Datasets:

---

hienbm

/

psychology_dataset

like 0

his or her life. During a recent plane trip, the woman next to me described her relationship with her husband, problems in raising her child, her opinionson drugs, sex education, and abortion —all without a single bit of encourage- ment or comparable disclosure from me. What is notable about this “stranger on the bus ”phenomenon is that it violates society’ s rules for the way social interaction is supposed to progress. Like many social behaviors, the way we reveal information about ourselves isgoverned by a set of unstated but understood rules. Occasionally, parents teachus these rules directly ( “Don’t stare at people ”), but more often we can ’t say how we learned what is expected and what is inappropriate when interactingwith others. One of these social rules is known as disclosure reciprocity . According to this rule, people involved in a get-acquainted conversation reveal information about themselves at roughly the same level of intimacy: I reveal personal information to you as long as you continue to match that level of inti-macy with personal information about yourself. Investigators have demonstrated the rule of disclosure reciprocity in labo- ratory research (Davis, 1977; Taylor & Belgrave, 1986). Undergraduate stu- dents in one study were randomly paired with a member of the same gender whom they did not know (Davis, 1976). The students took turns getting to know one another by volunteering information about themselves. They were given a list of 72 discussion topics, previously ranked for level of intimacy, ranging from fairly trivial to extremely revealing. The winner of a coin toss began by talking for 1 minute on any one of the topics. The partner then talked for 1 minute on any one of the remaining topics. This procedure con- tinued until both partners had spoken 12 times. As shown in Figure 12.1, the students selected increasingly intimate topics as the interaction progressed. They typically began with something safe, perhaps discussing their favorite movies or foods. But they soon moved to more personal areas, such as pro- blems with their parents or ways in which they felt personally inadequate.Moreover, participants tended to match their partners ’intimacy levels. That is, if one person chose an intimate topic, the partner usually responded by se-lecting a similarly intimate topic. In other words, the students in this experi-ment followed the rule of disclosure reciprocity. Other studies show that children as young as 8 years of age seem to understand and follow the reci- procity rule (Cohn & Strassberg, 1983). Why do we reciprocate disclosure intimacy? One reason is that self- disclosure leads to feelings of attraction and trust (Derlega, Winstead, &Greene, 2008). When people disclose information about themselves to us, weare attracted to them, and feelings of trust follow. We respond by disclosing personal information back, thus creating the reciprocity effect. Consistent with this explanation, studies find that we disclose to people we like and we likethose who disclose to us (Collins & Miller, 1994). But disclosure alone does not lead to intimacy and liking. Relationships also require a responsive partner (Laurenceau, Feldman Barrett, & Pietromonaco, 1998; Reis & Patrick, 1996).When partners respond to personal disclosure with signs of caring and by revealing their own feelings, intimacy develops. Failure to respond appropri- ately likely ends the chance that the relationship will blossom.310 CHAPTER 12 The Humanistic Approach

Self-Disclosure Among Friends and Romantic Partners If you apply the disclosure reciprocity rule to recent conversations you ’ve had with friends, you may find that it doesn ’t always work. It ’s quite possible that one of you did most of the talking while the other one just listened. When a friend calls and says “I need to talk, ”we usually don ’t interrupt with per- sonal examples of our own. Researchers find the reciprocity rule doesn ’t al- ways apply to good friends. After a certain level of intimacy is reached in arelationship, we feel free to disclose to friends without requiring reciprocaldisclosure (Altman & Taylor, 1973; Derlega, Wilson, & Chaikin, 1976). One researcher found the highest level of disclosure reciprocity among people who knew each other somewhat, but who were still in the process of develop-ing their relationship (Won-Doornink, 1985). Apparently these individuals had made a commitment to learn more about each other, but they didn ’t know each other well enough to assume that the trust would be there without some sign of assurance. However, these findings do not mean that strangers disclose more to each other than friends. On the contrary, friends are much more likely to talk about such intimate topics as their relationships, self-concepts, and sexual Mean Intimacy Scale Value Trials123456789 1 0 1 1 1 24.04.55.05.5 FIGURE 12.1 Progression of Intimacy During Dyad Conversation Source: From “Self-disclosure in an acquaintance exercise: Responsibility for level of intimacy,” by J. D. Davis, Journal of Personality and Social Psychology , 1976, 33, 787 –792.Self-Disclosure 311

experiences (Bauminger, Finzi-Dottan, Chason, & Har-Even, 2008). In one demonstration of this difference, researchers recorded (with permission) thetelephone conversations of female college students (Hornstein & Truesdell, 1988). The students talked about significantly more intimate information when they interacted with friends than when they spoke on the phone with someone they identified as only an acquaintance. Conversations among good friends also include many noticeable signs of intimacy that are lacking in con- versations with strangers (Hornstein, 1985). These signs include the use of fa-miliar terms, laughing at similar points, and understanding when to speak and when the conversation is coming to an end. Willingness to self-disclose is also related to how easily one makes friends. One team of researchers asked incoming college freshmen how will-ing they were to reveal negative emotions like anxiety, fear, and sadness to others (Graham, Huang, Clark, & Helgeson, 2008). Students who were rela-tively willing to disclose their emotions to others developed more and more intimate social relationships during their first semester on campus than stu- dents who were reluctant to reveal this information about themselves. Studies with couples in long-term romantic relationships find similar pat- terns. The amount of self-disclosure in a marriage is a strong predictor of re- lationship satisfaction (Farber & Sohn, 2007; Harvey & Omarzu, 1997; Sprecher & Hendrick, 2004). The more couples talk to one another about what’s personal and important to them, the better each of them feels about the marriage. Of course, it also may be that couples disclose because they feel good about each other. However, it is not the case that people who dis- close a lot necessarily have more success at romance. Rather, researchers find that couples in good relationships have selectively chosen one another to dis- close to rather than being high disclosers generally (Prager, 1986). And, as with good friends, married couples do not feel the need to reciprocate their partner ’s disclosure during every conversation (Morton, 1978). Disclosing Men and Disclosing Women Not long ago, my wife made an interesting observation about one of my male friends. “He interacts with people like a woman, ”she said. I immediately un- derstood her point. My friend ’s voice is deep and masculine, and he doesn ’t use feminine hand gestures. But he often fills our conversations with fairly re-vealing information about his thoughts and feelings. This behavior would be appropriate, my wife continued, if my friend were a woman. But high levelsof self-disclosure struck her as unusual for a man. Consistent with these observations, investigators find that women typi- cally disclose more intimately and to more people than do men (Dindia &Allen, 1992). Some psychologists argue that men learn as they grow up to limit personal disclosures (Jourard, 1971). They avoid talking about their true feelings out of fear of being ridiculed or rejected. Participants in onestudy read about someone who was either highly disclosing or not very dis- closing about personal problems (Derlega & Chaikin, 1976). Half the partici- pants thought they were reading about a man, and half thought the person was a woman. The participants who thought they were reading about a312 CHAPTER 12 The Humanistic Approach

female rated that person better adjusted when she was disclosing. However, when they thought the discloser was a male, the disclosure was seen as a signof poor psychological adjustment. Other studies suggest at least a few exceptions to this rule. The freedom women feel to disclose may be limited by the nature of what they are talkingabout. Highly disclosing women in one study were liked more when they talked about their parents or about their sexual attitudes. However, women who disclosed about their personal aggressiveness were liked less (Kleinke &Kahn, 1980). Similarly, self-disclosing men are seen as well adjusted as long as they talk about masculine topics (Cunningham, Strassberg, & Haan, 1986). In other words, men and women are more likely to be accepted when they dis-close within the appropriate societal roles for their gender. For men this usually means withholding information; for women it means being open and disclos- ing, but only on topics society deems appropriate. The result is an unfortunatelimitation on personal expression. American men have learned to be friendly but to avoid intimacy. American women feel freer expressing themselves with their friends, but within limits. Perhaps as traditional gender roles continue toerode, both men and women will feel free to interact with friends at whatever level of intimacy they choose. Disclosing Traumatic Experiences Students participating in a psychology experiment some years ago were asked to write anonymously about an upsetting or traumatic experience they once had, something they may have kept inside for years and told to no one(Pennebaker & Beall, 1986). One of the interesting findings from this and other studies like it is that nearly every participant is able to identify a secret trauma (Pennebaker, 1989, 2000). People write about personal failures andhumiliations, illegal activities, drug and alcohol problems, and experiences with sexual abuse. They often express guilt over regrettable actions or great sadness about a personal loss. About a quarter of the participants cry. Thestudents in this study wrote about themselves for 15 minutes each night for 4 consecutive November nights. Other students assigned to a control condi- tion were instructed to spend this same amount of time writing about rela-tively trivial topics (for example, a description of their living room). What impact did this writing exercise have on the students? Measures of blood pressure and self-reported mood indicated that writing about a trau- matic experience led to more stress and a more negative mood immediately after the disclosure. However, the investigators contacted the students again in May, 6 months after they had written about their experiences. Students were asked about their health during the 6 months and about how many days they had been restricted because of an illness during this period. In addi- tion, the number of visits each student had made to the campus health center was recorded. Some of the differences between the two groups are shown in Figure 12.2. Students in the trivial topic group showed a significant increase in the numberof days they were restricted by illness and the number of visits they made to the health center. But this was not the case for the students who had written aboutSelf-Disclosure 313

