more on the objective situations we face or how we construe them? Social beliefs can be self-fulfilling. For example, happily married people will attribute their spouse’s acid remark (“Can’t you ever put that where it belongs?”) to something external (“He must have had a frustrating day”). Unhappily married people will attribute the same remark to a mean disposition (“Is he ever hostile!”) and may respond with a coun­ terattack. Moreover, expecting hostility from their spouse, they may behave resentfully, thereby eliciting the hostility they expect.

• Would people be cruel if ordered? How did Nazi Germany conceive and implement the unconscionable slaughter of 6 million Jews? Those evil acts occurred partly because thousands of people followed orders. They put the prisoners on trains, herded them into crowded “showers,” and poisoned

Introducing Social Psychology

Social psychology is the scientific study of …

Social thinking

• How we perceive ourselves and others

• What we believe • Judgments we make • Our attitudes

Social influence

• Culture • Pressures to conform • Persuasion • Groups of people I

Social relations Prejudice

Aggression Attraction and intimacy Helping

Introducing Social Psychology Chapter 1 5

them with gas. How could people engage in such horrific actions? Were those individuals normal human beings? Stanley Milgram (1974) wondered. So he set up a situation in which people were ordered to administer increasing lev­ els of electric shock to someone who was having difficulty learning a series of words. As discussed in Chapter 6, nearly two-thirds of the participants fully complied.

• To help? Or to help oneself? As bags of cash tumbled from an armored truck one fall day, $2 million was scattered along a Columbus, Ohio, street. Some motorists stopped to help, returning $100,000. Judging from the $1,900,000 that dis­ appeared, many more stopped to help themselves. (What would you have done?) When similar incidents occurred several months later in San Francisco and Toronto, the results were the same: Passersby grabbed most of the money (Bowen, 1988). What situations trigger people to be helpful or greedy? Do some cultural contexts—perhaps villages and small towns—^breed greater helpfulness?

These questions all deal with how people view and affect one another. And that is what social psychology is all about. Social psy­ chologists study attitudes and beliefs, conformity and independence, love and hate.



Tired of looking at the stars. Professor Mueller takes up social psychology. Reprinted with permission of Jason Love at

Identify and describe the central concepts behind social psychology.

In many academic fields, the results of tens of thousands of studies, the conclu­ sions of thousands of investigators, and the insights of hundreds of theorists can be boiled down to a few central ideas. Biology offers us natural selection and adapta­ tion. Sociology builds on concepts such as social structure and organization. Music harnesses our ideas of rhythm, melody, and harmony.

Similarly, social psychology builds on a short list of fundamental principles that will be worth remembering long after you have forgotten most of the details. My short list of “great ideas we ought never to forget” includes these (Figure 1.2), each of which we will explore further in chapters to come.

We Construct Our Social Reality We humans have an irresistible urge to explain behavior, to attribute it to some cause, and therefore to make it seem orderly, predictable, and controllable. You and I may react differently to a situation because we think differently. How we react to a friend’s insult depends on whether we attribute it to hostility or to a bad day.

A 1951 Princeton-Dartmouth football game provided a classic demonstration of how we construct reality (Hastorf & Cantril, 1954; see also Toy & Andrews, 1981). The game lived up to its billing as a grudge match; it was rough and dirty. A Prince­ ton All-American was gang-tackled, piled on, and finally forced out of the game with a broken nose. Fistfights erupted, and there were further injuries on both sides. The whole performance hardly fit the Ivy League image of gentility.

Not long afterward, two psychologists, one from each school, showed films of the game to students on each campus. The students played the role of scientist- observer, noting each infraction as they watched and who was responsible for it.


6 Chapter 1 Introducing Social Psychology

Sott’® Big Ideas in Social Psychol^

yy .>*S5 ■

1. We construct our social ‘ reality

2. Our social intuitions are powerful, sometimes perilous

3. Attitudes shape, and are shaped by, behavior

FIGURE:: 1.2 Some Big Ideas in Soda! Psychology

But they could not set aside their loyalties. The Princeton students, for example, saw twice as many Dartmouth violations as the Dartmouth students saw. The con­ clusion: There is an objective reality out there, but we always view it through the lens of our beliefs and values.

We are all intuitive scientists. We explain people’s behavior, usually with enough speed and accuracy to suit our daily needs. When someone’s behavior is consistent and distinctive, we attribute that behavior to his or her personality. For example, if you observe someone who makes repeated snide comments, you may infer that this person has a nasty disposition, and then you might try to avoid the person.

Our beliefs about ourselves also matter. Do we have an optimistic outlook? Do we see ourselves as in control of things? Do we view ourselves as relatively supe­ rior or inferior? Our answers influence our emotions and actions. How we construe the world, and ourselves, matters.

Our Social Intuitions Are Often Powerful but Sometimes Perilous Our instant intuitions shape our fears (Is flying dangerous?), impressions (Can I trust him?), and relationships (Does she like me?). Intuitions influence presidents in times of crisis, gamblers at the table, jurors assessing guilt, and personnel directors screening applicants. Such intuitions are commonplace.

Indeed, psychological science reveals a fascinating unconscious mind—an intuitive backstage mind—that Freud never told us about. More than psychologists realized until recently, thinking occurs offstage, out of sight. Our intuitive capacities are revealed by studies of what later chapters will explain: “automatic processing,” “implicit memory,” “heuristics,” “spontaneous trait inference,” instant emotions, and nonver­ bal communication. Thinl^g, memory, and attitudes all operate on two levels—one

Introducing Social Psychology Chapter 1 7

conscious and deliberate, the other unconscious and automatic. Today’s researchers call it “dual processing.” We know more than we know we know. We think on two levels—”intuitive” and “deliberate” (Kruglanski & Gigerenzer, 2011). A book title by Nobel laureate psychologist Daniel Kahneman (2011) captures the idea: We do Think­ ing, Fast and Slow.

Intuition is huge, but intuition is also perilous. For example, as we cruise through life, mostly on automatic pilot, we intuitively judge the like­ lihood of things by how easily various instances come to mind. We carry readily available mental images of plane crashes. Thus, most people fear flying more than driving, and many will drive great distances to avoid risking the skies. Actu­ ally, we are many times safer (per mile traveled) in a commercial plane than in a motor vehicle (in the United States, air travel was 170 times safer between 2005 and 2007, reports the National Safety Council [2010]).

Even our intuitions about ourselves often err. We intuitively trust our memories more than we should. We misread our own minds; in experiments, we deny being affected by things that do influence us. We mispredict our own feelings—how bad we’ll feel a year from now if we lose our job or our romance breaks up, and how good we’ll feel a year from now, or even a week from now, if we win our state’s lottery. And we often mispredict our own future. When selecting clothes, people approaching middle age will still buy snug (“1 anticipate shedding a few pounds”); rarely does anyone say, more realistically, “I’d better buy a relatively loose fit; people my age tend to put on pounds.”

Our social intuitions, then, are noteworthy for both their powers and their per­ ils. By reminding us of intuition’s gifts and alerting us to its pitfalls, social psy­ chologists aim to fortify our thinking. In most situations, “fast and frugal” snap judgments serve us well. But in others, in which accuracy matters—such as when needing to fear the right things and spend our resources accordingly—we had best restrain our impulsive intuitions with critical thinking. Our intuitions and uncon­ scious information processing are routinely powerful and sometimes perilous.

Social Influences Shape Our Behavior We are, as Aristotle long ago observed, social animals. We speak and think in words we learned from others. We long to connect, to belong, and to be well thought of. Matthias Mehl and James Pennebaker (2003) quantified their University of Texas students’ social behavior by inviting them to wear microcassette record­ ers and microphones. Once every 12 minutes during their waking hours, the computer-operated recorder would imperceptibly record for 30 seconds. Although the observation period covered only weekdays (including class time), almost 30 percent of the students’ time was spent in conversation. Relationships are a big part of being human.

As social creatures, we respond to our immediate contexts. Sometimes the power of a social situation leads us to act contrary to our expressed attitudes. Indeed, pow­ erfully evil situations sometimes overwhelm good intentions, inducing people to agree with falsehoods or comply with cruelty. Under Nazi influence, many decent people became instruments of the Holocaust. Other situations may elicit great gen­ erosity and compassion. After a major earthquake and tsunami in 2011, Japan was overwhelmed with offers of assistance.

“He didn’t actually threaten me, but Iperceived him as a threat. *

Social cognition matters. Our behavior is influenced not just by the objec­ tive situation but also by how we construe it. © Lee Lorenz/The New Yorker Collection/

8 Chapter 1 Introducing Social Psychology

The power of the situation is also dramatically evident in varying attitudes regarding same-sex relationships. Tell me whether you live in Africa or the Middle East (where most oppose such relationships) or in western Europe, Canada, or Australia/New Zealand, and I will make a reasonable guess as to what your atti­ tude is about these relationships. I will become even more confident in my guess if I know your educational level, the age of your peer group, and the media you watch. Our situations matter.

Our cultures help define our situations. For example, our standards regarding promptness, frankness, and clothing vary with our culture.

• Whether you prefer a slim or a voluptuous body depends on when and where in the world you live.

• Whether you define social justice as equality (all receive the same) or as equity (those who earn more receive more) depends on whether your ideol­ ogy has been shaped more by socialism or by capitalism.

• Whether you tend to be expressive or reserved, casual or formal, hinges partly on your culture and your ethnicity.

• Whether you focus primarily on yourself—your personal needs, desires, and morality—or on your family, clan, and communal groups depends on how much you are a product of modern Western individualism.

Social psychologist Hazel Markus (2005) sums it up: “People are, above all, mal­ leable.” Said differently, we adapt to our social context. Our attitudes and behavior are shaped by external social forces.

Personal Attitudes and Dispositions Also Shape Behavior Internal forces also matter. We are not passive tumbleweeds, merely blown this way and that by the social winds. Our inner attitudes affect our behavior. Our political attitudes influence our voting behavior. Our smoking attitudes influence our sus­ ceptibility to peer pressure to smoke. Our attitudes toward the poor influence our willingness to help them. (As we will see, our attitudes also follow our behavior, which leads us to believe strongly in those things we have committed ourselves to or suffered for.)

Personality dispositions also affect behavior. Facing the same situation, differ­ ent people may react differently. Emerging from years of political imprisonment, one person exudes bitterness and seeks revenge. Another, such as South Africa s Nelson Mandela, seeks reconciliation and unity with his former enemies. Attitudes and personality influence behavior.

Social Behavior Is Biologically Rooted Twenty-first-century social psychology is providing us with ever-growing insights into our behavior’s biological foundations. Many of our social behaviors reflect a deep biological wisdom.

Everyone who has taken introductory psychology has learned that nature and nurture together form who we are. As the area of a rectangle is determined by both its length and its width, so do biology and experience together create us. As evolu- tionary psychologists remind us (see Chapter 5), our inherited human nature predis­ poses us to behave in ways that helped our ancestors survive and reproduce. We carry the genes of those whose traits enabled them and their children to survive and reproduce. Our behavior, too, aims to send our DNA into the future. Thus, evo­ lutionary psychologists ask how natural selection might predispose our actions and reactions when dating and mating, hating and hurting, caring and sharing. Nature also endows us with an enormous capacity to learn and to adapt to varied environ­ ments. We are sensitive and responsive to our social context.

9Introducing Social Psychology Chapter 1

If every psychological event (every thought, every emotion, every behavior) is simultaneously a biological event, then we can also examine the neurobiology that underlies social behavior. What brain areas enable our experiences of love and contempt, helping and aggression, perception and belief? Do extraverts, as some research suggests, require more stimulation to keep their brain aroused? When shown a friendly face, do socially secure people, more than shy people, respond in a brain area concerned with reward? How do brain, mind, and behavior function together as one coordinated system? What does the timing of brain events reveal about how we process information? Such questions are asked by those in social neuroscience (Cacioppo & others, 2010; Klein & others, 2010).

Social neuroscientists do not reduce complex social behaviors, such as help­ ing and hurting, to simple neural or molecular mechanisms. Their point is this: To understand social behavior, we must consider both under-the-skin (biological) and between-skins (social) influences. Mind and body are one grand system. Stress hormones affect how we feel and act: A testosterone dose decreases trust, oxytocin increases it (Bos & others, 2010). Social ostracism elevates blood pressure. Social support strengthens the disease-fighting immune system. Wc are bio-psycho-social organisms. We reflect the interplay of our biological, psychological, and social influ­ ences. And that is why today’s psychologists study behavior from these different levels of analysis.

Social Psychology’s Principles Are Applicable in Everyday Life Social psychology has the potential to illuminate your life, to make visible the sub­ tle influences that guide your thinking and acting. And, as we will see, it offers many ideas about how to know ourselves better, how to win friends and influence people, how to transform closed fists into open arms.

^holars are also applying social psychological insights. Principles of social think­ ing, social influence, and social relations have implications for human health and well-being, for judicial procedures and juror decisions in courtrooms, and for influ­ encing behaviors that will enable an environmentally sustainable human future.

As but one perspective on human existence, psychological science does not answer life’s ultimate questions: What is the meaning of human life? What should be our purpose? What is our ultimate destiny? But social psychology does give us a method for asking and answering some exceedingly interesting and important questions. Social psychology is all about life—your life: your beliefs, your attitudes, your relationships.

The rest of this chapter takes us inside social psychology. Let’s first consider how social psychologists’ own values influence their work in obvious and subtle ways. And then let’s focus on this chapter’s biggest task: glimpsing how we do social psy­ chology. How do social psychologists search for explanations of social thinking, social influence, and social relations? And how might you and I use these analytical tools to think smarter?

social neuroscience An interdisciplinary field that explores the neural bases of social and emotional processes and behaviors, and how these processes and behaviors affect our brain and biology.

Throughout this book, a brief summary will conclude each major section. I hope these summaries will help you assess how well you have learned the material in each section.

SUMMING UP: What Are Social Psychology’s Big Ideas? Social psychology is the scientific study of how people think about, influence, and relate to one another. Its central themes include the following:

• How we construe our social worlds • How our social intuitions guide and sometimes

deceive us

• How our social behavior is shaped by other peo­ ple, by our attitudes and personalities, and by our biology

• How social psychology’s principles apply to our everyday lives and to various other fields of study

10 Chapter 1 Introducing Social Psychology


Identify the ways that values penetrate the work of social psychologists.

Social psychology is less a collection of findings than a set of strategies for answer­ ing questions. In science, as in courts of law, personal opinions are inadmissible. When ideas are put on trial, evidence determines the verdict.

But are social psychologists really that objective? Because they are human beings, don’t their values—their personal convictions about what is desirable and how people ought to behave—seep into their work? If so, can social psychology really be scientific?

There are two general ways that values enter psychology: the obvious and the subtle.

Different sciences offer different perspectives.

Obvious Ways Values Enter Psychology Values enter the picture when social psychologists choose research topics. These choices typically reflect social history (Kagan, 2009). It was no accident that the study of prejudice flourished during the 1940s as fascism raged in Europe; that the 1950s, a time of look-alike fashions and intolerance of differing views, gave us stud­ ies of conformity; that the 1960s saw interest in aggression increase with riots and rising crime rates; that the feminist movement of the 1970s helped stimulate a wave of research on gender and sexism; that the 1980s offered a resurgence of attention to psychological aspects of the arms race; and that the 1990s and the early twenty-first century were marked by heightened interest in how people respond to diversity in culture, race, and sexual orientation. Susan Fiske {2011a) suggests that we can expect future research to reflect today’s and tomorrow’s issues, including immigra­

tion, income inequality, and aging. Values differ not only across time but also across cul­

tures. In Europe, people take pride in their nationalities. The Scots are more self-consciously distinct from the En­ glish, and the Austrians from the Germans, than are simi­ larly adjacent Michiganders from Ohioans. Consequently, Europe has given us a major theory of “social identity,” whereas American social psychologists have focused more on individuals—how one person thinks about others, is influenced by them, and relates to them (Fiske, 2004; Tajfel, 1981; Turner, 1984). Australian social psychologists have drawn theories and methods from both Europe and North America (Feather, 2005).

Values also influence the types of people who are attracted to various disciplines (Campbell, 1975a; Moynihan, 1979). At your school, do the students majoring in the humanities, the arts, the natural sciences, and the social sciences differ noticeably from one another? Do social psychology and sociology attract people who are—for example—relatively eager to challenge tradition, people more inclined to shape the future than preserve the past? And does social science study enhance such inclinations (Dambrun & others, 2009)? Such factors explain why, when psychologist Jonathan Haidt (2011) asked approximately 1000 social psycholo­

gists at a national convention about their politics, 80 to 90 percent raised their hands to indicate they were “liberal.” When he asked for those who were “conservative,”

Introducing Social Psychology

three hands raised. (Be assured that most topics covered in this text—from “How do our attitudes influence our behavior?” to “Does TV violence influence aggres­ sive behavior?”—are not partisan.)

Finally, values obviously enter the picture as the object of social psychological analysis. Social psychologists investigate how values form, why they change, and how they influence attitudes and actions. None of that, however, tells us which values are “right.”

Not-So-Obvious Ways Values Enter Psychology We less often recognize the subtle ways in which value commitments masquerade as objective truth. What are three not-so-obvious ways values enter psychology?

THE SUBJECTIVE ASPECTS OF SCIENCE Scientists and philosophers agree: Science is not purely objective. Scientists do not simply read the book of nature. Rather, they interpret nature, using their own mental categories. In our daily lives, too, we view the world through the lens of our precon­ ceptions. Whether we see a moving light in the sky as a flying saucer or see a face in a pie crust depends on our perceptual set. While reading these words, you have been unaware that you are also looking at your nose. Your mind blocks from awareness something that is there, if only you were predisposed to perceive it. This tendency to prejudge reality based on our expectations is a basic fact about the human mind.

Because scholars at work in any given area often share a common viewpoint or come from the same culture, their assumptions may go unchallenged. What we take for granted—the shared beliefs that some European social psychologists call our social representations (Augoustinos & Innes, 1990; Moscovici, 1988,2001)—are often our most important yet most unexamined convictions. Sometimes, however, some­ one from outside the camp will call attention to those assumptions. During the 1980s, feminists and Marxists exposed some of social psychology’s unexamined assump­ tions. Feminist critics called attention to subtle biases—for example, the political conservatism of some scientists who favored a biological interpretation of gender dif­ ferences in social behavior (Unger, 1985). Marxist critics called attention to competi­ tive, individualist biases—for example, the assumption that conformity is bad and that individual rewards are good. Marxists and feminists, of course, make their own assumptions, as critics of academic “political correctness” are fond of noting. Social psychologist Lee Jussim (2005), for example, argues that progressive social psycholo­ gists sometimes feel compelled to deny group differences and to assume that stereo­ types of group difference are never rooted in reality but always in racism.

In Chapter 3, we will discuss more ways in which our preconceptions guide our interpretations. As those Princeton and Dartmouth football fans remind us, what guides our behavior is less the situation-as-it-is than the situation-as-we-construe-it.

PSYCHOLOGICAL CONCEPTS CONTAIN HIDDEN VALUES Implicit in our understanding that psychology is not objective is the realization that psychologists’ own values may play an important part in the theories and judg­ ments they support. Psychologists may refer to people as mature or immature, as well adjusted or poorly adjusted, as mentally healthy or mentally ill. They may talk as if they were stating facts, when they are really making value judgments. The fol­ lowing are examples: DEFINING THE GOOD LIFE Values influence our idea of how best to live. The personality psychologist Abraham Maslow, for example, was known for his sensitive descriptions of “self-actualized” people—people who, with their needs for survival, safety, belonging, and self-esteem satisfied, go on to fulfill their human potential. He described, among other individuals, Thomas Jefferson, Abraham Lincoln, and Eleanor Roosevelt. Few readers noticed that Maslow, guided by his own values, selected his sample of self-actualized people himself. The resulting description of self-actualized

Chapter 1 11










culture The enduring behaviors, ideas, attitudes, and traditions shared by a large group of people and transmitted from one generation to the next

social representations A society’s widely held ideas and values, including assumptions and cultural ideologies. Our social representations help us make sense of our world.

12 Chapter 1 Introducing Social Psychology

Hidden (and not-so-hidden) values seep into psycho­ logical advice. They permeate popular psychology books that offer guidance on living and loving.

personalities—as spontaneous, autonomous, mystical, and so forth—reflected Maslow’s personal values. Had he begun with some­ one else’s heroes—say, Napoleon, Alexander the Great, and John D. Rockefeller—his resulting description of self-actualization would have differed (Smith, 1978).

PROFESSIONAL ADVICE Psychological advice also reflects the advice giver’s personal values. When mental health profession­ als advise us how to get along with our spouse or our co-workers, when child-rearing experts tell us how to handle our children, and when some psychologists advocate living free of concern for others’ expectations, they are expressing their personal values. (In Western cultures, those values usually will be individualistic—encouraging what feels best for “me.” Non-Western cultures more often encour­ age what is best for “we.”) Unaware of those hidden values, many people defer to the “professional.” But professional psychologists cannot answer questions of ultimate moral obligation, of purpose and direction, and of life’s meaning.

FORMING CONCEPTS Hidden values even seep into psychol­ ogy’s research-based concepts. Pretend you have taken a personality test and the psychologist, after scoring your answers, announces: “You scored high in self-esteem. You are low in anxiety. And you have exceptional ego-strength.” “Ah,” you think, “1 suspected as much, but it feels good to know that.” Now another psychologist gives you

a similar test, which asks some of the same questions. Afterward, the psychologist informs you that you seem defensive, for you scored high in “repressiveness.” “How could this be?” you wonder. “The other psychologist said such nice things about me.” It could be because all these labels describe the same set of responses (a tendency to say nice things about oneself and not to acknowledge problems). ShaU we call it high self-esteem or defensiveness? The label reflects the judgment.

LABELING Value judgments, then, are often hidden within our social psycho­ logical language—but that is also true of everyday language:

• Whether we label a quiet child as “bashful” on “cautious,” as “holding back or as “an observer,” conveys a judgment.

• Whether we label someone engaged in guerrilla warfare a “terrorist” or a “freedom fighter” depends on our view of the cause.

• Whether we view wartime civilian deaths as “the loss of innocent lives” or as “collateral damage” affects our acceptance of such.

• Whether we call public assistance “welfare” or “aid to the needy” reflects our political views.

• When “they” exalt their country and people, it is nationalism; when “we” do it, it is patriotism.

• Whether someone involved in an extramarital affair is practicing “open mar­ riage” or “adultery” depends on one’s personal values.

• “Brainwashing” is social influence we do not approve of. • “Perversions” are sex acts we do not practice.

As these examples indicate, values lie hidden within our cultural definitions of mental health, our psychological advice for living, our concepts, and our psycho­ logical labels. Throughout this book, 1 will call your attention to additional exam­ ples of hidden values. The point is never that the implicit values are necessarily bad. The point is that scientific interpretation, even at the level of labeling phenom­ ena, is a human activity. It is therefore inevitable that prior beliefs and values will influence what social psychologists think and write.

Introducing Social Psychology Chapter 1 13

Should we dismiss science because it has its subjective side? Quite the contrary: The realization that human thinking always involves interpretation is precisely why we need researchers with varying biases to undertake scientific analysis. By constantly checking our beliefs against the facts, we restrain our biases. System­ atic observation and experimentation help us clean the lens through which we see reality.

SUMMING UP: How Do Human Values Influence Social Psychology?

• Social psychologists’ values penetrate their work in obvious ways, such as their choice of research topics and the types of people who are attracted to various fields of study.

• They also do this in subtler ways, such as their hid­ den assumptions when forming concepts, choosing labels, and giving advice.

• This penetration of values into science is not a rea­ son to fault social psychology or any other science. That human thinking is seldom dispassionate is precisely why we need systematic observation and experimentation if we are to check our cherished ideas against reality.


Explore how social psychology’s theories provide new insight into the human condition.

Many of the conclusions presented in this book may already have occurred to you, for social psychological phenomena are all around you. We constantly observe people thinking about, influencing, and relating to one another. It pays to discern what a facial expression predicts, how to get someone to do something, or whether to regard another as friend or foe. For centuries, philosophers, novelists, and poets have observed and commented on social behavior.

Does this mean that social psychology is just common sense in fancy words? Social psychology faces two contradictory criticisms: first, that it is trivial because it documents the obvious; second, that it is dangerous because its findings could be used to manipulate people.

Chapter 7 explores the second criticism. Here, let’s examine the first objection. Do social psychology and the other social sciences simply formalize what any

amateur already knows intuitively? Writer Cullen Murphy (1990) took that view: “Day after day social scientists go out into the world. Day after day they discover that people’s behavior is pretty much what you’d expect.” Nearly a half-century earlier, historian Arthur Schlesinger, Jr. (1949), reacted with similar scorn to social scientists’ studies of American World War II soldiers. Sociologist Paul Lazarsfeld (1949) reviewed those studies and offered a sample with interpretive comments, a few of which I paraphrase:

1. Better-educated soldiers suffered more adjustment problems than did less- educated soldiers. (Intellectuals were less prepared for battle stresses than were street-smart people.)

2. Southern soldiers coped better with the hot South Sea Island climate than did Northern soldiers. (Southerners are more accustomed to hot weather.)

14 Chapter 1 Introducing Social Psychology

hindsight bias The tendency to exaggerate, after learning an outcome, one’s ability to have foreseen how something turned out. Also known as the l-knew-it- all-along phenomenon.

3. White privates were more eager for promotion than were Black privates. (Years of oppression take a toll on achievement motivation.)

4. Southern Blacks preferred Southern to Northern White officers. (Southern officers were more experienced and skilled in interacting with Blacks.)

As you read those findings, did you agree that they were basically common sense? If so, you may be surprised to learn that Lazarsfeld went on to say, “Every one of these statements is the direct opposite of what was actually found.” In reality, the studies found that less-educated soldiers adapted more poorly. Southerners were not more likely than northerners to adjust to a tropical climate. Blacks were more eager than Whites for promotion, and so forth. “If we had mentioned the actual results of the investigation first [as Schlesinger experienced], the reader would have labeled these ‘obvious’ also.

One problem with common sense is that we invoke it after we know the facts. Events are far more “obvious” and predictable in hindsight than beforehand. Exper­ iments reveal that when people learn the outcome of an experiment, that outcome sudderUy seems unsurprising—much less surprising than it is to people who are simply told about the experimental procedure and the possible outcomes (Slovic & Fischhoff, 1977).

Likewise, in everyday life we often do not expect something to happen until it does. Then we suddenly see clearly the forces that brought the event about and feel unsurprised. Moreover, we may also misremember our earlier view (Blank & others, 2008; Nestler & others, 2010). Errors in judging the future’s foreseeability and in remembering our past combine to create hindsight bias (also called the I-knew-it-all-along phenomenon).

Thus, after elections or stock market shifts, most commentators find the turn of events unsurprising: “The market was due for a correction.” After the 2010 Gulf oil disaster, it seemed obvious—in hindsight—that BP employees had taken some shortcuts and ignored warnings, and that government oversight was lax. As the Danish philosopher-theologian Soren Kierkegaard put it, “Life is lived forwards, but understood backwards.”

If hindsight bias is pervasive, you may now be feeling that you already knew about this phenomenon. Indeed, almost any conceivable result of a psychological experiment can seem like common sense—after you know the result.

You can demonstrate the phenomenon yourself. Take a group of people and tell half of them one psychological finding and the other half the opposite result. For example, tell half as follows:

Social psychologists have found that, whether choosing friends or falling in love, we are most attracted to people whose traits are different from our own. There seems to be wisdom in the old saying “Opposites attract.”

In hindsight, events seem obvious and predictable.

Tell the other half: Social psychologists have found that, whether choosing friends or falling in love, we are most attracted to people whose traits are similar to our own. There seems to be wisdom in the old saying “Birds of a feather flock together.”

Ask the people first to explain the result. Then ask them to say whether it is “surprising” or “not surprising.” Virtually all will find a good explanation for whichever result they were given and will say it is “not surprising.”

Indeed, we can draw on our stockpile of proverbs to make almost any result seem to make sense. If a social psycholo­ gist reports that separation intensifies romantic attraction, John Q. Public responds, “You get paid for this? Everybody knows that ‘absence makes the heart grow fonder.'” Should

it turn out that separation weakens attraction, John will say, “My grandmother could have told you, ‘Out of sight, out of mind.'”

Introducing Social Psychology Chapter 1 15

focus ON I Knew It All Along

Cullen Murphy (1990), managing editor of the At/ant/c, faulted “sociology, psychology, and other social sciences for too often merely discerning the obvious or confirming the commonplace.” His own casual survey of social science findings “turned up no ideas or conclusions that can’t be found in Bart/ett’s or any other encyclopedia of quo­ tations.” Nevertheless, to sift through competing sayings, we need research. Consider some dueling proverbs:

1$ It more true that… We should keep our eye on the

prize. Too many cooks spoil the broth. The pen is mightier than the sword. You can’t teach an old dog new

tricks. Blood is thicker than water. He who hesitates is lost. Forewarned is forearmed.

Karl Teigen (1986) must have had a few chuckles when he asked University of Leicester (England) students to evaluate actual proverbs and their opposites. When given the proverb “Fear is stronger than love,” most rated it as true. But so did students who were given its reversed form, “Love is stronger than fear.” Like­ wise, the genuine proverb “He that is fallen cannot help him who is down” was rated highly; but so too was “He that is fallen can help him who is down.” My favorites, however, were two highly rated proverbs: “Wise men make proverbs and fools repeat them” (authentic) and its made-up counterpart, “Fools make .. proverbs and wise men repeat them.” For more dueling proverbs, see “Focus On: I Knew It All Along.”

The hindsight bias creates a problem for many psychology students. Some­ times results are genuinely surprising (for example, that Olympic bronze med­ alists take more joy in their achievement than do silver medalists). More often, when you read the results of experiments in your textbooks, the material seems easy, even obvious. When you later take a multiple-choice test on which you must choose among several plausible conclusions, the task may become surprisingly difficult. “I don’t know what happened,” the befuddled student later moans. “I thought I knew the material.”

The 1-knew-it-all-along phenomenon can have unfortunate consequences. It is conducive to arrogance—an overestimation of our own intellectual powers. More­ over, because outcomes seem as if they should have been foreseeable, we are more likely to blame decision makers for what are in retrospect “obvious” bad choices than to praise them for good choices, which also seem “obvious.”

Starting after the morning of 9/11 and working backward, signals pointing to the impending disaster seemed obvious. A U.S. Senate investigative report listed the missed or misinterpreted clues (Gladwell, 2003): The CIA knew that al Qaeda operatives had entered the country. An FBI agent sent a memo to head­ quarters that began by warning “the Bureau and New York of the possibility of a coordinated effort by Osama bin Laden to send students to the United States to attend civilian aviation universities and colleges.” The FBI ignored that accu­ rate warning and failed to relate it to other reports that terrorists were planning to use planes as weapons. The president received a daily briefing titled “Bin Laden Determined to Strike Inside the United States” and stayed on holiday. “The dumb fools!” it seemed to hindsight critics. “Why couldn’t they connect the dots?”

But what seems clear in hindsight is seldom clear on the front side of his­ tory. The intelligence community is overwhelmed with “noise”—piles of useless information surrounding the rare shreds of useful information. Analysts must therefore be selective in deciding which to pursue, and only when a lead is pur­ sued does it stand a chance of being connected to another lead. In the 6 years

Or that… We should keep our nose to the

grindstone. Two heads are better than one, Actions speak louder than words. You’re never too old to learn.

Many kinfolk, few friends. Look before you leap. Don’t cross the bridge until you

come to it.

16 Chapter 1 Introducing Social Psychology











before 9/11, the FBI’s counterterrorism unit could never have pursued all 68,000 uninvestigated leads. In hindsight, the few useful ones are now obvious.

In the aftermath of the 2008 world financial crisis, it seemed obvious that govern­ ment regulators should have placed safeguards against the ill-fated bank lending practices. But what was obvious in hindsight was unforeseen by the chief American regulator, Alan Greenspan, who found himself “in a state of shocked disbelief” at the economic collapse.

We sometimes blame ourselves for “stupid mistakes”—perhaps for not having handled a person or a situation better. Looking back, we see how we should have handled it. “I should have known how busy I would be at the semester’s end and started that paper earlier.” But sometimes we are too hard on ourselves. We forget that what is obvious to us now was not nearly so obvious at the time.

Physicians who are told both a patient’s symptoms and the cause of death {as determined by autopsy) sometimes wonder how an incorrect diagnosis could have been made. Other physicians, given only the symptoms, do not find the diagnosis nearly so obvious (Dawson & others, 1988). Would juries be slower to assume mal­ practice if they were forced to take a foresight rather than a hindsight perspective?

What do we conclude—that common sense is usually wrong? Sometimes it is. At other times, conventional wisdom is right—or it falls on both sides of an issue. Does happiness come from knowing the truth, or from preserving illusions? From being with others, or from living in peaceful solitude? Opinions are a dime a dozen. No matter what we find, there will be someone who foresaw it. (Mark Twain jested that Adam was the only person who, when saying a good thing, knew that nobody had said it before.) But which of the many competing ideas best fit reality? Research can specify the circumstances under which a commonsense truism is valid.

The point is not that common sense is predictably wrong. Rather, common sense usually is right—after the fact. We therefore easily deceive ourselves into think­ ing that we know and knew more than we do and did. And that is precisely why we need science to help us sift reality from illusion and genuine predictions from easy hindsight.

SUMMING UP: l Knew It All Along: Is Social Psychology Simply Common Sense?

• Social psychology is criticized for being trivial • This hindsight bias (the l-knew-it-all-along phenom- because it documents things that seem obvious. enon) often makes people overconfident about the

. Experiments, however, reveal that outcomes are validity of their judgments and predictions, more “obvious” after the facts are known.


I Examine the methods that make social psychology a science. We have considered some of the intriguing questions social psychology seeks to answer. We have also seen how subjective, often unconscious, processes influence social psychologists’ work. Now let’s consider how social psychologists go about doing research.

Introducing Social Psychology Chapter 1 17

In their quest for insight, social psychologists propose theories that organize their observations and imply testable hypotheses and practical predictions. To test a hypothesis, social psychologists may do research that predicts behavior using cor­ relational studies, often conducted in natural settings. Or they may seek to explain behavior by conducting experiments that manipulate one or more factors under con­ trolled conditions. Then they may explore ways to apply their findings to improve people’s everyday lives.

We are all amateur social psychologists. People-watching is a universal hobby. As we observe people, we form ideas about how human beings think about, influ­ ence, and relate to one another. Professional social psychologists do the same, only more systematically (by forming theories) and painstakingly (often with experi­ ments that create miniature social dramas that pin down cause and effect).

Forming and Testing Hypotheses We social psychologists have a difficult time thinking of anything more fascinating than human existence. As we wrestle with human nature to pin down its secrets, we organize our ideas and findings into theories. A theory is an integrated set of prin­ ciples that explain and predict observed events. Theories are a scientific shorthand.

In everyday conversation, “theory” often means “less than fact”—a middle rung on a confidence ladder from guess to theory to fact. Thus, people may dismiss Charles Darwin’s theory of evolution as “just a theory.” Indeed, notes Alan Leshner (2005), chief officer of the American Association for the Advancement of Science, “Evolution is only a theory, but so is gravity.” People often respond that gravity is a fact—^but the fact is that your keys fall to the ground when dropped. Gravity is the theoretical explanation that accounts for such observed facts.

To a scientist, facts and theories are apples and oranges. Facts are agreed-upon statements about what we observe. Theories are ideas that summarize and explain facts. “Science is built up with facts, as a house is with stones,” wrote the French scientist Jules Henri Poincare, “but a collection of facts is no more a science than a heap of stones is a house.”

Theories not only summarize but also imply testable predictions, called hypotheses. Hypotheses serve several purposes. First, they allow us to test a theory by suggesting how we might try to falsify it. Second, predictions give direction to research and sometimes send investigators looking for things they might never have thought of. Third, the predictive feature of good theories can also make them practical. A complete theory of aggression, for example, would predict when to expect aggres­ sion and how to control it. As the pio­ neering social psychologist Kurt Lewin declared, “There is nothing so practical as a good theory.”

Consider how this works. Say we observe that people who loot, taunt, or attack often do so in groups or crowds. We might therefore theorize that being part of a crowd, or group, makes individuals feel anonymous and lowers their inhibitions. How could we test this theory? Perhaps (I’m playing with this theory) we could devise a laboratory experiment that simulates aspects of execution by electric chair. What if we asked individuals in groups to administer punishing shocks to a hapless victim









theory An integrated set of principles that explain and predict observed events.

hypothesis A testable proposition that describes a relationship that may exist between events.

For humans, the most fascinating subject is people. ® V/arren Mil[ef/The New Yorker Collection/www.cartoonbankxom

18 Chapter 1 Introducing Social Psychology

field research Research done in natural, real-life settings outside the laboratory.

correlational research The study of the naturally occurring relationships arr^ong variables.

experimental research Studies that seek clues to cause-effect relationships by manipulating one or more factors (independent variables) while controlling others (holding them constant).

without knowing which member of the group was actually shocking the victim? Would these individuals administer stronger shocks than individuals acting alone, as our theory predicts?

We might also manipulate anonymity: Would people deliver stronger shocks if they were wearing masks? If the results confirm our hypothesis, they might suggest some practical applications. Perhaps police brutality could be reduced by having officers wear large name tags and drive cars identified with large numbers, or by videotaping their arrests—all of which have, in fact, become common practice in many cities.

But how do we conclude that one theory is better than another? A good theory

• effectively summarizes many observations, and • makes clear predictions that we can use to

• confirm or modify the theory, • generate new exploration, and • suggest practical applications.

When we discard theories, usually it is not because they have been proved false. Rather, like old cars, they are replaced by newer, better models.

Correlational Research: Detecting Natural Associations Let’s now go backstage and see how social psychology is done. This glimpse behind the scenes should be just enough for you to appreciate findings discussed later. Understanding the logic of research can also help us think critically about everyday social events.

Social psychological research varies by location. It can take place as laboratory research {a controlled situation) or as field research (everyday situations). And it varies by method—whether correlaUonal (asking whether two or more factors are naturally associated) or experimental (manipulating some factor to see its effect on another). If you want to be a critical reader of psychological research reported in the media, it will pay you to understand the difference between correlational and experimental research.

Let’s first consider the advantages of correlational research (often involving important variables in natural settings) and its major disadvantage (ambiguous interpretation of cause and effect). As we will discuss in Chapter 14, today’s psy­ chologists relate personal and social factors to human health. In search of possible links between socioeconomic status and health, Douglas Carroll, George Davey Smith, and Paul Bennett (1994) ventured into Glasgow, Scotland’s old graveyards. As a measure of health, they noted from grave markers the life spans of 843 indi­ viduals. As an indication of status, they measured the height of the grave pillars, reasoning that height reflected cost and therefore affluence. As Figure 1.3 shows, taller grave markers were related to longer lives, for both men and women.