their traumatic secrets. Similarly, only the disclosing students showed a de- crease in the number of illnesses. In other words, although writing about theirproblems created some mild, short-term discomfort, it appears that the act of disclosing, even in the relatively mild form used in this study, improved the health of the already healthy college students. The health benefits of disclosing traumatic experiences have been found in numerous subsequent investigations using this same basic procedure and with many different kinds of participants (Frattaroli, 2006; Frisina, Borod, &ASSESSING YOUR OWN PERSONALITY Disclosure and Concealment Indicate the extent to which you agree with each of the following state- ments. Use a 5-point scale to indicate your response, with 1 ¼Strongly disagree and 5 ¼Strongly agree. 1. When I feel upset, I usually confide in my friends. 2. I prefer not to talk about my problems. 3. When something unpleasant happens to me, I often look for someone to talk to. 4. I typically don ’t discuss things that upset me. 5. When I feel depressed or sad, I tend to keep those feelings to myself. 6. I try to find people to talk with about my problems. 7. When I am in a bad mood, I talk about it with my friends. 8. If I have a bad day, the last thing I want to do is talk about it. 9. I rarely look for people to talk with when I am having a problem. 10. When I ’m distressed, I don’ t tell anyone. 11. I usually seek out someone to talk to when I am in a bad mood. 12. I am willing to tell others my distressing thoughts. To score, reverse the answer values for items 2, 4, 5, 8, 9, and 10 (that is, 1¼5, 2¼4, etc.). Then add all 12 answer values together. High scores indicate a tendency to disclose distressing experiences to others, whereaslow scores indicate a tendency to conceal information about distressingevents. You can compare your score with norms from an undergraduate student sample (Kahn & Hessling, 2001). Men in this sample had a mean score of 36.33 (standard deviation ¼8.98), and women had a mean score of 42.21 (standard deviation ¼9.16). Scale: The Distress Disclosure Index Source: From “Measuring the tendency to conceal versus disclose psychological distress, ”by J. H. Kahn and R. M. Hessling, Journal of Social and Clinical Psychology , 2001, 20, 41– 65. Copyright © 2001 Guilford Publications, Inc. Reprinted by permission.314 CHAPTER 12 The Humanistic Approach

Lepore, 2004; Kelly & McKillop, 1996). Putting into words what has been kept under wraps for sometimes years consistently leads to improvements in physical health. Moreover, the benefits are found regardless of whether the disclosure is hand-written or typed or even if the information is disclosed orally (Frattaroli, 2006; Harrist, Carlozzi, McGovern, & Harrist, 2007). Participants also need not disclose about something as emotionally intense as a personal trauma. One team of investigators asked freshmen to write about the problems and emotionsthey encountered leaving home and adjusting to college (Pennebaker, Colder, & Sharp, 1990). Students who wrote about these thoughts and feelings for 3 consecutive nights made fewer visits to the health center over the next severalmonths than those who wrote about trivial topics. The connection between self-disclosure and health is also found in studies that look at how victims of specific traumas react to their experiences.One team of researchers contacted people who had lost a spouse either through an accidental death or because of suicide (Pennebaker & O’ Heeron, 1984). The investigators asked how often the participants had discussed the experience with friends and about the participants ’health since the death. They found that the more people had talked about the tragedy, the fewer health problems they had. Another study found that World War II Holocaust survivors who spoke openly about their ordeal were in better health than those who were less willing to disclose about the experience (Finkelstein & Levy, 2006). Visits to Health Center Before StudyAfter Study.511.5 Traumatic Topics ConditionTrivial Topics Condition FIGURE 12.2 Mean Number of Visits to the Health Center as a Function of Experimental Condition Source: From Pennebaker and Beall (1986).Self-Disclosure 315

But the value of disclosure is not limited to physical health. Writing about previously undisclosed experiences also leads to better emotional and psycho- logical well-being several months later (Frattaroli, 2006). Even putting our feelings about everyday sources of stress into words often makes it easier to cope with those concerns. College students in one study were less emotionally upset about taking graduate school entrance exams when they wrote about their feelings concerning the upcoming exam (Lepore, 1997). Freshmen in another study who wrote about the problems they faced adjusting to collegehad higher grade point averages their first semester than students who wrote about trivial topics (Cameron & Nicholls, 1998). Researchers also find that individuals who typically conceal unpleasant personal information experiencemore distress and have a lower sense of well-being than those who tend to be more open (Kahn & Hessling, 2001; Larson & Chastain, 1990). But why does disclosure, even when written anonymously, result in better physical and psychological health? One reason is that actively inhibiting thoughts and feelings about unpleasant experiences requires a great deal of psychological and physiological work (Pennebaker, 1989). The impact of thisstress is both immediate and long term. One investigation found an increase in immune system strength immediately after participants wrote about traumatic experiences (Petrie, Booth, & Pennebaker, 1998). Another study found partici- pants slept better in the days following their disclosure (Mosher & Danoff- Burg, 2006). The cumulative effect of withholding secrets over time takes its toll in the form of increased illnesses and other stress-related problems. Expressing thoughts and feelings also provides disclosers insight into their feelings that they might not have recognized otherwise, and this insightmakes it easier to take steps to move beyond the experience (Kelly, Klusas, von Weiss, & Kenny, 2001; King, 2001; King & Miner, 2000; Langens & Schuler, 2007; Smyth, True, & Souto, 2001). As Rogers and other humanistic theorists argued, putting feelings into words allows us to “see” our emotions and thereby deal with them more effectively. Undergraduates in one study were asked to write about traumatic experiences during three 20-minute sessions (Hemenover, 2003). Three months later, these students scored higheron measures of mastery, personal growth, and self-acceptance than students who had written about trivial topics. In other words, the writing experience led to changes in the way participants thought about themselves. Examples of beneficial self-disclosure can be found all around us. People often turn to friends, bartenders, and clergy members when they need to talkabout their problems. Some people keep diaries, others write letters theynever intend to mail. Increasingly people are disclosing all kinds of personal shortcomings and ordeals on anonymous websites. The benefits of disclosing secrets can also be seen in psychotherapy (Cepeda-Benito & Short, 1998;Kelly, 1998; Kelly & Archer, 1995). Talking or writing about disturbing experiences may be an important step in working through trauma (Janoff- Bulman, 1992). Providing clients with an opportunity to openly discuss emo-tions, and thereby turning vague emotional images into coherent thoughts, is probably one reason psychotherapy works for some people (Donnelly & Murray, 1991; Murray, Lamnin, & Carver, 1989; Segal & Murray, 1993).316 CHAPTER 12 The Humanistic Approach

LONELINESS A few years ago a national survey asked Americans about the people with whom they discuss “matters important to you ”(McPherson, Smith-Lovin, & Brashears, 2006). The researchers were particularly interested in the number of people who fell into this “confidant ”category. In 1985, when the survey takers first asked the question, American adults reported an average of 2.94 friends and family members who they considered confidants. But by 2004 the number had dropped to 2.08. In two decades the typical American went from having three close friends to only two. During this same time, the num-ber who said they had no one to discuss important matters with rose from 10% to 24.6% of the population. In short, people appear to be lonelier today than they were just a few decades ago. From time to time we have all felt the pain of loneliness. Each of us has suf- fered through a period when there was no one to talk to, when everyone elseappeared to be with someone while we were alone, when all our relationshipsIN THE NEWS Keeping Secrets From the outside, Tom Paciorek seemed to have a life almost anyonewould envy. Paciorek is a formerall-star outfielder who played majorleague baseball for 18 years. He stands 6 foot 4 inches and was oncevoted the second most handsomeman in baseball. Upon retiring fromthe game, he became a televisionbroadcaster. He and his wife of many years raised six healthy children. Yet at age 55 Paciorek was not a happy man. In fact, he had spent most of his life hiding a dark, hu-miliating secret. When Paciorek wasa boy growing up in Detroit, he wasmolested repeatedly over a period ofseveral years by a teacher at hisCatholic high school. That teacherlater became a priest. Like manychildren who are victims of sexualabuse, Paciorek told no one. Theabuse did not stop until he wentaway to college, and the emotionalpain Paciorek carried with him as aresult of the molestation never ended. “You try to deny it ever hap- pened, to bury it in your mind, ” Paciorek said, “but you live with horrible emotions, with the loss of self-esteem, with a loss of trust forothers” (as cited in Berkow, 2002). As researchers have demon- strated in numerous studies, trau- matic experiences like the one Paciorek went through often taketheir toll emotionally and physi-cally. The memories haunted Paciorek throughout adulthood. Hedescribed his life as “chaos.” More than 20 years after the abuse ended,Paciorek sought out psychotherapy.For 15 years he tried to workthrough his emotional pain with hiscounselor. But when Paciorek readthat the priest who had oncemolested him had been assigned towork with students, he knew thetime had come to disclose his secret.Paciorek joined hundreds of othervictims who came forward during the wave of sexual abuse scandalsthat rocked the Catholic Church in2002. He was 55 years old andhad held onto his secret for nearly 40 years. Paciorek didn ’t sleep the night before his story appeared in theDetroit Free Press . But he soon discovered, as have other trauma victims, that talking about the ex- perience is often better than holdingemotions inside. Paciorek received aflood of supportive phone calls and e-mail messages. In an importantway, disclosing his secret had sethim free. On the day his story wasreported in the media, Paciorekwent jogging. Sometime during hisrun, while his thoughts were filledwith the events from so many yearsbefore, Paciorek heard a voice thatcaused him to burst into tears(Whitley, 2002). “It’s over, ”the voice said. “It took 40 years, but it ’s over. ”Loneliness 317