Carroll and colleagues report that other researchers, using contemporary data, have confirmed the status-longevity correlation. Scottish postal-code regions hav­ ing the least overcrowding and unemployment also have the greatest longevity. In the United States, income correlates with longevity (poor and lower-status peo­ ple are more at risk for premature death). In today’s Britain, occupational status correlates with longevity. One study followed 17,350 British civil service workers over 10 years. Compared with top-grade administrators, those at the professional- executive grade were 1.6 times more likely to have died. Clerical workers were 2.2 times and laborers 2.7 times more likely to have died (Adler & others, 1993,1994). Across times and places, the status-health correlation seems reliable.

CORRELATION AND CAUSATION The status-longevity question illustrates the most irresistible thinking error made by both amateur and professional social psychologists: When two factors such as

Introducing Social Psychology Chapter 1 19

Low Medium High Height of grave pillars

FIGURE :: 1.3 Correlating Status and Longevity Tall grave pillars commemorated people who also tended to live longer.

status and health go together, it is tempting to conclude that one is causing the other. Status, we might presume, somehow protects a person from health risks. But might it be the other way around? Could it be that health promotes vigor and suc­ cess? Perhaps people who live longer simply have more time to accumulate wealth (enabling them to have more expensive grave markers). Or might a third variable, such as diet, be involved (did wealthy and working-class people tend to eat differ­ ently)? Correlations indicate a relationship, but that relationship is not necessarily one of cause and effect. Correlational research allows us to predict, but it cannot tell us whether changing one variable (such as social status) will cause changes in another (such as health).

The correlation-causation confusion is behind much muddled thinking in popu­ lar psychology. Consider another very real correlation—between self-esteem and academic achievement. Children with high self-esteem tend also to have high aca­ demic achievement. (As with any correlation, we can also state this the other way around: High achievers tend to have high self-esteem.) Why do you suppose that is true (Figure 1.4)?

Some people believe a “healthy self- concept” contributes to achievement. Thus, boosting a child’s self-image may also boost school achievement. Believing so, 30 U.S. states have enacted more than 170 self-esteem-promoting statutes.

But other people, including psycholo­ gists William Damon (1995), Robyn Dawes (1994), Mark Leary (1999), Martin Seligman (1994, 2002), and Roy Baumeister with John Tierney (2011), doubt that self-esteem is really “the armor that protects kids” from under­ achievement (or drug abuse and delin­ quency). Perhaps it is the other way around: Perhaps problems and failures cause low self-esteem. Perhaps self­ esteem often reflects the reality of how

Commemorative markers in Glasgow Cathedral graveyard.

20 Chapter 1 Introducing Social Psychology

FIGURE:: 1.4 Correlation and Causations When two variables correlate, any combination of three expla­ nations is possible. Either one may cause the other, or both may be affected by an underlying “third factor.”

X Correlation


Social status Health

things are going for us. Perhaps self-esteem grows from hard-won achievements. Do well and you will feel good about yourself; goof off and fail and you will feel like a dolt. A study of 635 Norwegian schoolchildren showed that a (legitimately earned) string of gold stars by one’s name on the spelling chart and accompany­ ing praise from the admiring teacher can boost a child’s self-esteem (Skaalvik & Hagtvet, 1990). Or perhaps, as in a study of nearly 6,000 German seventh-graders, the traffic between self-esteem and academic achievements runs both ways (Traut- wein & Liidtke, 2006).

It is also possible that self-esteem and achievement correlate because both are linked to underlying intelligence and family social status. That possibility was raised in two studies—one a nationwide sample of 1,600 young American men and the other of 715 Minnesota youngsters (Bachman & O’Malley, 1977; Maruyama & others, 1981). When the researchers mathematically removed the predictive power of intelligence and family status, the relationship between self-esteem and achieve­ ment evaporated.

Correlations quantify, with a coefficient known as r, the degree of relationship between two factors—from -1.0 (as one factor score goes up, the other goes down) through 0 to -t-1.0 (the two factors’ scores rise and fall together). Scores on self­ esteem and depression tests correlate negatively (about -.6). Identical twins’ intel­ ligence scores correlate positively (above +.8). The great strength of correlational research is that it tends to occur in real-world settings where we can examine fac­ tors such as race, gender, and social status (factors that we cannot manipulate in the laboratory). Its great disadvantage lies in the ambiguity of the results. This point is so important that even if it fails to impress people the first 25 times they hear it, it is worth repeating a twenty-sixth time: Knowing that two variables change together (correlate) enables us to predict one when we know the other, but correlation does not specify cause and effect.

Advanced correlational techniques can, however, suggest cause-effect rela­ tionships. Time-lagged correlations reveal the sequence of events (for example, by indicating whether changed achievement more often precedes or follows changed self-esteem). Researchers can also use statistical techniques that extract the influ­ ence of “confounded” variables, as when the correlation between self-esteem and achievement evaporated after extracting intelligence and family status. Recall our earlier mention of a third variable, such as diet. Thus, the Scottish research team wondered whether the status-longevity relationship would survive their remov­ ing the effect of cigarette smoking, which is now much less common among those of higher status. It did, which suggested that some other factors, such as

increased stress and decreased feelings of control, may also account for poorer people’s earlier mortality.

SURVEY RESEARCH How do we measure variables such as status and health? One way is by surveying representative samples of people. If survey researchers want to describe a whole population (which for many psychology surveys is not the aim), then they will obtain a representative group by taking a random sample—one in which every person in the population being studied has an equal chance of inclusion. With this procedure any subgroup of people—^blondes, joggers, liberals—will tend to be represented in the survey to the extent that they are represented in the total population.

Whether we survey people in a city or in a whole country, 1,200 randomly selected participants will enable us to be 95 percent confident of describing the entire popu­ lation with an error margin of 3 percentage points or less. Imagine a huge jar filled with beans, 50 percent red and 50 percent white. Randomly sample 1,200 of these, and you will be 95 percent certain to draw out between 47 percent and 53 percent red beans—regardless of whether the jar contains 10,000 beans or 100 million beans. If we think of the red beans as supporters of one presidential candidate and the white beans as supporters of the other candidate, we can understand why, since 1950, the Gallup polls taken just before U.S. national elections have diverged from election results by an average of less than 2 percent. As a few drops of blood can speak for the whole body, so can a random sample speak for a population.

Bear in mind that polls do not literally predict voting; they only describe public opinion at the moment they are taken. Public opinion can shift. To evaluate sur­ veys, we must also bear in mind four potentially biasing influences: unrepresenta­ tive samples, question order, response options, and question wording.

UNREPRESENTATIVE SAMPLES How closely the sample represents the popu­ lation under study matters greatly. In 1984, columnist Ann Landers accepted a let­ ter writer’s challenge to poll her readers on the question of whether women find affection more important than sex. Her question: “Would you be content to be held close and treated tenderly and forget about ‘the act’?” Of the more than 100,000 women who replied, 72 percent said yes. An avalanche of worldwide publicity followed. In response to critics, Landers (1985, p. 45) granted that “the sampling may not be representative of all American women. But it does provide honest— valuable—insights from a cross section of the public. This is because my column is

Chapter 1 21

Even exit polls require a random (and therefore representative) sample of voters.

random sampling Survey procedure in which every person in the population being studied has an equal chance of inclusion.

22 Chapter 1 Introducing Social Psychology

SRC’s Survey Services Laboratory at the University of Michigan’s Institute for Social Research has inter­ viewing carrels with moni­ toring stations. Staff and visitors must sign a pledge to honor the strict confidentiality of all interviews.

read by people from every walk of life, approximately 70 million of them.” Still, one wonders, are the 70 million readers representative of the entire population? And are the 1 in 700 readers who took the trouble to reply to the survey representative of the 699 in 700 who did not?

The importance of representativeness was effectively demonstrated in 1936 when a weekly newsmagazine. Literary Digest, mailed a postcard presidential election poll to 10 million Americans. Among the more than 2 million returns, Alf Landon won by a landslide over Franklin D. Roosevelt. When the actual votes were counted a few days later, Landon carried only two states. The magazine had sent the poll only to people whose names it had obtained from telephone books and automobile registrations—thus ignoring the millions of voters who could afford neither a tele­ phone nor a car (Cleghom, 1980). ORDER OF QUESTIONS Given a representative sample, we must also contend with other sources of bias, such as the order of questions in a survey. Americans’ support for civil unions of gays and lesbians rises if they are first asked their opin­ ion of gay marriage, compared with which civil unions seem a more acceptable alternative (Moore, 2004a, 2004b). RESPONSE OPTIONS Consider, too, the dramatic effects of the response options. When Joop van der Plight and co-workers (1987) asked English voters what percent­ age of Britain’s energy they wished came from nuclear power, the average preference was 41 percent. They asked other voters what percentage they wished came from (1) nuclear, (2) coal, and (3) other sources. The average preference for nuclear power among these respondents was 21 percent. WORDING OF QUESTIONS The precise wording of questions may also influ­ ence answers. One poll found that only 23 percent of Americans thought their gov­ ernment was spending too much “on assistance to the poor.” Yet 53 percent thought the government was spending too much “on welfare” {Time, 1994). Likewise, most people favor cutting “foreign aid” and increasing spending “to help hungry people in other nations” (Simon, 1996).

Survey questioning is a very delicate matter. Even subtle changes in the tone of a question can have marked effects (Krosnick & Schuman, 1988; Schuman & Kalton, 1985). “Forbidding” something may be the same as “not allowing” it. But in 1940, 54 percent of Americans said the United States should “forbid” speeches against democracy, and 75 percent said the United States should “not allow” them. Even when people say they feel strongly about an issue, a question’s form and wording may affect their answer.

Introducing Social Psychology Chapter 1 23






by Garry Trudeau

Survey researchers must be sensitive to subtle and not-so-subtle biases. DOONESBURY © G. B. Trudeau. Reprinted with permission of Universal Press Syndicate. All rights reserved.

Order, response, and wording effects enable political manipulators to use sur­ veys to show public support for their views. Consultants, advertisers, and physi­ cians can have similar disconcerting influences upon our decisions by how they frame our choices. No wonder the meat lobby in 1994 objected to a new U.S. food labeling law that required declaring ground beef, for example, as “30 percent fat,” rather than “70 percent lean, 30 percent fat.” To 9 in 10 college students, a condom seems effective if its protection against the AIDS virus has a “95 percent success rate.” Told that it has a “5 percent failure rate,” only 4 in 10 students say they find it effective (Linville & others, 1992).

Framing research also has applications in the definition of everyday default options:

• Opting in or out of organ donation. In many countries, people decide, when renewing their drivers’ license, whether they want to make their body available for organ donation. In countries where the default option is yes but one can “opt out,” nearly 100 percent of people choose to be donors. In the United States, Britain, and Germany, where the default option is no but one can “opt in,” approximately 1 in 4 choose to be donors (Johnson & Goldstein, 2003).

• Opting in or out of retirement savings. For many years, American employees who wanted to defer part of their compensation to a 401(k) retirement plan had to elect to lower their take-home pay. Most chose not to do so. A 2006 pension law, influenced by framing research, reframed the choice. Now com­ panies are given an incentive to enroll their employees automatically in the plan and to allow them to opt out (and to raise their take-home pay). The choice was preserved. But one study found that with the “opt out” framing, enrollments soared from 49 to 86 percent (Madrian & Shea, 2001).

‘The lesson of framing research is told in the story of a sultan who dreamed he had lost all his teeth. Summoned to interpret the dream, the first interpreter said, “Alas! The lost teeth mean you will see your family members die.” Enraged, the sul­ tan ordered 50 lashes for this bearer of bad news. When a second dream interpreter heard the dream, he explained the sultan’s good fortune: “You will outlive your whole clan!” Reassured, the sultan ordered his treasurer to go and fetch 50 pieces of gold for this bearer of good news. On the way, the bewildered treasurer observed to the second interpreter, “Your interpretation was no different from that of the first interpreter.” “Ah yes,” the wise interpreter replied, “but remember: What matters is not only what you say, but how you say it.”

framing The way a question or an issue is posed; framing can influence people’s decisions and expressed opinions.

A young monk was once rebuffed when asking if he could smoke while he prayed. Ask a different question, advised a friend: Ask if you can pray while you smoke (Crossen, 1993).

24 Chapter 1

independent variable The experimental factor that a researcher manipulates.

Note: Obesity correlated with marital status and income.

Whom the men were shown—a normal or an overweight woman—was the independent variable.

Introducing Social Psychology

Experimental Research: Searching for Cause and Effect The difficulty of discerning cause and effect among naturally correlated events prompts most social psychologists to create laboratory simulations of everyday pro­ cesses whenever this is feasible and ethical. These simulations are akin to aeronau­ tical wind tunnels. Aeronautical engineers do not begin by observing how flying objects perform in various natural environments. The variations in both atmo­ spheric conditions and flying objects are too complex. Instead, they construct a sim­ ulated reality in which they can manipulate wind conditions and wing structures.

CONTROL: MANIPULATING VARIABLES Like aeronautical engineers, social psychologists experiment by constructing social situations that simulate important features of our daily lives. By varying just one or two factors at a time—called independent variables—the experimenter pinpoints their influence. As the wind tunnel helps the aeronautical engineer discover princi­ ples of aerodynamics, so the experiment enables the social psychologist to discover principles of social thinking, social influence, and social relations.

To illustrate the laboratory experiment, consider two experiments that typify research from upcoming chapters on prejudice and aggression. Each experiment suggests possible cause-effect explanations of correlational findings.

CORRELATIONAL AND EXPERIMENTAL STUDIES OF PREJUDICE AGAINST THE OBESE People often perceive the obese as slow, lazy, and sloppy (Roehling & others, 2007; Ryckman & others, 1989). Do such attitudes spawn dis­ crimination? In hopes of finding out, Steven Gortmaker and colleagues (1993) studied 370 obese 16- to 24-year-old women. When they restudied them 7 years later, two-thirds of the women were still obese and were less likely to be married and earning high salaries than a comparison group of approximately 5,000 other women. Even after correcting for any differences in aptitude test scores, race, and parental income, the obese women’s incomes were $7,000 a year below average.

Correcting for certain other factors makes it look as though discrimination might explain the correlation between obesity and lower status. But we cannot be sure. (Can you think of other possibilities?) Enter social psychologists Mark Snyder and Julie Haugen (1994,1995). They asked 76 University of Minnesota male students to have a get-acquainted phone conversation with 1 of 76 female students. Unknown to the women, each man was shown a photo said to picture his conversational part­ ner. Half were shown an obese woman (not the actual partner); the other half were shown a normal-weight woman. Later analysis of just the women’s side of the con­ versation revealed that they spoke less warmly and happily if they were presumed obese. Clearly, something in the men’s tone of voice and conversational content induced the supposedly obese women to speak in a way that confirmed the idea that obese women are undesirable. The men’s prejudice and discrimination were having an effect. Recalling the effect of the stepmother’s behavior, perhaps we should call this the “Cinderella effect.”

CORRELATIONAL AND EXPERIMENTAL STUDIES OF TV VIOLENCE VIEWING As a second example of how experiments clarify causation, consider the correlation between television viewing and children’s behavior. The more vio­ lent television children watch, the more aggressive they tend to be. Are children learning and reenacting what they see on the screen? As 1 hope you now recognize, this is a correlational finding. Figure 1.4 reminds us that there are two other cause-effect interpretations. (What are they?)

Social psychologists have therefore brought television viewing into the labo­ ratory, where they control the amount of violence the children see. By exposing children to violent and nonviolent programs, researchers can observe how the

Introducing Social Psychology Chapter 1 25

amount of violence affects behavior. Chris Boyatzis and colleagues (1995) showed some elementary schoolchil­ dren, but not others, an episode of the most popular—and violent—children’s television program of the 1990s, Power Rangers. Immediately after viewing the episode, the viewers committed seven times as many aggressive acts per 2-minute interval as the nonview­ ers. The observed aggressive acts we call the dependent variable. Such experiments indicate that television can be one cause of children’s aggres­ sive behavior.

So far we have seen that the logic of experimentation is simple: By creating and controlling a miniature reality, we can vary one factor and then another and discover how those factors, sepa­ rately or in combination, affect people. Now let’s go a little deeper and see how an experiment is done.

Every social psychological experiment has two essential ingredients. We have just considered one—control. We manipulate one or more independent variables while trying to hold everything else constant. The other ingredient is random assignment.

RANDOM ASSIGNMENT: THE GREAT EQUALIZER Recall that we were reluctant, on the basis of a correlation, to assume that obesity caused lower status (via discrimination) or that violence viewing caused aggressive­ ness (see Table 1.1 for more examples). A survey researcher might measure and statistically extract other possibly pertinent factors and see if the correlations sur­ vive. But one can never control for all the factors that might distinguish obese from non-obese, and viewers of violence from nonviewers. Maybe viewers of violence differ in education, culture, intelligence—or in dozens of ways the researcher has not considered.

Does viewing violence on TV or in other media lead to imitation, especially among children? Experiments sug­ gest that it does.

dependent variable The variable being measured, so called because it may depend on manipulations of the independent variable.

TABLE i* 1.1 Recognizing Correlational and Experimental Research

Can Participants Be Randomly Assigned to Condition? Independent Variable Dependent Variable

Are early-maturing children more confident? No -* Correlational Do students learn more in online or classroom YesExperimental Take class online or in Learning courses? classroom Do school grades predict vocational success? No —» Correlational Does playing violent video games increase Yes -» Experimental Play violent or Aggressiveness aggressiveness? Do people find comedy funnier when alone or (you answer) with others?

(you answer)

nonviolent game

Do higher-income people have higher self-esteem?

26 Chapter 1 Introducing Social Psychology

FIGURE:: 1.5 Random Assignment Experiments randomly assign people either to a condition that receivesthe experimental treatment or to a control condi­ tion that does not. This gives the researcher confidence that any later difference is somehow caused by the treatment.




Violent TV

Control Nonvioletit TV




random assignment The process of assigning participants to the conditions of an experiment such that all persons have the same chance of being in a given condition. (Note the distinction between random assignment in experiments and random sampling m surveys. Random assignment helps us infer cause and effect. Random sampling helps us generalize to a population.)

mundane realism Degree to which an experiment is superficially similarto everyday situations,

experimental realism Degree to which an experiment absorbs and involves its participants.

deception In research, an effect by which participants are misinformed or misled about the study’s methods and purposes.

In one fell swoop, random assignment eliminates all such extraneous factors. With random assignment, each person has an equal chance of viewing the violence or the nonviolence. Thus, the people in both groups would, in every conceivable way— family status, intelligence, education, initial aggressiveness, hair color—average about the same. Highly intelligent people, for example, are equally likely to appear in both groups. Because random assignment creates equivalent groups, any later aggression difference between the two groups will almost surely have something to do with the only way they differ—whether or not they viewed violence (Figure 1.5).

THE ETHICS OF EXPERIMENTATION Our television example illustrates why some conceivable experiments raise ethi­ cal issues. Social psychologists would not, over long periods, expose one group of children to brutal violence. Rather, they briefly alter people’s social experience and note the effects. Sometimes the experimental treatment is a harmless, perhaps even enjoyable, experience to which people give their knowing consent. Occasionally, however, researchers find themselves operating in a gray area between the harm­ less and the risky.

Social psychologists often venture into that ethical gray area when they design experiments that engage intense thoughts and emotions. Experiments need not have what Elliot Aronson, Marilynn Brewer, and Merrill Carlsmith (1985) called mundane realism. That is, laboratory behavior need not be like everyday behav­ ior, which is typically mundane, or unimportant. But the experiment should have experimental realism—it should engage the participants. Experimenters do not want their people consciously play-acting or ho-humming it; they want to engage real psychological processes. An example of such engagement would be delivering electric shocks as part of an experiment on aggression. Forcing people to choose whether to give intense or mild electric shock to someone else can be a realistic measure of aggression. It functionally simulates real aggression.

Achieving experimental realism sometimes requires deceiving people with a plausible cover story. If the person in the next room actually is not receiving the shocks, the experimenter does not want the participants to know that. That would destroy the experimental realism. Thus, approximately one-third of social psycho­ logical studies (though a decreasing number) have used deception (Korn & Nicks, 1993; Vitelli, 1988).

Experimenters also seek to hide their predictions lest the participants, in their eagerness to be “good subjects,” merely do what is expected or, in an ornery mood, do the opposite. Small wonder, says Ukrainian professor Anatoly Koladny, that only 15 percent of Ukrainian survey respondents declared themselves “religious” while under Soviet communism in 1990 when religion was oppressed by the

27Introducing Social Psychology

government—and that 70 percent declared themselves “reUgious” in post-communist 1997 (Nielsen, 1998). In subtle ways, too, the experimenter’s words, tone of voice, and gestures may call forth desired responses. Even search dogs trained to detect explo­ sives and drugs are more likely to bark false alerts in places where their handlers have been misled into thinking such illegal items are located (Lit & others, 2011). To minimize such demand characteristics—cues that seem to “demand” certain behavior—experimenters typically standardize their instructions or even use a computer to present them.

Researchers often walk a tightrope in designing experiments that will be involv­ ing yet ethical. To believe that you are hurting someone, or to be subjected to strong social pressure, may be temporarily uncomfortable. Such experiments raise the age- old question of whether ends justify means. The social psychologists’ deceptions are usually brief and mild compared with many misrepresentations in real life and in some of television’s reality shows. (One network reality TV series deceived women into competing for the hand of a handsome supposed millionaire, who turned out to be an ordinary laborer.)

University ethics committees review social psychological research to ensure that it will treat people humanely and that the scientific merit justifies any temporary deception or distress. Ethical principles developed by the American Psychological Association (2010), the Canadian Psychological Association (2000), and the British Psychological Society (2009) mandate investigators to do the following:

• Tell potential participants enough about the experiment to enable their informed consent.

• Be truthful. Use deception only if essential and justified by a significant purpose and not “about aspects that would affect their willingness to participate.”

• Protect participants (and bystanders, if any) from harm and significant discomfort.

• Treat information about the individual participants confidentially. • Debrief participants. Fully explain the experiment afterward, including any

deception. The only exception to this rule is when the feedback would be dis­ tressing, such as by making participants realize they have been stupid or cruel.

The experimenter should be sufficiently informative and considerate that people leave feeling at least as good about themselves as when they came in. Better yet, the participants should be compensated by having learned something (Sharpe & Faye, 2009). When treated respectfully, few participants mind being deceived (Epley & Huff, 1998; Kimmel, 1998). Indeed, say social psychology’s advocates, professors provoke far greater anxiety and distress by giving and returning course exams than researchers provoke in their experiments.

Generalizing from Laboratory to Life As the research on children, television, and violence illustrates, social psychology mixes everyday experience and laboratory analysis. Throughout this book, we do the same by drawing our data mostly from the laboratory and our illustrations mostly from life. Social psychology displays a healthy interplay between labora­ tory research and everyday life. Hunches gained from everyday experience often inspire laboratory research, which deepens our understanding of our experience.

This interplay appears in the children’s television experiment. What people saw in everyday life suggested correlational research, which led to experimental research. Network and government policymakers, those with the power to make changes, are now aware of the results. The consistency of findings on television’s effects—in the lab and in the field—is true of research in many other areas, includ­ ing studies of helping, leadership style, depression, and self-efficacy. The effects

Chapter 1

demand characteristics Cues in an experiment that tell the participant what behavior is expected.

informed consent An ethical principle requiring that research participants be told enough to enable them to choose whether they wish to participate.

debriefing In social psychology, the postexperimental explanation of a study to its participants. Debriefing usually discloses any deception and often queries participants regarding their understandings and feelings.

28 Chapter 1 Introducing Social Psychology

one finds in the lab have been mirrored by effects in the field. “The psychology lab­ oratory has generally produced psychological truths rather than trivialities,” note Craig Anderson and colleagues (1999).

We need to be cautious, however, in generalizing from laboratory to life. Although the laboratory uncovers basic dynamics of human existence, it is still a simplified, controlled reality. It tells us what effect to expect of variable X, all other things being equal—which in real life they never are. Moreover, as you will see, the participants in many experiments are college students. Although that may help you identify with them, college students are hardly a random sample of all humanity (Henry, 2008a, 2008b). And most participants are from WEIRD (Western, Educated, Industrialized, Rich, and Democratic) cultures that represent but 12 percent of humanity (Henrich & others, 2010). Would we get similar results with people of different ages, educational levels, and cultures? That is always an open question.

Nevertheless, we can distinguish between the content of people’s thinking and acting (for example, their attitudes) and the process by which they think and act (for example, how attitudes affect actions and vice versa). The content varies more from culture to culture than does the process. People from various cultures may hold dif­ ferent opinions yet form them in similar ways. Consider the following;

• College students in Puerto Rico have reported greater loneliness than do col­ legians on the U.S. mainland. Yet in the two cultures the ingredients of lone­ liness have been much the same—shyness, uncertain purpose in life, and low self-esteem (Jones & others, 1985).

* Ethnic groups differ in school achievement and delinquency, but the dif­ ferences are “no more than skin deep,” report David Rowe and colleagues (1994). To the extent that family structure, peer influences, and parental edu­ cation predict achievement or delinquency for one ethnic group, they do so for other groups.

Although our behaviors may differ, we are influenced by the same social forces. Beneath our surface diversity, we are more alike than different.

SUMMING UP: Research Methods: How Do We Do Social Psychology?

Social psychologists organize their ideas and find­ ings into theories. A good theory will distill an array of facts into a much shorter list of predictive prin­ ciples. We can use those predictions to confirm or modify the theory, to generate new research, and to suggest practical application. Most social-psychological research is either cor­ relational or experimental. Correlational studies, sometimes conducted with systematic survey meth­ ods, discern the relationship between variables, such as between amount of education and amount of income. Knowing two things are naturally related is valuable information, but it is not a reliable indi­ cator of what is causing what—or whether a third variable is involved. When possible, social psychologists prefer to con­ duct experiments that explore cause and effect. By constructing a miniature reality that is under their

control, experimenters can vary one thing and then another and discover how those things, separately or in combination, affect behavior. We randomly assign participants to an experimental condition, which receives the experimental treatment, or to a control condition, which does not. We can then attribute any resulting difference between the two conditions to the independent variable (Figure 1.6). In creating experiments, social psychologists some­ times stage situations that engage people’s emo­ tions. In doing so, they are obliged to follow professional ethical guidelines, such as obtaining people’s informed consent, protecting them from harm, and fully disclosing afterward any tempo­ rary deceptions. Laboratory experiments enable social psychologists to test ideas gleaned from life experience and then to apply the principles and findings to the real world.

Introducing Social Psychology Chapter 1 29

Research methods

Correlational Experimental

Advantage Often uses real- world settings

Causation often ambiguous

Advantage Can explore cause and effect by controlling variables and by random assignment

Disadvantage Some important variables cannot be studied with experiments

FIGURE:: 1.6 Two Methods of Doing Research: Correlational and Experimental

POSTSCRIPT: Why I Wrote This Book I write this text to offer social psychology’s powerful, hard-wrought principles. They have, I believe, the power to expand your mind and enrich your life. If you finish this book with sharpened critical thinking skills and with a deeper under­ standing of how we view and affect one another—and why we sometimes like, love, and help one another and sometimes dislike, hate, and harm one another— then I will be a satisfied author and you, I trust, will be a rewarded reader.

I write knowing that many readers are in the process of defining their life goals, identities, values, and attitudes. The novelist Chaim Potok recalls being urged by his mother to forgo writing: “Be a brain surgeon. You’ll keep a lot of people from dying; you’ll make a lot more money.” Potok’s response: “Mama, I don’t want to keep people from dying; I want to show them how to live” (quoted by Peterson, 1992, p. 47).

Many of us who teach and write psychology are driven not only by a love for giving psychology away but also by wanting to help students live better lives— wiser, more fulfilling, more compassionate lives. In this we are like teachers and writers in other fields. “Why do we write?” asks theologian Robert McAfee Brown. “I submit that beyond all rewards .. .we write because we want to change things. We write because we have this [conviction that we] can make a difference. The ‘differ­ ence’ may be a new perception of beauty, a new insight into self-understanding, a new experience of joy, or a decision to join the revolution” (quoted by Marty, 1988). Indeed, I write hoping to do my part to restrain intuition with critical thinking, refine judgmentalism with compassion, and replace illusion with understanding.

/ conclude each chapter with a brief reflection on social psychology’s human significance.

This book unfolds around its definition of social psychology: the scientific study of how we th/nk about (Part One), influ­ ence (Part Two), and relate to (Part Three) one another. Part Four offers additional, focused examples of how the research and the theories of social psychology are applied to real life.

Part One examines the scientific study of how we think about one another (also called social cognition). Each chapter con­ fronts some overriding questions; How reasonable are our social attitudes, expla­ nations, and beliefs? Are our impressions of ourselves and others generally accu­ rate? How does our social thinking form? How is it prone to bias and error, and how might we bring it closer to reality?

Chapter 2 explores the interplay between our sense of self and our social worlds. How do our social surroundings shape our self-identities? How does self- interest color our social judgments and motivate our social behavior?

Chapter 3 looks at the amazing and sometimes rather amusing ways we form beliefs about our social worlds. It also alerts us to some pitfalls of social think­ ing and suggests how to avoid them and think smarter.

Chapter 4 explores the links between our thinking and our actions, between our attitudes and our behaviors: Do our attitudes determine our behaviors, or vice versa? Or does it work both ways?


2 The Self ‘There are three things extremely hard, Steel, a Diamond, and to know one’s self.” ……………………………………………………….. … —,

At the center of our worlds, more pivotal for us than anything else, is ourselves. As we navigate our daily lives, our sense of self con­ tinually engages the world.

Consider this example: One morning, you wake up to find your

hair sticking up at weird angles on your head. It’s too late to jump in

the shower and you can’t find a hat, so you smooth down the random

spikes of your hair and dash out the door to class. All morning, you are

acutely self-conscious about your very bad hair day. To your surprise,

your friends in class don’t say anything. Are they secretly laughing to

themselves about how ridiculous you look, or are they too preoccu­

pied with themselves to notice your spiky hair?

* This 11th edition chapter is co-authored by Jean Twenge, professor of psychology at San Diego State University. Professor Twenge’s research on social rejection and on generational changes in personality and the self has been published in many articles and books, including Ceneralion Me: Why Today’s Young Americans Are More Confident, Assertive, Entitled—and More Miserable Than Ever Before (2006) and The Narcissism Epidemic: Living in the Age of Entitlement (with W. Keith Campbell, 2009).

34 Part One Social Thinking

The spotlight effect: Overes­ timating others’ noticing our behavior and appearance. FOR BEnER OR FOR V/ORSE © 2005 Lynn Johnston Prodjctions. Dist. by Universal Press Syndicate. Reprinted with permission. All rights reserved.

Spotlight effect The belief that others are paying more attention to our appearance and behavior than they really are.

illusion of transparency The illusion that our concealed emotions leak out and can be easily read by others.


Describe the spotlight effect and its relation to the illusion of transparency.

Why do we often feel that others are paying more attention to us than they really are? The spotlight effect means seeing ourselves at center stage, thus intuitively overestimating the extent to which others attention is aimed at us.

Timothy Lawson (2010) explored the spotlight effect by having college students change into a sweatshirt with “American Eagle” on the front before meeting a group of peers. Nearly 40 percent were sure the other students would remember what the shirt said, but only 10 percent actually did. Most observers did not even notice that the students changed sweatshirts after leaving the room for a few min­ utes. In another experiment, even noticeably embarrassing clothes, such as a T-shirt with singer Barry Manilow on it, provoked only 23 percent of observers to notice- many fewer than the 50 percent estimated by the unfortunate students sporting the 1970s soft rock warbler on their chests (Gilovich & others, 2000).

What’s true of our dorky clothes and bad hair is also true of our emotions: our anxiety, irritation, disgust, deceit, or attraction (Gilovich & others, 1998). Fewer peo­ ple notice than we presume. Keenly aware of our own emotions, we often suffer an illusion of transparency. If we’re happy and we know it, then our face will surely

research CLOSE-UP On Being Nervous About Looking Nervous

Have you ever felt self-conscious when approaching someone you felt attracted to, concerned that your ner­ vousness was obvious? Or have you felt yourself trem­ bling while speaking before an audience and presumed that everyone was noticing?

Kenneth Savitsky and Thomas Gilovich (2003) knew j from their own and others’ studies that people over- j estimate the extent to which their internal states “leak | out.” People asked to tell lies presume that others will | detect their deceit, which feels so obvious. People |

The Self in a Social World Chapter 2 35

asked to sample horrid-tasting drinks presume that others notice their disgust, which they can barely suppress.

Many people who give a presentation report not just feeling anxious, but anxious that others will notice their anxiety. And if they feel their knees shaking and hands trembling, their worry that others are noticing may com­ pound and perpetuate their anxiety. This is similar to fretting about not falling asleep, which further impedes falling asleep, or feeling anxious about stuttering, which worsens the stuttering. (As a former stutterer and speech therapy patient, I know this is true.)

Savitsky and Gilovich wondered whether an “illu­ sion of transparency” might surface among inexperi­ enced public speakers—and whether it might disrupt their performance. To find out, they invited 40 Cornell University students to their laboratory in pairs. One person stood at the podium and spoke for 3 minutes (on a topic such as “The Best and Worst Things About Life Today”) as the other sat and listened. Then the two switched positions and the other person gave a differ­ ent 3-minute impromptu talk. Afterward, each rated how nervous they thought they appeared while speak­ ing (from 0, not at all, to 10, very) and how nervous the other person seemed.

The results? People rated themselves as appearing relatively nervous (6.65, on average). But to their part­ ner they appeared not so nervous (5.25), a difference great enough to be statistically significant (meaning that a difference this great, for this sample of people, is very unlikely to have been due to chance varia­ tion). Twenty-seven of the 40 participants (68 percent) believed that they appeared more nervous than did their partner.

To check on the reliability of their finding, Savitsky and Gilovich replicated (repeated) and extended the experiment by having people speak before an audience of people who weren’t going to be giving

speeches themselves, to rule out the possibility that this might explain the previous results. Once again, speakers overestimated the transparency of their nervousness.

Savitsky and Gilovich next wondered whether inform­ ing speakers that their nervousness isn’t so obvious might help them relax and perform better. They invited 77 more Cornell students to come to the lab and, after 5 minutes’ preparation, give a 3-minute videotaped speech on race relations at their university. Those in one group—the control condition—were given no further instructions. Those in the reassured condition were told that it was naturaj to feel anxious but that “You shouldn’t worry much about what other people think…. With this in mind you should just relax and try to do your best. Know that if you become nervous, you probably shouldn’t worry about it.” To those in the informed condition he explained the illusion of transparency. After telling them it was natural to feel anxious, the experimenter added that “Research has found that audiences can’t pick up on your anxiety as well as you might expect,.., Those speaking feel that their nervousness is transparent, but in reality their feelings are not so apparent…. With this in mind, you should just relax and try to do your best. Know that if you become nervous, you’ll probably be the only one to know.”

After the speeches, the speakers rated their speech quality and their perceived nervousness (this time using a 7-point scale) and were also rated by the observers. As Table 2,1 shows, those informed about the illuslon-of- transparency phenomenon felt better about their speech and their appearance than did those in the control and reassurance conditions. What’s more, the observers con­ firmed the speakers’ self-assessments.

So, the next time you feel nervous about looking nervous, pause to remember the lesson of these experi­ ments; Other people are noticing less than you might suppose.

TABLE •• 2.1 Average Ratings of Speeches by Speakers and Observers on a 1 to 7 Scale

j . Speakers’ self-ratings Speech quality

■ ;.,;;,Relaxed



2.83 3.50*


Observers’ ratings

Composed appearance

*Each of these results differs by a statistically significant margin from those of the control and reassured condition.

36 Part One Social Thinking










show it. And others, we presume, will notice. Actually, we can be more opaque than we realize. (See “Research Close-Up: On Being Nervous About Looking Nervous” on pages 34-35.)

We also overestimate the visibility of our social blunders and public mental slips. When we trigger the library alarm or accidentally insult someone, we may be mor­ tified (“Everyone thinks Fm a jerk”). But research shows that what we agonize over, others may hardly notice and soon forget (Savitsky & others, 2001).

The spotlight effect and the related illusion of transparency are but two of many examples of the interplay between our sense of self and our social worlds. Here are more examples:

• Social surroundings affect our self-awareness. When we are the only member of our race, gender, or nationality in a group, we notice how we differ and how others are reacting to our difference. A White American friend once told me how self-consciously White he felt while living in a rural village in Nepal; an hour later, an African-American friend told me how self-consciously Ameri­ can she felt while in Africa.

• Self-interest colors our social judgment. When problems arise in a close relation­ ship such as marriage, we usually attribute more responsibility to our part­ ners than to ourselves. When things go well at home or work or play, we see ourselves as more responsible.

• Self-concern motivates our social behavior. In hopes of making a positive impression, we agonize about our appearance. Like savvy politicians, we also monitor others’ behavior and expectations and adjust our behavior accordingly.

• Social relationships help define our sense of self In our varied relationships, we have varying selves, note Susan Andersen and Serena Chen (2002). We may be one self with Mom, another with friends, another with teachers. How we think of ourselves is linked to the person we’re with at the moment. And when relationships change, our self-concepts can change as well. College students who recently broke up with a romantic partner shifted their self­ perceptions and felt less certain about who they were—one reason breakups can be so emotionally distressing (Slotter & others, 2010).

As these examples suggest, the traffic between ourselves and others runs both ways. Our ideas and feelings about ourselves affect how we respond to others. And others help shape our sense of self.

No topic in psychology today is more heavily researched than the self. In 2011, the word “self” appeared in 21,693 book and article summaries in PsycINFO (the online archive of psychological research)—more than 20 times the number that appeared in 1970. Our sense of self organizes our thoughts, feelings, and actions. Our sense of self enables us to remember our past, assess our present, and project our future—and thus to behave adaptively.

In later chapters, we will see that much of our behavior is not consciously con­ trolled but, rather, automatic and unself-conscious. However, the self does enable long-term planning, goal-setting, and restraint. It imagines alternatives, compares itself with others, and manages its reputation and relationships. Moreover, as Mark Leary (2004a) has noted, the self can sometimes be an impediment to a satisfying life. Its egocentric preoccupations are what religious meditation practices seek to prune, by quieting the self, reducing its attachments to material pleasures, and redi­ recting it. “Mysticism,” adds psychologist Jonathan Haidt (2006), “everywhere and always, is about losing the self, transcending the self, and merging with something larger than the self.”

In the remainder of this chapter, we examine our self-concept (how we come to know ourselves) and the self in action (how our sense of self drives our attitudes and actions).


The Self in a Social World Chapter 2 37

SUMMING UP: Spotlights and Illusions: What Do They Teach Us About Ourselves?

« Concerned with the impression we make on oth- ® We also tend to believe that our emotions are more ers, we tend to believe that others are paying more obvious than they are (the illusion of transparency). attention to us than they are (the spotlight effect).

SELF-CONCEPT: WHO AM I? Understand how, and how accurately, we know ourselves and what determines our self-concept.

You have many ways to complete the sentence “I am____ .” (What five answers might you give?) Your answers provide a glimpse of your self-concept.

At the Center of Our Worlds: Our Serv;e of Self The most important aspect of yourself is your self. To discover where this sense of self arises, neuroscientists are exploring the brain activity that underlies our con­ stant sense of being oneself. Some studies suggest an important role for the right hemisphere. Put yours to sleep (with an anesthetic to your right carotid artery) and you likely will have trouble recognizing your own face. One patient with right hemisphere damage failed to recognize that he owned and was controlling his left hand (Decety & Sommerville, 2003). The “medial prefrontal cortex,” a neuron path located in the cleft between your brain hemispheres just behind your eyes, seem­ ingly helps stitch together your sense of self. It becomes more active when you think about yourself (Farb & others, 2007; Zimmer, 2005).

The elements of your self-concept, the specific beliefs by which you define your­ self, are your self-schemas (Markus & Wurf, 1987). Schemas are mental templates by which we organize our worlds. Our self-schemas—our perceiving ourselves as athletic, overweight, smart, or whatever—powerfully affect how we perceive, remember, and evaluate other people and ourselves. If athletics is central to your self- concept (if being an athlete is one of your self-schemas), then you will tend to notice others’ bodies and skills. You will quickly recall sports-related experiences. And you will welcome information that is consis­ tent with your self-schema (Kihlstrom & Cantor, 1984). If your friend’s birth­ day is close to yours, you’ll be more likely to remember it (Kesebir & Oishi, 2010). The self-schemas that make up our self-concepts help us organize and retrieve our experiences.

POSSIBLE SELVES Our self-concepts include not only our self-schemas about who we currently are but also who we might become— our possible selves. Hazel Markus and colleagues (Inglehart Sz others, 1989; Markus & Nurius, 1986) note that our possible selves include our visions of the self we dream of becoming—the rich self, the thin self, the passionately

self-concept What we know and believe about ourselves.

self-schema Beliefs about self that organize and guide the processing of self-relevant information.

possible selves Images of what we dream of or dread becoming in the future.

Oprah Winfrey’s imagined possible selves, including the dreaded overweight self, the rich self, and the helpful self, motivated her to work to achieve the life she wanted.

38 Part One Social Thinking

social comparison Evaluating one’s abilities and opinions by comparing oneself with others.

FIGURE:: 2.1 The Self

loved and loving self. They also include the self we fear becoming—the imderem- ployed self, the unloved self, the academically failed self. Such possible selves moti­ vate us with a vision of the life we long for—or to avoid the one we dread.

Development of the Social Seif The self has become a major social psychological focus because it helps organize our thinking and guide our social behavior (Figure 2.1). But what determines our self- concepts? Studies of twins point to genetic influences on personality and self-concept, but social experience also plays a part. Among these influences are the following:

• The roles we play • The social identities we form • The comparisons we make with others • How other people judge us • The surrounding culture

THE ROLES WE PLAY As we enact a new role—college student, parent, salesperson—we initially feel self- conscious. Gradually, however, what begins as playacting in the theater of life is absorbed into our sense of self. For example, while playing our roles we may give lip service to something we haven’t really thought much about. After defending our group, we then justify our words by believing more strongly in it. Role playing becomes reality (see Chapter 4).

SOCIAL COMPARISONS How do we decide if we are rich, smart, or short? One way is through social comparisons (Festinger, 1954). Others around us help to define the standard by which we define ourselves as rich or poor, smart or dumb, tall or short: We compare ourselves with them and consider how we differ. Social comparison explains why students tend to have a higher academic self-concept if they attend a high school with mostly aver­ age students (Marsh & others, 2000), and how that self-concept can be threatened after graduation when a student who excelled in an average high school goes on to an aca­ demically selective university. The “big fish” is no longer in a small pond.

Much of life revolves around social comparisons. We feel handsome when others seem homely, smart when others seem dull, caring when others seem callous. When we witness a peer’s performance, we cannot resist implicitly comparing ourselves (Gilbert & others, 1995). We may, therefore, privately take some pleasure in a peer’s

Social self

My roles as a student, ‘J family member, and ’

friend; my group | identity y

‘ ‘V’ ‘■*

39The Self in a Social World

failure, especially when it happens to someone we envy and when we don’t feel vulnerable to such misfortune ourselves (Lockwood, 2002; Smith & others, 1996).

Social comparisons can also diminish our satisfaction. When we experience an increase in affluence, status, or achievement, we “compare upward”—we raise the standards by which we evaluate our attainments. When climbing the ladder of success, we tend to look up, not down; we compare ourselves with others doing even better (Gmder, 1977; Suls & Tesch, 1978; Wheeler & others, 1982). When facing competition, we often protect our shaky self-concept by perceiving the competitor as advantaged. For example, college swimmers believed that their competitors had better coaching and more practice time (Shepperd & Taylor, 1999).

OTHER PEOPLE’S JUDGMENTS When people think well of us, it helps us think well of ourselves. Children whom others label as gifted, hardworking, or helpful tend to incorporate such ideas into their self-concepts and behavior (see Chapter 3). If minority students feel threat­ ened by negative stereotypes of their academic ability, or if women feel threatened by low expectations for their math and science performance, they may “disiden- tify” with those realms. Rather than fight such prejudgments, they may identify their interests elsewhere (Steele, 2010; see Chapter 9).

The looking-glass self was how sociologist Charles H. Cooley (1902) described our use of how we think others perceive us as a mirror for perceiving ourselves. Fel­ low sociologist George Herbert Mead (1934) refined this concept, noting that what matters for our self-concepts is not how others actually see us but the way we imag­ ine they see us. People generally feel freer to praise than to criticize; they voice their compliments and restrain their gibes. We may, therefore, overestimate others’ appraisal, inflating our self-images (Shrauger & Schoeneman, 1979).

Self-inflation, as we will see, is found most strikingly in Western countries. Shi- nobu Kitayama (1996) reports that Japanese visitors to North America are routinely struck by the many words of praise that friends offer one another. When he and his colleagues asked people how many days ago they last complimented someone, the most common American response was 1 day. In Japan, where people are socialized less to feel pride in personal achievement and more to feel shame in failing others, the most common response was 4 days.

Our prehistoric ancestors’ fate depended on what others thought of them. Their survival was enhanced when protected by their group—in a time before grocery

Chapter 2

Private Pleasure in a Peer’s Pratfall In 2011, when powerful media magnates Rupert Murdoch and his son, James Murdoch, were embarrassed by illegal practices at one of their newspapers, some people felt schadenfreude (a German word for the pleasure felt over someone else’s misfortune).



-KING CHARLES 1,1600-1649

40 Part One

individualism The concept of giving priority to one’s own goals over group goals and defining one’s identity in terms of personal attributes rather than group identifications.

independent self Construing one’s identity as an autonomous self.

collectivism Giving priority to the goals of one’s group (often one’s extended family or work group) and defining one’s identity accordingly.

interdependent self Construing one’s identity in relation to others.

Social Thinking

stores, it was difficult for one person alone to hunt and gather enough food or to pro­ tect him- or herself from predators. When perceiving their group’s disapproval, there was biological wisdom to their feeling shame and low self-esteem. As their heirs, hav­ ing a similar deep-seated need to belong, we feel the pain of low self-esteem when we face social exclusion, notes Mark Leary {1998, 2004b). Self-esteem, he argues, is a psychological gauge by which we monitor and react to how others appraise us.

Self and Culture How did you complete the “I am____ ” statement on page 37? Did you give infor­ mation about your personal traits, such as “I am honest,” “I am tall,” or “I am outgo­ ing”? Or did you also describe your social identity, such as “I am a Pisces,” “I am a MacDonald,” or “I am a Muslim”?

For some people, especially those in industrialized Western cultures, individualism prevails. Identity is self-contained. Adolescence is a time of separating from parents, becoming self-reliant, and defining one’s personal, independent self. One’s identity— as a unique individual with particular abilities, traits, values, and dreams—remains fairly constant.

The psychology of Western cultures assumes that your life will be enriched by believing in your power of personal control. Western literature, from The Iliad to The Adventures of Huckleberry Finn, celebrates the self-reliant individual. Movie plots fea­ ture rugged heroes who buck the establishment. Songs proclaim “I Gotta Be Me,” declare that “The Greatest Love of All” is loving oneself (Schoeneman, 1994), and state without irony that “1 Believe the World Should Revolve Around Me.” Individualism flourishes when people experience affluence, mobility, urbanism, and mass media (Freeman, 1997; Marshall, 1997; Triandis, 1994).

Most cultures native to Asia, Africa, and Central and South America place a greater value on collectivism, by respecting one’s groups and identifying oneself accordingly. They nurture what Shinobu Kitayama and Hazel Markus (1995) call the interdependent self. In these cultures, people are more self-critical and have less need for positive self-regard (Heine & others, 1999). Malaysians, Indians, Koreans, Japanese, and traditional Kenyans such as the Maasai, for example, are much more likely than Australians, Americans, and the British to complete the “I am” statement with their group identities (Kanagawa & others, 2001; Ma & Schoeneman, 1997). When speaking, people using the languages of collectivist countries say “I” less often (Kashima & Kashima, 1998, 2003). A person might say “Went to the movie” rather than “I went to the movie.” Compared with U.S. church websites, Korean church websites place more emphasis on social connections and participation and less on personal spiritual growth and self-betterment (Sasaki & Kim, 2011).

Pigeonholing cultures as solely individualist or collectivist oversimplifies, because within any culture individualism varies from person to person (Oyserman & others, 2002a, 2002b). There are individualist Chinese and collectivist Americans, and most of us sometimes behave communally, sometimes individualistically (Bandura, 2004). Individualism-collectivism also varies across a country’s regions and political views. In the United States, Native Hawaiians and people living in the deep South exhibit greater collectivism than do those in Mountain West states such as Oregon and Mon­ tana (Plaut & others, 2002; Vandello & Cohen, 1999). Conservatives tend to be eco­ nomic individualists (“don’t tax or regulate me”) and moral collectivists (“legislate against immorality”). Liberals tend to be economic collectivists (supporting national health care) and moral individualists (“keep your laws off my body”). Despite indi­ vidual and subcultural variations, researchers continue to regard individualism and collectivism as genuine cultural variables (Schimmack & others, 2005).

GROWING INDIVIDUALISM WITHIN CULTURES Cultures can also change over time, and many seem to be growing more individu­ alistic. New economic opportunities have challenged traditional collectivistic ways

The Self in a Social World Chapter 2 41

in India. Chinese citizens younger than 25 are more likely than those older than 25 to agree with individualistic statements such as “make a name for yourself” and “live a life that suits your tastes” (Arora, 2005). Chinese citizens who are younger, more urban, more affluent, and only children—all mod­ em attributes—are also more likely to endorse self-centered statements (Cai & others, 2011). In the United States, younger generations report signifi­ cantly more positive self-feelings than young people did in the 1960s and 1970s (Gentile & others, 2010; Twenge & Campbell, 2008; Twenge & others, 2011; but for an opposing view, see Trzesniewski & Donnellan, 2010). One study found that popular song lyr­ ics became more likely to use “I” and “me” and less likely to use “we” and “us” between 1980 and 2007 (DeWall & others, 2011), with the norm shifting from the sappy love song of the 1980s (“Endless Love,” 1981) to the self-celebration of the 2000s (Justin Timberlake singlehandedly bringing “Sexy Back,” 2006).

Even your name might show the shift toward individualism: American parents are now less likely to give their children common names and more likely to help them stand out with an unusual name. While nearly 20 percent of boys born in 1990 received one of the 10 most common names, only 8 percent received such a common name by 2010, with the numbers similar for girls (Twenge & others, 2010). Today, you don’t have to be the child of a celebrity to get a name as unique as Shiloh, Suri, Knox, or Apple.

Americans and Australians, most of whom are descended from those who struck out on their own to emigrate, are more likely than Europeans to give their children imcommon names. Parents in the western United States and Canada, descended from independent pioneers, are also more likely than those in the more established East to give their children uncommon names (Vamum & Kitayama, 2011). The more individualistic the time or the place, the more children receive unique names.

These changes demonstrate something that goes deeper than a name: the interac­ tion between individuals and society. Did the culture focus on uniqueness first and cause the parents’ name choices, or did individual parents decide they wanted their children to be unique, thus creating the culture? A similar chicken-and-egg ques­ tion applies to the song lyrics: Did a more self-focused population listen to more self-focused songs, or did listening to more self-focused songs make people more self- focused? The answer, though not yet fuUy understood, is probably both (Markus & Kitayama, 2010).

® Jack Ziegler/The New Yorker Collection/www.cattoonbankxom

CULTURE AND COGNITION In his book The Geography of Thought (2003), social psychologist Richard Nisbett con­ tends that collectivism also results in different ways of thinking. Consider: Which two—of a panda, a monkey, and a banana—go together? Perhaps a monkey and a panda because they both fit the category “animal”? Asians more often than Americans see relationships: Monkey eats banana. When shown an animated underwater scene (Figure 2.2), Japanese spontaneously recalled 60 percent more background features than did Americans, and they spoke of more relationships (the frog beside the plant). Americans look more at the focal object, such as a single big fish, and less at the sur- roimdings (Chua & others, 2005; Nisbett, 2003), a result duplicated in studies exam­ ining activation in different areas of the brain (Goh & others, 2007; Lewis & others.

Social Thinking42 Part One

FIGURE :: 2.2 Asian and Western Thinking When shown an underwater scene, Americans focus on the biggest fish. Asians are more likely to reference the back­ ground, such as the plants, bub­ bles, and rocks (Nisbett, 2003).

FIGURE :: 2.3 Which Pen Would You Choose? When Heejung Kim and Hazel Markus <1999) invited people to choose one of these pens, 77 percent of Americans but only 31 percent of Asians chose the uncommon color (regardless of whether it was orange, as here, or green). This result illustrates differing cultural preferences for uniqueness and conformity, note Kim and Markus.

2008). When shown drawings of groups of children, Japa­ nese students took the facial expressions of all of the chil­ dren into account when rat­ ing the happiness or anger of an individual child, whereas Americans focused only on the child they were asked to rate (Masuda & others, 2008). Nisbett and Takahiko Masuda (2003) conclude from such studies that East Asians think more holistically—perceiving and thinking about objects and

people in relationship to one another and to their environment. If you grew up in a Western culture, you were probably told to “express yourself”—

through writing, the choices you make, the products you buy, and perhaps through your tattoos or piercings. When asked about the purpose of language, American students were more likely to explain that it allows self-expression, whereas Korean students focused on how language allows communication with others. American stu­ dents were also more likely to see their choices as expressions of themselves and to evaluate their personal choices more favorably (Kim & Sherman, 2007). The individu­ alized latte—”decaf, single shot, skinny, extra hot”—that seems just right at a North American coffee shop would seem strange in Seoul, note Kim and Hazel Markus (1999). In Korea, people place less value on expressing their uiriqueness and more on tradition and shared practices (Choi & Choi, 2002; Figure 2.3). Korean advertisements tend to feature people together, whereas American advertisements highlight personal choice or freedom (Markus, 2001; Morling «Sc Lamoreaux, 2008).

With an interdependent self, one has a greater sense of belonging. If they were uprooted and cut off from family, colleagues, and loyal friends, interdependent peo­ ple would lose the social connections that define who they are. When Chinese par­ ticipants were asked to think about their mothers, a brain region associated with the self became activated—an area that lit up for Western participants only when they thought about themselves (Zhu & others, 2007). Interdependent selves have not one self but many selves: self-with-parents, self-at-work, self-with-friends (Cross & others, 1992). As Figure 2.4 and Table 2.2 suggest, the interdependent self is embedded in social memberships. Conversation is less direct and more polite (Holtgraves, 1997), and people focus more on gaining social approval (Lalwani & others, 2006). In one study, 60 percent of American students said they had seriously dated someone even though their friends disliked him or her, compared to only 27 percent of Chinese stu­ dents. Half of the Chinese students said they would stop dating someone if their par­ ents disapproved, compared with less than one-third of American students (Zhang & Kline, 2009). In an interdependent culture, the goal of social life is to harmonize

with and support one’s communities, not—as it is in more individualistic societies—to enhance one’s individual self and make choices independently.

Even within one culture, personal history can influence self-views. People who have moved from place to place are happier when others understand their constant, personal selves; people who have always lived in the same town are more pleased when someone recognizes

43The Self in a Social World

Interdependent view of self

their collective identity (Oishi & others, 2007a, 2007b). Our self-concepts seem to adjust to our situation: If you interact with the same people all your life, they are more important to your identity than if you are uprooted every few years and must make new friends. Your self becomes your constant companion (“Wherever you go, there you are”).

CULTURE AND SELF-ESTEEM Self-esteem in collectivist cultures correlates closely with “what others think of me and my group.” Self-concept in these cultures is malleable (context-specific) rather than stable (enduring across situations). In one study, four in five Canadian stu­ dents but only one in three Chinese and Japanese students agreed that “the beliefs that you hold about who you are (your inner self) remain the same across different activity domains” (Tafarodi &: others, 2004).

For those in individualistic cultures, self-esteem is more personal and less rela­ tional. Threaten our personal identity and we’ll feel angrier and gloomier than when someone threatens our collective identity (Gaertner & others, 1999). Unlike Japanese, who persist more on tasks when they are failing (wanting not to fall short of others’ expectations), people in individualistic countries persist more when suc­ ceeding, because success elevates self-esteem (Heine & others, 2001). Western indi­ vidualists like to make comparisons with others that boost their self-esteem. Asian collectivists make comparisons (often upward, with those doing better) in ways that facilitate self-improvement (White & Lehman, 2005).

So when, do you suppose, are university students in collectivist Japan and indi­ vidualist United States most likely to report positive emotions such as happiness

TABLE :: 2.2 Seif-Concept: Independent or interdependetv.

Independent Interdependent … I ^’r..

Identity is

What matters

Personal, defined by individual traits and goals Me—personal achievement and fulfillment; my rights and liberties

Social, defined by connections with others We—group goals and solidar­ ity; our social responsibilities and relationships

Egotiatn “No one is an island’

Disapproves of Corxformity

Illustrative motto “To thine own self be true p’^-^ultures that Indi^udi^ C6&ci@visde^;^ih

FIGURE:: 2.4 Self-Construal as Independent or interdependent The independent self acknowl­ edges relationships with others. But the interdependent self is more deeply embedded in others (Markus & Kitayama, 1991).

Chapter 2






44 Part One Social Thinking

and elation? For Japanese students, happiness comes with positive social engagement—with feeling close, friendly, and respectful. For American students, it more often comes with disengaged emotions—with feeling effective, superior, and proud (Kitayama & Markus, 2000). Conflict in collectivist cultures often takes place between groups; individualist cultures breed more conflict (and crime and divorce) between individuals (Triandis, 2000).

When Kitayama (1999), after 10 years of teaching and researching in America, visited his Japanese alma mater, Kyoto University, graduate students were “astounded” when he explained the Western idea of the independent self. “I per­ sisted in explaining this Western notion of self-concept—one that my American students understood intuitively—and finally began to persuade them that, indeed, many Americans do have such a disconnected notion of self. Still, one of them, sighing deeply, said at the end, ‘Could this really be true?”‘

When East meets West—as happens, for example, thanks to Western influences in urban Japan and to Japanese exchange students visiting Western countries—does the self-concept become more individualized? Are the Japanese influenced when exposed to Western promotions based on individual achievement, with admoni­ tions to “believe in one’s own possibilities,” and with movies in which the heroic individual police officer catches the crook despite others’ interference? They seem to be, report Steven Heine and co-researchers (1999). Personal self-esteem increased among Japanese exchange students after spending 7 months at the University of British Columbia. In Canada, individual self-esteem is also higher among long­ term Asian immigrants than among more recent immigrants (and than it is among those living in Asia).

Self-Knowledge “Know thyself,” admonished an ancient Greek oracle. We certainly try. We readily form beliefs about ourselves, and we in Western cultures don’t hesitate to explain why we feel and act as we do. But how well do we actually know ourselves?

“There is one thing, and only one in the whole universe which we know more about than we could learn from external observation,” noted C. S. Lewis (1952, pp. 18-19). “That one thing is [ourselves]. We have, so to speak, inside information;

Collectivism in action: Follow­ ing the 2010 earthquake in Haiti, people acted together to help one another.

The Self in a Social World Chapter 2 45

THE inside STORY Hazel Markus and Shinobu Kitayama on Cultural Psychology

I We began our collaboration by wondering out loud, i Japanese researcher Shinobu wondered why American i life was so weird. American researcher Hazel countered

with anecdotes about the strangeness of Japan. Cultural psychology is about making the strange familiar and the familiar strange. Our shared cultural encounters aston­ ished us and convinced us that when it comes to psycho­ logical functioning, culture matters.

After weeks of lecturing in Japan to students with a good command of English, Hazel wondered why the stu­ dents did not say anything—no questions, no comments. She assured students she was interested in ideas that were different from hers, so why was there no response? Where were the arguments, debates, and signs of critical thinking? Even if she asked a straightforward question,

i “Where is the best noodle shop?” the answer was invari- i ably an audible intake of air followed by, “It depends.”

Didn’t Japanese students have preferences, ideas, opin­ ions, and attitudes? What is inside a head if it isn’t these

I things? How could you know someone if she didn’t tell 1 you what she was thinking? I Shinobu was curious about why American students I shouldn’t just listen to a lecture and why they felt the i need to be constantly interrupting each other and talk- j ing over each other and the professor. Why did the com­

ments and questions reveal strong emotions and have a i competitive edge? What was the point of this arguing?

Why did intelligence seem to be associated with getting the best of another person, even within a class where people knew each other well?

Shinobu expressed his amazement at American hosts who bombard their guests with choices. Do you want wine or beer, or soft drinks or juice, or coffee or tea? Why burden the guest with trivial decisions? Surely the host knew what would be good refreshment on this occasion and could simply provide something appropriate.

Choice as a burden? Hazel wondered if this could i be the key to one particularly humiliating experience in I Japan. A group of eight—all native Japanese except for I Hazel—was in a French restaurant, and everyone was fol- j lowing the universal restaurant script and studying the I menu. The waiter approached and stood nearby. Hazel

announced her choice of appetizer and entree. Next was a tense conversation among the Japanese host and the Japanese guests. When the meal was served, it was not what she had ordered. Everyone at the table was served the same meal. This was deeply disturbing. If you can’t choose your own dinner, how could it be enjoyable? What was the point of the menu if everybody is served the same meal?

Could a sense of sameness be a good or a desir­ able feeling in Japan? When Hazel walked around the grounds of a temple in Kyoto, there was a fork in the path and a sign that read: “ordinary path.” Who would want to take the ordinary path? Where was the special, less traveled path? Choosing the non-ordinary path may be an obvious course for Americans, but in this case it led to the temple dump outside the temple grounds. The ordinary path did not denote the dull and unchallenging way, but meant the appropriate and the good way.

These exchanges inspired our experimental stud­ ies and remind us that there are ways of life beyond the ones that each of us knows best. So far, most of psychol­ ogy has been produced by psychologists in middle-class White American settings studying middle-class White American respondents. In other sociocultural contexts, there can be different ideas and practices about how to be a person and how to live a meaningful life, and these differences have an influence on psychological function­ ing. It is this realization that fuels our continuing interest in collaboration and in cultural psychology.

Hazel Rose Markus Stanford University


Shinobu Kitayama University of Michigan


46 Part One




planning fallacy The tendency to under­ estimate how long it will take to complete a task.

Social Thinking

we are in the know.” Indeed. Yet sometimes we think we know, but our inside information is wrong. That is the unavoidable conclusion of some fascinating research.

EXPLAINING OUR BEHAVIOR Why did you choose where to go to college? Why did you lash out at your room­ mate? Why did you fall in love with that special person? Sometimes we know. Some­ times we don’t. Asked why we have felt or acted as we have, we produce plausible answers. Yet, when causes are subtle, our self-explanations are often wrong. We may dismiss factors that matter and inflate others that don’t. People may misattrib- ute their rainy-day gloom to life’s emptiness (Schwarz & Clore, 1983). And people routinely deny being influenced by the media, which, they readily acknowledge, affects others.

Also thought provoking are studies in which people have recorded their moods every day for 2 or 3 months (Stone & others, 1985; Weiss & Brown, 1976; Wilson & others, 1982). They also recorded factors that might affect their moods: the day of the week, the weather, the amount they slept, and so forth. At the end of each study, the people judged how much each factor had affected their moods. Even with their attention on their daily moods, there was little relationship between their perceptions of how well a factor predicted their mood and how well it really did. For example, people thought they would experience more negative moods on Mondays, but in fact their moods were no more negative on Mondays than on other weekdays. This raises a disconcerting question: How much insight do we really have into what makes us happy or unhappy? As Daniel Gilbert (2007, 2011) notes, not much: We are remarkably bad predictors of what will make us happy. “We seem to know less about the worlds inside our heads than about the world our heads are inside.”

PREDICTING OUR BEHAVIOR People also err when predicting their behavior. Dating couples tend to predict the longevity of their relationships through rose-colored glasses. Their friends and family often know better, report Tara MacDonald and Michael Ross (1997). Among University of Waterloo students, their roommates were better predictors of whether their romances would survive than they were. Medical residents weren’t very good at predicting whether they would do well on a surgical skills exam, but their peers in the program predicted each other’s performance with startling accuracy (Lutsky & others, 1993). So if you’re in love and want to know whether it will last, don’t listen to your heart—ask your roommate. And if you want to predict your routine daily behaviors—how much time you will spend laughing, on the phone, or watch­ ing TV, for example—your close friends’ estimates will likely prove at least as accu­ rate as your own (Vazire & Mehl, 2008).

One of the most common errors in behavior prediction is underestimating how long it will take to complete a task (called the planning fallacy). The Big Dig free­ way construction project in Boston was supposed to take 10 years and actually took 20 years. The Sydney Opera House was supposed to be completed in 6 years; it took 16. In one study, college students writing a senior thesis paper were asked to pre­ dict when they would complete the project. On average, students finished 3 weeks later than their “most realistic” estimate—and a week later than their “worst-case scenario” estimate (Buehler Sz others, 2002). However, friends and teachers were able to predict just how late these papers would be. Just as you should ask your friends how long your relationship is likely to survive, if you want to know when you will finish your term paper, ask your roommate or your mom. You could also do what Microsoft does: Managers automatically add 30 percent onto a software developer’s estimate of completion—and 50 percent if the project involves a new operating system (Dunning, 2006).

The Self in a Social World Chapter 2 47

So, how can you improve your self-predictions? The best way is to be more realistic about how long tasks took in the past. Apparently people underestimate how long something will take because they misremember previous tasks as tak­ ing less time than they in fact did (Roy & others, 2005).

Are people equally bad at predicting how much money they will spend? Johanna Peetz and Roger Buehler (2009) found that the answer was yes. Canadian undergraduates predicted that they would spend $94 over the next week but actually spent $122. Considering they had spent $126 in the week before the study, their guess should have been more accurate. When they came back a week later, they still predicted they would spend only $85 in the coming week. Students who said they wanted to save money were more likely to predict they would spend less—^but ended up spending the same amount as everyone else. So just as we think we will complete tasks quickly, we think we will save our money. The difficulty lies in actually doing so. If Lao-tzu was right—”He who knows others is learned. He who knows himself is enlightened”—then most people, it would seem, are more learned than enlightened.

Predicting behavior, even one’s own, is no easy matter, which may be why some people go to tarot card readers in hope of help.

PREDICTING OUR FEELINGS Many of life’s big decisions involve predicting our future feelings. Would mar­ rying this person lead to lifelong contentment? Would entering this profes­ sion make for satisfying work? Would going on this vacation produce a happy experience? Or would the likelier results be divorce, job burnout, and holiday disappointment?

Sometimes we know how we will feel—if we fail that exam, win that big game, or soothe our tensions with a half-hour jog. We know what exhilarates us and what makes us anxious or bored. Other times we may mispredict our responses. Asked how they would feel if asked sexually harassing questions on a job interview, most women studied by Julie Woodzicka and Marianne LaFrance (2001) said they would feel angry. When actually asked such questions, however, women more often expe­ rienced fear.

Studies of “affective forecasting” reveal that people have greatest difficulty pre­ dicting the intensity and the duration of their future emotions (Wilson & Gilbert, 2003). People have mispredicted how they would feel some time after a romantic breakup, receiving a gift, losing an election, winning a game, and being insulted (Gilbert & Ebert, 2002; Loewenstein & Schkade, 1999). Some examples:

When young men are sexually aroused by erotic photographs, then exposed to a passionate date scenario in which their date asks them to “stop,” they admit that they might not stop. If not shown sexually arousing pictures first, they more often deny the possibility of being sexually aggressive. When not aroused, one easily mispredicts how one will feel and act when aroused—a phenomenon that leads to unexpected professions of love during lust, to unintended pregnancies, and to repeat offenses among sex abusers who have sincerely vowed “never again.” Hungry shoppers do more impulse buying (“Those doughnuts would be delicious!”) than do shoppers who have just enjoyed a quarter-pound blue­ berry muffin (Gilbert & Wilson, 2000). When we are hungry, we mispredict how gross those deep-fried doughnuts will seem when we are sated. When stuffed, we may underestimate how yummy a doughnut might be with a late-night glass of milk—a purchase whose appeal quickly fades when we have eaten one or two.










48 Part One Social Thinking

impact bias Overestimating the enduring impact of emotion-causing events.




• How much will you like the guy you’re about to speed date? Ask the woman who went before you. Female college students predicted their enjoyment of a date better when another woman who had speed-dated him clued them in than when relying on facts such as a picture and a profile. Yet at the end of the experiment, most women still said that relying on the profile would be a better predictor of their feelings than the subjective opinion of another speed-dater (Gilbert & others, 2009).

• When natural disasters like hurricanes occur, people predict that their sad­ ness will be greater if more people are killed. But after Hurricane Katrina struck in 2005, students’ sadness was similar when it was believed that 50 people had been killed or 1,000 had been killed (Dunn & Ashton-James, 2008). What did influence how sad people felt? Seeing pictures of victims. No wonder poignant images on TV have so much influence on us after disasters.

• People overestimate how much their well-being would be affected by both bad events (a romantic breakup, failing to reach an athletic goal [Eastwick &: others, 2007a; van Dijk & others, 2008]) and good events (warmer winters, weight loss, more television channels, more free time). Even extreme events, such as winning a state lottery or suffering a paralyzing accident, affect long­ term happiness less than most people suppose.

Our intuitive theory seems to be: We want. We get. We are happy. If that were true, this chapter would have fewer words. In reality, note Daniel Gilbert and Timothy Wilson (2000), we often “miswant.” People who imagine an idyllic des­ ert island holiday with sun, surf, and sand may be disappointed when they dis­ cover “how much they require daily structure, intellectual stimulation, or regular infusions of Pop Tarts.” We think that if our candidate or team wins we will be delighted for a long while. But study after study reveals our vulnerability to impact bias—overestimating the enduring impact of emotion-causing events. Faster than we expect, the emotional traces of such good tidings evaporate.

Moreover, we are especially prone to impact bias after negative events. When Gilbert and colleagues (1998) asked assistant professors to predict their happiness a few years after achieving tenure or not, most believed a favorable outcome was important for their future happiness: “Losing my job would crush my life’s ambi­ tions. It would be terrible.” Yet when surveyed several years after the event, those denied tenure were about as happy as those who received it. Impact bias is impor­ tant, say Wilson and Gilbert (2005), because people’s “affective forecasts”—their predictions of their future emotions—influence their decisions. If people overesti­ mate the intensity and the duration of the pleasure they will gain from purchasing a new car or undergoing cosmetic surgery, then they may make ill-advised invest­ ments in that new Mercedes or extreme makeover.

Let s make this personal. Gilbert and Wilson invite us to imagine how we might feel a year after losing our nondominant hands. Compared with today, how happy would you be?

Thinking about that, you perhaps focused on what the calamity would mean: no clapping, no shoe tying, no competitive basketball, no speedy keyboarding. Although you likely would forever regret the loss, your general happiness some time after the event would be influenced by “two things: (a) the event, and (b) every­ thing else” (Gilbert & Wilson, 2000). In focusing on the negative event, we discount the importance of everything else that contributes to happiness and so overpredict our enduring misery. “Nothing that you focus on will make as much difference as you think,” write researchers David Schkade and Daniel Kahneman (1998).

Moreover, say Wilson and Gilbert (2003), people neglect the speed and the power of their psychological immune system, which includes their strategies for rationaliz­ ing, discounting, forgiving, and limiting emotional trauma. Being largely ignorant-PSALM 30:5

The Self in a Social World Chapter 2 49

of the speed and strength of our psychological immune system (a phenomenon Gilbert and Wilson call immune neglect), we adapt to disabilities, romantic break­ ups, exam failures, tenure denials, and personal and team defeats more readily than we would expect. Ironically, as Gilbert and colleagues report (2004), major negative events (which activate our psychological defenses) can be less enduringly distress­ ing than minor irritations (which don’t activate our defenses). We are, under most circumstances, amazingly resilient.

THE WISDOM AND ILLUSIONS OF SELF-ANALYSIS To a striking extent, then, our intuitions are often dead wrong about what has influenced us and what we will feel and do. But let’s not overstate the case. When the causes of our behavior are conspicuous and the correct explanation fits our intuition, our self-perceptions will be accurate (Gavanski & Hoffman, 1987). When the causes of behavior are obvious to an observer, they are usually obvious to us as well.

As Chapter 3 will explore further, we are unaware of much that goes on in our minds. Perception and memory studies show that we are more aware of the results of our thinking than of its process. For example, we experience the results of our mind’s unconscious workings when we set a mental clock to record the passage of time or to awaken us at an appointed hour, or when we somehow achieve a spontaneous creative insight after a problem has unconsciously “incu­ bated.” Similarly, creative scientists and artists often cannot report the thought processes that produced their insights, although they have superb knowledge of the results.

Timothy Wilson (1985,2002) offers a bold idea: The mental processes that control our social behavior are distinct from the mental processes through which we explain our behavior. Our rational explanations may therefore omit the unconscious atti­ tudes that actually guide our behavior. In nine experiments, Wilson and colleagues (1989,2008) found that the attitudes people consciously expressed toward things or people usually predicted their subsequent behavior reasonably well. Their attitude reports became useless, however, if the participants were first asked to analyze their feelings. For example, dating couples’ level of happiness with their relationship accurately predicted whether they would still be dating several months later. But participants who first listed all the reasons they could think of why their relation­ ship was good or bad before rating their happiness were misled—their happiness ratings were useless in predicting the future of the relationship! Apparently, the process of dissecting the relationship drew attention to easily verbalized factors that were actually not as important as harder-to-verbalize happiness. We are often “strangers to ourselves,” Wilson concluded (2002).

Such findings illustrate that we have a dual attitude system, say Wilson and colleagues (2000). Our automatic implicit attitudes regarding someone or some­ thing often differ from our consciously controlled, explicit attitudes (Gawronski & Bodenhausen, 2006; Nosek, 2007). From childhood, for example, we may retain a habitual, automatic fear or dislike of people for whom we now consciously verbal­ ize respect and appreciation. Although explicit attitudes may change with relative ease, notes Wilson, “implicit attitudes, like old habits, change more slowly.” With repeated practice, however, new habitual attitudes can replace old ones.

This research on the limits of our self-knowledge has two practical implications. The first is for psychological inquiry. Self-reports are often untrustworthy. Errors in self-understanding limit the scientific usefulness of subjective personal reports.

The second implication is for our everyday lives. The sincerity with which people report and interpret their experiences is no guarantee of the validity of those reports. Personal testimonies are powerfully persuasive (as we will see in Chapter 15, Social Psychology in Court). But they may also be wrong. Keeping this potential for error in mind can help us feel less intimidated by others and be less gullible.

immune neglect The human tendency to underestimate the speed and the strength of the “psychological immune system,” which enables emotional recovery and resilience after bad things happen.






dual attitude system Differing implicit (automatic) and explicit (consciously controlled) attitudes toward the same object. Verbalized explicit attitudes may change with education and persuasion; implicit attitudes change slowly, with practice that forms new habit

50 Part One Social Thinking

SUMMING UP: Self-Concept: Who Am I? • Our sense of self helps organize our thoughts and

actions. When we process information with refer­ ence to ourselves, we remember it well (the self­ reference effect). Self-concqjt consists of two ele­ ments: the self-schemas that guide our processing of self-relevant information, and the possible selves that we dream of or dread.

• Cultures shape the self, too. Many people in indi­ vidualistic Western cultures assume an independent self. Others, often in collectivistic cultures, assume a more interdependent self As Chapter 5 will further explain, these contrasting ideas contribute to cul­ tural differences in social behavior.

• Our self-knowledge is curiously flawed. We often do not know why we behave the way we do. When influences upon our behavior are not conspicuous enough for any observer to see, we, too, can miss them. The unconscious, implicit processes that con­ trol our behavior may differ from our conscious, explicit explanations of it. We also tend to mispre­ dict our emotions. We underestimate the power of our psychological immune systems and thus tend to overestimate the durability of our emotional reactions to significant events.

self-esteem A person’s overall self- evaluation or sense of self-worth.


Understand self-esteem and its implications for behavior and cognition.

Everyone desires self-esteem, which we are motivated to enhance. But how can self-esteem sometimes be problematic?

Is self-esteem—our overall self-evaluation—the sum of all our self-schemas and possible selves? If we see ourselves as attractive, athletic, smart, and destined to be rich and loved, will we have high self-esteem? Yes, say Jennifer Crocker and Connie Wolfe (2001)—when we feel good about the domains (looks, smarts, or whatever) important to our self-esteem. “One person may have self-esteem that is highly con­ tingent on doing well in school and being physically attractive, whereas another may have self-esteem that is contingent on being loved by God and adhering to moral standards.” Thus, the first person will feel high self-esteem when made to feel smart and good-looking, the second person when made to feel moral.