seemed superficial. Paradoxically, loneliness has become epidemic on college campuses. Despite the presence of people seemingly everywhere, 75% of thefreshmen contacted at a large university 2 weeks into the school year said they had experienced loneliness since school began, and more than 40% said their loneliness had been either moderate or severe (Cutrona, 1982). Humanistic psychologists are concerned with loneliness for a number of reasons. Some have argued that humanistic psychology ’s rise in popularity in the 1960s can be attributed to feelings of alienation and loneliness that had begun to creep into many American lives (Buhler & Allen, 1972). People faced with an increasingly dehumanized, mechanistic society welcomed the humanists ’emphasis on the individual with his or her unique potential. Some psychologists believe feelings of loneliness reflect existential anxiety and a need to find meaning in one ’s life (Sadler & Johnson, 1980). And humanistic therapists often help clients develop meaningful encounters to overcome lone- liness (Moustakas, 1968). Perhaps the most notable development in this area was the growth of encounter groups in the late 1960s and early 1970s. Within the safe confines of the group, humanistic therapists helped membersdiscover the richness of intimate interpersonal encounters with others and thereby learn something about themselves (Rogers, 1970). Loneliness is a common problem on college campuses, but some people are more prone to bouts of loneliness than others.318 CHAPTER 12 The Humanistic Approach

Defining and Measuring Loneliness Loneliness is not the same as isolation. Some of the loneliest individuals are surrounded by people most of the day. Rather, loneliness concerns our per- ception of how much social interaction we have and the quality of that inter- action. As one team of investigators explained, “Loneliness occurs when a person ’s network of social relationships is smaller orless satisfying than the person desires” (Peplau, Russell, & Heim, 1979, p. 55, italics added). You can have very little contact with people, but if you are satisfied with that con-tact, you won ’t feel lonely. On the other hand, you may have many friends, yet still feel a need for more or deeper friendships and thus become lonely. Thinking of loneliness in terms of personal satisfaction with one ’ss o c i a lr e - lationships helps explain why some people who live in virtual isolation find the solitude enjoyable, whereas other individuals surrounded by people feel lonely. I commonly hear college students complain that, although they have a lot ofacquaintances and people to hang around with, they don ’t have many real friends. For these students, the unmet need to interact with that special person in an intimate and honest way can create intense feelings of being alone. Loneliness is often caused by the circumstances people find themselves in, such as moving to a new city or attending a new school. Moreover, the kindsof relationships we desire change as we pass through the life cycle (Green, Richardson, Lago, & Schatten-Jones, 2001; Pinquart & Sorensen, 2001). Young adults often require a larger number of friends to fend off loneliness, whereas older adults prefer fewer but closer friends. The causes and conse- quences of loneliness also vary as a function of culture (Anderson, 1999). The absence of an intimate friend or romantic partner often contributes to loneliness in Western societies. In fact, when people in individualistic cultures think of loneliness, they often imagine someone without a spouse or romantic partner. However, this source of loneliness is less common in Asian cultures, which emphasize instead associations with family members and the commu- nity (Rokach, 1998). Because collectivist cultures emphasize one ’s place in a larger social network, feeling alone in these cultures is more likely to lower aperson ’s sense of well-being (Goodwin, Cook, & Yung, 2001). Although feelings of loneliness come and go as circumstances change, re- searchers also find loneliness can be conceived of as a fairly stable personalitytrait. That is, although everyone feels lonely on occasion, some people are highlyvulnerable to feelings of loneliness and seem to chronically suffer from not having enough close friends. Other people are relatively immune from these experiences. Several personality inventories have been developed to assess individual differ-ences in our tendency to feel lonely (Cramer, Ofosu, & Barry, 2000; Rubenstein & Shaver, 1980; Russell, Peplau, & Cutrona, 1980; Schmidt & Sermat, 1983). Like other personality variables, our vulnerability to loneliness is relativelystable over time (Segrin, 1999; Weeks, Michela, Peplau, & Bragg, 1980). Chronically Lonely People Correlations between measures of loneliness and other personality variables paint a drab and sullen picture of lonely people (Ernst & Cacioppo, 1999).Loneliness 319

High scores on loneliness scales are related to high scores on social anxiety and self-consciousness and low levels of self-esteem and assertiveness (Bruch,Kaflowitz, & Pearl, 1988; Jones, Freemon, & Goswick, 1981; Solano & Koester, 1989). Lonely people are more likely to be introverted, anxious, and sensitive to rejection (Russell et al., 1980) and more likely to suffer from de- pression (Joiner, Catanzaro, Rudd, & Rajab, 1999; Wei, Russell, & Zakalik, 2005). High loneliness scores also are associated with pessimism and negative mood (Cacioppo et al., 2006). Not surprisingly, lonely people have more than their share of social diffi- culties (Heinrich & Gullone, 2006). They have a hard time trusting other peo-ple (Rotenberg, 1994) and are often uncomfortable when others open up tothem (Rotenberg, 1997). Lonely people spend less time with friends, date less frequently, attend fewer parties, and have fewer close friends than nonlonely people (Archibald, Bartholomew, & Marx, 1995). They have difficulty initiat-ing social activity and participating in groups (Horowitz & de Sales French, 1979). Acquaintances of lonely people confirm the accuracy of these assess- ments. College students say their relationships with lonely people are noticeablyless intimate than they are with nonlonely people (Williams & Solano, 1983). In addition to the emotional and social toll that comes with feeling alone, a growing body of research indicates that loneliness also may be hazardous toyour health (Cacioppo & Patrick, 2008; Cacioppo, Hawkley, Crawford et al., 2002; Cohen & Janicki-Devers, 2009; Uchino, Cacioppo, & Kiecolt-Glaser, 1996). Compared to those who have few social contacts, people with a large and diverse social network have a decreased risk of cancer recurrence (Helgeson, Cohen, & Fritz, 1998), stroke (Rutledge et al., 2008), and heart disease (Kop et al., 2005). Not surprisingly, people with large social networks also live longer (Berkman, 1995). Healthy adults in one study were— with permission —deliberately exposed to a cold virus (Cohen, Doyle, Turner, Alper, & Skoner, 2003). Researchers found that the more social the participants,the less likely they were to come down with a cold. Lonely students in another study had a poorer antibody response to r eceiving a flu shot than did nonlonely students (Pressman, Cohen, Miller, Barkin, Rabin, & Treanor, 2005). But why does loneliness affect health? As shown in Table 12.1, researchers have identified five possible pathways (Hawkley & Cacioppo, 2007). First, lonely people often have poorer health habits than nonlonely people. Mostnoteworthy, they tend to be less active physically (Hawkley, Thisted, & Cacioppo, 2009). Whereas nonlonely people are out hiking, dancing, golfing, etc. with friends, lonely people are staying home watching TV. This lack of ac-tivity combined with a poorer diet makes lonely people more prone to obesity than nonlonely individuals (Lauder, Mummery, Jones, & Caperchione, 2006). Lonely people also are more likely to smoke (Lauder et al., 2006). Second,lonely people are subject to more sources of stress than nonlonely individuals. People who suffer from chronic loneliness tend to experience stress in more areas of their lives (e.g., financial, social, employment) than nonlonely people(Hawkley, Burleson, Berntson, & Cacioppo, 2003). Third, lonely people do not cope with this stress as well as nonlonely people. Seeking emotional support from friends when times are rough is an effective coping strategy. However,320 CHAPTER 12 The Humanistic Approach