But Jonathon Brown and Keith Dutton (1994) argue that this “bottom-up” view of self-esteem is not the whole story. The causal arrow, they believe, also goes the other way. People who value themselves in a general way—those with high self-esteem—are more likely to value their looks, abilities, and so forth. They are like new parents who, loving their infant, delight in the baby’s fingers, toes, and hair: The parents do not first evaluate their infant’s fingers or toes and then decide how much to value the whole baby.

Specific self-perceptions do have some influence, however. If you think you’re good at math, you will be more likely to do well at math. Although general self­ esteem does not predict academic performance very well, academic self-concept— whether you think you are good in school—does predict performance (Marsh & O’Mara, 2008). Of course, each causes the other: Doing well at math makes you think you are good at math, which then motivates you to do even better. So if you want to encourage someone (or yourself!), it’s better if your praise is specific (“You’re good at math”) instead of general (“You’re great”) and if your kind words reflect true ability and performance (“You really improved on your last test”) rather than unrealistic optimism (“You can do anything”). Feedback is best when it is true and specific (Swann & others, 2007).

The Self in a Social World Chapter 2 51

Imagine you’re getting your grade back for the first test in a psychology class. When you see your grade, you groan—you’re hovering somewhere between a D and an F. But then you get an encouraging e-mail with some review questions for the class and this message: “Students who have high self-esteem not only get bet­ ter grades, but they remain self-confident and assured…. Bottom line: Hold your head—and your self-esteem—high.” Another group of students instead get a mes­ sage about taking personal control of their performance, or receive review ques­ tions only. So how would each group do on the final exam? To the surprise of the researchers, the students whose self-esteem was boosted did by far the worst on the final—in fact, they flunked it (Forsyth & others, 2007). Struggling students told to feel good about themselves, the researchers muse, may have thought, “I’m already great—why study?”

Self-Esteem Motivation Most people are extremely motivated to maintain their self-esteem. In fact, a study found that college students preferred a boost to their self-esteem to eating their favorite food, engaging in their favorite sexual activity, seeing a best friend, drink­ ing alcohol, or receiving a paycheck (Bushman & others, 2011). So, somewhat incredibly, self-esteem was more important than sex, pizza, and beer!

What happens when your self-esteem is threatened—for example, by a failure or an unflattering comparison with someone else? When brothers have markedly dif­ ferent ability levels—for example, one is a great athlete and the other is not—they report not getting along well (Tesser, 1988).

Self-esteem threats also occur among friends, whose success can be more threat­ ening than that of strangers (Zuckerman & Jost, 2001). Your level of self-esteem also makes a difference: High self-esteem people usually react to a self-esteem threat by compensating for it (blaming someone else or trying harder next time). These reac­ tions help them preserve their positive feelings about themselves. Low self-esteem people, however, are more likely to “break” by blaming themselves or giving up (VanDellen & others, 2011).

What underlies the motive to maintain or enhance self-esteem? Mark Leary (1998, 2004b, 2007) believes that our self-esteem feelings are like a fuel gauge. Rela­ tionships enable surviving and thriving. Thus, the self-esteem gauge alerts us to threatened social rejection, motivating us to act with greater sensitivity to others’ expectations. Studies confirm that social rejection lowers our self-esteem and makes

Among sibling relationships, the threat to self-esteem is greatest for an older child with a highly capable younger brother or sister.

52 Part One Social Thinking

terror management theory Proposes that people exhibit self-protective emotional and cognitive responses (including adhering more strongly to their cultural worldviews and prejudices) when confronted with reminders of their mortality.

us more eager for approval. Spurned or jilted, we feel unattractive or inadequate. Like a blinking dashboard light, this pain can motivate action—self-improvement and a search for acceptance and inclusion elsewhere.

Jeff Greenberg (2008) offers another perspective, called “terror management theory/’ which argues that humans must find ways to manage their overwhelm­ ing fear of death. If self-esteem were only about acceptance, he counters, why do “people strive to be great rather than to just be accepted”? The reality of our own death, he argues, motivates us to gain recognition from our work and values. There’s a worm in the apple, however: Not everyone can achieve such recogni­ tion, which is exactly why it is valuable, and why self-esteem can never be wholly unconditional (“You’re special just for being you” is an example of self-esteem being granted unconditionally). To feel our lives are not in vain, Greenberg main­ tains, we must continually pursue self-esteem by meeting the standards of our societies.

The “Dark Side” of Self-Esteem People with low self-esteem often have problems in life—they make less money, abuse drugs, and are more likely to be depressed (Salmela-Aro & Nurmi, 2007; Trzesniewski &: others, 2006). As you learned in Chapter 1, though, a correlation between two variables is sometimes caused by a third factor. Maybe people low in self-esteem also faced poverty as children, experienced sexual abuse, or had par­ ents who used drugs—all possible causes of later struggling. Sure enough, a study that controlled for these factors found that the link between self-esteem and nega­ tive outcomes disappeared (Boden & others, 2008). In other words, low self-esteem was not the cause of these young adults’ problems—the seeming cause, instead, was that many could not escape their tough childhoods.

High self-esteem does have some benefits—it fosters initiative, resilience, and pleasant feelings (Baumeister & others, 2003). Yet teen males who engage in sex­ ual activity at an “inappropriately young age” tend to have higher than average self-esteem. So do teen gang leaders, extreme ethnocentrists, terrorists, and men in prison for committing violent crimes (Bushman & Baumeister, 2002; Dawes, 1994, 1998). “Hitler had very high self-esteem,” note Baumeister and co-authors (2003).

NARCISSISM: SELF-ESTEEM’S CONCEITED SISTER High self-esteem becomes especially problematic if it crosses over into narcissism, or having an inflated sense of self. Most people with high self-esteem value both individual achievement and relationships with others. Narcissists usually have high self-esteem, but they are missing the piece about caring for others (Campbell & others, 2002). Although narcissists are often outgoing and charming early on, their self-centeredness often leads to relationship problems in the long run (Campbell, 2005). The link between narcissism and problematic social relations led Delroy Paulhus and Kevin Williams (2002) to include narcissism in “The Dark Triad” of negative traits, along with Machiavellianism (manipulativeness) and antisocial psychopathy.

In a series of experiments conducted by Brad Bushman and Roy Baumeister (1998), undergraduate volunteers wrote essays and received rigged feedback that said, “This is one of the worst essays I’ve read!” Those who scored high on narcis­ sism were much more likely to retaliate, blasting painful noise into the headphones of the student they believed had criticized them. Narcissists weren’t aggressive toward someone who praised them (“great essay!”). It was the insult that set them off. But what about self-esteem? Maybe only the “insecure” narcissists—those low in self-esteem—would lash out. But that’s not how it turned out—instead, the stu­ dents high in both self-esteem and narcissism were the most aggressive. The same was true in a classroom setting—those who were high in both self-esteem and nar­ cissism were the most likely to retaliate against a classmate’s criticism by giving him

53The Self in a Social World

C •a y.

or her a bad grade (Bushman & others, 2009; Figure 2.5). Narcissists can be charming and entertaining. But as one wit has said, “God help you if you cross them.”

It’s also possible to have too much narcissistic pride in your group, not just yourself. Polish undergraduates who displayed a “collective narcissism,” believ­ ing their country was superior to others, were more prejudiced against Jewish people. Mexican undergraduates high in collective narcissism were more likely to view the construction of a U.S.-Mexico border wall as an insult and to endorse a boycott of U.S. products in retaliation (Golec deZavala & others, 2009). So whether someone has excessive pride in themselves or their group, others may end up suffering.

Some studies have found small correlations between low self-esteem and antiso­ cial behavior, even when IQ and family income were taken into account (Donnellan & others, 2005; Trzesniewski & others, 2006). However, another study found that the link between low self-esteem and antisocial behavior disappeared when sexual abuse and earlier behavioral problems were considered (Boden & others, 2007). Kids aren’t acting aggressively because they have low self-esteem, it seems, but because they were hurt in the past. “The enthusiastic claims of the self-esteem movement mostly range from fantasy to hogwash,” says Baumeister (1996), who suspects he has “probably published more studies on self-esteem than anybody else…. The effects of self-esteem are small, limited, and not all good.” Folks with high self-esteem, he reports, are more likely to be obnoxious, to interrupt, and to talk at people rather than with them (in contrast to the more shy, modest, self-folks with low self-esteem). “My conclusion is that self-control is worth 10 times as much as self-esteem.”

What about the idea that an overinflated ego is just a cover for deep-seated inse­ curity? Do narcissistic people actually hate themselves “deep down inside”? Recent studies show that the answer is no. People who score high on measures of narcissis­ tic personality traits also score high on measures of self-esteem. In case narcissists were claiming high self-esteem just for show, researchers also asked undergradu­ ates to play a computer game where they had to press a key as quickly as possible to match the word “me” with words such as “good,” “wonderful,” “great,” and “right,” and words such as “bad,” “awful,” “terrible,” and “wrong.” High scorers on the narcissism scale were faster than others to associate themselves with good words, and slower than others to pair themselves with bad words (Campbell & others, 2007). And narcissists were even faster to identify with words such as “out­ spoken,” “dominant,” and “assertive.” Although it might be comforting to think

FIGURE :: 2.5 Narcissism, Self-Esteem, and Aggression Narcissism and self-esteem interact to influence aggression. In an experiment by Brad Bushman and colleagues (2009), the recipe for retaliation against a critical classmate required both narcissism and high self-esteem.

Chapter 2














54 Part One Social Thinking

that an arrogant classmate is just covering for his insecurity, chances are that deep down inside he thinks he’s awesome.

NARCISSISM ON THE RISE After tracking self-importance across the past several decades, psychologist Jean Twenge {2006; Twenge &: others, 2008) reports that today’s young generation— Generation Me, she calls it—express more narcissism (by agreeing with statements such as “If I ruled the world, it would be a better place” or “I think I am a special person”). Narcissism scores rose over time on college campuses from Alabama to Maryland to California {Stewart & Bernhardt, 2010; Twenge & Foster, 2008, 2010). Narcissism correlates with materialism, the desire to be famous, inflated expecta­ tions, fewer committed relationships and more “hooking up,” more gambling, and more cheating, all of which have also risen as narcissism has increased. Narcissism is also linked to a lack of empathy—the ability to take someone else’s perspective and be concerned about their problems—and empathy has dropped precipitously among college students {Konrath & others, 2011). The researchers speculate that today’s generation may be so wrapped up in online interaction that their in-person interaction skills have atrophied. Or, they say, empathy might have declined because young people today are “feeling too busy on their paths to success,” single-mindedly concentrating on their own achievement because the world is now so competitive. Yet, ironically, those high in narcissism and low in empathy are less—not more— successful in the long run, making lower grades in college and performing poorly at work {Judge & others, 2006; Robins & Beer, 2001).

LOW VERSUS SECURE SELF-ESTEEM The findings linking a highly positive self-concept with negative behavior exist in tension with the findings that people expressing low self-esteem are more vulnera­ ble to assorted clinical problems, including anxiety, loneliness, and eating disorders. When feeling bad or threatened, low-self-esteem people often take a negative view of everything. They notice and remember others’ worst behaviors and think their partners don’t love them {Murray & others, 1998, 2002; Ybarra, 1999). Although there is no evidence that low-self-esteem people choose less desirable partners, they are quick to believe that their partners are criticizing or rejecting them. Per­ haps as a result, low-self-esteem people are less satisfied with their relationships {Fincham & Bradbury, 1993). They may also be more likely to leave those relation­ ships. Low-self-esteem undergraduates decided not to stay with roommates who saw them in a positive light {Swann & Pelham, 2002).

Secure self-esteem—one rooted more in feeling good about who one is than in grades, looks, money, or others’ approval—is conducive to long-term well-being {Kemis, 2003; Schimel & others, 2001). Jennifer Crocker and colleagues {Crocker, 2002; Crocker and Luhtanen, 2003; Crocker and Park, 2004; Crocker and Knight, 2005) confirmed this in studies with University of Michigan students. Those whose self-worth was most fragile—most contingent on external sources—experienced more stress, anger, relationship problems, drug and alcohol use, and eating disor­ ders than did those whose sense of self-worth was rooted more in internal sources, such as personal virtues.

Ironically, note Crocker and Lora Park (2004), those who pursue self-esteem, perhaps by seeking to become beautiful, rich, or popular, may lose sight of what really makes for quality of life. Moreover, if feeling good about ourselves is our goal, then we may become less open to criticism, more likely to blame than empa­ thize with others, and more pressured to succeed at activities rather than enjoy them. Over time, such pursuit of self-esteem can fail to satisfy our deep needs for competence, relationship, and autonomy, note Crocker and Park. To focus less on one’s self-image, and more on developing one’s talents and relationships, eventu­ ally leads to greater well-being. Kristin Neff (2011) suggests we label this approach

The Self in a Social World Chapter 2 55

self-compassion—leaving behind comparisons with others and instead treating ourselves with kindness. As an Indian proverb puts it, “There is nothing noble in being superior to some other person. The true nobility is in being superior to your previous self.”

SUMMING UP: What Is the Nature and Motivating Power of Self-Esteem?

• Self-esteem is the overall sense of self-worth we use to appraise our traits and abilities. Our self- concepts are determined by multiple influences, including the roles we play, the comparisons we make, our social identities, how we perceive others appraising us, and our experiences of success and failure.

• Self-esteem motivation influences our cognitive processes: Facing failure, high-self-esteem people

sustain their self-worth by perceiving other people as failing, too, and by exaggerating their superior­ ity over others.

• Although high self-esteem is generally more ben­ eficial than low, researchers have found that peo­ ple high in both self-esteem and narcissism are the most aggressive. Someone with a big ego who is threatened or deflated by social rejection is poten­ tially aggressive.


Understand se/f-concepf through examination of the self in action.

We have considered what a self-concept is, how it develops, and how well {or poorly) we know ourselves. Now let’s see why our self-concepts matter, by viewing the self in action.

The Self’s Energy The self’s capacity for action has limits, note Roy Baumeister and colleagues {1998, 2000; Muraven & others, 1998). Consider:

• People who exert self-control—by forcing themselves to eat radishes rather than chocolates, or by suppressing forbidden thoughts—subsequently quit faster when given unsolvable puzzles.

• People who have tried to control their emotional responses to an upsetting movie exhibit decreased physical stamina.

• People who have spent their willpower on tasks such as controlling their emotions during an upsetting film later become more aggressive and more likely to fight with their partners {DeWall & others, 2007; Finkel & Campbell, 2001). They also become less restrained in their sexual thoughts and behav­ iors. When asked to express intimacy with their partner, those with depleted willpower were more likely to passionately kiss their partner and even remove some clothing right in the lab {Gailliot & Baumeister, 2007).

Effortful self-control depletes our limited willpower reserves. Our brain’s “cen­ tral executive” consumes available blood sugar when engaged in self-control (Gailliot, 2008). Self-control therefore operates similarly to muscular strength, conclude Baumeister and Julie Exline (2000): Both are weaker after exertion, replen­ ished with rest, and strengthened by exercise.

56 Part One

self-efficacy A sense that one is competent and effective, distinguished from self-esteem, which is one’s sense of self-worth. A sharpshooter in the military might feel high self-efficacy and low self-esteem.

Social Thinking

Although the self’s energy can be temporarily depleted, our self-concepts do influence our behavior (Graziano & others, 1997). Given challenging tasks, people who imagine themselves as hardworking and successful outperform those who imagine themselves as failures (Ruvolo & Markus, 1992). Envision your positive possibilities and you become more likely to plan and enact a successful strategy.

Self-Efficacy Stanford psychologist Albert Bandura (1997, 2000, 2008) captured the power of positive thinking in his research and theorizing about self-efficacy (how compe­ tent we feel on a task). Believing in our own competence and effectiveness pays dividends (Bandura & others, 1999; Maddux and Gosselin, 2003). Children and adults with strong feelings of self-efficacy are more persistent, less anxious, and less depressed. They also live healthier lives and are more academically successful.

In everyday life, self-efficacy leads us to set challenging goals and to per­ sist. More than 100 studies show that self-efficacy predicts worker productivity (Stajkovic & Luthans, 1998). When problems arise, a strong sense of self-efficacy leads workers to stay calm and seek solutions rather than ruminate on their inade­ quacy. Competence plus persistence equals accomplishment. And with accomplish­ ment, self-confidence grows. Self-efficacy, like self-esteem, grows with hard-won achievements.

Even subtle manipulations of self-efficacy can affect behavior. Becca Levy (1996) discovered this when she subliminally exposed 90 older adults to words that evoked (primed) either a negative or a positive stereotype of aging. Some subjects viewed .066-second presentations of negative words such as “decline,” “forgets,” and “senile,” or of positive words such as “sage,” “wise,” and “learned.” At the conscious level, the participants perceived only a flash of light. Yet being given the positive words led to heightened “memory self-efficacy” (confidence in one’s mem­ ory) and better memory performance. Viewing the negative words had the opposite effect. We can observe a similar phenomenon outside the laboratory: Older adults in China, where positive images of aging prevail and memory self-efficacy may be greater, seem to suffer less memory decline than is commonly observed in Western countries (Schacter & others, 1991).

If you believe you can do something, will that belief necessarily make a differ­ ence? That depends on a second factor: Do you have control over your outcomes? You may, for example, feel like an effective driver (high self-efficacy) yet feel endan­ gered by drunken drivers (low control). You may feel like a competent student or worker but, fearing discrimination based on your age, gender, or appearance, you may think your prospects for success are dim.

Many people confuse self-efficacy with self-esteem. If you believe you can do something, that’s self-efficacy. If you like yourself overall, that’s self-esteem. When you were a child, your parents may have encouraged you by saying things such as “You’re special!” (intended to build self-esteem) or “I know you can do it!” (intended to build self-efficacy). One study showed that self-efficacy feedback (“You tried really hard”) led to better performance than self-esteem feedback (“You’re really smart”). Children told they were smart were afraid to try again—maybe they wouldn’t look so smart next time. Those praised for working hard, however, knew they could exert more effort again (Mueller & Dweck, 1998). If you want to encour­ age someone, focus on their self-efficacy, not their self-esteem.

Locus of Control “I have no social life,” complained a 40-something single man to student thera­ pist Jerry Phares. At Phares’s urging, the patient went to a dance, where several women danced with him. “I was just lucky,” he later reported. “It would never happen again.” When Phares reported this to his mentor, Julian Rotter, it crys­ tallized an idea he had been forming. In Rotter’s experiments and in his clinical

The Self in a Social World

practice, some people seemed to persistently “feel that what happens to them is governed by external forces of one kind or another, while others feel that what happens to them is governed largely by their own efforts and skills” (quoted by Hunt, 1993, p. 334).

What do you think about your own life? Are you more often in charge of your destiny or a victim of circumstance? Rotter called this dimension locus of control. With Phares, he developed 29 paired statements to measure a person’s locus of control. Imagine taking this test. Which statements do you more strongly believe?

a. In the long run, people get the respect they deserve in this world.

a. What happens to me is my own doing.

a. The average person can have an influence in government decisions.

or b. Unfortunately, people’s worth passes unrecognized no matter how hard they try.

or b. Sometimes I feel that I don’t have enough control over the direction my life is taking.

or b. This world is run by the few people in power, and there is not much the little guy can do about it,

If your answers to these questions (from Rotter, 1973) were mostly “a,” you prob­ ably believe you control your own destiny {internal locus of control). If your answers were mostly “b,” you probably feel chance or outside forces determine your fate {external locus of control, as in Figure 2.6). Those who see themselves as internally controlled are more likely to do well in school, be more productive at work, make more money, successfully stop smoking, maintain a healthy weight, deal with marital problems directly, be more satisfied with life, and achieve long-term goals (Findley & Cooper, 1983; Gale & others, 2008; Miller & others, 1986; Wang & others, 2010).

How much control we feel is related to how we explain setbacks. Perhaps you have known students who view themselves as victims—who blame poor grades on things beyond their control, such as their “bad” teachers, texts, or tests. If such stu­ dents are coached to adopt a more hopeful sense of personal control—to believe that

7 ^ st#vrt – matteH

I do, things don’t go my way. She%

goingto get ahead

control my my. With enougfi ft, I could win ■.

Chapter 2 57

locus of control The extent to which people perceive outcomes as internally controllable by their own efforts or as externally controlled by chance or outside forces.

FIGURE :: 2.6 Locus of Control

58 Part One Social Thinking






learned helplessness The sense of hopelessness and resignation learned when a human or animal perceives no control over repeated bad events.

FIGURE:: 2.7 Learned Helplessness When animals and people experi­ ence uncontrollable bad events, they learn to feel helpless and resigned.

effort, good study habits, and self-discipline can make a difference—their academic performance tends to go up (Noel & others, 1987; Peterson & Barrett, 1987). They are also less likely to cheat: Students who were told that free will is an illusion—that what happened to them is outside their control—peeked at answers and paid them­ selves more money for mediocre work (Vohs & Schooler, 2008).

When rating employees’ job performance, bosses gave significantly higher rat­ ings to those with stronger free will beliefs, probably because these employees believed they could control their actions (Stillman & others, 2010). New life insur­ ance sales representatives who view failures as controllable (“It’s difficult, but with persistence I’ll get better”) sell more policies. They are only half as likely as their more pessimistic colleagues to quit during their first year (Seligman & Schulman, 1986). Among college swim team members, those with an optimistic “explanatory style” are more likely than pessimists to perform beyond expectations (Seligman & others, 1990). As the Roman poet Virgil said in the Aeneid, “They can because they think they can.”

Some people, however, have taken these ideas a little too far. The popular book The Secret, for example, claims that thinking positive thoughts causes positive things to happen to you (“The only reason any person does not have enough money is because they are blocking money from coming to them with their thoughts”). So should we conclude that we need not help those poor Somalis in Africa—all they need to do is think happy thoughts? And if you are sick, they say, your thoughts just aren’t positive enough—despite the thousands of cancer patients who desper­ ately want to get well. Obviously, there are limits to the power of positive thinking. Being optimistic and feeling in control can reap great benefits, but poverty and sick­ ness can happen to anyone.

Learned Helplessness Versus Self-Determination The benefits of feelings of control also appear in animal research. In research dorte before today’s greater concern for animal welfare, dogs confined in a cage and taught that they could not escape shocks learned a sense of helplessness. Later, these dogs cowered passively in other situations when they could escape pun­ ishment. Dogs that learned personal control (by successfully escaping their first shocks) adapted easily to a new situation. Researcher Martin Seligman (1975,1991) noted similarities to this learned helplessness in human situations. Depressed or oppressed people, for example, become passive because they believe their efforts have no effect. Helpless dogs and depressed people both suffer paralysis of the will, passive resignation, and even motionless apathy (Figure 2.7).

On the other hand, people benefit by training their self-control “muscles.” Col­ lege students who practiced self-control by sticking with an exercise program or reducing their impulse-buying also ate less jimk food, cut down on alcohol, and studied more (Oaten & Cheng, 2006a, 2006b). So if you learn how to exert willpower in one area of your life, resisting temptation in other areas becomes easier too.

Ellen Langer and Judith Rodin (1976) tested the importance of personal con­ trol by treating elderly patients in a highly rated Connecticut nursing home in one of two ways. With one group, the benevolent caregivers emphasized “our responsibility to make this a home you can be proud of and happy in.” They gave the patients their normal well-intentioned, sympathetic care and allowed them to assume a passive care-receiving role. Three weeks later, most of these patients were rated by themselves, by interviewers, and by nurses as further debilitated.

Uncontrollable i bad events

Perce!ve3^ lack of T • control m (



The Self in a Social World Chapter 2 59

Langer and Rodin’s other treatment promoted personal control. It emphasized opportunities for choice, the possibilities for influencing nursing-home policy, and the person’s responsibility “to make of your life whatever you want.” These patients were given small decisions to make and responsibilities to fulfill. During the ensuing 3 weeks, 93 percent of this group showed improved alertness, activity, and happiness.

Studies confirm that systems of governing or managing people that promote personal control will indeed promote health and happiness (Deci & Ryan, 1987). Here are some additional examples:

• Prisoners given some control over their environments—by being able to move chairs, control TV sets, and operate the lights—experience less stress, exhibit fewer health problems, and commit less vandalism (Ruback & others, 1986; Wener & others, 1987).

• Workers given leeway in carrying out tasks and making decisions experience improved morale (Miller & Monge, 1986). So do telecommuting workers who have more flexibility in balancing their work and personal life (Valcour, 2007).

• In aU countries studied, people who perceive themselves as having free choice experience greater satisfaction with their lives. And countries where people experience more freedom have more satisfied citizens (Inglehart & others, 2008).

THE COSTS OF EXCESS CHOICE Can there ever be too much of a good thing such as freedom and self-determination? Barry Schwartz (2000, 2004) contends that individualistic modern cultures indeed have “an excess of freedom,” causing decreased life satisfaction and increased rates of clinical depression. Too many choices can lead to paralysis, or what Schwartz calls “the tyranny of freedom.” After choosing from among 30 kinds of jams or chocolates, people express less satisfaction with their choices than those choosing from among 6 options (Iyengar & Lepper, 2000). Making choices is also tiring. Stu­ dents who read the catalog and chose which classes they would take during the upcoming semester—versus those who simply read it but made no choices—were later less likely to study for an important test and more likely to procrastinate by playing video games and reading magazines. In another study, students who chose among an array of consumer products were later less able to consume an unsa­ vory but healthy drink (Vohs «Sc others, 2008). So after choosing among the 19,000 possible beverage combina­ tions at Starbucks or the 40,000 items at the average super­ market, you might be less satisfied with your choices and more likely to go home and eat the ice cream straight from the container.

Christopher Hsee and Reid Hastie (2006) illustrate how choice may enhance regret. Give employees a free trip to either Paris or Hawaii and they will be happy. But give them a choice between the two and they may be less happy. People who choose Paris may regret that it lacks the warmth and the ocean. Those who choose Hawaii may regret the lack of great museums.

In other experiments, people have expressed greater sat­ isfaction with irrevocable choices (such as those made in an “all purchases final” sale) than with reversible choices (as when allowing refunds or exchanges). Ironically, people like and will pay for the freedom to reverse their choices. Vet, note Daniel Gilbert and Jane Ebert (2002), that same freedom “can inhibit the psychological processes that man­ ufacture satisfaction.”

Personal control: Inmates of Spain’s modern Valencia prison have, with work and appropriate behavior, gained access to classes, sports facilities, cultural opportunities, and money in an account that can be charged for snacks.

60 Part One Social Thinking

Confidence and feelings of self-efficacy grow from successes. © Edward Koren/The New Yorker Collection/www.cartoonbank.corr

That principle may help explain a curious social phenomenon (Myers, 2000a): National surveys show that people expressed more satisfaction with their marriages several decades ago when marriage was more irrevocable (“all purchases final”). Today despite greater freedom to escape bad marriages and try new ones, people tend to express somewhat less satisfaction with the marriage that they have.

Research on self-control gives us greater confidence in traditional vir­ tues such as perseverance and hope. Bandura (2004) acknowledges that self- efficacy is fed by social persuasion (“You have what it takes to succeed”) and by self-persuasion (“I think I can, I think I can”). Modeling—seeing similar others succeed with effort—helps, too. But the biggest source of self-efficacy, he says, is mastery experiences. “Successes build a robust belief in one’s efficacy.” If your initial efforts to lose weight, stop

smoking, or improve your grades succeed, your self-efficacy increases. A team of researchers led by Roy Baumeister (Baumeister & others, 2003) concurs

with Bandura’s conclusion about mastery experiences. “Praising all the children just for being themselves,” they contend, “simply devalues praise.” Better to praise and bolster self-esteem “in recognition of good performance…. As the person per­ forms or behaves better, self-esteem is encouraged to rise, and the net effect will be to reinforce both good behavior and improvement. Those outcomes are conducive to both the happiness of the individual and the betterment of society.”

*‘This give! my confidence a real boost.’’

THE inside STORY Daniel Gilbert on the Benefits of Irrevocable Commitments

In 2002 I changed my mind about the benefit of being able to change my mind.

Jane Ebert and I discovered that people are gen­ erally happier with decisions when they can’t undo them. When participants in our experiments were abl« to undo their decisions they tended to consider both the positive and negative features of the decisions they had made. When they couldn’t undo their decisions they tended to concentrate on the good features and ignore the bad. As such, they were more satisfied when they made irrevocable than revocable decisions. Ironi­ cally, subjects did not realize this would happen and strongly preferred to have the opportunity to change their minds.

Now, up until this point I had always believed that love causes marriage. But these experiments suggested to me that marriage would also cause love. If you take data seri­ ously you act on it, so when these results came in I went home and proposed to the woman I was living with. She said yes, and it turned out that the data were right: I love my wife more than I loved my girl­ friend, (Excerpted with per­ mission from

Daniel Gilbert Harvard University

The Self in a Social World Chapter 2 61

SUMMING UP: What Does It Mean to Have “Perceived Self-Control”?

• Several lines of research show the benefits of a sense of self-efficacy and feelings of control. People who believe in their own competence and effec­ tiveness, and who have an internal locus of control, cope better and achieve more than others.

• Learned helplessness often occurs when attempts to improve a situation have proven fruitless;

self-determination, in contrast, is bolstered by experiences of successfully exercising control and improving one’s situation.

• When people are given too many choices, they may be less satisfied with what they have than when offered a smaller range of choices.

WHAT IS SELF-SERVING BIAS? Explain self-serving bias and its adaptive and maladaptive aspects.

Most of us have a good reputation with ourselves. In studies of self-esteem, even low-scoring people respond in the midrange of possible scores. (A low-self-esteem person responds to statements such as “1 have good ideas” with a qualifying adjective, such as “somewhat” or “sometimes.”) In a study of self-esteem across 53 nations, the average self-esteem score was above the midpoint in every country (Schmitt & Allik, 2005). In recent samples of U.S. college students, the most common score on a self-esteem measure was the maximum—in effect, “perfect” self-esteem (Gentile & others, 2010). One of social psychology’s most provocative yet firmly established conclusions concerns the potency of self-serving bias—a tendency to perceive oneself favorably.

Explaining Positive and Negative Events Many dozens of experiments have found that people accept credit when told they have succeeded. They attribute the success to their ability and effort, but they attri­ bute failure to external factors such as bad luck or the problem’s inherent “impos­ sibility” (Campbell & Sedikides, 1999). Similarly, in explaining their victories, athletes commonly credit themselves, but they attribute losses to something else: bad breaks, bad referee calls, or the other team’s super effort or dirty play (Grove & others, 1991; Lalonde, 1992; Mullen & Riordan, 1988). And how much responsibility do you suppose car drivers tend to accept for their accidents? On insurance forms, drivers have described their accidents in words such as these: “An invisible car came out of nowhere, struck my car, and vanished”; “As I reached an intersection, a hedge sprang up, obscuring my vision, and I did not see the other car”; and “A pedestrian hit me and went under my car” {Toronto Neivs, 1977).

Situations that combine skill and chance (games, exams, and job applications) are especially prone to the phenomenon. When I win at Scrabble, it’s because of my verbal dexterity; when I lose, it’s because “Who could get anywhere with a Q but no U?” Politicians similarly tend to attribute their wins to themselves (hard work, constituent service, reputation, and strategy) and their losses to factors beyond their control (their district’s party makeup, their opponent’s name, and political trends) (Kingdon, 1967). When corporate profits are up, the CEOs welcome big bonuses for their managerial skill. When profits turn to losses, well, what could you expect in a down economy? This phenomenon of self-serving attributions (attrib­ uting positive outcomes to oneself and negative outcomes to something else) is one of the most potent of human biases (Mezulis & others, 2004). That might be for a

self-serving bias The tendency to perceive oneself favorably.

self-serving attributions A form of self-serving bias; the tendency to attribute positive outcomes to oneself and negative outcomes to other factors.

62 Part One Social Thinking

DILBERT © Scott Adams. Distributed by United Feature Syndicate. Inc.


i i





H o o.D V

i $

good reason: Making self-serving attributions activates brain areas associated with reward and pleasure (Seidel & others, 2010).

Self-serving attributions contribute to marital discord, worker dissatisfaction, and bargaining impasses (Kruger & Gilovich, 1999). Small wonder that divorced people usually blame their partner for the breakup (Gray & Silver, 1990), or that managers often blame poor performance on workers’ lack of ability or effort (Imai, 1994; Rice, 1985). (Workers are more likely to blame something external—supplies, excessive workload, difficult co-workers, or ambiguous assignments.) Small won­ der, too, that people evaluate pay raises as fairer when they receive a bigger raise than most of their co-workers (Diekmann & others, 1997).

We help maintain our positive self-images by associating ourselves with success and distancing ourselves from failure. For example, “1 got an A on my econ test” versus “The prof gave me a C on my history exam.” Blaming failure or rejection on something external, even another’s prejudice, is less depressing than seeing oneself as undeserving (Major & others, 2003). We will, however, acknowledge our dis­ tant past failings—those by our “former” self, note Anne Wilson and Michael Ross (2001). Describing their old precollege selves, their University of Waterloo students offered nearly as many negative as positive statements. When describing their pres­ ent selves, they offered three times more positive statements. “I’ve learned and grown, and I’m a better person today,” most people surmise. Chumps yesterday, champs today.

Ironically, we are even biased against seeing our own bias. People claim they avoid self-serving bias themselves but readily acknowledge that others commit this bias (Pronin & others, 2002). This “bias blind spot” can have serious consequences during conflicts. If you’re negotiating with your roommate over who does house­ hold chores and you believe your roommate has a biased view of the situation, you’re much more likely to become angry (Pronin & Ross, 2006). Apparently we see ourselves as objective and everyone else as biased. No wonder we fight, because we’re each convinced we’re “right” and free from bias. As the T-shirt slogan says, “Everyone is entitled to my opinion.”

Is the self-serving bias universal, or are people in collectivistic cultures immune? People in collectivistic cultures associate themselves with positive words and valued traits (Gaertner & others, 2008; Yamaguchi & others, 2007). However, in some stud­ ies, collectivists are less likely to self-enhance by believing they are better than oth­ ers (Falk & others, 2009; Heine & Hamamura, 2007), particularly in individualistic domains (Sedikides & others, 2003, 2005).

Can We All Be Better Than Average? Self-serving bias also appears when people compare themselves with others. If the sixth-century b.c. Chinese philosopher Lao-tzu was right that “at no time in the world will a man who is sane over-reach himself, over-spend himself, over­ rate himself,” then most of us are a little insane. For on subjective, socially desirable, and common dimensions, most people see themselves as better than the average

The Self in a Social World Chapter 2 63

person. Compared with people in general, most people see themselves as more ethical, more competent at their job, friendlier, more intelligent, better looking, less prejudiced, healthier, and even more insightful and less biased in their self- assessments. (See “Focus On: Self-Serving Bias—How Do I Love Me? Let Me Count the Ways.”)

Every community, it seems, is like Garrison Keillor’s fictional Lake Wobegon, where “all the women are strong, all the men are good-looking, and all the children are above average.” Many people believe that they will become even more above average in the future—if I’m good now, I will be even better soon, they seem to think (Kanten & Teigen, 2008). One of Freud’s favorite jokes was the husband who told his wife, “If one of us should die, I think I would go live in Paris.”

The self-serving bias is also common in marriages. In a 2008 surv^ey, 49 percent of married men said they did half to most of the child care. But only 31 percent of wives said their husbands did this much. In the same survey, 70 percent of women said they do most of the cooking, but 56 percent of the men said they do most of the cooking (Galinsky & others, 2009). The general rule: Group members’ estimates of how much they contribute to a joint task typically sum to more than 100 percent (Savitsky & others, 2005).

focus Self-Serving Bias—How Do I Love Me? Let Me Count the Ways

“The one thing that unites all human beings, regard­ less of age, gender, religion, economic status, or ethnic background,” notes columnist Dave Barry (1998), “is that deep down inside, we all believe that we are above aver­ age drivers.” We also believe we are above average on most any other subjective and desirable trait Among the many faces of seif-serving bias are these:

• Ethics. Most businesspeople see themselves as more ethical than the average businessperson (Baumhart, 1968; Brenner & Molander, 1977). One national survey asked, “How would you rate your own mor­ als and values on a scale from 1 to 1(X) (100 being perfect)?” Fifty percent of people rated them­ selves 90 or above; only 11 percent said 74 or less (Lovett, 1997).

• Professional competence. In one survey, 90 percent of business managers rated their performance as supe­ rior to their average peer (French, 1968). In Australia, 86 percent of people rated their job performance as above average, and only 1 percent as below average (Headey & Wearing, 1987). Most surgeons believe their patients’ mortality rate to be lower than average (Gawande, 2002).

• Virtues. In The Netherlands, most high school stu­ dents rate themselves as more honest, persistent, original, friendly, and reliable than the average high

i school student (Hoorens, 1993, 1995).

Intelligence. Most people perceive themselves as more intelligent, better looking, and much less preju­ diced than their average peer (Public Opinion, 1984; Watt & Larkin, 2010; Wylie, 1979). When someone outperforms them, people tend to think of the other as a genius (Lassiter & Munhall, 2001). Parental support. Most adults believe they support their aging parents more than do their siblings (Lerner & others, 1991). Health. Los Angeles residents view themselves as healthier than most of their neighbors, and most col­ lege students believe they will outlive their actuarially predicted age of death by approximately 10 years (Larwood, 1978; Snyder, 1978). Attractiveness. Is it your experience, as it is mine, that most photos of you seem not to do you justice? One experiment showed people a lineup of faces— one their own, the others being their face morphed into those of less and more attractive faces (Epiey & Whitchurch, 2008). When asked which was their actual face, people tended to identify an attractively enhanced version of their face. Driving. Most drivers—even most drivers who have been hospitalized for accidents—believe themselves to be safer and more skilled than the average driver (Guerin, 1994; McKenna & Myers, 1997; Svenson, 1981). Dave Barry was right.

m •r»

64 Part One Social Thinking

® Jean Sorensen.

My wife and I used to pitch our laun­ dry on the floor next to our bedroom clothes hamper. In the morning, one of us would put it in. When she sug­ gested that 1 take more responsibility for this, I thought, “Huh? I already do it 75 percent of the time.” So I asked her how often she thought she picked up the clothes. “Oh,” she replied, “about 75 percent of the time.”

Within commonly considered domains, subjective behavioral dimen­ sions (such as “disciplined”) trigger even greater self-serving bias than observable behavioral dimensions (such as “punctual”). Subjective quali­ ties give us leeway in constructing our own definition of success (Dunning & others, 1989,1991). Rating my “athletic ability,” I ponder my basketball play, not the agonizing weeks 1 spent as a Little League baseball player hiding in right field. Assessing my “leader­

ship ability,” 1 conjure up an image of a great leader whose style is similar to mine. By defining ambiguous criteria in our own terms, each of us can see our­ selves as relatively successful. In one College Entrance Examination Board survey of 829,000 high school seniors, none rated themselves below average in “ability to get along with others” (a subjective, desirable trait), 60 percent rated themselves in the top 10 percent, and 25 percent saw themselves among the top 1 percent! In 2011, 77 percent of incoming college students described themselves as above average in their “drive to achieve,” another subjective and desirable trait (Pryor & others, 2012).