lonely individuals lack a network of friends they can turn to when feeling over- whelmed and helpless. Instead, they are more likely than most people to engage in less effective withdrawal strategies when experiencing stress (Cacioppo et al., 2000). Fourth, the excessive amount of stress experienced by lonely people leads to changes in physiological conditions that eventually affect their health. In particular, loneliness is associated with high blood pressure (Hawkley, Masi, Berry & Cacioppo, 2006). Fifth, loneliness interferes with some of the body ’s natural restorative processes. Most noteworthy, lonely people sleep less well than nonlonely people (Cacioppo, Hawkley, Berntson et al., 2002). In short, a lifetime of chronic loneliness can grind away at a person ’s health. The Causes of Loneliness What is it about lonely people that continually frustrates their need for mean-ingful social contact? Researchers have identified two characteristics that seem to contribute to chronic loneliness —negative expectations and poorly developed social skills. Lonely people often enter a social situation with the expectation that this encounter, like so many before, will not go well (Goswick & Jones, 1981; Hanley-Dunn, Maxwell, & Santos, 1985; Jones et al., 1981; Jones, Sansone, &Helm, 1983; Levin & Stokes, 1986). In one study, lonely and nonlonely college students were asked to participate in a series of group activities with three other students (Christensen & Kashy, 1998). The students discussed and solvedproblems together for 30 minutes. The participants were then separated and asked to rate the other members of the group in terms of their intelligence, friendliness, and so on. The participants also rated themselves on these dimen-sions and guessed what kind of ratings they would receive from the other group members. The researchers found that lonely participants evaluated themselves less favorably than they evaluated the other group members. The nonlonely students did not do this. Moreover, the lonely participants expected that the other three members of the group also would rate them poorly. However, they were wrong. Despite their low expectations for how the other students would see them, the lonely students were evaluated no differently from anyone else — with one exception. The lonely students were actually perceived as beingTABLE 12.1 Pathways Between Loneliness and Health Health Behaviors Lonely people have poorer health habits Stress Exposure Lonely people experience more chronic stressorsCoping Styles Lonely people use less effective coping strategies when dealing with stress Stress Physiology Lonely people experience unhealthy changes in physiology Recuperation Lonely people are less able to rely on natural restorative processes that improve health Source: Adapted from Hawkley and Cacioppo (2007).Loneliness 321

friendlier than most of the people in the group. In short, the lonely students thought the others would not like them, yet as it turned out they were greatlymistaken. These low expectations can be poisonous when trying to develop a friendship or romantic relationship. Lonely people doubt a new acquaintancewill enjoy talking with them and suspect the person will find them boring or stupid by the end of the conversation. Consequently, lonely people often show little interest in getting to know other people and are quick to end theconversation and move on to something else. These negative expectations may also lead lonely people to interpret any small sign as rejection. Participants in one experiment spent 5 minutes talking with a stranger (Frankel & Prentice-Dunn, 1990). Later, participants saw a videotape of their partner ’s evaluation of them. The videotape contained positive and negative comments. As ex-pected, the lonely people paid attention to and recalled the negative feedbackbetter than the nonlonely participants. Because they believe their interactions have gone worse than they probably have, lonely people are unlikely to pursue a friendship with someone they ’ve met or to seek out others to do things with. Given this negative approach to social interactions, it is not surprising that lonely people have such a difficult time making friends. This research also helps explain why loneliness is a problem for many students on crowded college campuses. With so many potential friends around, there is little reason to seek out and nurture the friendship of someone who appears to be unfriendly. Chronically lonely people also tend to have poorly developed social skills. Perhaps you are one of those lucky individuals for whom conversation comeseasily. You enjoy meeting people, effortlessly finding out about them, and oc- casionally talking about yourself. If this is you, then you are probably puzzled by people who have difficulty interacting with others. Even for people who are not shy and who would like to meet new friends, engaging in more than a short and trivial conversation can be a chore. What these people may lackare basic social skills, the knowledge of how to carry on a conversation that both you and the other person find valuable and enjoyable. Several studies implicate just such a lack of social skills as part of what keeps some people trapped in a cycle of loneliness (Segrin, 1999; Segrin &Flora, 2000; Vitkus & Horowitz, 1987). The best way to learn the art of con- versation is to talk with others. Yet people without social skills may havesuch a difficult time developing relationships that they have little opportunity to develop these skills. They never learn how to initiate an interaction or how to keep the conversation lively, so their difficulty making friends continues. Consider the interaction styles one team of researchers found when they examined conversations with lonely and nonlonely individuals (Jones,Hobbs, & Hockenbury, 1982). Lonely participants showed relatively little in-terest in their partners. They asked fewer questions, often failed to comment on what the other person said, and made fewer references to the partner. Instead, these lonely people were more likely to talk about themselves and in-troduce new topics unrelated to their partner ’s interests. Another study found lonely people were more likely to give advice to strangers and less likely toacknowledge what the other person said (Sloan & Solano, 1984). Little322 CHAPTER 12 The Humanistic Approach

wonder, then, that we often fail to enjoy conversations with lonely people. It’ s not that lonely people are intentionally rude, but rather that they don ’t under- stand how their interaction style turns away potential friends. Fortunately, some efforts to combat loneliness with social skills training have been promis- ing (Rook & Peplau, 1982; Young, 1982). Other researchers examine the way lonely and nonlonely people use self- disclosure. Studies find that lonely people generally reveal less about them- selves than their partners (Berg & Peplau, 1982; Sloan & Solano, 1984). In one study lonely people selected relatively nonintimate topics to talk about ina get-acquainted conversation (Solano, Batten, & Parish, 1982). Not surpris- ingly, the lonely participants ’partners reciprocated with nonintimate topics as well. Other studies find lonely people are often not aware of social rules about when and how much to disclose (Chelune, Sultan, & Williams, 1980; Solano & Koester, 1989; Wittenberg & Reis, 1986). They may disclose too much or fail to reveal enough about themselves when the other person ex- pects it. Consequently, others may see them as either weird or aloof, and re- spond accordingly. SELF-ESTEEM If there is a single concept that threads its way through the writings of the hu- manistic psychologists, it may be how people feel about themselves. A central goal of Rogerian psychotherapy is to get clients to accept and appreciate them- selves for who they are. Maslow wrote about the need for self-respect and the need to feel content about who we are and what we do with our lives. In short, humanistic personality theory is concerned with the individual ’ss e l f - e s t e e m .324 CHAPTER 12 The Humanistic Approach Text not available due to copyright restrictions

Most researchers draw a distinction between self-esteem and self-concept. Your self-concept is the cumulation of what you see as your personal char- acteristics —that is, the kind of person you believe yourself to be. Self-esteem refers to your evaluation of your self-concept. In essence, do you like this person? Although we often speak of self-esteem in our everyday conversa- tions, researchers face several challenges when trying to identify and measure this concept. One problem is that the way we feel about ourselves can change from one situation to the next. Most people get a little down on themselves whenthey act in ways they know they shouldn ’t, and most of us can ’t help but think well of ourselves when someone heaps praise on us for a job well done (Heatherton & Polivy, 1991). However, these fluctuations in feelings should not be confused with self-esteem. Rather, psychologists refer to these ups and downs as feelings of self-worth (Brown & Dutton, 1995). In contrast, self- esteem has to do with relatively stable self-evaluations. As with other person- ality variables, researchers find some people are prone to more positive self- evaluations than others. These individuals may have bad days and disappointthemselves on occasion, but in general they like themselves and feel good about who they are and what they do. Of course, we also can identify people who frequently experience negative self-evaluations. Although these low self- esteem people also have good days and feel good about much of what they do, compared to others they seem to lack a basic confidence in themselves or an appreciation for who they are. Self-Esteem and Reaction to Failure Evaluation is an unavoidable part of most of our lives. After only a few years of elementary school, most students become accustomed to having their schoolwork graded. Evaluation is commonplace in the business world, if not overtly in the form of an annual review, then implicitly in the size of one ’s raise. Any type of competition, from sports to chess to gardening, brings with it the possibility of both victory and defeat as we compare our abilities and accomplishments agai nst those of others. All of this evalu- ation means that each of us has experienced our share of successes andfailures. However, not all people react the same to these evaluations. Several labo- ratory experiments have looked at how high and low self-esteem people re-spond when told they have done well or poorly on a test (Brockner, 1979;Brown & Dutton, 1995; Kernis, Brockner, & Frankel, 1989; Stake, Huff, & Zand, 1995; Tafarodi & Vu, 1997). Participants in these studies usually take a test supposedly measuring some intellectual aptitude or work on a task call- ing for some specific ability. Researchers give bogus feedback to participants, indicating they have done either very well or rather poorly. Of particular in- terest is how people respond to failure. When told they have done poorly, low self-esteem people typically don ’t try as hard, perform more poorly, and are more likely to give up when given a follow-up test. In contrast, high self- esteem people work just as hard on the second test regardless of how they did on the initial test.Self-Esteem 325