Researchers have wondered: Do people really believe their above-average self-estimates? Is their self-serving bias partly a function of how the ques­ tions are phrased (Krizan & Suls, 2008)? When Elanor Williams and Thomas Gilovich (2008) had people bet real money when estimating their relative perfor­ mance on tests, they found that, yes, “people truly believe their self-enhancing self-assessments.”





Unrealistic Optimism Optimism predisposes a positive approach to life. “The optimist,” notes H. Jackson Brown (1990, p. 79), “goes to the window every morning and says, ‘Good morning, God.’ The pessimist goes to the window and says, ‘Good God, morning.'”

Studies of more than 90,000 people across 22 cultures reveal that most humans are more disposed to optimism than pessimism (Fischer & Chalmers, 2008). Indeed, many of us have what researcher Neil Weinstein (1980,1982) terms “an unrealistic optimism about future life events.” In a 2006-2008 worldwide poll, most people expected their lives to improve more in the next 5 years than they did in the past 5 years (Deaton, 2009)—an especially striking expectation considering the world­ wide recession that followed. Partly because of their relative pessimism about oth­ ers’ fates (Hoorens & others, 2008; Shepperd, 2003), students perceive themselves as far more likely than their classmates to get a good job, draw a good salary, and own a home. They also see themselves as far less likely to experience nega­ tive events, such as developing a drinking problem, having a heart attack before age 40 years, or being fired. Adult women are much more likely to be unduly

65The Self in a Social World

optimistic than pessimistic about their relative risk of breast cancer (Waters & others, 2011). Football fans believe their favorite team has a 77 percent chance of winning their first game. Even after 4 months when (on average) their team won only half the time, they still hold out hope and predict a 70 percent chance of their team winning (Massey & others, 2011).

Parents extend their unrealistic optimism to their children, assuming their child is less likely to drop out of college, become depressed, or get lung can­ cer than the average child. According to a study (Lench & others, 2006), par­ ents assumed their children would be more likely to complete college, remain healthy, and stay happy.

Illusory optimism increases our vulnerability. Believing ourselves immune to misfortune, we do not take sensible precautions. Sexually active undergraduate women who don’t consistently use contraceptives perceived themselves, com­ pared with other women at their university, as much less vulnerable to unwanted pregnancy (Burger & Burns, 1988). People trying to quit smoking who believe they are above average in willpower are more likely to keep cigarettes around and stand near others who are smoking—^behaviors likely to lead to a relapse into smoking (Nordgren & others, 2009). Elderly drivers who rated themselves as “above aver­ age” were four times more likely than more modest drivers to flunk a driving test and be rated “unsafe” (Freund & others, 2005). Students who enter university with inflated assessments of their academic ability often suffer deflating self-esteem and well-being and are more likely to drop out (Robins & Beer, 2001). In perhaps the most wide-ranging example, many home buyers, mortgage lenders, and inves­ tors in the mid-2000s displayed unrealistic optimism in their belief that “housing never goes down,” accumulating large amounts of debt. The eventual result was a wave of home foreclosures that spawned the 2007-2009 recession, the most severe economic downturn since the Great Depression. For illusory optimism, we often pay a price.

Those who cheerfully deny the effects of smoking or stumble into ill-fated rela­ tionships remind us that blind optimism, like pride, may go before a fall. When gambling, optimists persist longer than pessimists, even when piling up losses (Gibson & Sanbonmatsu, 2004). If those who deal in the stock market or in real estate perceive their business intuition as superior to that of their competitors, they, too, may be in for disappointment. Even the seventeenth-century economist Adam Smith, a defender of human economic rationality, foresaw that people would over­ estimate their chances of gain. This “absurd presumption in their own good for­ tune,” he said, arises from “the overweening conceit which the greater part of men have of their own abilities” (Spiegel, 1971, p. 243).

Unrealistic optimism appears to be on the rise. In the 1970s, half of American high school seniors predicted that they would be “very good” workers as adults—the highest rating available, and thus the equivalent of giving themselves five stars out of five. By 2006, two-thirds of teens believed they would achieve this stellar out­ come (Twenge & Campbell, 2008). Even more striking, half of high school seniors in 2000 believed that they would earn a graduate degree—even though only 9 percent

Chapter 2

NON SEQUITUR © 1999 Wiley Miller. Distributed by Universal Press Syndicate. Reprinted with permission. All rights reserved.










66 Part One Social Thinking

Illusory optimism: Most couples marry feeling confident of long-term love. Actually, in individualistic cultures, half of marriages fail.

were likely to actually do so (Reynolds & others, 2006). Although aiming high has benefits for success, those who aim too high may struggle with depression as they learn to adjust their goals to more realistic heights (Wrosch & Miller, 2009).

Optimism definitely beats pessimism in promoting self-efficacy, health, and well-being (Armor & Taylor, 1996; Segerstrom, 2001). As natural optimists, most people believe they will be hap­ pier with their lives in the future—a belief that surely helps create happiness in the present (Robinson & Ryff, 1999). If our optimistic prehistoric ancestors were more likely than their pessimistic neighbors to surmount challenges and survive, then small wonder that we are disposed to optimism (Haselton &

defensive pessimism The adaptive value of anticipating problems and harnessing one’s anxiety to motivate effective action.

Nettle, 2006). Yet a dash of realism—or what Julie Norem (2000) calls defensive pessimism—

can sometimes save us from the perils of unrealistic optimism. Defensive pessi­ mism anticipates problems and motivates effective coping. As a Chinese proverb says, “Be prepared for danger while staying in peace.” Students who exhibit excess optimism (as many students destined for low grades do) benefit from some self-doubt, which motivates study (Prohaska, 1994; Sparrell & Shrauger, 1984). Students who are overconfident tend to underprepare, whereas their equally able but less confident peers study harder and get higher grades (Good- hart, 1986; Norem & Cantor, 1986; Showers & Ruben, 1987). Viewing things in a more immediate, realistic way often helps. Students in one experiment were wildly optimistic in predicting their test performance when the test was hypo­ thetical, but they were surprisingly accurate when the test was imminent (Armor & Sackett, 2006). Believing you’re great when nothing can prove you wrong is one thing, but with an evaluation fast approaching, best not to look like a bragging fool.

It’s also important to listen to criticism. “One gentle rule I often tell my students,” writes David Dunning (2006), “is that if two people independently give them the same piece of negative feedback, they should at least consider the possibility that it might be true.” So, there is a power to negative as well as positive thinking. The moral: Success in school and beyond requires enough optimism to sustain hope and enough pessimism to motivate concern.

false consensus effect The tendency to overestimate the commonality of one’s opinions and one’s undesirable or unsuccessful behaviors.

False Consensus and Uniqueness We have a curious tendency to enhance our self-images by overestimating or underestimating the extent to which others think and act as we do. On mat­ ters of opinion, we find support for our positions by overestimating the extent to which others agree—a phenomenon called the false consensus effect (Krueger & Clement, 1994b; Marks & Miller, 1987; Mullen & Goethals, 1990). Sharad Goel, Win­ ter Mason, and Duncan Watts (2010) found that Facebook users were 90 percent accurate in guessing when they agreed with their friends on political and other issues, but they were only 41 percent accurate in guessing disagreement. In other words, most of the time they thought their friends agreed with them when they didn’t. Business students asked to make decisions about ethical dilemmas overesti­ mated how many other students made the same choice (Flynn & Wiltermuth, 2010).

The Self in a Social World

Self-serving bias Example

Attributing one’s success to ability and effort, failure to luck and things external

Comparing oneself favorably to others

got the A in history because I studied hard. I got the D in sociology becau^j exams were unfair.

I do more for my parents than my sister does.

Unrealistic optimism Even though 50% of marriages fail, I know mine will be enduring joy.

I i

Ealse consensus and uniquene know most people agree with me that global warming threatens our futur^K^ White Australians prejudiced against Aborigines were more likely to believe that other Whites were also prejudiced (Watt & Larkin, 2010). The sense we make of the world seems like common sense.

When we behave badly or fail in a task, we reassure ourselves by thinking that such lapses also are common. After one person lies to another, the liar begins to perceive the other person as dishonest (Sagarin & others, 1998). If we feel sexual desire toward another, we may overestimate the other’s reciprocal desire. We guess that others think and act as we do: “I lie, but doesn’t everyone?” If we cheat on our income taxes, smoke, or enhance our appearance, we are likely to overestimate the number of other people who do likewise. As former Baywatch actor David Hassel- hoff said, “I have had Botox. Everyone has!” “We don’t see things as they are,” says a proverb. “We see things as we are.”

Dawes (1990) proposes that this false consensus may occur because we general­ ize from a limited sample, which prominently includes ourselves. Lacking other information, why not “project” ourselves; why not impute our own knowledge to others and use our responses as a clue to their likely responses? Most people are in the majority; so when people assume they are in the majority they are usually right. Also, we’re more likely to spend time with people who share our attitudes and behaviors and, consequently, to judge the world from the people we know. Small wonder that Germans tend to think that the typical European looks rather German, whereas the Portuguese see Europeans as looking more Portuguese (Imhoff & others, 2011).

On matters of ability or when we behave well or successfully, however, a false uniqueness effect more often occurs (Goethals & others, 1991). We serve our self- image by seeing our talents and moral behaviors as relatively unusual. Dutch col­ lege students preferred being part of a larger group in matters of opinion such as politics (false consensus) but wanted to be part of a smaller group in matters of taste such as musical preferences (false uniqueness; Spears & others, 2009). After all, a band isn’t cool anymore if too many people like it. Female college students who protect themselves while drinking by, for example, designating a driver or drinking only with a meal underestimate how many other women do the same (Benton Sz others, 2008). Thus, we may see our failings as relatively normal and our virtues as relatively exceptional.

To sum up, self-serving bias appears as self-serving attributions, self-congratulatory comparisons, illusory optimism, and false consensus for one’s failings (Figure 2.8).

FIGURE:: 2.8 How Seif-Serving Bias Works

Chapter 2 67








TALBERT, 1997)

false uniqueness effect The tendency to underestimate the commonality of one’s abilities and one’s desirable or successful behaviors.

68 Part One Social Thinking

Explaining Self-Serving Bias Why do people perceive themselves in self-enhancing ways? One explanation sees the self-serving bias as a by­ product of how we process and remember information about ourselves. Comparing ourselves with others requires us to notice, assess, and recall their behavior and ours. Thus, there are multiple opportunities for flaws in our information processing (Chambers & Windschitl, 2004). In one study, married people gave themselves credit for doing more housework than their spouses did. Might that not be due, as Michael Ross and Fiore Sicoly (1979) believed, to our greater recall for what weVe actively done and our lesser recall for what we’ve not done or merely observed our partner doing? I could easily picture myself picking up the laundry off the bedroom floor, but I was less aware of

the times when I absentmindedly overlooked it. Are the biased perceptions, then, simply a perceptual error, an emotion-free

glitch in how we process information? Or are self-serving motives also involved? It’s now clear from research that we have multiple motives. Questing for self- knowledge, we’re motivated to assess our competence (Dunning, 1995). Questing for self-confirmation, we’re motivated to verify our self-conceptions (Sanitioso & others, 1990; Swann, 1996,1997). Questing for self-affirmation, we’re especially motivated to enhance our self-image (Sedikides, 1993). Self-esteem motivation, then, helps power our self-serving bias. As social psychologist Daniel Batson (2006) surmises, “The head is an extension of the heart.”

Can we all be better than average? © William Haeleli/The New Yorker Collection/www

Reflections on Self-Esteem and Self-Serving Bias If you are like some readers, by now you are finding self-serving bias either depress­ ing or contrary to your own occasional feelings of inadequacy. Even people who exhibit the self-serving bias may feel inferior—to those who are a step or two higher on the ladder of success, attractiveness, or skill. Moreover, not everyone operates with a self-serving bias. Some people do suffer from low self-esteem. Positive self­ esteem does have some benefits.

THE SELF-SERVING BIAS AS ADAPTIVE Self-esteem has its dark side, but also its bright side. When good things happen, people with high self-esteem are more likely to savor and sustain the good feel­

ings (Wood & others, 2003). “Believing one has more talents and positive qualities than one’s peers allows one to feel good about oneself and to enter the stressful circumstances of daily life with the resources conferred by a positive sense of self,” noted Shelley Taylor and co­ researchers (2003b).

Self-serving bias and its accompanying excuses also help protect people from depres­ sion (Snyder & Higgins, 1988; Taylor & others, 2003a). Nonde- pressed people usually exhibit self-serving bias. They excuse

© Mike Twohy/The New Yorker Collection/

69The Self in a Social World

their failures on laboratory tasks or perceive themselves as being more in control than they are. Depressed people’s self-appraisals and their appraisals of how others really view them are not inflated (more on this in Chapter 14).

Self-serving bias additionally helps buffer stress. George Bonanno and colleagues (2005) assessed the emotional resiliency of workers who escaped from the World Trade Center or its environs on September 11,2001. They foimd that those who dis­ played self-enhancing tendencies were the most resilient.

In their “terror management theory,” Jeff Greenberg, Sheldon Solomon, and Tom Pyszczynski (1997; Greenberg, 2008) propose another reason why positive self-esteem is adaptive: It buffers anxiety, including anxiety related to our certain death. In childhood, we learn that when we meet the standards taught us by our parents, we are loved and protected; when we don’t, love and protection may be withdrawn. We therefore come to associate viewing ourselves as good with feeling secure. Greenberg and colleagues argue that positive self-esteem—viewing oneself as good and secure—even protects us from feeling terror over our eventual death. Their research shows that reminding people of their mortality (for example, by writing a short essay on dying) motivates them to affirm their self-worth. When fac­ ing such threats, self-esteem buffers anxiety. In 2004, a year after the U.S. invasion, Iraqi teens who felt their country was under threat reported the highest self-esteem (Carlton-Ford & others, 2008).

As research on depression and anxiety suggests, there is practical wisdom in self-serving perceptions. It may be strategic to believe we are smarter, stronger, and more socially successful than we are. Cheaters may give a more convincing display of honesty if they believe themselves honorable. Belief in our superiority can also motivate us to achieve—creating a self-fulfilling prophecy—and can sustain our hope through difficult times (Willard & Gramzow, 2009).

THE SELF-SERVING BIAS AS MALADAPTIVE Although self-serving pride may help protect us from depression, it can also be maladaptive. People who blame others for their social difficulties are often unhap- pier than people who can acknowledge their mistakes (Anderson & others, 1983; Newman & Langer, 1981; Peterson & others, 1981).

Research by Barry Schlenker (1976; Schlenker & Miller, 1977a, 1977b) has also shown how self-serving perceptions can poison a group. As a rock band guitar­ ist during his college days, Schlenker noted that “rock band members typically overestimated their contributions to a group’s success and underestimated their contributions to failure. I saw many good bands disintegrate from the problems caused by these self-glorifying tendencies.” In his later life as a University of Flor­ ida social psychologist, Schlenker explored group members’ self-serving percep­ tions. In nine experiments, he had people work together on some task. He then falsely informed them that their group had done either well or poorly. In every one of those studies, the members of successful groups claimed more responsibil­ ity for their group’s performance than did members of groups that supposedly failed at the task.

If most group members believe they are underpaid and underappreciated rela­ tive to their better-than-average contributions, disharmony and envy are likely. College presidents and academic deans will readily recognize the phenomenon. Ninety percent or more of college faculty members have rated themselves as superior to their average colleague (Blackburn & others, 1980; Cross, 1977). It is therefore inevitable that when merit salary raises are announced and half receive an average raise or less, many will feel themselves victims of injustice.

Self-serving biases also inflate people’s judgments of their groups, a phenom­ enon called group-serving bias. When groups are comparable, most people con­ sider their own group superior (Codol, 1976; Jourden & Heath, 1996; Taylor & Doria, 1981).

Chapter 2








—SENECA, DE IRA. a.d. 43

group-serving bias Explaining away outgroup members’ positive behaviors; also attributing negative behaviors to their dispositions (while excusing such behavior by one’s own group).

70 Part One Social Thinking

Seif-sen/ing pride in group settings can become especially dangerous. © Dana Fradon/The New Yorker Colleclion/wvw








• Most university sorority members perceive those in their sorority as far less likely to be conceited and snobbish than those in other sororities (Biemat & others, 1996).

• Stanford University intramural volleyball players who won their game attributed the success to their team; those who lost blamed other factors (Sherman & Kim, 2005).

• Fifty-three percent of Dutch adults rate their marriage or partnership as better than that of most others; only 1 percent rate it as worse than most (Buunk & van der Eijnden, 1997).

• Sixty-six percent of Americans give their oldest child’s public school a grade of A or B. But nearly as many—64 percent—give the nation’s public schools a grade of C or D (Whitman, 1996).

That people see themselves and their groups with a favorable bias is hardly new. The tragic flaw portrayed in ancient Greek drama was hubris, or pride. Like the sub­ jects of our experiments, the Greek tragic figures were not self-consciously evil; they merely thought too highly of themselves. In literature, the pitfalls of pride are por­ trayed again and again. In theology, pride has long been first among the “seven deadly sins.”

If pride is akin to the self-serving bias, then what is humility? Is it self-contempt? Humility is not handsome people believing they are ugly and smart people trying to believe they are slow-witted. False modesty can actually be a cover for pride in one’s better-than-average humility. (James Friedrich [1996] reports that most stu­ dents congratulate themselves on being better than average at not thinking them­ selves better than average!) True humility is more like self-forgetfulness than false modesty. It leaves us free to recognize accurately and rejoice in our special talents and, with the same honesty, to recognize the talents of others.

SUMMING UP: What Is Self-Serving Bias? • Contrary to the presumption that most people

suffer from low self-esteem or feelings of inferior­ ity, researchers consistently find that most people exhibit a self-serving bias. In experiments and every­ day life, we often take credit for our successes while blaming failures on the situation.

• Most people rate themselves as better than average on subjective, desirable traits and abilities.

• We exhibit unrealistic optimism about our futures.

• We overestimate the commonality of our opinions and foibles {false consensus) while underestimating

the commonality of our abilities and virtues {false uniqueness).

• Such perceptions arise partly from a motive to maintain and enhance self-esteem—a motive that protects people from depression but contributes to misjudgment and group conflict.

• Self-serving bias can be adaptive in that it allows us to savor the good things that happen in our lives. When bad things happen, however, self-serving bias can have the maladaptive effect of causing us to blame others or feel cheated out of something we “deserved.”

71The Self in a Social World


Identify self-presentation and understand how impression management can expiain behavior.

So far, we have seen that the self is at the center of our social worlds, that self­ esteem and self-efficacy pay some dividends, and that self-serving bias influences self-evaluations. Perhaps you have wondered: Are self-enhancing expressions always sincere? Do people have the same feelings privately as they express pub­ licly? Or are they just putting on a positive face even while living with self-doubt?

Self-Handicapping Sometimes people sabotage their chances for success by creating impediments that make success less likely. Far from being deliberately self-destructive, such behav­ iors typically have a self-protective aim (Arkin & others, 1986; Baumeister & Scher, 1988; ^odewalt, 1987): “Tm really not a failure—I would have done well except for this problem.”

Why would people handicap themselves with self-defeating behavior? Recall that we eagerly protect our self-images by attributing failures to external factors. Can you see why, fearing failure, people might handicap themselves by partying half the night before a job interview or playing video games instead of studying before a big exam? When self-image is tied up with performance, it can be more self-deflating to try hard and fail than to procrastinate and have a ready excuse. If we fail while handicapped in some way, we can cling to a sense of competence; if we succeed under such conditions, it can only boost our self-image. Handicaps protect both self-esteem and public image by allowing us to attribute failures to something temporary or external (“I was feeling sick”; “I was out too late the night before”) rather than to lack of talent or ability.

Steven Berglas and Edward Jones (1978) confirmed this analysis of self­ handicapping. One experiment was announced as concerning “drugs and intel­ lectual performance.” Imagine yourself in the position of their Duke University participants. You guess answers to some difficult aptitude questions and then are told, “Yours was one of the best scores seen to date!” Feeling incredibly lucky, you are then offered a choice between two drugs before answering more of these items. One drug will aid intellectual performance and the other will inhibit it. Which drug do you want? Most students wanted the drug that would suppos­ edly disrupt their thinking, thus providing a handy excuse for anticipated poorer performance.

Researchers have documented other ways people self-handicap. Fearing failure, people will

• reduce their preparation for important individual athletic events (Rhodewalt & others, 1984).

• give their opponent an advantage (Shepperd & Arkin, 1991). • perform poorly at the beginning of a task in order not to create unreachable

expectations (Baumgardner & Brownlee, 1987). • not try as hard as they could during a tough, ego-involving task (Hormuth,

1986; Pyszczynski & Greenberg, 1987; Riggs, 1992; Turner & Pratkanis, 1993).

Impression Management Self-serving bias, false modesty, and self-handicapping reveal the depth of our con­ cern for self-image. To varying degrees, we are continually managing the impressions

Chapter 2

self-handicapping Protecting one’s self-image with behaviors that create a handy excuse for later failure.




After losing to some younger rivals, tennis great Martina Navratilova confessed that she was “afraid to play my best…. I was scared to find out if they could beat me when I’m playing my best because if they can, then 1 am finished” (Frankel & Snyder 1987).

72 Part One

self-presentation The act of expressing oneself and behaving in ways designed to create a favorable impression or an impression that corresponds to one’s ideals.

©2008 by P. S. Mueller.

Social Thinking

we create. Whether we wish to impress, intimidate, or seem helpless, we are social animals, playing to an audience. So great is the human desire for social acceptance that it can lead people to risk harming themselves through smoking, binge eating, pre­ mature sex, or drug and alcohol abuse {Rawn & Vohs, 2011).

Self-presentation refers to our wanting to present a desired image both to an external audience (other people) and to an internal audience (ourselves). We work at managing the impressions we create. We excuse, jus­ tify, or apologize as necessary to shore up our self-esteem and verify our self- images (Schlenker & Weigold, 1992). Just as we preserve our self-esteem, we also must make sure not to brag too much and risk the disapproval of oth­ ers (Anderson & others, 2006). Social interaction is a careful balance of look­ ing good while not looking too good.

In familiar situations, self-presentation happens without conscious effort. In unfamiliar situations, perhaps at a party with people we would like to impress or in conversation with someone we have romantic interest in, we are acutely self- conscious of the impressions we are creating and we are therefore less modest than when among friends who know us well (Leary & others, 1994; Tice & others, 1995). Preparing to have our photographs taken, we may even try out different faces in a mirror. We do this even though active self-presentation depletes energy, which often leads to diminished effectiveness—for example, to less persistence on a tedious experimental task or more difficulty stifling emotional expressions (Vohs & others, 2005). The upside is that self-presentation can unexpectedly improve mood. People felt significantly better than they thought they would after doing their best to “put their best face forward” and concentrate on making a positive impression on their boyfriend or girlfriend. Elizabeth Dunn and colleagues (2008) conclude that “date nights” for long-term couples work because they encourage active self-presentation, which improves mood.

Social networking sites such as Facebook provide a new and sometimes intense venue for self-presentation. They are, says communications professor Joseph Wal- ther, “like impression management on steroids” (Rosenbloom, 2008). Users make careful decisions about which pictures, activities, and interests to highlight in their profiles. Some even think about how their friends will affect the impression they make on others; one study found that those with more attractive friends were per­ ceived as more attractive themselves (Walther & others, 2008). Given the concern with status and attractiveness on social networking sites, it is not surprising that people high in narcissistic traits thrive on Facebook, tallying up more friends and choosing more attractive pictures of themselves (Buffardi & Campbell, 2008).

Given our concern for self-presentation, it’s no wonder that people will self­ handicap when failure might make them look bad. It’s no wonder that people take health risks—tanning their skin with wrinkle- and cancer-causing radiation; having piercings or tattoos done without proper hygiene; becoming anorexic; or yielding to peer pressures to smoke, get drunk, and do drugs (Leary & others, 1994). It’s no wonder that people express more modesty when their self-flattery is vulnerable to being debunked, perhaps by experts who will be scrutinizing

73The Self in a Social World

their self-evaluations (Arkin & others, 1980; Riess & others, 1981; Weary & others, 1982). Professor Smith will likely express more modesty about the significance of her work when presenting it to professional colleagues than when presenting it to students.

For some people, conscious self-presentation is a way of life. They continually monitor their own behavior and note how others react, then adjust their social per­ formance to gain a desired effect. Those who score high on a scale of self-monitoring (who, for example, agree that “1 tend to be what people expect me to be”) act like social chameleons—they adjust their behavior in response to external situations (Gangestad & Snyder, 2000; Snyder, 1987). Having attuned their behavior to the situation, they are more likely to espouse attitudes they don’t really hold (Zanna & Olson, 1982). Being conscious of others, they are less likely to act on their own attitudes. As Mark Leary (2004b) observed, the self they know often differs from the self they show. As social chameleons, those who score high in self-monitoring are also less committed to their relationships and more likely to be dissatisfied in their marriages (Leone & Hawkins, 2006).

Those who score low in self-monitoring care less about what others think. They are more internally guided and thus more likely to talk and act as they feel and believe (McCann & Hancock, 1983). For example, if asked to list their thoughts about gay couples, they simply express what they think, regardless of the attitudes of their anticipated audience (Klein & others, 2004). As you might imagine, someone who is extremely low in self-monitoring could come across as an insensitive boor, whereas extremely high self-monitoring could result ^ dishonest behavior worthy of a con art­ ist. Most of us fall somewhere between those hvo extremes.

Presenting oneself in ways that create a desired impression is a delicate balancing set. People want to be seen as able but also

Chapter 2

Group identity. In Asian countries, self-presentation is restrained. Children learn to identify themselves with their groups.

self-monitoring Being attuned to the way one presents oneself in social situations and adjusting one’s performance to create the desired impression.










(Si Mike Marland.

74 Part One Soda! Thinking

as modest and honest (Carlston & Shovar, 1983). In most social situations, modesty creates a good impression, unsolicited boasting a bad one. Hence the false modesty phenomenon: We often display lower self-esteem than we privately feel (Miller & Schlenker, 1985). But when we have obviously done extremely well, the insincerity of a disclaimer (‘T did well, but it’s no big deal”) may be evident. To make good impressions—to appear modest yet competent—requires social skill.

SUMMING UP: How Do People Manage Their Self-Presentation?

• As social animals, we adjust our words and actions to suit our audiences. To varying degrees, we note our performance and adjust it to create the impres­ sions we desire.

• Such tactics explain examples of false modesty, in which people put themselves down, extol future competitors, or publicly credit others while pri­ vately crediting themselves.

• Sometimes people will even self-handicap with self- defeating behaviors that protect self-esteem by providing excuses for failure.

Self-presentation refers to our wanting to present a favorable image both to an external audience (other people) and to an internal audience (our­ selves). With regard to an external audience, those who score high on a scale of self-monitoring adjust their behavior to each situation, whereas those low in self-monitoring may do so little social adjusting that they seem insensitive.

POSTSCRIPT: Twin Truths—The Perils of Pride, the Powers of Positive Thinking This chapter offered two memorable truths—the truth of self-efficacy and the truth of self-serving bias. The truth concerning self-efficacy encourages us not to resign ourselves to bad situations. We need to persist despite initial failures and to exert effort without being overly distracted by self-doubts. Secure self-esteem is likewise adaptive. When we believe in our positive possibilities, we are less vulnerable to depression and we feel less insecure.

Thus, it’s important to think positively and try hard but not to be so self- confident that our goals are illusory or we alienate others with our narcissism. Taking self-efficacy too far leads to blaming the victim: If positive thinking can accomplish anything, then we have only ourselves to blame if we are unhappily married, poor, or depressed. For shame! If only we had tried harder, been more disciplined, less stupid. This viewpoint fails to acknowledge that bad things can happen to good people. Life’s greatest achievements, but also its greatest disap­ pointments, are bom of the highest expectations.

These twin truths—self-efficacy and self-serving bias—remind me of what Pascal taught 300 years ago: No single truth is ever sufficient, because the world is complex. Any truth, separated from its complementary truth, is a half-truth.

Social Beliefs and

Judgments How do we perceive our sodj^ worlds?

How do we judge our social worlds?

How do we explain our social worlds?

How do our expectations of our social worlds matter?

What can we conclude about social beliefs and judgments?

There is curious power to partisanship. Consider American politics:

• In the late 1980s, most Democrats believed inflation had risen

under Republican president Ronald Reagan (it had dropped).

• In 2010, most Republicans believed that taxes had increased

under Barack Obama (for most Americans, taxes had decreased)

(Cooper, 2010; Douthat, 2010).

• Obama is Muslim, agreed 31 percent of Republicans and

10 percent of Democrats surveyed by Pew (2010a). He is not.

• And he was born outside the United States, said 43 percent of

Republicans and 9 percent of Democrats surveyed by Gallup

shortly before the release of his long-form Hawaiian birth

certificate (Morales, 2011a).

Such “motivated reasoning” transcends political parties. Feelings—

such as a gut-level liking or disliking of certain politicians—powerfully

influence how we interpret evidence and view reality. Partisanship pre­

disposes perceptions. As an old Chinese proverb says, “Two-thirds of

v^hat we see is behind our eves.”

Postscript: Reflecting on Illusory thinking

78 Part One Social Thinking

priming Activating particular associations in memory.

Such differing responses to public evidence, which have been replicated in politi­

cal perceptions throughout the world, illustrate the extent to which we construct

social perceptions and beliefs as we

• perceive and recall events through the filters of our own assumptions;

• judge events, informed by our intuition, by implicit rules that guide our snap

judgments, and by our moods;

• explain events by sometimes attributing them to the situation, sometimes to

the person; and

• expect certain events, thereby sometimes helping bring them about.

This chapter explores how we perceive, judge, and explain our social worlds and

how—and to what extent—our expectations matter.


Understand the extent to which our assumptions and prejudgments guide our perceptions, interpretations, and recall.

Chapter 1 noted a significant fact about the human mind: Our preconceptions guide how we perceive and interpret information. We construe the world through belief-tinted glasses. “Sure, preconceptions matter,” people will agree; yet they fail to realize how great the effect is on themselves.

Let’s consider some provocative experiments. The first group of experiments examines how predispositions and prejudgments affect how we perceive and interpret information. The second group plants a judgment in people’s minds after they have been given information to see how after-the-fact ideas bias recall. The overarching point: We respond not to reality as it is but to reality as we construe it.

Priming Unattended stimuli can subtly influence how we interpret and recall events. Imagine yourself, during an experiment, wearing earphones and concentrating on ambiguous spoken sentences such as “We stood by the bank.” When a perti­ nent word (river or money) is simultaneously sent to your other ear, you don’t con­ sciously hear it. Yet the word “primes” your interpretation of the sentence {Baars & McGovern, 1994).

Our memory system is a web of associations, and priming is the awakening or activating of certain associations. Experiments show that priming one thought, even without awareness, can influence another thought, or even an action. John Bargh and colleagues (1996) asked people to complete a written sentence contain­ ing words such as “old,” “wise,” and “retired.” Shortly afterward, they observed these people walking more slowly to the elevator than did those not primed with aging-related words. Moreover, the slow walkers had no awareness of their walk­ ing speed or of having just viewed words that primed aging.

Social Beliefs and Judgments

Often our thinking and acting are sub­ tly primed by unnoticed events. Rob Hol­ land and colleagues (2005) observed that Dutch students exposed to the scent of an all-purpose cleaner were quicker to iden­ tify cleaning-related words. In follow-up experiments, other students exposed to a cleaning scent recalled more cleaning- related activities when describing their day’s activities and even kept their desk cleaner while eating a crumbly cookie. Moreover, all these effects occurred with­ out the participants’ conscious aware­ ness of the scent and its influence.

Priming experiments (Bargh, 2006) have their counterparts in everyday life:

• Watching a scary movie alone at home can activate emotions that, without our realizing it, cause us to interpret furnace noises as a possible intruder.

• Depressed moods, as this chapter explains later, prime negative associations. Put people in a good mood and suddenly their past seems more wonderful, their future brighter.

• Watching violence will prime people to interpret ambiguous actions (a shove) and words (“punch”) as aggressive.

• For many psychology students, reading about psychological disorders primes how they interpret their own anxieties and gloomy moods. Reading about disease symptoms similarly primes medical students to worry about their congestion, fever, or headache.

In a host of studies, priming effects surface even when the stimuli are presented subliminally—too briefly to be perceived consciously. What’s out of sight may not be completely out of mind. An electric shock that is too slight to be felt may increase the perceived intensity of a later shock. An imperceptibly flashed word, “bread,” may prime people to detect a related word, such as “butter,” more quickly than they detect an unrelated word, such as “bottle” or “bubble.” A subliminal color name facilitates speedier identification when the color appears on the computer screen, whereas an unseen wrong name delays color identification (Epley & others, 1999; Merikle & others, 2001). In each case, an invisible image or word primes a response to a later task.

Studies of how implanted ideas and images can prime our interpretations and recall illustrate one of this book’s take-home lessons: Much of our social information processing is automatic. It is unintentional, out of sight, and happens without our conscious awareness.

Even physical sensations, thanks to our embodied cognition, prime our social judgments and vice versa. After holding a warm drink, people become more likely to rate someone more warmly and behave more generously (Ijzerman & Semin, 2009; Williams & Bargh, 2008). After receiving a cold shoulder treatment, people judge the experimental room as colder than do those treated warmly (Zhong & Leonardelli, 2008). Physical warmth accentuates social warmth, and social exclu­ sion literally feels cold.

Perceiving and Interpreting Events I^spite some startling and oft-confirmed biases and logical flaws in how we per­ ceive and understand one another, we’re mostly accurate (Jussim, 2005). Our first

Chapter 3 79

Posting the second sign may prime customers to be dissatisfied with the handling of their complaints at the first window.

embodied cognition The mutual influence of bodily sensations on cognitive preferences and social judgments.

80 Part One Social Thinking

FIGURE:: 3.1 Pro-Israeli and pro-Arab students who viewed network news descriptions of the “Beirut massacre” believed the coverage was biased against their point of view.

Source: Data from Vallone & others (1985).

Perception of media bias

Pro-Israel 9 i———- — Members of each side perceived bias against their view



Anti-Israel Pro-Israeli Pro-Arab students students









impressions of one another are more often right than wrong. Moreover, the better we know people, the more accurately we can read their minds and feelings.

But on occasion, our prejudgments err. The effects of prejudgments and expecta­ tions are standard fare for psychology’s introductory course. Consider this phrase:



Did you notice anything wrong with it? There is more to perception than meets the eye.

POLITICAL PERCEPTIONS The same is true of social perception. Because social perceptions are very much in the eye of the beholder, even a simple stimulus may strike two people quite differently. Saying Britain’s David Cameron is “an okay prime minister” may sound like a put-down to one of his ardent admirers and like praise to someone who regards him with contempt. When social information is sub­ ject to multiple interpretations, preconceptions matter (Hilton & von Hippel, 1990).

An experiment by Robert Vallone, Lee Ross, and Mark Lepper (1985) reveals just how powerful preconceptions can be. They showed pro-Israeli and pro-Arab students six network news segments describing the 1982 killing of civilian refugees at two camps in Beirut, Lebanon. As Figure 3.1 illustrates, each group perceived the networks as hostile to its side.

The phenomenon is commonplace: Sports fans perceive referees as partial to the other side. Political candidates and their supporters nearly always view the news media as unsympathetic to their cause (Richardson & others, 2008).

It’s not just fans and politicians. People everywhere perceive mediators and media as biased against their position. “There is no subject about which people are less objective than objectivity,” noted one media commentator (Poniewozik, 2003). Indeed, people’s perceptions of bias can be used to assess their attitudes (Saucier & Miller, 2003). Tell me where you see bias, and you will signal your attitudes.

Our assumptions about the world can even make contradictory evidence seem supportive. For example, Ross and Lepper assisted Charles Lord (1979) in asking two groups of students to evaluate the results of two supposedly new research studies. Half the students favored capital punishment and half opposed it. Of the studies they

Social Beliefs and Judgments Chapter 3 81

evaluated, one confirmed and the other disconfirmed the students’ beliefs about the deterrent effect of the death penalty. The results: Both proponents and oppo­ nents of capital punishment readily accepted evidence that confirmed their belief but were sharply critical of dis- confirming evidence. Showing the two sides an identical body of mixed evi­ dence had not lessened their disagree­ ment but increased it. Likewise, when Anthony Bastardi and co-researchers (2011) showed people mixed evidence about the effects of day care on chil­ dren, those planning to use day care found the evidence more supportive of their plans.

Is that why, in politics, religion, and science, ambiguous information often fuels conflict? Presidential debates in the United States have mostly rein­ forced predebate opinions. By nearly a 10-to-l margin, those who already favored one candidate or the other perceived their candidate as having won (Kinder & Sears, 1985). Thus, report Geoffrey Munro and colleagues (1997), people on both sides may become even more supportive of their respective candidates after viewing a presidential debate.

Td like your honest, unbiased and possibly career-ending opinion on something. ”

Some circumstances make It difficult to be unbiased. © Alex Gfegory/The New Yorker Collection/www.cartoonbankxom

OUR PERCEPTIONS OF OTHERS In addition to these studies of people’s pre­ existing social and political attitudes, researchers have manipulated people’s preconceptions—with astonishing effects upon their interpretations and recollections.

Myron Rothbart and Pamela Birrell (1977) had University of Oregon students assess the facial expression of a man (Figure 3.2). Those told he was a Gestapo leader responsible for barbaric medical experiments on concentration camp inmates intui­ tively judged his expression as cruel. (Can you see that barely suppressed sneer?) Those told he was a leader in the anti-Nazi underground movement whose courage saved thousands of Jewish lives judged his facial expression as warm and kind. (Just look at those caring eyes and that almost smiling mouth.)

Filmmakers control people’s perceptions of emotion by manipulating the set­ ting in which they see a face. They call this the “Kulechov effect” after a Rus­ sian film director who would skillfully guide viewers’ inferences by manipulat­ ing their assumptions. Kulechov dem­ onstrated the phenomenon by creating three short films that presented identi­ cal footage of the face of an actor with a neutral expression after viewers had first been shown one of three different scenes: a dead woman, a bowl of soup, or a girl playing. As a result, in the first film the actor seemed sad, in the second thought­ ful, and in the third happy.