The importance of these findings for academic settings is obvious. Consider one study that examined college students ’reactions to a midterm exam grade (Brockner, Derr, & Laing, 1987). The students took their first exam for the class 5 weeks into the term and received their grade 1 week later. The investigators found that high and low self-esteem students per- formed almost identically on the midterm test. The researchers then divided the students into those who had done well on the test (received an A or a B) and those who had not done as well (received a C or lower). As shown inFigure 12.3, like the laboratory participants who received false feedback, low self-esteem students who did well on the first test continued to perform well. However, low self-esteem students who had not done well on the first testperformed significantly worse on the second exam. Another study found that low self-esteem people do not have to actually experience failure to show these negative effects; rather, they only have toimagine that they have failed (Campbell & Fairey, 1985). Participants in this investigation were asked to imagine they had done well or poorly on a 25- item anagram test. Low self-esteem people who imagined failing said they ex-pected to do poorly on a subsequent test, and indeed they performed more poorly on the test than low self-esteem people who first imagined they had done well. How can we explain these reactions? One possibility is that people are more likely to accept feedback consistent with their self-concept (Story,1998; Wood, Heimpel, Manwell, & Whittington, 2009; Wood, Heimpel, Newby-Clark, & Ross, 2005). People with low self-esteem probably accept the fact that they fail more than others. Consequently, it is easier for them to Grade on Second Test Did Well on First TestLow Self-EsteemHigh Self-Esteem Did Poorly on First Test60708090100 FIGURE 12.3 Performance on Second Test as a Function of Self-Esteem and Performance on First Test Source: Adapted from Brockner et al. (1987).326 CHAPTER 12 The Humanistic Approach

believe feedback confirming their negative self-image than information that violates their expectations. Another way to look at this is to say that the neg-ative feedback reminds low self-esteem people of the low evaluations they have of themselves (Dutton & Brown, 1997; Tafarodi & Vu, 1997). The feedback triggers associations with related negative thoughts, reminding low self-esteem people of other faults and weaknesses. This interpretation helps us understand why low self-esteem people perform more poorly on a task even when they have just imagined what it would be like to fail. But we can also turn this question around. What is it about high self- esteem people that prevents them from becoming discouraged after failure?Why don ’t they give up when they fail a test or do poorly at work? The an- swer appears to be that high self-esteem people develop personal strategies for blunting the effects of negative feedback (Heimpel, Wood, Marshall, & Brown, 2002). Included in this arsenal is a tendency to respond to failure byfocusing attention on their good qualities rather than on what they have done wrong. Whereas negative feedback causes people low in self-esteem to think about their faults and failures, this same feedback leads high self-esteem peo-ple to think about their abilities and achievements. The high self-esteem strategy for blunting the effects of failure has been demonstrated in several investigations (Brown & Gallagher, 1992; Dodgson &Wood, 1998; Greve & Wentura, 2003; Schlenker, Weigold, & Hallam, 1990). Participants in one study received feedback indicating they had performed either well or poorly on an achievement test (Brown & Smart, 1991). Participants were then asked how well a series of adjectives described them. Some adjectives were related to achievement situations (for example, compe- tent, intelligent ), and others were relevant for social situations (for example, sincere, kind ). As shown in Figure 12.4, low self-esteem participants rated 5.05.56.06.5Rating of Social Attributes High Self-EsteemFailureSuccess Low Self-Esteem FIGURE 12.4 Ratings of Social Attributes Following Success and Failure Source: Adapted from Brown and Smart (1991).Self-Esteem 327

themselves poorly on their social skills after discovering they had failed the achievement test. However, high self-esteem participants actually rated them-selves higher on their social attributes after failing. These results demonstrate one tactic high self-esteem people use to main- tain their feelings of high self-worth even in the face of negative feedback.When told they did not do well on one task, they simply remind themselves of how well they do in other areas. In another study high self-esteem people quickly abandoned a task they were having trouble with when given an op-portunity to work on something they could do well (Di Paula & Campbell, 2002). It ’s not that high self-esteem people think they ’re perfect; they just don’t dwell on their failures. If they mess up at work, they might remind themselves that they have a lot of friends. If they lose badly at handball, theymight recall how well they play chess. This strategy keeps high self-esteem people feeling good about themselves even when faced with life ’s inevitable downturns. Contingencies of Self-Worth To this point, we’ ve concentrated on what researchers refer to as global self- esteem; that is, the overall evaluation we have about ourselves. But very fewpeople feel entirely good or bad about themselves. Even the most contentamong us can point to deficiencies and weaknesses, areas where we feel less confident than others. Thus researchers sometimes find it useful to examine self-esteem within specific domains. For example, investigators might ask par-ticipants how they feel about themselves in terms of academic performance, personal ethics, or physical appearance. This strategy raises some interesting questions. Is global self-esteem simply the sum of how we feel about ourselvesin these specific areas? Is high self-esteem restricted to those who feel good about themselves in most, if not all, major domains? Put another way, is it necessary to feel competent and virtuous in many different areas before wecan feel good about who we are? Fortunately, the answer to all of these questions is “No.” Researchers find that global self-esteem seems to result from a two-step process (Crocker, Brook, Niiya, & Villacorta, 2006; Crocker & Park, 2003; Crocker & Wolfe, 2001). First, each of us identifies domains that we consider important to us; that is, areas we use to determine our self-worth. For one person these domains might be academic performance and acting in ethically sound ways. For someone else physical appearance and acceptance from family and friends might be most important. Researchers refer to these areas we use to evaluate ourselves as contingencies of self-worth . Second, we form evalua- tions of ourselves —that is, our global self-esteem —based on how we do in these selected areas. The woman who bases her self-esteem on academic per-formance will feel good about who she is as long as she performs well in class. The man whose contingencies of self-worth include his physical appear- ance will enjoy high self-esteem as long as he receives feedback that he is good-looking and desirable. Looking at self-esteem in terms of contingencies of self-worth helps us understand why people with limitations and deficiencies can still feel good328 CHAPTER 12 The Humanistic Approach

about who they are. You may be terrible at sports or math, but if you don’ t base your self-worth on how you perform in these areas, they probably will not affect your overall self-esteem. One team of researchers identified seven areas college students typically use as contingencies of self-worth (Crocker, Luhtanen, Cooper, & Bouvrette, 2003). As shown in Table 12.2, these con- tingencies range from some obvious areas (competence, appearance) to some you might not have considered (God ’s love). Where do contingencies of self-worth come from? That is, why does one person base her self-esteem on competence, whereas another bases his self-esteem on God ’s love? For starters, people tend to select contingency areas in which they typically excel. Athletes often base their self-esteem on their ability to perform well in sports, whereas good students tend to rely on their grades. But this is only part of the story. We all know people who select contingen- cies of self-worth that are difficult to attain. Sometimes parents or peers influ-ence these choices, such as a father who says he loves his little girl because of how pretty she is and admires his son for his athletic skills. Culture also plays a role, placing different values on different personal attributes. For example,our society communicates in many ways that certain standards are more rele- vant for one gender than for the other (Fredrickson & Roberts, 1997; Strahan et al., 2008). Girls are often reminded by peers and media messages that their value depends on how they look. Boys are told that beating others in various competitions is a sign of their self-worth. Using contingencies of self-worth to determine self-esteem has advan- tages and disadvantages. On the plus side, we don’ t have to be competent at everything —or even very many things —to feel good about who we are. The TABLE 12.2 Contingencies of Self-Worth for College Students Contingency Description Competencies Abilities and performance in various areas, particularly academic performance for college students Competition Outperforming other people in various compet- itive situations Approval from GeneralizedOthersApproval and acceptance from other people,based on what we believe they think of us Family Support Receiving approval and affection from the people closest to us, particularly from family members Appearance How physically attractive we believe others find us God’s Love The belief that we are loved, valued, and unique in the eyes of God Virtue Adhering to personal ethical standards and judging ourselves as good and moral individuals Source: Adapted from Crocker, Luhtanen, et al. (2003).Self-Esteem 329