OTHERS’ PERCEPTIONS OF US Construal processes also color others’ per­ ceptions of us. When we say something good or bad about another, people spon­ taneously tend to associate that trait with





CRESSIDA, 1601-1602

FIGURE :: 3.2 Judge for yourself: Is this person’s expression cruel or kind? If told he was a Nazi, would your reading of his face differ?

82 Part One Social Thinking

us, report Lynda Mae, Donal Carlston, and John Skowronski (1999; Carlston & Skowronski, 2005)—a phenomenon they call spontaneous trait transference. If we go around talking about others being gos- sipy, people may then unconsciously associate “gossip” with us. Call some­ one a jerk and folks may later construe you as one. Describe someone as sensi­ tive, loving, and compassionate, and you may seem more so. There is, it appears, intuitive wisdom in the childhood taunt, “I’m rubber, you’re glue; what you say bounces off me and sticks to you.”

The bottom line: We view our social worlds through the spectacles of our beliefs, attitudes, and values. That is one reason our beliefs are so important; they shape our interpretation of every­ thing else.

Belief Perseverance Partisan perceptions. Supporters media as favoring the other side.

belief perseverance Persistence of one’s initial conceptions, such as when the basis for one’s belief is discredited but an explanation of why the belief might be true survives.

Imagine a grandparent who decides, dur- of a particular candidate or cause tend to see the evening with a crying infant, that

bottle feeding produces colicky babies: “Come to think of it, cow’s milk obviously suits calves better than babies.” If the infant turns out to be suffering a high fever, will the sitter nevertheless persist in believing that bottle feeding causes colic (Ross & Anderson, 1982)? To find out, Lee Ross, Craig Anderson, and colleagues planted a falsehood in people’s minds and then tried to discredit it.

Their research reveals that it is surprisingly difficult to demolish a falsehood, once the person conjures up a rationale for it. Each experiment first implanted a belief, either by proclaiming it to be true or by showing the participants some anec­ dotal evidence. Then the participants were asked to explain why it is true. Finally, the researchers totally discredited the initial information by telling the participants the truth: The information was manufactured for the experiment, and half the par­ ticipants in the experiment had received opposite information. Nevertheless, the new belief survived approximately 75 percent intact, presumably because the par­ ticipants still retained their invented explanations for the belief. This phenomenon, called belief perseverance, shows that beliefs can grow their own legs and survive the discrediting of the evidence that inspired them.

An example: Anderson, Lepper, and Ross (1980) asked participants to decide whether individuals who take risks make good or bad firefighters. One group con­ sidered a risk-prone person who was a successful firefighter and a cautious person who was unsuccessful. The other group considered cases suggesting the opposite conclusion. After forming their theory that risk-prone people make better or worse firefighters, the participants wrote explanations for it—for example, that risk-prone people are brave or that cautious people have fewer accidents. Once each explana­ tion was formed, it could exist independently of the information that initially cre­ ated the belief. When that information was discredited, the participants still held their self-generated explanations and therefore continued to believe that risk-prone people really do make better or worse firefighters.

These experiments suggest that the more we examine our theories and explain how they might be true, the more closed we become to information that challenges our beliefs. Once we consider why an accused person might be guilty, why an

83Social Beliefs and Judgments

offending stranger acts that way, or why a favored stock might rise in value, our explanations may survive challenges (Davies, 1997; Jelalian & Miller, 1984).

The evidence is compelling: Our beliefs and expectations powerfully affect how we mentally construct events. Usually, we benefit from our preconceptions, just as scientists benefit from creating theories that guide them in noticing and interpret­ ing events. But the benefits sometimes entail a cost: We become prisoners of our own thought patterns. Thus, the supposed Martian “canals” that twentieth-century astronomers delighted in spotting (in fact, just dust or craters) turned out to be the product of intelligent life—an intelligence on Earth’s side of the telescope.

Is there a remedy for belief perseverance? There is: Explain the opposite. Charles Lord, Mark Lepper, and Elizabeth Preston (1984) repeated the capital punishment study described previously and added two variations. First, they asked some of their participants, when evaluating the evidence, to be “as objective and unbiased as possible.” T^at instruction accomplished nothing; whether for or against capital punishment, those who had received the plea made evaluations as biased as those who had not.

The researchers asked a third group to consider the opposite—to ask themselves “whether you would have made the same high or low evaluations had exactly the same study produced results on the other side of the issue.” After imagining an opposite finding, these people were much less biased in their evaluations of the evidence for and against their views. In his experiments, Craig Anderson (1982; Anderson & Sechler, 1986) consistently found that explaining why an opposite theory might be true—why a cautious rather than a risk-taking person might be a better firefighter—reduces or eliminates belief perseverance. Indeed, explaining any alternative outcome, not just the opposite, drives people to ponder various pos­ sibilities (Hirt & Markman, 1995).

Constructing Memories of Ourselves and Our Worlds Do you agree or disagree with this statement?

Memory can be likened to a storage chest in the brain into which we deposit material and from which we can withdraw it later if needed. Occasionally, something is lost from the “chest,” and then we say we have forgotten.

Approximately 85 percent of college students said they agreed (Lamal, 1979). As one magazine ad put it, “Science has proven the accumulated experience of a life­ time is preserved perfectly in your mind.”

Actually, psychological research has proved the opposite. Our memories are not exact copies of experiences that remain on deposit in a memory bank. Rather, we construct memories at the time of withdrawal. Like a paleontologist inferring the appearance of a dinosaur from bone fragments, we reconstruct our distant past by using our current feelings and expectations to combine information fragments. Thus, we can easily (although unconsciously) revise our memories to suit our cur­ rent knowledge. When one of my sons complained, “The June issue of Cricket never came,” and was then shown where it was, he delightedly responded, “Oh good, I knew Td gotten it.”

When an experimenter or a therapist manipulates people’s presumptions about their past, a sizable percentage of people will construct false memories. Asked to imagine a made-up childhood experience in which they ran, tripped, fell, and stuck their hand through a window, or knocked over a punch bowl at a wedding, approx­ imately one-fourth will later recall the fictitious event as something that actually happened (Loftus & Bernstein, 2005). In its search for truth, the mind sometimes constructs a falsehood.

In experiments involving more than 20,000 people, Elizabeth Loftus (2003, 2007, 2011a) and collaborators have explored our mind’s tendency to construct memories.

Chapter 3










84 Part One Social Thinking

misinformation effect Incorporating “misinformation” into one’s memory of the event after witnessing an event and receiving misleading information about it.










In the typical experiment, people witness an event, receive misleading information about it (or not), and then take a memory test. The repeated finding is the misinformation effect. People incorporate the misinformation into their memo­ ries: They recall a yield sign as a stop sign, hammers as screwdrivers. Vogue maga­ zine as Mademoiselle, Dr. Henderson as “Dr. Davidson,” breakfast cereal as eggs, and a clean-shaven man as a fellow with a mustache. Suggested misinformation may even produce false memories of supposed child sexual abuse, argues Loftus.

This process affects our recall of social as well as physical events. Jack Croxton and colleagues (1984) had students spend 15 minutes talking with someone. Those who were later informed that this person liked them recalled the person’s behavior as relaxed, comfortable, and happy. Those informed that the person disliked them recalled the person as nervous, uncomfortable, and not so happy.

RECONSTRUCTING OUR PAST ATTITUDES Five years ago, how did you feel about nuclear power? About your country’s presi­ dent or prime minister? About your parents? If your attitudes have changed, what do you think is the extent of the change?

Experimenters have explored such questions, and the results have been unnerving. People whose attitudes have changed often insist that they have always felt much as they now feel. Daryl Bern and Keith McConnell (1970) conducted a survey among Carnegie Mellon University students. Buried in it was a ques­ tion concerning student control over the university curriculum. A week later, the students agreed to write an essay opposing student control. After doing so, their attitudes shifted toward greater opposition to student control. When asked to recall how they had answered the question before writing the essay, the students “remembered” holding the opinion that they now held and denied that the experi­ ment had affected them.

After observing Clark University students similarly denying their former atti­ tudes, researchers D. R. Wixon and James Laird (1976) commented, “The speed, magnitude, and certainty” with which the students revised their own histories “was striking.” As George Vaillant (1977) noted after following adults through time, “It is all too common for caterpillars to become butterflies and then to maintain that in their youth they had been little butterflies. Maturation makes liars of us all.”

The construction of positive memories brightens our recollections. Terence Mitchell, Leigh Thompson, and colleagues (1994, 1997) report that people often exhibit rosy retrospection—they recall mildly pleasant events more favorably than they experienced them. College students on a 3-week bike trip, older adults on a guided tour of Austria, and undergraduates on vacation all reported enjoying their experiences as they were having them. But they later recalled such experiences even more fondly, minimizing the unpleasant or boring aspects and remembering the high points. Thus, the pleasant times during which I have sojourned in Scotland I now (back in my office facing deadlines and interruptions) romanticize as pure bliss. The drizzle and the pesky midges are but dim memories. The spectacular scenery and the fresh sea air and the favorite tea rooms are still with me. With any positive experience, some of our pleasure resides in the anticipation, some in the actual experience, and some in the rosy retrospection.

Cathy McFarland and Michael Ross (1985) found that as our relationships change, we also revise our recollections of other people. They had university stu­ dents rate their steady dating partners. Two months later, they rated them again. Students who were more in love than ever had a tendency to overestimate their first impressions—it was “love at first sight.” Those who had broken up were more likely to underestimate their earlier liking—recalling the partner as somewhat selfish and bad-tempered.

Diane Holmberg and John Holmes (1994) discovered the phenomenon also oper­ ating among 373 newlywed couples, most of whom reported being very happy. When resurveyed 2 years later, those whose marriages had soured recalled that

Social Beliefs and Judgments Chapters 85

things had always been bad. The results are “frightening,” say Holmberg and Holmes: “Such biases can lead to a dangerous downward spiral. The worse your current view of your partner is, the worse your memories are, which only further confirms your negative attitudes.”

It’s not that we are totally unaware of how we used to feel, just that when memo­ ries are hazy, current feelings guide our recall. When widows and widowers try to recall the grief they felt on their spouse’s death 5 years earlier, their current emo­ tional state colors their memories (Safer & others, 2001). When patients recall their previous day’s headache pain, their current feelings sway their recollections (Eich & others, 1985). Parents of every generation bemoan the values of the next genera­ tion, partly because they misrecall their youthful values as being closer to their cur­ rent values. And teens of every generation recall their parents as—depending on their current mood—wonderful or woeful (Bomstein & others, 1991).

RECONSTRUCTING OUR PAST BEHAVIOR Memory construction enables us to revise our own histories. Michael Ross, Cathy McFarland, and Garth Fletcher (1981) exposed some University of Waterloo stu­ dents to a message convincing them of the desirability of toothbrushing. Later, in a supposedly different experiment, these students recalled brushing their teeth more often during the preceding 2 weeks than did students who had not heard the message. Likewise, judging from surveys, people report smoking many fewer ciga­ rettes than are actually sold (Hall, 1985). And they recall casting more votes than were actually recorded (Bureau of the Census, 1993).

Social psychologist Anthony Greenwald (1980) noted the similarity of such find­ ings to happenings in George Orwell’s novel 1984—in which it was “necessary to remember that events happened in the desired manner.” Indeed, argued Green­ wald, we all have “totalitarian egos” that revise the past to suit our present views. Thus, we underreport bad behavior and overreport good behavior.

Sometimes our present view is that we’ve improved—in which case we may misrecall our past as more unlike the present than it actually was. This tendency resolves a puzzling pair of consistent findings: Those who participate in psycho­ therapy and self-improvement programs for weight control, antismoking, and exer­ cise show only modest improvement on average. Yet they often claim considerable benefit. Michael Conway and Michael Ross (1986) explain why: Having expended so much time, effort, and money on self-improvement, people may think, “I may not be perfect now, but I was worse before; this did me a lot of good.”

In Chapter 14, we will see that psychiatrists and clinical psychologists are not immune to these human tendencies. We all selectively notice, interpret, and recall events in ways that sustain our ideas. Our social judgments are a mix of observa­ tion and expectation, reason and passion.





SUMMING UP: How Do We Perceive Our Social Worlds? • Our preconceptions strongly influence how we inter­

pret and remember events. In a phenomenon called priming, people’s prejudgments have striking effects on how they perceive and interpret information.

• Other experiments have planted judgments or false ideas in people’s minds after they have been given information. These experiments reveal that as before- the-fact judgments bias our perceptions and interpre­ tations, so after-the-fact judgments bias our recall.

• Belief perseverance is the phenomenon in which peo­ ple cling to their initial beliefs and the reasons why a belief might be true, even when the basis for the belief is discredited.

• Far from being a repository for facts about the past, our memories are actually formed when we retrieve them, and they are subject to strong influ­ ence by the attitudes and feelings we hold at the time of retrieval.

86 Part One Social Thinking


rUnderstand how we form social judgments.

As we have already noted, our cognitive mechanisms are efficient and adaptive, yet occasionally error-prone. Usually they serve us well. But sometimes clinicians mis­ judge patients, employers misjudge employees, people of one race misjudge people of another, and spouses misjudge their mates. The results can be misdiagnoses, labor strife, racial prejudices, and divorces. So, how—and how well—do we make intuitive social judgments?

When historians describe social psychology’s first century, they will surely record 1980 to the present as the era of social cognition. By drawing on advances in cogni­ tive psychology—in how people perceive, represent, and remember events—social psychologists have shed welcome light on how we form judgments. Let’s look at what that research reveals of the marvels and mistakes of our social intuition.

Intuitive Judgments What are our powers of intuition—of immediately knowing something without reasoning or analysis? Advocates of “intuitive management” believe we should tune into our hunches. When judging others, they say, we should plug into the non- logical smarts of our “right brain.” When hiring, firing, and investing, we should listen to our premonitions. In making judgments, we should trust the force within.

Are the intuitionists right that important information is immediately available apart from our conscious analysis? Or are the skeptics correct in saying that intu­ ition is “our knowing we are right, whether we are or not”?

Priming research hints that the unconscious indeed controls much of our behav­ ior. As John Bargh and Tanya Chartrand (1999) explain, “Most of a person’s every­ day life is determined not by their conscious intentions and deliberate choices but by mental processes that are put into motion by features of the environment and that operate outside of conscious awareness and guidance.” When the light turns red, we react and hit the brake before consciously deciding to do so. Indeed, reflect Neil Macrae and Lucy Johnston (1998), “to be able to do just about anything at all (e.g., driving, dating, dancing), action initiation needs to be decoupled from the inefficient (i.e., slow, serial, resource-consuming) workings of the conscious mind, otherwise inaction inevitably would prevail.”

controlled processing “Explicit” thinking that is deliberate, reflective, and conscious.

automatic processing “Implicit” thinking that is effortless, habitual, and without awareness; roughly corresponds to “intuition.”

THE POWERS OF INTUITION “The heart has its reasons which reason does not know,” observed seventeenth- century philosopher-mathematician Blaise Pascal. Three centuries later, scientists have proved Pascal correct. We know more than we know we know. Studies of our unconscious information processing confirm our limited access to what’s going on in our minds (Bargh & Ferguson, 2000; Greenwald & Banaji, 1995; Strack & Deutsch, 2004). Our thinking is partly controlled (reflective, deliberate, and con­ scious) and—more than psychologists once supposed—partly automatic (impul­ sive, effortless, and without our awareness). Automatic, intuitive thinking occurs not “on-screen” but off-screen, out of sight, where reason does not go. Consider these examples of automatic thinking:

• Schemas are mental concepts or templates that intuitively guide our percep­ tions and interpretations. Whether we hear someone speaking of religious sects or sex depends not only on the word spoken but also on how we auto­ matically interpret the sound.

Social Beliefs and Judgments Chapter 3 87

• Emotional reactions are often nearly instantaneous, happening before there is time for deliberate thinking. One neural shortcut takes information from the eye or the ear to the brain’s sensory switchboard (the thalamus) and out to its emotional control center (the amygdala) before the thinking cortex has had any chance to intervene (LeDoux, 2002). Our ancestors who intuitively feared a sound in the bushes were usually fearing nothing. But when the sound was made by a dangerous predator, they became more likely to survive to pass their genes down to us than their more deliberative cousins.

• Given sufficient expertise, people may intuitively know the answer to a problem. Many skills, from piano playing to swinging a golf club, begin as a controlled, deliberate process of following rules and gradually become automatic and intuitive (Kruglanski & Gigerenzer, 2011). Master chess play­ ers intuitively recognize meaningful patterns that novices miss and often make their next move with only a glance at the board, as the situation cues information stored in their memory. Similarly, without knowing quite how, we recognize a friend’s voice after the first spoken word of a phone conversation.

• Faced with a decision but lacking the expertise to make an informed snap judgment, our unconscious thinking may guide us toward a satisfying choice. That’s what Dutch psychologist Ap Dijksterhuis and co-workers (Dijkster- huis & Nordgren, 2006; Dijksterhuis & others, 2006; Strick & others, 2010) discovered after showing people, for example, a dozen pieces of information about each of four potential apartments. Compared with people who made instant decisions or were given time to analyze the information, the most satisfying decisions were made by those who were distracted and unable to focus consciously on the problem. Although these findings are controversial (Gonzalez-Vallejo & others, 2008; Lassiter & others, 2009; Newell & others, 2008), this much seems true: When facing a tough decision, it often pays to take our time—even to sleep on it—and to await the intuitive result of our out-of-sight information processing (Sio & Ormerod, 2009).

Some things—facts, names, and past experiences—we remember explicitly (con­ sciously). But other things—skills and conditioned dispositions—we remember implicitly, without consciously knowing or declaring that we know. It’s true of us all but most strikingly evident in people with brain damage who cannot form new explicit memories. One such person never could learn to recognize her physician, who would need to reintroduce himself with a handshake each day. One day, the physician affixed a tack to his hand, causing the patient to jump with pain. When the physician next returned, he was still unrecognized (explicitly). But the patient, retaining an implicit memory, would not shake his hand.

Equally dramatic are the cases of blindsight. Having lost a portion of the visual cortex to surgery or stroke, people may be functionally blind in part of their field of vision. Shown a series of sticks in the blind field, they report seeing nothing. After correctly guessing whether the sticks are vertical or horizontal, the patients are astounded when told, “You got them all right.” Like the patient who “remem­ bered” the painful handshake, these people know more than they know they know.

Consider your own taken-for-granted capacity to recognize a face. As you look at it, your brain breaks the visual information into subdimensions such as color, depth, movement, and form and works on each aspect simultaneously before reas­ sembling the components. Finally, using automatic processing, your brain com­ pares the perceived image with previously stored images. Voila! Instantly and effortlessly, you recognize your grandmother. If intuition is immediately knowing something without reasoned analysis, then perceiving is intuition par excellence.

So, many routine cognitive functions occur automatically, unintentionally, with­ out awareness. We might remember how automatic processing helps us get through

88 Part One Social Thinking

life by picturing our minds as functioning like large corporations. Our CEO—our controlled consciousness—attends to many of the most important, complex, and novel issues, while subordinates deal with routine affairs and matters requiring instant action. Like a CEO, consciousness sets goals and priorities, often with little knowledge of operational activities in the underlying departments. This delegation of resources enables us to react to many situations quickly and efficiently. The bot­ tom line; Our brain knows much more than it tells us.

THE LIMITS OF INTUITION We have seen how automatic, intuitive thinking can “make us smart” (Gigerenzer, 2007, 2010). Elizabeth Loftus and Mark Klinger (1992) nevertheless spoke for other cognitive scientists in having doubts about the brilliance of intuition. They reported “a general consensus that the unconscious may not be as smart as pre­ viously believed.” For example, although subliminal stimuli can trigger a weak, fleeting response—enough to evoke a feeling if not conscious awareness—thexe is no evidence that commercial subliminal tapes can “reprogram your unconscious mind” for success. In fact, a significant body of evidence indicates that they can t (Greenwald, 1992).

Social psychologists have explored not only our error-prone hindsight judgments but also our capacity for illusion—for perceptual misinterpretations, fantasies, and constructed beliefs. Michael Gazzaniga (1992, 1998, 2008) reports that patients whose brain hemispheres have been surgically separated will instantly fabricate and believe—explanations of their own puzzling behaviors. If the patient gets up and takes a few steps after the experimenter flashes the instruction “walk” to the patient’s nonverbal right hemisphere, the verbal left hemisphere will instantly pro­ vide the patient with a plausible explanation (“I felt like getting a drink”).

Illusory intuition also appears in the vast new literature on how we take in, store, and retrieve social information. As perception researchers study visual illusions for what they reveal about our normal perceptual mechanisms, social psychologists study illusory thinking for what it reveals about normal information processing. These researchers want to give us a map of everyday social thinking, with the haz­ ards clearly marked.

As we examine some of these efficient thinking patterns, remember this: Dem­ onstrations of how people create counterfeit beliefs do not prove that all beliefs are counterfeit (although to recognize counterfeiting, it helps to know how it’s done).

overconfidence phenomenon The tendency to be more confident than correct—to overestimate the accuracy of one’s beliefs.

The air distance between New Delhi and Beijing is 2,500 miles.

So far we have seen that our cognitive systems process a vast amount of informa­ tion efficiently and automatically. But our efficiency has a trade-off; as we interpret our experiences and construct memories, our automatic intuitions sometimes err. Usually, we are unaware of our flaws.

To explore this overconfidence phenomenon, Daniel Kahneman and Amos Tversky (1979) gave people factual statements and asked them to fill in the blanks, as in the following sentence: “1 feel 98 percent certain that the air distance between New Delhi and Beijing is more than____ miles but less than_____ miles.” Most individuals were overconfident: Approximately 30 percent of the time, the correct answers lay outside the range they felt 98 percent confident about.

To find out whether overconfidence extends to social judgments, David Dunning and associates (1990) created a game show. They asked Stanford University stu­ dents to guess a stranger’s answers to a series of questions, such as “Would you prepare for a difficult exam alone or with others?” and “Would you rate your lecture notes as neat or messy?” Knowing the type of question but not the actual questions, the participants first interviewed their target person about background, hobbies, academic interests, aspirations, astrological sign—anything they thought might be helpful. Then, while the targets privately answered 20 of the two-choice

Social Beliefs and Judgments Chapter 3 89

DOONESBURY by Garry Trudeau

DOONESBURY © 2000 G. B. Trudeau.

questions, the interviewers predicted their target’s answers and rated their own Reprmted with permission of Universal confidence in the predictions.

The interviewers guessed right 63 percent of the time, beating chance by 13 percent. But, on average, they felt 75 percent sure of their predictions. When guess­ ing their own roommates’ responses, they were 68 percent correct and 78 percent confident. Moreover, the most confident people were most likely to be overconfi­ dent. People also are markedly overconfident when judging whether someone is telling the truth or when estimating things such as the sexual history of their dat­ ing partner or the activity preferences of their roommates (DePaulo & others, 1997; Swann & Gill, 1997).

Ironically, incompetence feeds overconfidence. It takes competence to recognize what competence is, note Justin Kruger and David Dunning (1999). Students who score at the bottom on tests of grammar, humor, and logic are most prone to over­ estimating their gifts at such. Those who don’t know what good logic or grammar is are often unaware that they lack it. If you make a list of all the words you can form out of the letters in “psychology,” you may feel brilliant—but then stupid when a friend starts naming the ones you missed. Deanna Caputo and Dunning (2005) re-created this phenomenon in experiments, confirming that our ignorance of our ignorance sustains our self-confidence. Follow-up studies indicate that this “ignorance of one’s incompetence” occurs mostly on relatively easy-seeming tasks. On very difficult tasks, poor performers more often appreciate their lack of skill (Burson & others, 2006).

Ignorance of one’s incompetence helps explain David Dunning’s (2005) startling conclusion from employee assessment studies that “what others see in us … tends to be more highly correlated with objective outcomes than what we see in ourselves.” In one study, participants watched someone walk into a room, sit, read a weather report, and walk out (Borkenau & Liebler, 1993). Based on nothing more than that, their esti­ mate of the person’s intelligence correlated with the person’s intelligence score about as well as did the person’s own self-estimate (.30 vs. .32)! If ignorance can beget false confidence, then—yikes!—where, we may ask, are you and I unknowingly deficient?

In Chapter 2, we noted that people overestimate their long-term emotional responses to good and bad happenings. Are people better at predicting their own behavior? To find out, Robert Vallone and colleagues (1990) had college students predict in September whether they would drop a course, declare a major, elect to

90 Part One Social Thinking












live off campus next year, and so forth. Although the students felt, on average, ^ percent sure of those self-predictions, they were wrong nearly twice as often as they expected to be. Even when feeling 100 percent sure of their predictions they erred 15 percent of the time. ^

In estimating their chances for success on a task, such as a major exam, people’s confidence runs highest when the moment of truth is off in the future. By exam day, the possibility of failure looms larger and confidence typically drops (Gilovich & others, 1993; Shepperd & others, 2005). Roger Buehler and colleagues (1994, 2010) report that most students also confidently underestimate how long it will take them to complete papers and other major assignments. They are not alone:

• The “planning fallacy.” How much free time do you have today? How much free time do you expect you will have a month from today? Most of us over­ estimate how much we’ll be getting done, and therefore how much free time we will have (Zauberman & Lynch, 2005). Professional planners, too, rou­ tinely underestimate the time and expense of projects. In 1969, Montreal Mayor Jean Drapeau proudly announced that a $120 million stadium with a retractable roof would be built for the 1976 Olympics. The roof was com­ pleted in 1989 and cost $120 million by itself. In 1985, officials estimated that Boston’s “Big Dig” highway project would cost $2.6 billion and take until 1998. The cost ballooned to $14.6 billion and the project took until 2006. Stockbroker overconfidence. Investment experts market their services with the confident presumption that they can beat the stock market average, forget- Hng that for every stockbroker or buyer saying “Sell!” at a given price, there IS another saying “Buy!” A stock’s price is the balance point between those mutually confident judgments. Thus, incredible as it may seem, economist Burton Malkiel (2011) reports that mutual fund portfolios selected by invest­ ment analysts have not outperformed randomly selected stocks.

• Political overconfidence. Overconfident decision makers can wreak havoc. It was a confident Adolf Hitler who from 1939 to 1945 waged war against the rest of Europe. It was a confident Lyndon Johnson who in the 1960s invested U.S. weapons and soldiers in the effort to salvage democracy in South Vietnam.

The perils of overconfidence. Before its exploded drilling platform spewed oil into the Gulf of Mexico, BP downplayed safety concerns, and then was overconfident that the spill would be modest (Mohr & others, 2010; Urbina, 2010).

What produces overconfidence? Why doesn’t experience lead us to a more real­ istic self-appraisal? For one thing, people tend to recall their mistaken judgments as times when they were almost right. Philip Tetlock (1998a, 1998b, 1999,2005) observed this after inviting various academic and government experts to project—from their viewpoint in the late 1980s—the future governance of the Soviet Union, South Africa, and Canada. Five years later, com­ munism had collapsed. South Africa had become a multiracial democracy, and Canada’s French-speaking minority had not seceded. Experts who had felt more than 80 percent confident were right in predicting these turns of events less than 40 percent of the time. Yet, reflecting on their judgments, those who erred believed they were still basically right. I was almost right,” said many. “The hardliners

almost succeeded in their coup attempt against Gorbachev.” “The Quebecois

91Social Beliefs and Judgments

separatists almost won the secessionist referendum.” “But for the coincidence of de Klerk and Mandela, there would have been a much bloodier transition to black major­ ity rule in South Africa.” The Iraq War was a good idea, just badly executed, excused many of those who had supported it. Among political experts—and also stock market forecasters, mental health workers, and sports prognosticators—overconfidence is hard to dislodge.

CONFIRMATION BIAS People also tend not to seek information that might disprove what they believe. P. C. Wason (1960) demonstrated this, as you can, by giving participants a sequence of three numbers—2, 4, 6—that conformed to a rule he had in mind. (The rule was simply any three ascending numbers.) To enable the participants to discover the rule, Wason invited each person to generate additional sets of three numbers. Each time, Wason told the person whether or not the set conformed to his rule. As soon as participants were sure they had discovered the rule, they were to stop and announce it.

The result? Seldom right but never in doubt: 23 of the 29 participants convinced themselves of a wrong rule. They typically formed some erroneous belief about the rule (for example, counting by two’s) and then searched for confirming evidence (for example, by testing 8,10,12) rather than attempting to disconfirm their hunches. We are eager to verify our beliefs but less inclined to seek evidence that might disprove them, a phenomenon called the confirmation bias.

Confirmation bias helps explain why our self-images are so remarkably stable. In experiments at the University of Texas at Austin, William Swann and Stephen Read (1981; Swann & others, 1992a, 1992b, 2007) discovered that students seek, elicit, and recall feedback that confirms their beliefs about themselves. People seek as friends and spouses those who bolster their own self views—even if they think poorly of themselves (Swann & others, 1991,2003).

Swann and Read (1981) liken this self-verification to how someone with a domi­ neering self-image might behave at a party. Upon arriving, the person seeks those guests whom she knows will acknowledge her dominance. In conversation, she then presents her views in ways that elicit the respect she expects. After the party, she has trouble recalling conversations in which her influence was minimal and more easily recalls her persuasiveness in the conversations that she dominated. Thus, her experience at the party confirms her self-image.

REMEDIES FOR OVERCONFIDENCE What lessons can we draw from research on overconfidence? One lesson is to be wary of other people’s dogmatic statements. Even when people are sure they are right, they may be wrong. Confidence and competence need not coincide.

Three techniques have successfully reduced the overconfidence bias. One is prompt feedback (Lichtenstein &: Fischhoff, 1980). In everyday life, weather forecast­ ers and those who set the odds in horse racing both receive clear, daily feedback. And experts in both groups do quite well at estimating their probable accuracy (Fischhoff, 1982).

To reduce “planning fallacy” overconfidence, people can be asked to unpack a task—to break it down into its subcomponents—and estimate the time required for each. Justin Kruger and Matt Evans (2004) report that doing so leads to more realis­ tic estimates of completion time.

When people think about why an idea might be true, it begins to seem true (Koehler, 1991). Thus, a third way to reduce overconfidence is to get people to think of one good reason zvhy their judgments might be wrong; that is, force them to consider disconfirming information (Koriat & others, 1980). Managers might fos­ ter more realistic judgments by insisting that all proposals and recommendations include reasons why they might not work.

Chapter 3








confirmation bias A tendency to search for information that confirms one’s preconceptions.

92 Part One

heuristic A thinking strategy that enables quick, efficient judgments.

representativeness heuristic The tendency to presume, sometimes despite contrary odds, that someone or something belongs to a particular group if resembling (representing! a typical member.

Social Thinking

Still, we should be careful not to undermine people’s reasonable self-confidence or to destroy their decisiveness. In times when their wisdom is needed, those lack­ ing self-confidence may shrink from speaking up or making tough decisions. Over- confidence can cost us, but realistic self-confidence is adaptive.

Heuristics: Mental Shortcuts With precious little time to process so much information, our cognitive system is fast and frugal. It specializes in mental shortcuts. With remarkable ease, we form impressions, make judgments, and invent explanations. We do so by using heuristics—simple, efficient thinking strategies. Heuristics enable us to live and make routine decisions with minimal effort {Shah & Oppenheimer, 2008). In most situations, our snap generalizations—”That’s dangerous!”—are adaptive. The speed of these intuitive guides promotes our survival. The biological purpose of thinking is less to make us right than to keep us alive. In some situations, however, haste makes error.

THE REPRESENTATIVENESS HEURISTIC University of Oregon students were told that a panel of psychologists interviewed a sample of 30 engineers and 70 lawyers and summarized their impressions in thumbnail descriptions. The following description, they were told, was drawn at random from the sample of 30 engineers and 70 lawyers:

Twice divorced, Frank spends most of his free time hanging around the country club. His clubhouse bar conversations often center on his regrets at having tried to follow his esteemed father’s footsteps. The long hours he had spent at academic drudgery would have been better invested in learning how to be less quarrelsome in his relations with other people. Question: What is the probability that Frank is a lawyer rather than an engineer?

Asked to guess Frank’s occupation, more than 80 percent of the students surmised he was one of the lawyers (Fischhoff & Bar-Hillel, 1984). Fair enough. But how do you suppose those estimates changed when the sample description was given to another group of students, modified to say that 70 percent were engineers? Not in the slightest. The students took no account of the base rate of engineers and law­ yers; in their minds, Frank was more representative of lawyers, and that was all that seemed to matter.

To judge something by intuitively comparing it to our mental representation of a category is to use the representativeness heuristic. Representativeness (typical­ ness) usually is a reasonable guide to reality. But, as we saw with “Frank” above, it doesn’t always work. Consider Linda, who is 31, single, outspoken, and very bright. She majored in philosophy in college. As a student, she was deeply con­ cerned with discrimination and other social issues, and she participated in anti­ nuclear demonstrations. Based on that description, would you say it is more likely that

a. Linda is a bank teller. b. Linda is a bank teller and active in the feminist movement.

Most people think h is more likely, partly because Linda better represents their image of feminists (Mellers & others, 2001). But ask yourself: Is there a better chance that Linda is both a bank teller and a feminist than that she’s a bank teller (whether feminist or not)? As Amos Tversky and Daniel Kahneman (1983) reminded us, the conjunction of two events cannot be more likely than either one of the events alone.

THE AVAILABILITY HEURISTIC Consider the following: Do more people live in Iraq or in Tanzania? (See page 94 for the answer.)

93Social Beliefs and Judgments

table •• 3.1 Fast and Frugal Heuristics

Heuristic ‘ Definition ‘ Example But May Lead to

V Representativeness Snap judgments of whether someone or something fits a category

Deciding that Carlos is a librarian rather than a trucker because he better represents one’s image of librarians

Discounting other important information

Availability Quick judg­ ments of likeli­ hood of events (how available in memory)

Estimating teen violence after school shootings

Overweighting vivid instances and thus, for example, fearing the wrong things

You probably answered according to how readily Iraqis and Tanzanians come to mind. If examples are readily available in our memory—as Iraqis tend to be— then we presume that other such examples are commonplace. Usually this is true, so we are often well served by this cognitive rule, called the availability heuristic (Table 3.1). Said simply, the more easily we recall something, the more likely it seems.

But sometimes the rule deludes us. If people hear a list of famous people of one sex (Oprah Winfrey, Lady Gaga, and Hillary Clinton) intermixed with an equal- size list of unfamous people of the other sex (Donald Scarr, William Wood, and Mel Jasper), the famous names will later be more cognitively available. Most peo­ ple will subsequently recall having heard more (in this instance) women’s names (McKelvie, 1995, 1997; Tversky & Kahneman, 1973). Likewise, media attention to gay-lesbian issues makes gays and lesbians cognitively available. Thus, the average U.S. adult in a 2011 Gallup poll estimated that 25 percent of Americans are gay or lesbian (Morales, 2011b)—some seven times the number who, in surveys, actually self-identify as gay, lesbian, or bisexual (Gates, 2011).

Even fictional happenings in novels, television, and movies leave images that later penetrate our judgments (Gerrig & Prentice, 1991; Green & others, 2002; Mar & Oatley, 2008). The more absorbed and “transported” the reader (“I could easily picture the events”), the more the story affects the reader’s later beliefs (Diekman & others, 2000). Readers who are captivated by romance novels, for example, may gain readily available sexual scripts that influence their own sexual attitudes and behaviors.

Or consider this: Order these four cities according to their crime rates: Atlanta, Los Angeles, New York, St. Louis. If, with available images from TV crime dra­ mas in mind, you thought New York and Los Angeles are the most crime-ridden, guess again; they’re the least crime-ridden of the four (Federal Bureau of Investi­ gation, 2011).

Our use of the availability heuristic highlights a basic principle of social thinking: People are slow to deduce particular instances from a general truth, but they are remarkably quick to infer general truth from a vivid instance. No wonder that after hearing and reading stories of rapes, robberies, and beatings, 9 out of 10 Canadians overestimated—usually by a considerable margin—the percentage of crimes that involved violence (Doob & Roberts, 1988). And no wonder that South Africans, after a series of headline-grabbing gangland robberies and slayings, estimated that violent crime had almost doubled between 1998 and 2004, when actually it had decreased substantially (Wines, 2005).

The availability heuristic explains why vivid, easy-to-imagine events, such as shark attacks or diseases with easy-to-picture symptoms, may seem more likely to occur than harder-to-picture events (MacLeod & Campbell, 1992; Sherman & others, 1985).


availability heuristic A cognitive rule that judges the likelihood of things in terms of their availability in memory. If instances of something come readily to mind, we presume it to be commonplace.



HOLMES, JR„ 1841-1935

94 Part One Social Thinking

Answer to Question on page 92: Tanzania’s 43 million people greatly outnumber Iraq’s 31 million. Most people, having more vivid images of Iraqis, guess wrong.

Likewise, powerful anecdotes can be more compelling than statistical information. We fret over extremely rare child abduction, even if we don’t buckle our children in the backseat. We dread terrorism but are indifferent to global climate change— “Armageddon in slow moHon.” Especially after the 2011 Japanese tsunami and nuclear power catastrophe, we fear nuclear power, with little concern for the many more deaths related to coal mining and burning (von Hippel, 2011). In short, we worry about remote possibilities while ignoring higher probabilities, a phenomenon that Cass Sunstein (2007b) calls our “probability neglect.”

Because news footage of airplane crashes is a readily available memory for most of us especially since September 11, 2001—we often suppose we are more at risk traveling in commercial airplanes than in cars. Actually, from 2003 to 2005, U.S. travelers were 230 times more likely to die in a car crash than on a commercial flight covering the same distance (National Safety Council, 2008). In 2006, reports the Flight Safety Foundation, there was one airliner accident for every 4.2 million flights by Western-built commercial jets (Wald, 2008). For most air travelers, the most dangerous part of the journey is the drive to the airport.

Soon after 9/11, as many people abandoned air travel and took to the roads, I estimated that if Americans flew 20 percent less and instead drove those unflown miles, we could expect an additional 800 traffic deaths in the ensuing year (Myers, 2001). It took a curious German researcher (why didn’t I think of this?) to check that prediction against accident data, which confirmed an excess of some 350 deaths in the last 3 months of 2001 compared with the 3-month average in the preceding 5 years (Gigerenzer, 2004). The 9/11 terrorists appear to have killed more people unnoticed—on America’s roads—than they did with the 266 fatalities on those four planes.

By now it is clear that our naive statistical intuitions, and our resulting fears, are driven not by calculation and reason but by emotions attuned to the availability heuristic. After this book is published, there likely will be another dramatic natural or terrorist event, which will again propel our fears, vigilance, and resources in a new direction. Terrorists, aided by the media, may again achieve their objective of

Vivid, memorable—and therefore cognitively available—events influence our perception of the social world. The resulting “probability neglect often leads people to fear the wrong things, such as fearing flying or terrorism more than smoking, driving, or climate change If four jumbo jets filled with children crashed every day—approximating the number of childhood diarrhea deaths resulting from the rotavirus—something would have been done about it. Reprinted courtesy of Dave 8ohn.

Social Beliefs and Judgments Chapter 3 95

capturing our attention, draining our resources, and distracting us from the mundane, undramatic, insidious risks that, over time, devastate lives, such as the rotavirus that each day claims the equivalent of four 747s filled with children (Parashar & others, 2006). But then again, dramatic events can also serve to awaken us to real risks. That, say some scientists, is what happened when the extreme floods, droughts, snows, and tornadoes of 2011 raised concern that global climate change, by raising sea levels and spawning extreme weather, is destined to become nature’s own weapon of mass destruction. For Australians and Americans, a temporary hot day can prime people to believe more in global warming (Li & others, 2011). Even feeling hot in an indoor room increases people’s belief in global warming (Risen & Critcher, 2011).

Counterfactual Thinking Easily imagined (cognitively available) events also influence our experiences of guilt, regret, frustration, and relief. If our team loses (or wins) a big game by one point, we can easily imagine the other outcome, and thus we feel regret (or relief). Imagining worse alternatives helps us feel better. Imagining better alternatives, and pondering what we might do differently next time, helps us prepare to do better in the future (Epstude & Roese, 2008).

In Olympic competition, athletes’ emotions after an event reflect mostly how they did relative to expectations, but also their counterfactual thinking—their mentally simulating what might have been (McGraw & others, 2005; Medvec & others, 1995). Bronze medalists (for whom an easily imagined alternative was finishing without a medal) exhibit more joy than silver medalists (who could more easily imagine having won the gold). On the medal stand, it has been said, happiness is as simple as 1-3-2. Similarly, the higher a student’s score within a grade category (such as B-l-), the worse they feel (Medvec & Savitsky, 1997). The B+ stu­ dent who misses an A- by a point feels worse than the B-K student who actually did worse and just made a B-l- by a point.

Such counterfactual thinking occurs when we can easily picture an alternative outcome (Kahneman & Miller, 1986; Markman & McMullen, 2003; Petrocelli & others, 2011):

• If we barely miss a plane or a bus, we imagine making it if only we had left at our usual time, taken our usual route, not paused to talk. If we miss our connection by a half hour or after taking our usual route, it’s harder to simulate a different outcome, so we feel less frustration.

• If we change an exam answer, then get it wrong, we will inevitably think “If only . .and will vow next time to trust our immediate intuition—although, contrary to student lore, answer changes are more often from incorrect to correct (Kruger & others, 2005).

• The team or the political candidate who barely loses will simulate over and over how they could have won (Sanna & others, 2003).

Counterfactual thinking imderlies our feelings of luck. When we have barely escaped a bad event—avoiding defeat with a last-minute goal or standing near a falling icicle—we easily imagine a negative counterfactual (losing, being hit) and therefore feel “good luck” (Teigen & others, 1999). “Bad luck” refers to bad events that did hap­ pen but easily might not have.

The more significant and unlikely the event, the more intense the counterfactual thinking (Roese & Hur, 1997). Bereaved people who have lost a spouse or a child in a vehicle accident, or a child to sudden infant death syndrome, commonly report replaying and










counterfactual thinking Imagining alternative scenarios and outcomes that might have happened, but didn’t.

Counterfactual thinking. When Deal or No Deal game show contestants dealt too late (walking away with a lower amount than they were previ­ ously offered) or too early (foregoing their next choice, which would have led to more money), they likely experienced counterfactual thinking— imagining what might have been.

96 Part One

People are more often apologetic about actions than inactions (Zeelenberg & others, 1998).

illusory correlation Perception of a relationship where none exists, or perception of a stronger relationship than actually exists.

illusion of control Perception of uncontrollable events as subject to one’s control or as more controllable than they are.

FAMILY CIRCUS © 1998 Bil Keane. Inc. King Features Syndicate.

Social Thinking

undoing the event (Davis & others, 1995, 1996). One friend of mine survived a head-on collision with a drunk driver that killed his wife, daughter, and mother. He recalled, “For months I turned the events of that day over and over in my mind. I kept reliving the day, changing the order of events so that the accident wouldn’t occur” (Sittser, 1994).

Across Asian and Western cultures, most people, however, live with less regret over things done than over things they failed to do, such as, “I wish I had been more serious in college” or “I should have told my father I loved him before he died” (Gilovich & Medvec, 1994; Rajagopal & others, 2006). In one survey of adults, the most common regret was not taking their education more seriously (Kinnier & Metha, 1989). Would we live with less regret if we dared more often to reach beyond our comfort zone—to venture out, risking failure, but at least hav­ ing tried?

Illusory Thinking Another influence on everyday thinking is our search for order in random events, a tendency that can lead us down all sorts of wrong paths.

ILLUSORY CORRELATION It is easy to see a correlation where none exists. When we expect to find significant relationships, we easily associate random events, perceiving an illusory correlation. William Ward and Herbert Jenkins (1965) showed people the results of a hypotheti­ cal 50-day cloud-seeding experiment. They told participants which of the 50 days the clouds had been seeded and which days it rained. The information was nothing more than a random mix of results: Sometimes it rained after seeding; sometimes it didn’t. Participants nevertheless became convinced—in conformity with their ideas about the effects of cloud seeding—that they really had observed a relationship between cloud seeding and rain.

Other experiments confirm the principle: People easily misperceive random events as confirming their beliefs (Crocker, 1981; Jennings & others, 1982; Trolier & Hamilton, 1986). If we believe a correlation exists, we are more likely to notice and recall con­ firming instances. If we believe that premonitions correlate with events, we notice

and remember any joint occur­ rence of the premonition and the event’s later occurrence. If we believe that overweight women are unhappier, we perceive that we have witnessed such a cor­ relation even when we have not (Viken & others, 2005). We ignore or forget all the times unusual events do not coincide. If, after we think about a friend, the friend calls us, we notice and remember that coincidence. We don’t notice all the times we think of a friend without any ensuing call, or receive a call from a friend about whom we’ve not been thinking.

ILLUSION OF CONTROL Our tendency to perceive ran­ dom events as related feeds an illusion of control—the idea that


sign so much! Every time they do, it gets bumpy.”

97Social Beliefs and Judgments Chapter 3

chance events are subject to our influence. This keeps gamblers going and makes the rest of us do all sorts of unlikely things.

gambling Ellen Langer (1977) demonstrated the illusion of control in betting experiments. Compared with those given an assigned lottery number, people who chose their own number demanded four times as much money when asked if they would sell their ticket. When playing a game of chance against an awkward and nervous person, they bet significantly more than when playing against a dapper, confident opponent. Being the person who throws the dice or spins the wheel increases people’s confidence (Wohl & Enzle, 2002). In these and other ways, more than 50 experiments have consistently found people acting as if they can predict or control chance events (Presson & Benassi, 1996; Thompson & others, 1998).

Observations of real-life gamblers confirm these experimental findings. Dice play­ ers may throw softly for low numbers and hard for high numbers (Henslin, 1967). The gambling industry thrives on gamblers’ illusions. Gamblers attribute wins to their skill and foresight. Losses become “near misses” or “flukes,” or for the sports gambler, a bad call by the referee or a freakish bounce of the ball (Gilovich & Douglas, 1986).

Stock traders also like the “feeling of empowerment” that comes from being able to choose and control their own stock trades, as if their being in control can enable them to outperform the market average. One ad declared that online investing “is about control.” Alas, the illusion of control breeds overconfidence and frequent losses after stock market trading costs are subtracted (Barber & Odean, 2001a, 2001b).

People like feeling in control, and so when experiencing a lack of control, will act to create a sense of predictability. In experiments, loss of control has led people to form illusory correlations in stock market information, to perceive nonexistent conspiracies, and to develop superstitions (Whitson & Galinsky, 2008).

REGRESSION TOWARD THE AVERAGE Tversky and Kahneman (1974) noted another way by which an illusion of control may arise: We fail to recognize the statis­ tical phenomenon of regression toward the average. Because exam scores fluctuate partly by chance, most students who get extremely high scores on an exam will get lower scores on the next exam. If their first score is at the ceiling, their second score is more likely to fall back (“regress”) toward their own average than to push the ceiling even higher. That is why a student who does consistently good work, even if never the best, will sometimes end a course at the top of the class. Conversely, the lowest scoring students on the first exam are likely to improve. If those who scored lowest go for tutoring after the first exam, the tutors are likely to feel effective when the student improves, even if the tutoring had no effect.

Indeed, when things reach a low point, we will try anything, and whatever we try—going to a psychotherapist, starting a new diet-exercise plan, reading a self- help book—is more likely to be followed by improvement than by further dete­ rioration. Sometimes we recognize that events are not likely to continue at an unusually good or bad extreme. Experi­ ence has taught us that when everything is going great, something will go wrong, and that when life is dealing us terrible blows, we can usually look forward to things getting better. Often, though, we fail to recognize this regression effect.

regression toward the average The statistical tendency for extreme scores or extreme behavior to return toward one’s average.

Regression to the average. When we are at an extremely low point, anything we try will often seem effective. “Maybe a yoga class will improve my life,” Events seldom continue at an abnormal low.

98 Part One Social Thinking

We puzzle at why baseball’s rookie of the year often has a more ordinary second year—did he become overconfident? Self-conscious? We forget that exceptional performance tends to regress toward normality.

By simulating the consequences of using praise and punishment, Paul Schaffner (1985) showed how the illusion of control might infiltrate human rela­ tions. He invited Bowdoin College students to train an imaginary fourth-grade boy, “Harold,” to come to school by 8:30 each morning. For each school day during a 3-week period, a computer displayed Harold’s arrival time, which was always between 8:20 and 8:40. The students would then select a response to Harold, ranging from strong praise to strong reprimand. As you might expect, they usu­ ally praised Harold when he arrived before 8:30 and reprimanded him when he arrived after 8:30. Because Schaffner had programmed the computer to display a random sequence of arrival times, Harold’s arrival time tended to improve (to regress toward 8:30) after he was reprimanded. For example, if Harold arrived at 8:39, he was almost sure to be reprimanded, and his randomly selected next-day arrival time was likely to be earlier than 8:39. Thus, even though their reprimands were having no effect, most students ended the experiment believing that their reprimands had been effective.

This experiment demonstrates Tversky and Kahneman’s provocative conclu­ sion: Nature operates in such a way that we often feel punished for rewarding oth­ ers and rewarded for punishing them. In actuality, as every student of psychology knows, positive reinforcement for doing things right is usually more effective and has fewer negative side effects.

Moods and Judgments Social judgment involves efficient information processing. It also involves our feelings: Our moods infuse our judgments. Some studies compare happy and sad individuals (Myers, 1993, 2000b). Unhappy people—especially those bereaved or depressed—tend to be more self-focused and brooding. A depressed mood moti­ vates intense thinking—a search for information that makes one’s environment more understandable and controllable (Weary & Edwards, 1994).

Happy people, by contrast, are more trusting, more loving, more responsive. If people are made temporarily happy by receiving a small gift while mall-shopping, they will report, a few moments later on an unrelated survey, that their cars and TV sets are working beautifully—better, if you took their word for it, than those belonging to folks who replied after not receiving gifts.

Moods pervade our thinking. To West Germans enjoying their team’s World Cup soccer victory (Schwarz & others, 1987) and to Australians emerging from a heart­ warming movie (Forgas & Moylan, 1987), people seem good-hearted, life seems wonderful. After (but not before) a 1990 football game between rivals Alabama and Auburn, victorious Alabama fans deemed war less likely and potentially devastat­ ing than did the gloomier Auburn fans (Schweitzer & others, 1992). When we are in a happy mood, the world seems friendlier, decisions are easier, and good news more readily comes to mind (DeSteno & others, 2000; Isen & Means, 1983; Stone & Glass, 1986).

Let a mood turn gloomy, however, and thoughts switch onto a different track. Off come the rose-colored glasses; on come the dark glasses. Now the bad mood primes our recollections of negative events (Bower, 1987; Johnson & Magaro, 1987). Our relationships seem to sour. Our self-images take a dive. Our hopes for the future dim. And other people’s behavior seems more sinister (Brown & Taylor, 1986; Mayer & Salovey, 1987).

University of New South Wales social psychologist Joseph Forgas (1999, 2008, 2010, 2011) had often been struck by how moody people’s “memories and judg­ ments change with the color of their mood.” To understand this “mood infusion,” he began to experiment. Imagine yourself in one such study. Using hypnosis.

99Social Beliefs and Judgments Chapter 3

Percent perceived behaviors

detected detected

FIGURE :: 3.3 A temporary good or bad mood strongly influenced people’s rat­ ings of their videotaped behavior. Those in a bad mood detected far fewer positive behaviors.

Source: Forgas & others (1984).

Forgas and colleagues (1984) put you in a good or a bad mood and then have you watch a videotape (made the day before) of yourself talking with someone. If made to feel happy, you feel pleased with what you see, and you are able to detect many instances of your poise, interest, and social skill. If you’ve been put in a bad mood, viewing the same tape seems to reveal a quite different you—one who is stiff, ner­ vous, and inarticulate (Figure 3.3). Given how your mood colors your judgments, you feel relieved at how things brighten when the experimenter switches you to a happy mood before leaving the experiment. Curiously, note Michael Ross and Garth Fletcher (1985), we don’t attribute our changing perceptions to our mood shifts. Rather, the world really seems different.

Our moods color how we judge our worlds partly by bringing to mind past experiences associated with the mood. When we are in a bad mood, we have more depressing thoughts. Mood-related thoughts may distract us from complex think­ ing about something else. Thus, when emotionally aroused—when angry or even in a very good mood—we become more likely to make snap judgments and evaluate others based on stereotypes (Bodenhausen & others, 1994; Paulhus & Lim, 1994).

SUMMING UP: How Do We Judge Our Social Worlds? • We have an enormous capacity for automatic, effi­

cient, intuitive thinking. Our cognitive efficiency, although generally adaptive, comes at the price of occasional error. Because we are generally unaware of those errors entering our thinking, it is useful to iden­ tify ways in which we form and sustain false beliefe.

• First, we often overestimate our judgments. This overconfidence phenomenon stems partly from the much greater ease with which we can imagine why we might be right than why we might be wrong. Moreover, people are much more likely to search for information that can confirm their beliefs than for information that can disconfirm them.

• Second, when given compelling anecdotes or even useless information, we often ignore useful

base-rate information. This is partly due to the later ease of recall of vivid information (the availability heuristic).

• Third, we are often swayed by illusions of correla­ tion and personal control. It is tempting to perceive correlations where none exist {illusory correlation) and to think we can predict or control chance events (the illusion of control).

• Finally, moods infuse judgments. Good and bad moods trigger memories of experiences associated with those moods. Moods color our interpreta­ tions of current experiences. And by distracting us, moods can also influence how deeply or superfi­ cially we think when making judgments.

100 Part One Social Thinking


Recognize how—and how accurately—we explain others’ behavior.

People make it their business to explain other people, and social psychologists make it their business to explain people’s explanations.

Our judgments of people depend on how we explain their behavior. Depending on our explanation, we may judge killing as murder, manslaughter, self-defense, or heroism. Depending on our explanation, we may view a homeless person as lack­ ing initiative or as victimized by job and welfare cutbacks. Depending on our expla­ nation, we may interpret someone’s friendly behavior as genuine warmth or as ingratiation. Attribution theory helps us make sense of how this explanation works.

misattribution Mistakenly attributing a behavior to the wrong source.

Attributing Causality: To the Person or the Situation We endlessly analyze and discuss why things happen as they do, especially when we experience something negative or unexpected (Weiner, 1985, 2008, 2010). If worker productivity declines, do we assume the workers are getting lazier? Or has their workplace become less efficient? Does a young boy who hits his classmates have a hostile personality? Or is he responding to relentless teasing? Researchers found that married people often analyze their partners’ behaviors, especially their negative behaviors. Cold hostility, more than a warm hug, is likely to leave the partner won­ dering zvhi/? (Holtzworth-Munroe & Jacobson, 1985; Holtzworth & Jacobson, 1988).

Spouses’ answers correlate with marriage satisfaction. Unhappy couples usu­ ally offer distress-maintaining explanations for negative acts (“She was late because she doesn’t care about me”). Happy couples more often externalize (“She was late because of heavy traffic”). Explanations for positive acts similarly work either to maintain distress (“He brought me flowers because he wants sex”) or to enhance the relationship (“He brought me flowers to show he loves me”) (Hewstone & Fincham, 1996; McNulty & others, 2008; Weiner, 1995).

Antonia Abbey (1987, 1991, 2011; Abbey & others, 1998) and colleagues have repeatedly found that men are more likely than women to attribute a woman’s friendliness to mild sexual interest. (Men’s romantic interest is easier to read [Place & others, 2009]). Men’s misreading of women’s warmth as a sexual come-on— an example of misattribution—can con­ tribute to behavior that women regard as sexual harassment or even rape (Farris & others, 2008; Kolivas & Gross, 2007; Pryor & others, 1997). Many men believe women are flattered by repeated requests for dates, which women more often view as harassment (Rotundo & others, 2001).

Misattribution is particularly likely when men are in positions of power. A manager may misinterpret a subordinate woman’s submissive or friendly behav­ ior and, full of himself, see her in sexual terms (Bargh & Raymond, 1995). Men

A misattribution? Date rape sometimes begins with a man’s misreading a woman’s warmth as a sexual come-on.

101Social Beliefs and Judgments

more often than women think about sex (see Chapter 5). Men also are more likely than women to assume that others share their feelings (recall from Chapter 2 the “false consensus effect”). Thus, a man may greatly overestimate the sexual significance of a woman’s courtesy smile (Levesque & others, 2006; Nelson & LeBoeuf, 2002).

Such misattributions help explain the greater sexual assertiveness exhibited by men throughout the world and the greater tendency of men in various cultures, from Boston to Bombay, to justify rape by arguing that the victim consented or implied consent (Kanekar & Nazareth, 1988; Muehlenhard, 1988; Shotland, 1989). Women more often judge forcible sex as meriting conviction and a stiff sentence (Schutte & Hosch, 1997). Misattributions also help explain why, in one national survey, the 23 percent of American women who said they had been forced into unwanted sex­ ual behavior was eight times greater than the 3 percent of American men who said they had ever forced a woman into a sexual act (Laumann & others, 1994).

Attribution theory analyzes how we explain people’s behavior and what we infer from it. The variations of attribution theory share some common assumptions. As Daniel Gilbert and Patrick Malone (1995) explain, each “construes the human skin as a special boundary that separates one set of ‘causal forces’ from another. On the sunny side of the epidermis are the external or situational forces that press inward upon the person, and on the meaty side are the internal or personal forces that exert pressure outward. Sometimes these forces press in conjunction, sometimes in opposition, and their dynamic interplay manifests itself as observable behavior.”

Attribution theory pioneer Fritz Heider (1958) and others after him analyzed the “commonsense psychology” by which people explain everyday events. They con­ cluded that when we observe someone acting intentionally, we sometimes attribute that person’s behavior to internal causes (for example, the person’s disposition or mental state) and sometimes to external causes (for example, something about the person’s situation). A teacher may wonder whether a child’s underachievement is due to lack of motivation and ability (a dispositional attribution) or to physical and social circumstances (a situational attribution). Also, some of us are more inclined to attribute behavior to stable personality; others tend more to attribute behavior to situations (Bastian & Haslam, 2006; Robins & others, 2004).

Chapter 3

attribution theory The theory of how people explain others’ behavior—for example, by attributing it eitherto internal dispositions (enduring traits, motives, and attitudes) or to external situations.

dispositional attribution Attributing behavior to the person’s disposition and traits.

INFERRING TRAITS Edward Jones and Keith Davis (1965) noted that we often infer that other people’s actions are indicative of their intentions and dispositions. If I observe Rick making a sarcastic comment to Linda, I infer that Rick is a hostile person. Jones and Davis’s “theory of correspondent inferences” specified the conditions under which people

situational attribution Attributing behavior to the environment.

To what should we attribute a student’s sleepiness? To lack of sleep? To boredom? Whether we make internal or external attributions depends on whether we notice her con­ sistently sleeping in this and other classes, and on whether other students react as she does to this particular class.

102 Part One Social Thinking

spontaneous trait inference An effortless, automatic inference of a trait after exposure to someone’s behavior.

An exception: Asians are less likely to attribute people’s behavior to their personality traits (Na & Kitayama, 2011).

FIGURE:: 3.4 Harold Kelley’s Theory of Attributions Three factors—consistency, distinctiveness, and consensus— influence whether we attribute someone’s behavior to internal or external causes. Try creating your own examples, such as the following; If Mary and many others criticize Steve (with con­ sensus), and if Mary isn’t critical of others (high distinctiveness), then we make an external attribu­ tion (it’s something about Steve). If Mary alone (low consensus) criticizes Steve, and if she criti­ cizes many other people, too (low distinctiveness), then we are drawn to an internal attribution (it’s something about Mary).

infer traits. For example, normal or expected behavior tells us less about the person than does unusual behavior. If Samantha is sarcastic in a job interview, where a person would normally be pleasant, that tells us more about Samantha than if she is sarcastic with her siblings.

The ease with which we infer traits—a phenomenon called spontaneous trait inference—is remarkable. In experiments at New York University, James Uleman (1989; Uleman & others, 2008) gave students statements to remember, such as “The librarian carries the old woman’s groceries across the street.” The students would instantly, unintentionally, and unconsciously infer a trait. When later they were helped to recall the sentence, the most valuable clue word was not “books” (to cue librarian) or “bags” (to cue groceries) but “helpful”—the inferred trait that I sus­ pect you, too, spontaneously attributed to the librarian. Given even just 1 /10th of a second exposure to someone’s face, people will spontaneously infer some personal­ ity traits (Willis & Todorov, 2006).

COMMONSENSE ATTRIBUTIONS As the theory of correspondent inferences suggests, attributions often are ratio­ nal. Pioneering attribution theorist Harold Kelley (1973) described how we explain behavior by using information about “consistency,” “distinctiveness,” and “consensus” (Figure 3.4).

Consistency: How consistent is the person’s behavior in this situation? Distinctiveness: How specific is the person’s behavior to this particular situation? Consensus: To what extent do others in this situation behave similarly?

When explaining why Edgar is having trouble with his computer, most people would use information concerning consistency (Is Edgar usually unable to get his com­ puter to work?), distinctiveness (Does Edgar have trouble with other computers, or only this one?), and consensus (Do other people have similar problems with this make of computer?). If we learn that Edgar alone consistently has trouble with tiiis and other computers, we likely will attribute the troubles to Edgar, not to defects in this computer.

So our commonsense psychology often explains behavior logically. But Kelley also found that people often discount a contributing cause of behavior if other plau­ sible causes are already known. If we can specify one or two sufficient reasons a student might have done poorly on an exam, we often ignore or discount alterna­ tive possibilities (McClure, 1998). When given information about people’s college grade average and asked to judge their suitability for graduate school, people dis­ count the school’s grading leniency (Moore & others, 2010).

Consistency: Does this person usually behave this way

in this situation? (If yes, we seek an explanation.)

External attribution

(to the person’s situation)

Distinctiveness: Does this person behave

differently in this situation than in others?

Consensus: Do others behave

similarly in this situation?


Internal attribution

(to the person’s


Social Beliefs and Judgments Chapter 3 103

The Fundamental Attribution Error Social psychology’s most important lesson concerns the influence of our social envi­ ronment. At any moment, our internal state, and therefore what we say and do, depends on the situation as well as on what we bring to the situation. In experi­ ments, a slight difference between two situations sometimes greatly affects how peo­ ple respond. As a professor, I have seen this when teaching the same subject at both 8:30 a.m. and 7:00 p.m. Silent stares would greet me at 8:30; at 7:00,1 had to break up a party. In each situation, some individuals were more talkative than others, but the difference between the two situations exceeded the individual differences.

Attribution researchers have found a common problem with our attributions. When explaining someone’s behavior, we often underestimate the impact of the situation and overestimate the extent to which it reflects the individual’s traits and attitudes. Thus, even knowing the effect of the time of day on classroom conversa­ tion, I found it terribly tempting to assume that the people in the 7:00 p.m. class were more extraverted than the “silent types” who came at 8:30 a.m. Likewise, we may infer that people fall because they’re clumsy rather than because they were tripped; that people smile because they’re happy rather than faking friendliness, and that people speed past us on the highway because they’re aggressive rather than late for an important meeting.

This discounting of the situation, dubbed by Lee Ross (1977) the fundamental attribution error, appears in many experiments. In the first such study, Edward Jones and Victor Harris (1967) had Duke University students read debaters’ speeches supporting or attacking Cuba’s leader, Fidel Castro. When told that the debater chose which position to take, the students logically enough assumed it reflected the person’s own attitude. But what happened when the students were told that the debate coach had assigned the position? People who are merely feigning a position write more forceful statements than you’d expect (Allison & others, 1993; Miller & others, 1990). Thus, even knowing that the debater had been told to take a pro- or anti-Castro position did not prevent students from inferring that the debater in fact had the assigned leanings (Figure 3.5). People seemed to think, “Yeah, I know he was assigned that position, but, you know, I think he really believes it.”

Even when people know they are causing someone else’s behavior, they still underestimate external influences. If individuals dictate an opinion that someone

Attitude attributed

Pro-Castro 80

fundamental attribution error The tendency for observers to underestimate situational influences and overestimate dispositional influences upon others’ behavior. (Also called correspondence bias because we so often see behavior as corresponding to a disposition.)

FIGURE :: 3.5 The Fundamental Attribution Error When people read a debate speech supporting or attack­ ing Fidel Castro, they attributed corresponding attitudes to the speechwriter, even when the debate coach assigned the writer’s position.

Source: Data from Jones & Harris (1967).

Chose to give a Castro speech

Assigned to give a Castro speech

104 Part One Social Thinking

When viewing a movie actor playing a “good-guy” or a “bad-guy” role, we find it difficult to escape the illusion thatthe scripted behavior reflects an inner disposition. Perhaps that is why Leonard Nimoy, who played Mr. Spock in the original “Star Trek” series, titled one of his books I Am Not Spock.

else must then express, they still tend to see the person as actually holding that opinion (Gilbert & Jones, 1986). If people are asked to be either self-enhancing or self-deprecating during an interview, they are very aware of why they are acting so. But they are wnaware of their effect on another person. If Juan acts modestly, his naive partner Bob is likely to exhibit modesty as well. Juan will easily understand his own behavior, but he will think that poor Bob suffers from low self-esteem (Baumeister & others, 1988). In short, we tend to presume that others are the way they act. Observing Cinderella cowering in her oppressive home, people (ignoring the situation) infer that she is meek; dancing with her at the ball, the prince sees a suave and glamorous person.

The discounting of social constraints was evident in a thought-provoking experiment by Lee Ross and collaborators (Ross & others, 1977). The experiment re-created Ross’s firsthand experience of moving from graduate student to profes­ sor. His doctoral oral exam had proved a humbling experience as his apparently brilliant professors quizzed him on topics they specialized in. Six months later. Dr. Ross was himself an examiner, now able to ask penetrating questions on his favorite topics. Ross’s hapless student later confessed to feeling exactly as Ross had a half-year before—dissatisfied with his ignorance and impressed with the appar­ ent brilliance of the examiners.

In the experiment, with Teresa Amabile and Julia Steinmetz, Ross set up a simu­ lated quiz game. He randomly assigned some Stanford University students to play the role of questioner, some to play the role of contestant, and others to observe. The researchers invited the questioners to make up difficult questions that would demonstrate their wealth of knowledge. Any one of us can imagine such questions using one’s own domain of competence: “Where is Bainbridge Island?” “How did Mary, Queen of Scots, die?” “Which has the longer coastline, Europe or Africa?” If even those few questions have you feeling a little uninformed, then you will appre­ ciate the results of this experiment.*

Everyone had to know that the questioners would have the advantage. Yet both contestants and observers (but not the questioners) came to the erroneous conclusion that the questioners really were more knowledgeable than the contestants (Figure 3.6).

* Bainbridge Island is across Puget Sound from Seattle. Mary was ordered beheaded by her cousin Queen Elizabeth I. Although the African continent is more than double the area of Europe, Europe’s coastline is longer. (It is more convo­ luted, with many harbors and inlets, a geographical fact that contributed to its role in the history of maritime trade.)

Social Beliefs and Judgments Chapter 3 105

Rating of general knowledge

100 r—————————-












Contestants’ ratings

Observers’ ratings

FIGURE:: 3.6 Both contestants and observers of a simulated quiz game assumed that a person who had been randomly assigned the role of questioner was far more knowledgeable than the contes­ tant. Actually, the assigned roles of questioner and contestant simply made the questioner seem more knowledgeable. The failure to appreciate this illustrates the fundamental attribution error. Source; Data from Ross & others (1977).

Follow-up research shows that these misimpressions are hardly a reflection of low social intelligence. If anything, college students and other intelligent and socially competent people are more likely to make the attribution error (Bauman & Skitka, 2010; Block & Funder, 1986).

In real life, those with social power usually initiate and control conversations, which often leads underlings to overestimate their knowledge and intelligence. Medical doctors, for example, are often presumed to be experts on all sorts of questions unrelated to medicine. Similarly, students often overestimate the brilliance of their teachers. (As in the experiment, teachers are questioners on subjects of their special expertise.) When some of these stu­ dents later become teachers, they are usually amazed to dis­ cover that teachers are not so brilliant after all.

To illustrate the fundamental attribution error, most of us need look no further than our own experiences. Deter- iTiined to make some new friends, Bev plasters a smile on her face and anxiously plunges into a party. Everyone else seems quite relaxed and happy as they laugh and talk with one another. Bev wonders to herself, “Why is everyone always so at ease in groups like this while I’m feeling shy and tense?” Actually, everyone else is feeling nervous, too, and making the same attribution error in assuming that Bev and the others are as they appear—confidently convivial.

Why do we make the attribution error? So far we have seen a bias in the way we explain other peo­ ple’s behavior: We often ignore powerful situational deter- ^^^ants. Why do we tend to underestimate the situational determinants of others’ behavior but not of our own?

People often attribute keen intelligence to those, such as teachers and quiz show hosts, who test others’ knowledge.

106 Part One Social Thinking

The fundamental attribution error: observers underesti­ mating the situation. Driving into a gas station, we may think the person parked at the second pump (blocking access to the first) is incon­ siderate. That person, having arrived when the first pump was in use, attributes her behavior to the situation.

PERSPECTIVE AND SITUATIONAL AWARENESS Attribution theorists have pointed out that we observe others from a different perspective than we observe ourselves (Jones, 1976; Jones & Nisbett, 1971). When we act, the environment com­ mands our attention. When we watch another person act, that person occupies the center of our attention and the environment becomes relatively invisible. If Tm mad, it’s the situation that’s making me angry. But someone I see getting mad may seem like an ill-tempered person.

From his analysis of 173 studies, Bertram Malle (2006) concluded that the actor- observer difference is actually minimal. When our action feels intentional and admirable, we attribute it to our own good reasons, not to the situation. It’s only when we behave badly that we’re more likely to attribute our behavior to the situa­ tion, while someone observing us may spontaneously infer a trait.

In some experiments, people have viewed a videotape of a suspect confessing during a police interview. If they viewed the confession through a camera focused on the suspect, they perceived the confession as genuine. If they viewed it through a camera focused on the detective, they perceived it as more coerced (Lassiter & Irvine, 1986; Lassiter & others, 2005, 2007). The camera perspective influenced people’s guilt judgments even when the judge instructed them not to allow this to happen (Lassiter & others, 2002).

In courtrooms, most confession videotapes focus on the confessor. As we might expect, noted Daniel Lassiter and Kimberly Dudley (1991), such tapes yield a nearly 100 percent conviction rate when played by prosecutors. Aware of this research, reports Lassiter, New Zealand has made it a national policy that police interroga­ tions be filmed with equal focus on the officer and the suspect, such as by filming them with side profiles of both.

As the once-visible person recedes in their memory, observers often give more and more credit to the situation. As we saw previously in the groundbreaking attri­ bution error experiment by Edward Jones and Victor Harris (1967), immediately after hearing someone argue an assigned position, people assume that’s how the person really felt. Jerry Burger and M. L. Palmer (1991) found that a week later they are much more ready to credit the situational constraints. The day after a pres­ idential election. Burger and Julie Pavelich (1994) asked voters why the election turned out as it did. Most attributed the outcome to the candidates’ personal traits and positions (the winner from the incumbent party was likable). When they asked

Social Beliefs and Judgments Chapter 3 107

other voters the same question a year later, only a third attributed the verdict to the candidates. More people now credited circumstances, such as the country’s good mood and the robust economy.

Let’s make this personal: Are you gen­ erally quiet, talkative, or does it depend on the situation? “Depends on the situ­ ation” is a common answer. Likewise, when asked to predict their feelings 2 weeks after receiving grades or learn­ ing the outcome of their country’s national election, people expect the situ­ ation to rule their emotions; they under­ estimate the importance of their own sunny or dour dispositions (Quoidbach & Dunn, 2010). But when asked to describe a friend—or to describe what they were like 5 years ago—people more often ascribe trait descriptions. When recalling our past, we become like observers of someone else, note researchers Emily Pronin and Lee Ross (2006). For most of us, the “old you” is someone other than today’s “real you.” We regard our distant past selves (and our distant future selves) almost as if they were other people occupying our body.

All these experiments point to a reason for the attribution error: Wc find causes where we look for them. To see this in your own experience, consider: Would you say your social psychology instructor is a quiet or a talkative person?

My guess is you inferred that he or she is fairly outgoing. But consider: Your attention focuses on your instructor while he or she behaves in a public context that demands speaking. The instructor also observes his or her own behavior in many different situations—in the classroom, in meetings, at home. “Me talkative?” your instructor might say. “Well, it all depends on the situation. When I’m in class or with good friends, Tm rather outgoing. But at conventions and in unfamiliar situ­ ations I feel and act rather shy.” Because we are acutely aware of how our behav­ ior varies with the situation, we see ourselves as more variable than other people (Baxter & Goldberg, 1987; Kammer, 1982; Sande & others, 1988). “Nigel is uptight, Fiona is relaxed. With me it varies.”

CULTURAL DIFFERENCES Cultures also influence attribution error (Ickes, 1980; Watson, 1982). A Western worldview predisposes people to assume that people, not situations, cause events. Internal explanations are more socially approved (Jellison & Green, 1981). “You can do it!” we are assured by the pop psychology of positive-thinking Western culture. You get what you deserve and deserve what you get.

As children grow up in Western culture, they learn to explain behavior in terms of the other’s personal characteristics (Rholes & others, 1990; Ross, 1981). As a first- grader, one of my sons unscrambled the words “gate the sleeve caught Tom on his” into “The gate caught Tom on his sleeve.” His teacher, applying the Western cultural assumptions of the curriculum materials, marked that wrong. The “right” ^swer located the cause within Tom: “Tom caught his sleeve on the gate.”

The fundamental attribution error occurs across varied cultures (Krull & others, 1999). Yet people in Eastern Asian cultures are somewhat more sensitive than West­ erners are to the importance of situations. Thus, when aware of the social context, they are less inclined to assume that others’ behavior corresponds to their traits (Choi & others, 1999; Farwell & Weiner, 2000; Masuda & Kitayama, 2004).

Focusing on the person. Would you infer that your professor for this course, or the professor shown here, is naturally outgoing?










Under alcohol’s influence, people’s attentional focus narrows and they become more likely to attribute someone’s action—perhaps a bump at a bar—to intentionality (Begue & others, 2010). Thinking that a jolt or seeming insult was intentional may then trigger an aggravated reaction.

108 Part One

Whether conservatives or liberals offer more situational attributions depends on the topic. When explaining poverty, liberals offer stronger situational attributions. When explaining U.S. Marines’ killing of Iraqi civilians, conservatives offer stronger situational attributions (Morgan & others, 2010).

FIGURE:: 3.7 Attributions and Reactions How we explain someone’s nega­ tive behavior determines how we feel about it.

Social Thinking

Some languages promote external attributions. Instead of “1 was late/’ Spanish idiom allows one to say, “The clock caused me to be late.” In collectivist cultures, people less often perceive others in terms of personal dispositions (Lee & others, 1996; Zebrowitz-McArthur, 1988). They are less likely to spontaneously interpret a behavior as reflecting an inner trait (Newman, 1993). When told of someone’s actions, Hindus in India are less likely than Americans to offer dispositional explanations (“She is kind”) and more likely to offer situational explanations (“Her friends were with her”) (Miller, 1984).

The fundamental attribution error is fundamental because it colors our explana­ tions in basic and important ways. Researchers in Britain, India, Australia, and the United States have found that people’s attributions predict their attitudes toward the poor and the unemployed (Furnham, 1982; Pandey & others, 1982; Skitka, 1999; Wagstaff, 1983; Weiner & others, 2011). Those who attribute poverty and unem­ ployment to personal dispositions (“They’re just lazy and undeserving”) tend to adopt political positions unsympathetic to such people (Figure 3.7). This disposi­ tional attribution ascribes behavior to the person’s disposition and traits. Those who make situational attributions (“If you or I were to live with the same overcrowding, poor education, and discrimination, would we be any better off?”) tend to adopt political positions that offer more direct support to the poor. Tell me your attribu­ tions for poverty and I will guess your politics.

Can we benefit from being aware of the attribution error? I once assisted with some interviews for a faculty position. One candidate was interviewed by six of us at once; each of us had the opportunity to ask two or three questions. I came away thinking, “What a stiff, awkward person he is.” The second candidate I met pri­ vately over coffee, and we immediately discovered we had a close, mutual friend. As we talked, I became increasingly impressed by what a “warm, engaging, stimu­ lating person she is.” Only later did I remember the fundamental attribution error and reassess my analysis. I had attributed his stiffness and her warmth to their dis­ positions; in fact, I later realized, such behavior resulted partly from the difference in their interview situations.

WHY WE STUDY ATTRIBUTION ERRORS This chapter, like the one before it, explains some foibles and fallacies in our social thinking. Reading about these may make it seem, as one of my students put it.

Dispositional attribution (The man is a

hostile person.)

Unfavorable reaction

(I don’t like this man.)

Negative behavior^ (A man is rude to his


Situational attribution (The man was unfair


Sympathetic reaction

(I can understand.)

109Social Beliefs and Judgments

dtat “social psychologists get their kicks out of playing tricks on people.” Actu­ ally/ the experiments are not designed to demonstrate “what fools these mortals be” (although some of the experiments are rather amusing). Rather, their purpose is to reveal how we think about ourselves and others.