student who breezes through classes and the one who struggles can both have high self-esteem. Similarly, the star athlete and the ultimate physical klutz canboth feel good about themselves. Returning to humanistic personality theory, it is ultimately up to us to choose which standards we use for this evaluation. On the down side, people sometimes select contingencies that are difficult to achieve. A young woman loved and admired by family and friends still might not feel good about herself if she feels academically inferior to some of her peers. Thinking of self-esteem in terms of contingencies of worth also helps us understand why some people have relatively stable feelings of self-worthwhereas others fluctuate wildly between liking and not liking themselves.That’s because some contingencies leave us at the mercy of other people and unmanageable forces. We can ’t always control whether we will be accepted by others, retain good looks, or succeed in competition. As a result, peoplewho tie their self-esteem to uncontrollable forces may be more prone to bouts of anxiety and depression (Crocker & Park, 2004). Undergraduates who base their self-worth largely on academic performance often experience depressionand a drop in self-esteem when they receive a low grade (Crocker, Karpinski, Quinn, & Chase, 2003; Park, Crocker & Kiefer, 2007) or a rejection letter from graduate school (Crocker, Sommers, & Luhtanen, 2002). In one study students who used academic performance as a contingency of self-worth ex- perienced more stress —but did not obtain higher grades —than students who relied on other contingencies (Crocker & Luhtanen, 2003). Making one ’s self-esteem contingent on physical appearance or the ap- proval of others also can lead to an emotional roller coaster. A compliment or a pleasant conversation might lead to pride and self-liking, but a rude re- mark or a broken date can trigger self-doubt. People who make their self- esteem contingent on the state of their romantic relationships may be subject to strong emotional reactions when inevitable relationship issues surface(Knee, Canevello, Bush, & Cook, 2008). Participants in one study who based their self-esteem on physical appearance felt alone and rejected when simply made aware of flaws in their appearance (Park, 2007). Another investigationfound that college freshmen who based their self-worth on their appearance were more prone to drinking alcohol, presumably because their self-esteem was threatened in the kinds of situations (parties and social events, for exam-ple) in which drinking occurs (Luhtanen & Crocker, 2005). In short, people who rely on uncontrollable self-esteem contingencies may be putting their emotional well-being on the line every day. On the otherhand, people who base their self-esteem on contingencies largely under their control (such as virtue or God ’s love) are less prone to depression and anxiety than those who rely on more externally based contingencies (Sargent, Crocker, & Luhtanen, 2006). Self-Esteem and Culture People growing up in Western culture often assume that everyone wants to excel, to stand out from the crowd, to be recognized for personal accomplish- ments. Teachers and parents foster high self-esteem in children by identifying330 CHAPTER 12 The Humanistic Approach

the child ’s unique strengths and helping the child develop and excel in these areas. Adolescents who say they are “no better than average” at anything might be readily labeled as poorly adjusted. Indeed, youth from disadvan- taged backgrounds are encouraged to believe in themselves, to believe they can achieve whatever they set their minds to. In short, in most Western socie- ties, the recipe for high self-esteem is feeling good about who you are and what you do to distinguish yourself. However, some researchers have challenged the universality of these no- tions (Heine, 2001; Kitayama & Markus, 1994; Markus & Kitayama, 1991,1994; Triandis, 1989, 2001). Recall from Chapter 1 that Western conceptions of the self are not shared by all cultures. People in collectivist cultures are moreconcerned with interdependence than with independence. Whereas individual- istic countries like the United States emphasize the uniqueness of the individ- ual, people in collectivist countries see themselves as part of a larger culturalunit. One implication of these different views is that we may need to rethink the way we conceptualize self-esteem when working with people from differ- ent cultures. Self-esteem scales developed primarily for American research participants often ask test takers about feelings of competence and about how much they value their unique attributes. Such items make little sense to people who see their value in terms of belongingness and cooperation. Researchers sometimes illustrate the difference between an individualistic country like the United States and a collectivist country like Japan by pointing to a pair of expressions from these cultures. In the United States we some- times say, “The squeaky wheel gets greased, ”meaning that one has to stand up and assert oneself to get ahead. In Japan one often hears, “The nail that stands up is the one that gets hammered, ”meaning that asserting one’ s indi- viduality is unacceptable and is likely to result in negative consequences. These different perspectives on the self also mean that people from the two types of cultures have different ideas about what leads to self-satisfaction and feeling good (Kang, Shaver, Sue, Min, & Jing, 2003; Tafarodi, Marshall, & Katsura, 2004). People in individualistic cultures typically feel good aboutthemselves when they think about their unique value and personal accomplish- ments. In contrast, people from collectivist cultures derive self-satisfaction from their perceived relationships with others. People from collectivist culturesfeel good when they obtain a sense of belonging within the culture, of occupy- ing their appropriate place. Fitting in and doing one ’s duty are sources of pride in collectivist cultures. Personal achievements and independence are valued in individualistic cultures. Consistent with these observations, research with American students finds a nearly universal tendency to see oneself in a better light than objective datamight suggest. When American college students are asked to compare them- selves to their peers on a variety of skills and aptitudes, they almost always report their superiority over those around them (Taylor, 1989). In this coun-try, it seems, we are all better than average. However, when researchers pres- ent these same questions to students from collectivist cultures, they find relatively little evidence for such a bias (Heine & Hamamura, 2007). TypicalSelf-Esteem 331

citizens in a collectivist culture simply do not see themselves as any better than other members of society. In the United States these feelings of averageness mightbe taken as evidence of poor self-esteem. However, in other countries such a self- evaluation is considered quite healthy. In a collectivist culture, it is the person full of self-importance who is cause for concern, the nail that gets hammered down. Needless to say, this cultural difference can be a source of conflict for people who move from one culture to another. American baseball players, known for their elevated sense of self, often have a very difficult time playing for Japaneseclubs, which emphasize the team above the individual (Whiting, 1989). In an interesting demonstration of the relation between culture and self- esteem, one group of researchers compared average self-esteem scores forAsians as a function of their exposure to North American culture (Heine, Lehman, Markus, & Kitayama, 1999). Remember, the self-esteem scales were designed so that high scores reflect Western notions of individual ac-complishment and pride in personal achievements. As shown in Figure 12.5, the participants ’self-esteem scores changed with their amount of contact with individualistic cultures. Three generations after the participants ’families had immigrated to Canada, the Asian Canadian self-esteem scores were nodifferent from those of the European Canadians. Culture also affects the standards people use to decide whether they are satisfied with their lives (Kuppens, Realo, & Diener, 2008; Oishi, Diener,Choi, Kim-Prieto, & Choi, 2007; Steger, Kawabata, Shimai, & Otake, 2008). Most Americans assume that happiness is the key to life satisfaction. 30 35 40Never Been AbroadBeen AbroadRecent ImmigrantsLong-Term ImmigrantsSecond GenerationThird Generation Self-Esteem Score FIGURE 12.5 Average Self-Esteem Score of Asians and Asian Canadians as a Function of Exposure to Western Culture Source: Adapted from “Is there a universal need for positive self-regard?, ”by S. J. Heine, et al., Psychological Review , 1999, 106, 766 –794. Copyright © 1999 by the American Psychological Association. Adapted by permission of the American Psychological Association.332 CHAPTER 12 The Humanistic Approach

That is, I will be satisfied with my life to the extent that I feel good about my- self and experience positive emotions, such as happiness. Indeed, studies findthis to be the case in individualistic cultures (Diener & Diener, 1995; Oishi & Diener, 2001; Suh, Diener, Oishi, & Triandis, 1998). However, this path to life satisfaction does not apply in collectivist cultures. Instead, how well people meet the culturally defined standard of proper behavior predicts life satisfac- tion in collectivist nations. Whereas feeling good is the key to a good life in individualistic cultures, fitting into the role prescribed by society is the key incollectivist cultures. In short, the theory and research on self-esteem presented in this chapter is probably applicable only to people living in individualistic cultures.However, we should not take this to mean that people in collectivist cultures do not have self-esteem. Rather, we should recognize that concepts like the self and self-esteem have different meanings in different cultures. Personalityresearchers have come to appreciate these differences and are actively con- ducting research on these questions. SOLITUDE In many ways Naomi is different from most people. Although she could easily join her coworkers in the company cafeteria, Naomi often chooses to have lunch alone. She ’ll eat a sandwich in the nearby park or sometimes spend the lunch hour taking a solitary walk around the neighborhood. When friends ask her to drop by on the weekend or to join them after work, she frequently declines even when she has no other plans. Surprisingly, most people who know her describe Naomi as a warm and engaging person. And Naomi verymuch enjoys her friends and coworkers. Still, compared to most people, Naomi spends a significant amount of time by herself. What might a personality psychologist say about Naomi? Research indi- cates that our interpersonal relationships are among our most importantsources of happiness (Diener & Seligman, 2002; Myers, 1992). So why would someone frequently turn down opportunities to socialize? Several possible ex-planations can be found in earlier sections of this book. Perhaps Naomi is in- troverted (Chapter 10). She may not find the stimulation of social activity as attractive as more extraverted people do. On the other hand, Naomi might avoid people because she suffers from social anxiety (Chapter 8). Perhaps she is afraid that others will evaluate her negatively, so she reduces her anxiety by simply avoiding social interactions whenever possible. Then again, in some ways Naomi ’s behavior is similar to what Karen Horney called the neurotic style of “moving away from people ”(Chapter 5). According to this analysis, Naomi may have adopted her avoidance style as a way to protect herselffrom anxiety when she was a child. Yet another possibility is that Naomi suf- fers from loneliness. As described earlier in this chapter, she may lack some basic social skills and spends time alone because she has difficulty interacting with people and developing relationships. Although each of these explanations can account for a person ’s desire to spend time alone, there is at least one more interpretation, one that casts“A desire for mean- ingful solitude is by no means neurotic; on the contrary, an incapacity for con-structive solitude is itself a sign of neurosis. ” Karen HorneySolitude 333

Naomi ’s quest for solitude in a different light. When Abraham Maslow stud- ied psychologically healthy people, he found a curious similarity among mem- bers of this select group. Although these self-actualized people possessed characteristics that made them the warmest of friends, they also spent a sur- prisingly large amount of time by themselves. “For all my subjects it is true that they can be solitary without harm to themselves and without discom- fort,”Maslow (1970) observed. “Furthermore, it is true for almost all that they positively like solitude and privacy to a definitely greater degree than the average person ”(p. 160). Maslow ’s observations thus provide another explanation for Naomi ’s preference for solitude. It is possible that she is not introverted, socially anx-ious, or lonely. Perhaps Naomi ’s desire to spend time by herself is something positive. Her preference for solitude may be both a reflection of and a con-tributor to her personal growth and development. Maslow was quick to pointout that psychologically healthy people also tend to express a great deal of in- terpersonal warmth and have especially close relationships with their dearest friends. People with a high desire for solitude are not necessarily trying to es-cape from relationships. Rather, some people who spend a great deal of time by themselves have come to recognize the benefits of solitude. Time Alone Although most of us live in social worlds, in truth we also spend a good deal of our day alone. To determine how often we spend time by ourselves, investiga-tors sometimes use a procedure known as the Experience Sampling Method (Larson & Csikszentmihalyi, 1980; Larson, Csikszentmihalyi, & Graef, 1982; Larson & Richards, 1991; Larson, Zuzanek, & Mannell, 1985). Participants in these studies carry electronic pagers or handheld computers with them 24 hours a day for about a week. At random intervals throughout each day, the researchers signal participants that it is time to fill out a quick report on whatthey are doing and how they feel at that moment. The results of this research, shown in Figure 12.6, confirm that Americans spend a significant amount of their waking hours alone. Moreover, solitude becomes a more common experi-ence as we age. How do people react to time alone? As suggested by conventional wis- dom, most of us find time by ourselves less pleasant than time spent withothers (Larson, 1990). People typically complain of loneliness and boredom when they are by themselves. After spending a long period of time alone, most people eagerly seek out social interactions (O ’Connor & Rosenblood, 1996). Emotional difficulties are often made worse when there is no one around. People prone to depression often experience negative emotion when alone. One study found that bulimics had a particularly negative reaction to solitude (Larson & Johnson, 1985). The researchers speculate that the loneli- ness and confusion experienced by bulimics when alone contributes to their eating disorder. Clearly, time by oneself can be unpleasant and unsatisfying for many people. But consider the results of one national survey (Crossen, 1996). Only6% of American adults in that survey said they wanted less time by334 CHAPTER 12 The Humanistic Approach

themselves. In contrast, 31% wished they had more time alone in their lives. These numbers are consistent with observations of several researchers who argue that isolation from other people also has benefits (Buchholz & Helbraun, 1999; Burger, 1995; Larson, 1990; Long & Averill, 2003; Storr, 1988). To better understand the advantages of time alone, investigators some- times divide solitude into three kinds of experiences (Burger, 1998). We can look at the effects of short periods of solitude during the day, usually mea-sured in minutes. We also can study the effects of longer, planned time by oneself, typically measured in hours. Or we can examine the impact of ex- tended periods of solitude, such as those measured in days. Each of thesekinds of solitude has the potential to contribute to our well-being. Even short periods of solitude spaced throughout the day can make a rough day go better. Sometimes we just need a break from constant social ac-tivity to organize our thoughts and psychologically prepare for future activi- ties. Some writers have referred to this as a “self-restoration ”process in which we reestablish a sense of who we are separate from the “social ”self we present to others (Altman, 1975). Other psychologists describe these mo- ments of solitude as a kind of emotional renewal. Adults and adolescents in the pager studies said they felt more cheerful and more alert after spending short periods of time by themselves (Larson et al., 1982). Not surprisingly, a common complaint among people experiencing stress is that they have too lit- tle time to themselves (Webb, 1978).1020304050Percentage of Day Spent Alone Children 9–12High School StudentsAdults Retired Adults FIGURE 12.6 Percentage of Day Spent Alone as a Function of Age Source: Adapted from Larson (1990).Solitude 335

But sometimes people need more than a few minutes alone. Occasionally, we need a more extended amount of time by ourselves to work through per- sonal problems and make important decisions. Although consulting with others can be useful, many times people need extended time alone to think things through. Time for contemplation may be especially valuable for adoles- cents as they address personal questions about religion, values, personal iden- tity, and life goals. To test this possibility, adolescents in one study were followed for one week through the Experience Sampling Procedure to deter-mine how much time they spent by themselves (Larson, 1997). The investiga- tor found that teenagers who spent a moderate amount of time by themselves, roughly between 25% and 45% of their nonclass hours, tended to be betteradjusted and less depressed than either those who spent very little time in sol- itude or those who spent a great deal of their time alone. Teenagers who spent a moderate amount of time alone also had better grades than the otherstudents. These findings also make it clear that although some time alone has positive benefits, more solitude is not necessarily better. Teenagers who spend an excessive amount of time away from others may fail to accrue some of thebenefits that come from social contact. For some, the benefits of solitude may be found in extended periods of isolation. Long periods of solitude —days or perhaps even weeks alone —can provide the opportunity to develop oneself spiritually, intellectually, and crea- tively. One psychologist found examples of several influential people whose contributions could be traced back to an extended period of isolation and in- trospection (Storr, 1988). The works of several famous writers, such as Beatrix Potter and Rudyard Kipling, are the result of inspiration that evolved during extended solitude. Similarly, many religious leaders, including Jesus and Buddha, are said to have come to their insights during extended periods Some people find time by themselves lonely and painful, but for others time alone is precious.336 CHAPTER 12 The Humanistic Approach

alone (Storr, 1988). Even psychologists have been known to take advantage of extended solitude. You may recall from Chapter 5 that Carl Jung deliber-ately isolated himself for the better part of 7 years while he explored the con- tents of his own unconscious. Isolation from others can also be used during psychotherapy (Suedfeld, 1980, 1982). Volunteers who go through extended periods of social isolation and sensory restriction often describe the experi- ence as pleasant and rewarding. Individual Differences in Preference for Solitude It seems clear that spending time alone can have both positive and negative consequences. Solitude can be boring and lonely, or it can bring insight and a sense of restoration. Whether people dislike or enjoy their time alone maybe a function of their preference for solitude . As with other personality vari- ables, researchers find people exhibit relatively stable patterns in the extentto which they seek out and enjoy time by themselves (Burger, 1995; Cramer &Lake, 1998; Haans, Kaiser, & de Kort, 2007; Larson & Lee, 1996; Pedersen, 1999). On one end of this individual difference dimension, we have people who avoid solitude whenever possible and get eaten up by loneliness and sad-ness when forced to spend even a few hours by themselves. People on the other end of this dimension are more likely to resemble the self-actualized individuals Maslow described. They appreciate the benefits that come from solitude andprobably arrange their days so they have at least a little time to themselves to collect their thoughts and reflect on matters that concern them. In one demonstration of these individual differences, college students were asked to complete daily reports of their activities for 7 consecutive days (Burger, 1995). Students filled out a 24-hour report sheet indicating what they had done each hour of the day, whether they had been alone or with others, and whether they had found the experience enjoyable. The researcher also measured students ’preference for solitude. After eliminating time spent in class, at work, and sleeping, students with a high and low preference for solitude were compared for how they spent their free time. As shown in Table 12.3, virtually all the students spent most of their free time with other people. Thus it is not the case that people who enjoy solitude are hermits who avoid contact with others. However, the students with a high preference TABLE 12.3 Free Time Spent by Students with High and Low Preference for Solitude Preference for Solitude High Low Percentage of time spent alone 19.80 11.00 Percentage of time alone rated as pleasant 74.50 55.80Percentage of time with others rated as pleasant 87.30 92.90 Source: From Burger (1995).Solitude 337

for solitude did manage to find more time for solitude in their days than the typical student. Moreover, whereas almost all the students said their timewith others was pleasant, the students with a high preference for solitude were significantly more likely to report that their time alone was enjoyable. In short, people with a high preference for solitude do not avoid social encounters and, in fact, enjoy their time with others quite a lot. Moreover, people who enjoy time alone are not simply seeking an escape from social anxiety (Leary, Herbst, & McCrary, 2003). Rather, these individuals findsomething positive in being by themselves. What are these positive features of solitude? Undergraduate students in one study rated the importance of seven positive aspects of solitude they sometimes experience when by them-selves (Long, Seburn, Averill, & More, 2003). As shown in Table 12.4, the list runs from solving problems to a spiritual experience. These findings are consistent with Maslow ’s observations of psychologically healthy people. Not only are a preference for solitude and well-being compatible, they may actually go hand in hand. SUMMARY 1. Although humanistic psychologists sometimes shun empirical studies, re- search on topics introduced or promoted by these psychologists has pro- vided insight into some important aspects of humanistic personality theory. 2. Research on self-disclosure finds that people follow social rules con- cerning when and how to reveal information about themselves. Foremost among these is the rule of disclosure reciprocity. People ina get-acquainted situation tend to match the intimacy level of their conversational partner. However, friends who have already shared338 CHAPTER 12 The Humanistic Approach TABLE 12.4 Seven Positive Aspects of Solitude Problem Solving Opportunity to think about specific problems or decisions you are facing Inner Peace Feel calm and relaxed, away from the pressures of everyday life Self-Discovery Gain insight into your fundamental values and goals, come to realize your unique strengths and weaknesses Creativity Stimulates novel ideas or innovative ways of expressing yourselfAnonymity Act in whatever ways you feel like at the moment, without concern for social niceties or what others might think Intimacy Although alone, you feel especially close to someone you care about Spirituality A sense of transcending everyday concerns, of being a part of something grander than yourself Source: From Long et al. (2003).