If our capacity for illusion and self-deception is shocking, remember that our modes of thought are generally adaptive. Illusory thinking is often a by-product of our mind’s strategies for simplifying complex information. It parallels our perceptual mechanisms, which generally give us useful images of the world but sometimes lead us astray.

A second reason for focusing on thinking biases such as the fundamental attri­ bution error is humanitarian. C^e of social psychology’s “great humanizing mes­ sages,” note Thomas Gilovich and Richard Eibach (2001), is that people should not always be blamed for their problems. “More often than people are willing to acknowledge,” they conclude, “failure, disability, and misfortune are … the prod­ uct of real environmental causes.”

A third reason for focusing on biases is that we are mostly unaware of them and can benefit from greater awareness. As with other biases, such as the self- serving bias (see Chapter 2), people see themselves as less susceptible than oth­ ers to attribution errors (Pronin, 2008). My hunch is that you will find more surprises, more challenges, and more benefit in an analysis of errors and biases than you would in a string of testimonies to the human capacity for logic and intellectual achievement. That is also why world literature so often portrays pride and other human failings. Social psychology aims to expose us to fallacies in our thinking in the hope that we will become more rational, more in touch with reality.

Chapter 3










SUMMING UP: How Do We Explain Our Social Worlds? • Attribution theory involves how we explain people’s

behavior. Misattribution—attributing a behavior I to the wrong source—is a major factor in sexual ■ harassment, as a person in power (typically male) K interprets friendliness as a sexual come-on.

• Although we usually make reasonable attributions, we often commit the fundamental attribution error (also called correspondence bias) when explaining other people’s behavior. We attribute their behavior

so much to their inner traits and attitudes that we discount situational constraints, even when those are obvious. We make this attribution error partly because when we watch someone act, that person is the focus of our attention and the situation is rela­ tively invisible. When we act, our attention is usu­ ally on what we are reacting to—the situation is more visible.


l^ain insight into how our social beliefs matter.

Having considered how we explain and judge others—efficiently, adaptively, but sometimes erroneously—we conclude this chapter by pondering the effects of our social judgments. Do our social beliefs matter? Can they change reality?

Our social beliefs and judgments do matter. They influence how we feel and ^ct, and by so doing may help generate their own reality. When our ideas lead us *0 act in ways that produce their apparent confirmation, they have become what sociologist Robert Merton (1948) termed self-fulfilling prophecies—^beliefs that

self-fulfilling prophecy A belief that leads to its own fulfillment.

110 Part One Social Thinking

focus ON The Self-Fulfilling Psychology of the Stock Market

On the evening of January 6, 1981, Joseph Granville, a popular Florida investment adviser, wired his clients; “Stock prices will nosedive; sell tomorrow.” Word of Granville’s advice soon spread, and January 7 became the heaviest day of trading in the previous history of the New York Stock Exchange. All told, stock values lost $40 billion.

Nearly a half-century ago, John Maynard Keynes likened such stock market psychology to the popular beauty contests then conducted by London newspapers. To win, one had to pick the six faces out of a hundred that were, in turn, chosen most frequently by the other newspaper contestants. Thus, as Keynes wrote, “Each competitor has to pick not those faces which he himself finds prettiest, but those which he thinks likeliest to catch the fancy of the other competitors.”

Investors likewise try to pick not the stocks that touch their fancy but the stocks that other investors will favor. The name of the game is predicting others’ behavior. As one Wall Street fund manager explained, “You may or may not agree with Granville’s view—but that’s usually beside the point.” If you think his advice will cause others

to sell, then you want to sell quickly, before prices drop more. If you expect others to buy, you buy now to beat the rush.

The self-fulfilling psychology of the stock market worked to an extreme on Monday, October 19, 1987, when the Dow Jones Industrial Average lost 20 percent. Part of what happens during such crashes is that the media and the rumor mill focus on whatever bad news is available to explain them. Once reported, the explana­ tory news stories further diminish people’s expectations, causing declining prices to fall still lower. The process also works in reverse by amplifying good news when stock prices are rising.

In April of 2000, the volatile technology market again demonstrated a self-fulfilling psychology, now called “momentum investing.” After 2 years of eagerly buying stocks (because prices were rising), people started fran­ tically selling them (because prices were falling). Such wild market swings—”irrational exuberance” followed by a crash—are mainly self-generated, noted economist Robert Shiller (2005). In 2008 and 2009, the market psy­ chology headed south again as another bubble burst.

…. . ……. . ” …………….. ………- • …………………………………………………………………….. •” •’ ……….

Rosenthal (2008) recalls submitting a paper describing his early experiments on experimenter bias to a leading journal and to an American Association for the Advancement of Science prize competition. On the same day, some weeks later, he received a letter from the journal rejecting his paper, and from the association naming it the year’s best social science research. In science, as in everyday life, some people appreciate what others do not, which is why it often pays to try and, when rebuffed, to try again.

lead to their own fulfillment. If, led to believe that their bank is about to crash, its customers race to withdraw their money, then their false perceptions may create reality, noted Merton. If people are led to believe that stocks are about to soar, they will indeed. (See “Focus On: The Self-Fulfilling Psychology of the Stock Market.”)

In his well-known studies of experimenter bias, Robert Rosenthal (1985, 2006) found that research participants sometimes live up to what they believe experi­ menters expect of them. In one study, experimenters asked individuals to judge the success of people in various photographs. The experimenters read the same instruc­ tions to all their participants and showed them the same photos. Nevertheless, experimenters who expected their participants to see the photographed people as successful obtained higher ratings than did those who expected their participants to see the people as failures. Even more startling—and controversial—are reports that teachers’ beliefs about their students similarly serve as self-fulfilling prophe­ cies. If a teacher believes a student is good at math, will the student do well in the class? Let’s examine this.

Teacher Expectations and Student Performance Teachers do have higher expectations for some students than for others. Perhaps you have detected this after having a brother or sister precede you in school, after receiving a label such as “gifted” or “learning disabled,” or after being tracked with “high-ability” or “average-ability” students. Perhaps conversation in the teachers’ lounge sent your reputation ahead of you. Or perhaps your new teacher scrutinized

Social Beliefs and Judgments Chapter 3 111


i your school file or discovered your family’s social status. It’s clear that teachers’ evaluations correlate with student achievement: Teachers think well of students who do well. That’s mostly because teachers accurately perceive their students’ abilities and achievements. “About 75 percent of the correlation between teacher expectations and student future achievement reflects accuracy,” report Lee Jussim, Stacy Robustelli, and Thomas Cain (2009).

But are teachers’ evaluations ever a cause as well as a consequence of student performance? One correlational study of 4300 British schoolchildren by William Crano and Phyllis Mellon (1978) suggested yes. Not only is high performance fol­ lowed by higher teacher evaluations but also the reverse is true as well.

Could we test this “teacher-expectations effect” experimentally? Pretend we gave a teacher the impression that Dana, Sally, Todd, and Manuel—four randomly selected students—are unusually capable. Will the teacher give special treatment to these four and elicit superior performance from them? In a now-famous experi­ ment, Rosenthal and Lenore Jacobson (1968) reported precisely that. Randomly selected children in a San Francisco elementary school who were said (on the basis of a fictitious test) to be on the verge of a dramatic intellectual spurt did then spurt ahead in IQ score.

That dramatic result seemed to suggest that the school problems of “disadvan­ taged” children might reflect their teachers’ low expectations. The findings were soon publicized in the national media as well as in many college textbooks. How­ ever, further analysis—which was not as highly publicized—revealed the teacher- expectations effect to be not as powerful and reliable as this initial study had led many people to believe (Jussim & others, 2009; Spitz, 1999). By Rosenthal’s own count, in only approximately 4 in 10 of the nearly 500 published experiments did expectations significantly affect performance (Rosenthal, 1991, 2002). Low expecta­ tions do not doom a capable child, nor do high expectations magically transform a slow learner into a valedictorian. Human nature is not so pliable.

High expectations do, however, seem to boost low achievers, for whom a teach­ er’s positive attitude may be a hope-giving breath of fresh air (Madon & others, 1997). How are such expectations transmitted? Rosenthal and other investigators report that teachers look, smile, and nod more at “high-potential students.” Teach­ ers also may teach more to their “gifted” students, set higher goals for them, call on them more, and give them more time to answer (Cooper, 1983; Harris & Rosenthal, 1985,1986; Jussim, 1986).

In one study, Elisha Babad, Frank Bernieri, and Rosenthal (1991) videotaped teachers talking to, or about, unseen students for whom they held high or low expectations. A random 10-second clip of either the teacher’s voice or the teacher’s face was enough to tell viewers—^both children and adults—whether this was a good or a poor student and how much the teacher liked the student. (You read that right: 10 seconds.) Although teachers may think they can conceal their feelings and behave impartially toward the class, students are acutely sensitive to teachers’ facial expressions and body movements (Figure 3.8).

Self-presumed expectations associated with one’s gender (“women are bad at math”) or race (“Blacks don’t do so well on aptitude tests”) can create anxiety that suppresses test scores. Remove the “stereotype threat” (see Chapter 9) and performance may improve.

Teacher’s expectation Teacher’s behavior Student’s behavior I “Rena’s older brother was brilliant. I bet she is. too.”

Smiling more at Rena, teaching her more, calling on her more, giving more time to answer.

Rena responds enthusiastically.

FIGURE :: 3.8 Self-Fulfilling Prophecies Teacher expectations can become self-fulfilling prophecies. But for the most part, teachers’ expectations accurately reflect reality (Jussim & Harber, 2005).


112 Part One Social Thinking

Reading the experiments on teacher expectations makes me wonder about the effect of students’ expectations upon their teachers. You no doubt begin many of your courses having heard “Professor Smith is interesting” and “Professor Jones is a bore.” Robert Feldman and Thomas Prohaska (1979; Feldman & Theiss, 1982) found that such expectations can affect both student and teacher. Students in a learning experiment who expected to be taught by an excellent teacher perceived their teacher (who was unaware of their expectations) as more competent and inter­ esting than did students with low expectations. Furthermore, the students actually learned more. In a later experiment, women who were led to expect their male instructor to be sexist had a less positive experience with him, performed worse, and rated him as less competent than did women not given the sexist expectation (Adams & others, 2006).

Were these results due entirely to the students’ perceptions or also to a self- fulfilling prophecy that affected the teacher? In a follow-up experiment, Feldman and Prohaska videotaped teachers and had obser\’ers rate their performances. Teachers were judged most capable when assigned a student who nonverbally con­ veyed positive expectations.

To see whether such effects might also occur in actual classrooms, a research team led by David Jamieson (Jamieson & others, 1987) experimented with four Ontario high school classes taught by a newly transferred teacher. During individ­ ual interviews, they told students in two of the classes that both other students and the research team rated the teacher very highly. Compared with the control classes, students who were given positive expectations paid better attention during class. At the end of the teaching unit, they also got better grades and rated the teacher as clearer in her teaching. The attitudes that a class has toward its teacher are as important, it seems, as the teacher’s attitude toward the students.

Getting from Others What We Expect So the expectations of experimenters and teachers, although usually reasonably accurate, occasionally act as self-fulfilling prophecies. How widespread are self- fulfilling prophecies? Do we get from others what we expect of them? Studies show that our perceptions of others are more accurate than biased (Jussim, 2012). Self- fulfilling prophecies have “less than extraordinary power.” Yet sometimes, self- fulfilling prophecies do operate in work settings (with managers who have high or low expectations), in courtrooms (as judges instruct juries), and in simulated police contexts (as interrogators with guilty or innocent expectations interrogate and pres­ sure suspects) (Kassin & others, 2003; Rosenthal, 2003,2006).

Do self-fulfilling prophecies color our personal relationships? There are times when negative expectations of someone lead us to be extra nice to that person, which induces him or her to be nice in return—thus rfisconfirming our expectations. But a more common finding in studies of social interaction is that, yes, we do to some extent get what we expect (Olson &c others, 1996).

In laboratory games, hostility nearly always begets hostility: People who per­ ceive their opponents as noncooperative will readily induce them to be noncooper­ ative (Kelley & Stahelski, 1970). Each party’s perception of the other as aggressive, resentful, and vindictive induces the other to display those behaviors in self- defense, thus creating a vicious self-perpetuating circle. In another experiment, people anticipated interacting with another person of a different race. When led to expect that the person disliked interacting with someone of their race, they felt more anger and displayed more hostility toward the person (Butz & Plant, 2006). Likewise, whether I expect my wife to be in a bad mood or in a loving mood may affect how I relate to her, thereby inducing her to confirm my belief.

So, do intimate relationships prosper when partners idealize each other? Are positive illusions of the other’s virtues self-fulfilling? Or are they more often self- defeating, by creating high expectations that can’t be met? Among University of

To judge a teacher or professor’s overall warmth and enthusiasm also takes but a thin slice of behavior— mere seconds (Ambady & Rosenthal, 1992, 1993).

Social Beliefs and Judgments Chapter 3 113

Waterloo dating couples followed by Sandra Murray and associates (1996a, 1996b, 2000), positive ideals of one’s partner were good omens. Idealization helped buffer conflict, bolster satisfaction, and turn self- perceived frogs into princes or princesses. When some­ one loves and admires us, it helps us become more the person he or she imagines us to be.

When dating couples deal with conflicts, hope­ ful optimists and their partners tend to perceive each other as engaging constructively. Compared to those with more pessimistic expectations, they then feel more supported and more satisfied with the outcome (Srivastava & others, 2006). Among married couples, too, those who worry that their partner doesn’t love and accept them interpret slight hurts as rejections, which motivates them to devalue the partner and dis­ tance themselves. Those who presume their partner’s love and acceptance respond less defensively, read less into stressful events, and treat the partner better (Murray & others, 2003). Love helps create its pre­ sumed reality.

Several experiments conducted by Mark Snyder (1984) at the University of Minnesota show how, once formed, erroneous beliefs about the social world can induce others to confirm those beliefs, a phenomenon called behavioral confirmation. In a classic study, Snyder, Elizabeth Tanke, and Ellen Berscheid (1977) had male students talk on the telephone with women they thought (from having been shown a picture) were either attractive or unattractive. Analysis of just the women’s comments during the conversations revealed that the supposedly attractive women spoke more warmly than the supposedly unattract­ ive women. The men’s erroneous beliefs had become a self-fulfilling prophecy by leading them to act in a way that influenced the women to fulfill the men’s stereo­ type that beautiful people are desirable people.

Behavioral confirmation also occurs as people interact with partners holding mistaken beliefs. People who are believed lonely behave less sociably (Rotenberg & others, 2002). People who believe they are accepted and liked (rather than disliked) then behave warmly—and do get accepted and liked (Stinson & others, 2009). Men who are believed sexist behave less favorably toward women (Pinel, 2002). Job interviewees who are believed to be warm behave more warmly.

Imagine yourself as one of the 60 young men or 60 young women in an experi­ ment by Robert Ridge and Jeffrey Reber (2002). Each man is to interview one of the Women to assess her suitability for a teaching assistant position. Before doing so, he IS told either that she feels attracted to him (based on his answers to a biographical questionnaire) or not attracted. (Imagine being told that someone you were about to meet reported considerable interest in getting to know you and in dating you, or none whatsoever.) The result was behavioral confirmation: Applicants believed to feel an attraction exhibited more flirtatiousness (without being aware of doing so). Ridge and Reber believe that this process, like the misattribution phenomenon dis­ cussed previously, may be one of the roots of sexual harassment. If a woman’s behavior seems to confirm a man’s beliefs, he may then escalate his overtures until they become sufficiently overt for the woman to recognize and interpret them as ^appropriate or harassing.

Expectations influence children’s behavior, too. After observing the amount of litter in three classrooms, Richard Miller and colleagues (1975) had the teacher and others repeatedly tell one class that they should be neat and tidy. This persuasion ^creased the amount of litter placed in wastebaskets from 15 to 45 percent, but ®rdy temporarily. Another class, which also had been placing only 15 percent of

Behavioral confirmation. If each of these people feels attracted to the other, but presumes that feeling isn’t reciprocated, they may each act cool to avoid feeling rejected—and decide that the other’s coolness confirms the presumption. Danu Stinson and colleagues (2009) note that such “self-protective inhibition of warmth” dooms some would-be relationships.

behavioral confirmation A type of self-fulfilling prophecy whereby people’s social expectations lead them to behave in ways that cause others to confirm their expectations.













114 Part One Social Thinking

its litter in wastebaskets, was repeatedly congratulated for being so neat and tidy. After 8 days of hearing this, and still 2 weeks later, these children were fulfilling the expectation by putting more than 80 percent of their litter in wastebaskets. Tell children they are hardworking and kind (rather than lazy and mean), and they may live up to their labels.

These experiments help us understand how social beliefs, such as stereotypes about people with disabilities or about people of a particular race or sex, may be self-confirming. How others treat us reflects how we and others have treated them.

SUMMING UP: How Do Our Expectations of Our Social Worlds Matter?

• Our beliefs sometimes take on lives of their own. Usually, our beliefs about others have a basis in reality. But studies of experimenter bias and teacher expectations show that an erroneous belief that cer­ tain people are unusually capable (or incapable) can lead teachers and researchers to give those people special treatment. This may elicit superior

(or inferior) performance and, therefore, seem to confirm an assumption that is actually false.

• Similarly, in everyday life we often get behavioral confirmation of what we expect. Told that someone we are about to meet is intelligent and attractive, we may come away impressed with just how intel­ ligent and attractive he or she is.


View human nature through cognitive social psychology.

Social cognition studies reveal that our information-processing powers are impres­ sive for their efficiency and adaptiveness (“in apprehension how like a god!” exclaimed Shakespeare’s Hamlet). Yet we are also vulnerable to predictable errors and misjudgments (“headpiece filled with straw,” said T. S. Eliot). What practical lessons, and what insights into human nature, can we take home from this research?

We have reviewed reasons why people sometimes form false beliefs. We cannot easily dismiss these experiments: Most of their participants were intelligent people, often students at leading universities. Moreover, people’s intelligence scores are uncorrelated with their vulnerability to many different thinking biases (Stanovich & West, 2008). One can be very smart and exhibit seriously bad judgment.

Trying hard also doesn’t eliminate thinking biases. These predictable distortions and biases occurred even when payment for right answers motivated people to think optimally. As one researcher concluded, the illusions “have a persistent qual­ ity not unlike that of perceptual illusions” (Slovic, 1972).

Research in cognitive social psychology thus mirrors the mixed review given humanity in literature, philosophy, and religion. Many research psychologists have spent lifetimes exploring the awesome capacities of the human mind. We are smart enough to have cracked our own genetic code, to have invented talking computers, and to have sent people to the moon. Three cheers for human reason.

Well, two cheers—because the mind’s premium on efficient judgment makes our intuition more vulnerable to misjudgment than we suspect. With remarkable ease, we form and sustain false beliefs. Led by our preconceptions, feeling over­ confident, persuaded by vivid anecdotes, perceiving correlations and control even

115Social Beliefs and Judgments

where none may exist, we construct our social beliefs and then influence others to confirm them. “The naked intellect,” observed novelist Madeleine L’Engle, “is an extraordinarily inaccurate instrument.”

But have these experiments just been intellectual tricks played on hapless par­ ticipants, thus making them look worse than they are? Richard Nisbett and Lee Ross (1980) contended that, if anything, laboratory procedures overestimate our intuitive powers. The experiments usually present people with clear evidence and warn them that their reasoning ability is being tested. Seldom does real life say to us: “Here is some evidence. Now put on your intellectual Sunday best and answer these questions.”

Often our everyday failings are inconsequential, but not always so. False impres­ sions, interpretations, and beliefs can produce serious consequences. Even small biases can have profound social effects when we are making important social judg­ ments: Why are so many people homeless? Unhappy? Homicidal? Does my friend love me or my money? Cognitive biases even creep into sophisticated scientific thinking. Human nature has hardly changed in the 3000 years since the Old Testa­ ment psalmist noted that “no one can see his own errors.”

Is this too cynical? Leonard Martin and Ralph Erber (2005) invite us to imagine that an intelligent being swooped down and begged for information that would help it understand the human species. When you hand it this social psychology text, the alien says “thank you” and zooms back off into space. How would you feel about having offered social psychology’s analysis of human life? Joachim Krueger and David Funder (2003a, 2003b) wouldn’t feel too good. Social psychology’s pre­ occupation with human foibles needs balancing with “a more positive view of human nature,” they argue.

Fellow social psychologist Lee Jussim (2005, 2012) agrees, adding, “Despite the oft demonstrated existence of a slew of logical flaws and systematic biases in lay judgment and social perception, such as the fundamental attribution error, false consensus, over-reliance on imperfect heuristics, self-serving biases, etc., people’s perceptions of one another are surprisingly (though rarely perfectly) accurate.” The elegant analyses of the imperfections of our thinking are themselves a trib­ ute to human wisdom. Were one to argue that all human thought is illusory, the assertion would be self-refuting, for it, too, would be but an illusion. It would be logically equivalent to contending “All generalizations are false, including this one.”

As medical science assumes that any given body organ serves a function, so behavioral scientists find it useful to assume that our modes of thought and behav­ ior are adaptive. The rules of thought that produce false beliefs and deficient intu­ ition usually serve us well. Frequently, the errors are a by-product of our mental shortcuts that simplify the complex information we receive.

Nobel laureate psychologist Herbert Simon (1957) was among the modem researchers who first described the bounds of human reason. Simon contends that to cope with reality, we simplify it. Consider the complexity of a chess game: The number of possible games is greater than the number of particles in the universe. How do we cope? We adopt some simplifying rules—heuristics. These heuristics sometimes lead us to defeat. But they do enable us to make efficient snap judgments.

Illusory thinking can likewise spring from useful heuristics that aid our survival. In many ways, heuristics do make us smart (Gigerenzer & Gaissmaier, 2011). The belief in our power to control events helps maintain hope and effort. If things are sometimes subject to control and sometimes not, we maximize our outcomes by positive thinking. Optimism pays dividends. We might even say that our beliefs are like scientific theories—sometimes in error yet useful as generalizations. As social psychologist Susan Fiske (1992) says, “Thinking is for doing.”

Might we reduce errors in our social thinking? In school, math teachers teach, teach, teach until the mind is finally trained to process numerical information accu­ rately and automatically. We assume that such ability does not come naturally;

Chapter 3



















116 Part One Social Thinking













otherwise, why bother with the years of training? Research psychologist Robyn Dawes (1980a, 1980b)—who was dismayed that “study after study has shown [that] people have very limited abilities to process information on a conscious level, par­ ticularly social information”—suggested that we should also teach, teach, teach how to process social information.

Richard Nisbett and Lee Ross (1980) have agreed that education could indeed reduce our vulnerability to certain types of error. They offer the following recommendations:

• Train people to recognize likely sources of error in their own social intuition. • Set up statistics courses geared to everyday problems of logic and social

judgment. Given such training, people do in fact reason better about every­ day events (Lehman & others, 1988; Nisbett & others, 1987).

• Make such teaching more effective by illustrating it richly with concrete, vivid anecdotes and examples from everyday life.

• Teach memorable and useful slogans, such as “It’s an empirical question,” “Which hat did you draw that sample out of?” or “You can lie with statistics, but a well-chosen example does the job better.”

SUMMING UP: What Can We Conclude About Social Beliefs and Judgments?

Research on social beliefs and judgments reveals how psychology will therefore appreciate both the powers we form and sustain beliefs that usually serve us and the perils of social thinking, well but sometimes lead us astray. A balanced social





DUCK, 1884

POSTSCRIPT: Reflecting on Illusory Thinking Is research on pride and error too humbling? Surely we can acknowledge the hard truth of our human limits and still sympathize with the deeper message that people are more than machines. Our subjective experiences are the stuff of our humanity— our art and our music, our enjoyment of friendship and love, our mystical and reli­ gious experiences.

The cognitive and social psychologists who explore illusory thinking are not out to remake us into unfeeling logical machines. They know that emotions enrich human experience and that intuitions are an important source of creative ideas. They add, however, the humbling reminder that our susceptibility to error also makes clear the need for disciplined training of the mind. The American writer Norman Cousins (1978) called this “the biggest truth of all about learning: that its purpose is to unlock the human mind and to develop it into an organ capable of thought—conceptual thought, analytical thought, sequential thought.”

Research on error and illusion in social judgment reminds us to “judge not”—to remember, with a dash of humility, our potential for misjudgment. It also encour­ ages us not to feel intimidated by the arrogance of those who cannot see their own potential for bias and error. We humans are wonderfully intelligent yet fallible creatures. We have dignity but not deity.

Social Beliefs and Judgments Chapter 3 117

Such humility and distrust of human authority is at the heart of both religion and science. No wonder many of the founders of modem science were religious people whose convictions predisposed them to be humble before nature and skeptical of human authority (Hooykaas, 1972; Merton, 1938). Science always involves an inter­ play between intuition and rigorous test, between creative hunch and skepticism. To sift reality from illusion requires both open-minded curiosity and hard-headed rigor. This perspective could prove to be a good attitude for approaching all of life: to be critical but not cynical, curious but not gullible, open but not exploitable.

“The ancestor of every action is a thought.” ………………………………… —Ralph .Waldo Emerson, .Essays, f/rs.t Series, .184.1.

How well do our attitudes predict our behavior?

When does our behavior affect our attitudes?

Why does our behavior affect our attitudes?

Postscript: Changing ourselves through action

What is the relationship between what we are (on the inside) and what we do (on the outside)? Philosophers, theologians, and educators speculate about the connections between attitude

and action, character and conduct, and private word and public deed.

Underlying most teaching, counseling, and child rearing is an assump­

tion: Our private beliefs and feelings determine our public behav­

ior; so if we want to change behavior, we must first change hearts

and minds.

In the beginning, social psychologists agreed; To know people’s

attitudes is to predict their actions. As demonstrated by genocidal kill­

ers and by suicide terrorists, extreme attitudes can produce extreme

behavior. Countries whose people detest another country’s lead­

ers are more likely to produce terrorist acts against them (Krueger &

Maleckova, 2(X)9). Hateful attitudes spawn violent behavior.

But in 1964, Leon Festinger concluded that the evidence showed

that changing people’s attitudes hardly affects their behavior.

Festinger believed the attitude-behavior relation works the other way

around. As Robert Abelson (1972) put it, we are “very well trained and

very good at finding reasons for what we do, but not very good at

doing what we find reasons for.” This chapter explores the interplay of

attitudes and behavior.

120 Part One Social Thinking

attitude A favorable or unfavorable evaluative reaction toward something or someone (often rooted in one’s beliefs, and exhibited in one’s feelings and intended behavior).



—BUDDHA, 563 B.C.-483 B.C,





GRAY, 1926

FIGURE:: 4.1 The ABCs of Attitudes

When social psychologists talk about someone’s attitude, they refer to beliefs and

feelings related to a person or an event and the resulting behavior tendency. Taken

together, favorable or unfavorable evaluative reactions toward something—often

rooted in beliefs and exhibited in feelings and inclinations to act—define a person’s

attitude {Eagly & Chaiken, 2(X)5). Thus, a person may have a negative attitude toward

coffee, a neutral attitude toward the French, and a positive attitude toward the next-

door neighbor.

Attitudes efficiently size up the world. When we have to respond quickly to some­

thing, the way we feel about it can guide how we react. For example, a person who

believes a particular ethnic group is lazy and aggressive may feel dislike for such peo­

ple and therefore intend to act in a discriminatory manner. You can remember these

three dimensions as the ABCs of attitudes; Affect (feelings), Behavior tendency, and

Cognition (thoughts) (Figure 4.1).

The study of attitudes is central to social psychology and was one of its first con­

cerns. For much of the last century, researchers wondered how much our attitudes

affect our actions.


State the extent to which, and under what conditions, our inner attitudes drive our outward actions.

A blow to the supposed power of attitudes came when social psychologist Allan Wicker (1969) reviewed several dozen research studies covering a variety of people, attitudes, and behaviors. Wicker offered a shocking conclusion: People’^s expressed attitudes hardly predicted their varying behaviors.

• Student attitudes toward cheating bore little relation to the likelihood of their actually cheating.

• Attitudes toward the church were only modestly linked with worship atten­ dance on any given Sunday.

• Self-described racial attitudes provided little clue to behaviors in actual situa­ tions. Many people say they express being upset with someone making racist


remarks; yet, when they hear racism (such as someone using the N-word) respond indifferently (Kawakami & others, 2009).

The disjuncture between attitudes and actions is what Daniel Batson and his colleagues (1997, 2001, 2002; Valdesolo & DeSteno, 2007, 2008) call “moral hypocrisy” (appearing moral while avoiding the costs of being so). Their studies presented people with

Behavior and Attitudes Chapter 4 121

an appealing task with a possible $30 prize and a dull task with no rewards. The participants had to assign themselves to one of the tasks and a supposed second participant to the other. Only 1 in 20 believed that assigning the positive task to themselves was the more moral thing to do, yet 80 percent did so. In follow-up experiments, participants were given coins they could flip pri­ vately if they wanted. Even if they chose to flip, 90 percent assigned themselves to the positive task! (Was that because they could specify the consequences of heads and tails after the coin toss?) In another experiment, Batson put a sticker on each side of the coin, indicating what the flip outcome would signify. Still, 24 of 28 people who made the toss assigned themselves to the posi­ tive task. When morality and greed were put on a collision course, greed usually won.

If people don’t walk the same line that they talk, it’s little wonder that attempts to change behavior by changing attitudes often fail. Warnings about the dangers of smoking affect only minimally those who already smoke. Increasing public aware­ ness of the desensitizing and brutalizing effects of television violence has stimu­ lated many people to voice a desire for less violent programming—yet they still watch media murder as much as ever. Sex education programs have often influ­ enced attitudes toward abstinence and condom use without affecting long-term abstinence and condom use behaviors. We are, it seems, a population of hypocrites.

All in all, the developing picture of what controls behavior emphasized external social influences, such as others’ behavior and expectations, and played down inter­ nal factors, such as attitudes and personality. Thus, the original thesis that attitudes determine actions was countered during the 1960s by the antithesis that atti­ tudes determine virtually nothing.

Thesis. Antithesis. Is there a synthesis? The surprising finding that what people say often differs from what they do sent social psychologists scurrying to find out why. Surely, we reasoned, convictions and feelings sometimes make a difference.

Indeed. In fact, what I am about to explain now seems so obvious that I wonder why most social psychologists (myself included) were not thinking this way before the early 1970s. I must remind myself, however, that truth seldom seems obvious until it is known.

Attitudes and behavior misaligned. After former U.S. congressman Mark Souder and staff member Tracey Jackson together recorded a pro-abstinence video, news broke that the two had been having an affair outside of their own marriages. “You’ll go crazy If you don’t have some sense of irony,” the family values advocate told a local newspaper (Elliott 2010).







When Attitudes Predict Behavior The reason—now obvious—why our behavior and our expressed attitudes differ is that both are subject to other influences. Many other influences. One social psy­ chologist counted 40 factors that complicate their relationship (Triandis, 1982; see also Kraus, 1995). Our attitudes do predict our behavior when these other influences on ivhat we say and do are minimal, when the attitude is specific to the behavior, and when the attitude is potent.

WHEN SOCIAL INFLUENCES ON WHAT WE SAY ARE MINIMAL Unlike a physician measuring heart rate, social psychologists never get a direct reading on attitudes. Rather, we measure expressed attitudes. Like other behaviors, expressions are subject to outside influences. Sometimes, for example, we say what we think oth­ ers want to hear. In late 2002, many U.S. legislators, sensing their country’s post-9/11 fear, anger, and patriotic fervor, publicly voted to support President Bush’s planned War against Iraq while privately having reservations (Nagourney, 2002). On the roll- call vote, strong social influence—fear of criticism—had distorted the true sentiments.

122 Part One Social Thinking

THE inside STORY Mahzarin R. Banaji on Discovering Experimental Social Psychology

Graduating from high school in India at age 15,1 had but a single goal—to leave my well-adjusted and secure fam­ ily to live the patently more daring and exciting life of a secretarial assistant. Proficient at typing scores of words a minute, I looked forward to a life of independence that involved living a block away from my parents. My mother, despite not having attended college, persuaded me to try college—but only for a semester, we agreed, after which I would be free to choose my path.

The end of my first semester at Nizam College came and went. Mother didn’t ask about my plans. I didn’t have to swallow and tell. Just before one holiday trip home, I bought the five volumes of the 1968 Handbook of Social Psycho/ogy for the equivalent of a dollar apiece (it seemed like a lot of book for the money). By the end of a 24-hour train ride home, I had polished off one vol­ ume and knew with blunt clarity that this science, which studied social processes experimentally, was something I had to do.

Doctoral and postdoctoral fellowships enabled me to work with three remarkable people early in my career: Tony Greenwald at Ohio State, and Claude Steele and Elizabeth Loftus at the University of Washington. At Yale, while still interested in human memory researchers, I dis­ covered that memories come in both explicit (conscious)

and implicit (unconscious) forms. Might this also be true of attitudes, beliefs, and values? Hesitantly, I wrote the words “Implicit Attitudes” as the title of a grant proposal, not knowing it would become such a central part of what my students and I would study for the next two decades.

With Tony Greenwald and Brian Nosek, I have enjoyed an extended collaboration on implicit social cognition that few scientists are blessed with. From the hundreds of studies that have used the Implicit Association Test ( and the millions of tests taken, we now know that people carry knowledge (stereotypes) and feelings (attitudes) of which they are unaware, and which often contrast with their conscious expressions. We know that subcortical brain activity can be an inde­ pendent marker of implicit atti­ tudes, that people differ in their implicit attitudes, and that such attitudes and stereotypes pre­ dict real-life behavior. Most opti­ mistically, we know that implicit attitudes, even old ones, can be modified by experience.

Mahzarin Banaji

Harvard University

implicit association test (lAT) A computer-driven assessment of implicit attitudes. The test uses reaction times to measure people’s automatic associations between attitude objects and evaluative words. Easier pairings (and faster responses)are taken to indicate stronger unconscious associations.

Today’s social psychologists have some clever means at their disposal for mini­ mizing social influences on people’s attitude reports. Some of these complement traditional self-report measures of explicit (conscious) attitudes with measures of implicit (unconscious) attitudes. One such test measures facial muscle responses to various statements (Cacioppo & Petty, 1981). Those measurements, the researchers hope, can reveal enough of a microsmile or a microfrown to indicate the partici­ pant’s attitude about a given statement.

A newer and widely used attitude measure, the implicit association lest (lAT), uses reaction times to measure how quickly people associate concepts (Greenwald & others, 2002, 2003). One can, for example, measure implicit racial attitudes by assessing whether White people take longer to associate positive words with Black faces than with White faces. Implicit attitude researchers have offered various lAT assessments online ( The some 5 million completed tests since 1998 have, they report, shown that

• Implicit biases are pervasive. For example, 80 percent of people show more implicit negativity toward the elderly compared with the young.

• People differ in implicit bias. Depending on their group memberships, their conscious attitudes, and the bias in their immediate environment, some people exhibit more implicit bias than others.

123Behavior and Attitudes

• People are often unaware of their implicit biases. Despite thinking themselves unprejudiced, even the researchers exhibit some implicit biases (negative associations with various social groups).

Do implicit biases predict behavior? A review of the available research (now over 200 investigations) reveals that both explicit (self-report) and implicit attitudes help predict people’s behaviors and judgments (Greenwald & others, 2008; Nosek & others, 2011). Thus, explicit and implicit attitudes may together predict behav­ ior better than either alone (Spence & Townsend, 2007). The behavior predictions range from dental flossing to the fate of romantic relationships to suicide attempts (Lee & others, 2010; Millar, 2011; Nock & others, 2010). In one study, hiring manag­ ers received job applications that were matched on credential strength, but with one, the applicants’ photos were digitally altered to make them appear obese. Sev­ eral months later, when 153 of the managers completed an lAAT, their automatic anti-obesity bias score predicted which applicants they had invited for interviews (Agerstrom & Rooth, 2011).

For attitudes formed early in life—such as racial and gender attitudes— implicit and explicit attitudes frequently diverge, with implicit attitudes often being the bet­ ter predictor of behavior. For example, implicit racial attitudes have successfully predicted interracial roommate relationships (Towles-Schwen & Fazio, 2006). For other attitudes, such as those related to consumer behavior and support for politi­ cal candidates, explicit self-reports are the better predictor. (See “The Inside Story: Mahzarin R. Banaji on Discovering Experimental Social Psychology.”)

Recent neuroscience studies have identified brain centers that produce our auto­ matic, implicit reactions (Stanley & others, 2008). One area deep in the brain (the amygdala, a center for threat perception) is active as we automatically evaluate social stimuli. For example. White people who show strong unconscious racial bias on the lAT also exhibit high amygdala activation when viewing unfamiliar Black faces rather than White faces. Other frontal lobe areas are involved in detecting and regulating implicit attitudes.

A word of caution: Despite much excitement over these recent studies of implicit attitudes hiding in the mind’s basement, the implicit associations test has detractors (Arkes & Tetlock, 2004; Blanton & others, 2006, 2007, 2009). They note that, unlike an aptitude test, the lAT is not reliable enough for use in assessing and comparing individuals. Moreover, a score that suggests some relative bias doesn’t distinguish a positive bias for one group (or greater familiarity with one group) from a negative bias against another. The critics also wonder whether compassion and guilt rather than latent hostility might slow one’s speed in associating Blacks with positive words. Regardless, the existence of distinct explicit and implicit attitudes confirms one of twenty-first-century psychology’s biggest lessons: our “dual processing” capacity for both controlled (deliberate, conscious, explicit) and automatic (effortless, habitual, implicit) thinking.

WHEN OTHER INFLUENCES ON BEHAVIOR ARE MINIMAL On any occasion, it’s not only our inner attitudes that guide us but also the situ­ ation we face. As Chapters 5 to 8 will illustrate again and again, social influences can be enormous—enormous enough to induce people to violate their deepest con­ victions. So, would averaging many occasions enable us to detect more clearly the impact of our attitudes? Predicting people’s behavior is like predicting a baseball or cricket player’s hitting. ‘The outcome of any particular turn at bat is nearly impos­ sible to predict. But when we aggregate many times at bat, we can compare their approximate batting averages.

To use a research example, people’s general attitude toward religion poorly pre­ dicts whether they will go to worship services during the coming week (because attendance is also influenced by the weather, the worship leader, how one is feel­ ing, and so forth). But religious attitudes predict quite well the total quantity of


Request for a custom paper or place a new order

Assignment Solutions For You


Forget All Your Assignment & Essay Related Worries By Simply Filling Order Form