intimate information in a reciprocal manner need not always return to this pattern. Other studies find that men and women are restricted inwhat they disclose by what societ y deems appropriate. Holding trau- matic secrets inside may take its toll on a person ’sh e a l t h . 3. Loneliness is not the same as isolation. Researchers define loneliness as a discrepancy between the amount and quality of social contact we desire and the amount and quality we receive. Although loneliness is influenced by social situations, people tend to suffer from loneliness at a fairly stablelevel. Research on chronically lonely people indicates they approach con- versations with negative expectations and lack some basic social skills. Because of this tendency, they inadvertently stifle social interactions anddiscourage potential friends. 4. High and low self-esteem people react differently to failure. Low self- esteem people become discouraged and unmotivated when they receivenegative feedback, whereas high self-esteem people employ tactics to blunt the effects of failure. Studies suggest that people base their self-esteem on how they perform in selected domains. Although people typically use areasin which they excel for their contingencies of self-worth, some individuals select contingencies that make it difficult for them to feel good about themselves. Research indicates that notions about the self and self-esteem taken from individualistic cultures may not apply to collectivist cultures. 5. Maslow observed that virtually all of his psychologically healthy people reported a high preference for solitude. Subsequent research finds thatmost people spend a large percentage of their time in solitude. Although some people typically find this time unpleasant, others seek out and enjoy their time alone. Researchers find that people differ in the extent to which they prefer solitude. People with a high preference for solitude enjoy their time alone but also enjoy time spent with others. KEY TERMS contingencies of self-worth (p. 328) self-disclosure (p. 308) self-esteem (p. 325) MEDIA RESOURCES Visit our website. Go to www.cengage.com/ psychology/burger to find online resources directly linked to your book, including quizzing, glossary, flashcards, and more.Summary 339

CHAPTER13 The Behavioral/Social Learning Approach Theory, Application, and Assessment Behaviorism Basic Principles of Conditioning Social Learning Theory Social-Cognitive TheoryApplication: Behavior Modification and Self-Efficacy TherapyAssessment: Behavior Observation MethodsStrengths and Criticisms of the Behavioral/ Social Learning Approach Summary 340

What do the following scenes have in common? A hospital patient suffering from depression makes her bed in the morning, dresses herself, and shows upto breakfast on time. A staff member hands the patient three tickets for com- pleting these three acts. A middle-aged man attends a workshop to overcome a fear of snakes. He watches another middle-aged man, who shows no out- ward signs of fear, pick up a snake. A college student declines her friends ’re- quest to join them at a party. She knows people will be smoking there and istrying to break her habit. In each of these scenes, someone is attempting to modify behavior by ap- plying basic principles of learning. Psychologists have studied how people andanimals learn for almost as long as there have been psychologists. The list oftopics examined from the behaviorist ’s perspective includes attitude change, language acquisition, psychotherapy, student –teacher interactions, problem solving, gender roles, and job satisfaction. Naturally, such a far-reachingapproach to the understanding of human behavior also provides a model for explaining why people engage in consistent behavior patterns —that is, a model for personality. Behavioral accounts of personality have gone through a slow but steady transition over the years. Early behaviorists limited their descriptions to ob- servable behaviors. Later, social learning theorists expanded the scope of the approach to include nonobservable concepts like thoughts, values, expectan- cies, and individual perceptions. Social learning psychologists also recognized that people can learn simply by watching someone else or even hearing about another person’ s behavior. More recently, many behavioral psychologists have moved toward invoking more cognitive explanations of behavior, so that today the line between behavioral and cognitive approaches to personal- ity is sometimes blurred. It is not uncommon these days for counselors to refer to themselves as “cognitive-behavioral ”therapists. Nonetheless, both traditional behaviorism —presented next —and the cognitive approaches to personality described in later chapters have a lot to tell us about the causesof personality and avenues for behavior change. BEHAVIORISM In 1913 a young and brash psychologist named John B. Watson published an article titled “Psychology as the Behaviorist Views It. ”This article signaled the beginning of a new movement in psychology called behaviorism. By 1924, with the publication of his book Behaviorism, Watson had made signifi- cant progress in his effort to redefine the discipline. He argued that if psychol- ogy were to be a science, psychologists must stop examining mental states. Researchers who concerned themselves with consciousness, the mind, andthoughts were not engaging in legitimate scientific study. Only the observable was reasonable subject matter for a science. Because our subjective inner feel- ings cannot be observed or measured in an agreed-upon, accurate manner,they have no place in an objective science. The sooner psychology abandons these topics, Watson maintained, the sooner it can become a respectable member of the scientific community.Behaviorism 341

What, then, was the appropriate subject matter for psychology? Watson ’s answer was overt behavior —that which can be observed, predicted, and even- tually controlled by scientists. We should appreciate just how much of psy- chology Watson was ready to jettison in his quest. Emotions, thoughts, expectancies, values, reasoning, insight, the unconscious, and the like were of interest to behaviorists only if they could be defined in terms of observable behaviors. Thus, according to Watson, thinking was simply a variant of ver- bal behavior, a “subvocal speech, ”as evidenced by the small vocal-cord movements he claimed accompanied it. John B. Watson 1878 –1958 As a child growing up in Greenville, South Carolina,John Broadus Watsonexhibited two characteristicsthat would later come to shape his career —he was a fighter, and he was a builder. He once wrote that his favorite activity inelementary school was fighting with classmates “until one or the other drew blood. ”But by age 12 he had also become something of a master carpenter. Later,during his first few years as a psychology professor,he built his own 10-room house virtually by himself. Watson ’s lack of enthusiasm for contemporary standards also surfaced early. In grammar school, “I was lazy, somewhat insubordinate, and, so far as I know, I never made above a passing grade. ”He also found that “little of my college life interested me. …I was unsocial and had few close friends ”(1936, p. 271). Watson bragged about being the only student to passthe Greek exam his senior year at Furman University.His secret was to cram the entire day before the test,powered only by a quart of Coca-Cola. “Today,” he reported years later, “…I couldn ’t to save my life write the Greek alphabet or conjugate a verb ”(1936, p. 272). Watson began his doctoral work in philosophy at the University of Chicago (in part because Princetonrequired a reading knowledge of Greek). He soonswitched to psychology, where, unlike his classmates,he preferred working with rats instead of humans.“Can’ t I find out by watching their behavior, ”he asked, “everything the other students are finding out? ” (1936, p. 276). Watson joined the faculty at Johns Hopkins University in 1908, where he began his quest to replace the psychology of the day with his newbehavioral approach. His views received a surprisinglywarm welcome from many scholars and academics,and in 1912 he was invited to give a series of publiclectures on his theory at Columbia University. Hepublished an influential paper, “Psychology as the Behaviorist Views It, ”in 1913 and his first book in 1914. Within a few years, behaviorism swept over thediscipline. Watson was elected president of theAmerican Psychological Association in 1915. Watsonthe fighter had taken on contemporary psychology and won, whereas Watson the builder had constructedan approach to understanding human behavior thatwould change the discipline of psychology for decades to come. But his academic career was cut short in 1920. Watson suddenly divorced his wife of 17 years and married Rosalie Rayner, the research assistant withwhom he had conducted the Little Albert experiments.The scandal that surrounded these actions forcedWatson out of an intolerant Johns Hopkins andinto the business world, where he eventually settledinto a successful career in advertising. After writing a few popular articles and a book in 1925, Watson severed his ties with psychology while still in his late40s. Several decades later, the foundation he built forthe behavioral approach to personality still stands. Underwood & Underwood/Bettmann/CORBIS342 CHAPTER 13 The Behavioral/Social Learning Approach

At about this same time, other researchers were beginning to study the basic processes of conditioning, or learning. Watson embraced these princi- ples as the key to understanding human behavior. Like Watson, these re- searchers focused their efforts on predicting overt behaviors without introducing inner mental states to explain their findings. The famous Russian physiologist Ivan Pavlov demonstrated that animals could be made to re- spond to stimuli in their environment by pairing these stimuli with events that already elicited a response. This process soon became known as classical conditioning. At the same time, other psychologists were exploring what to- day is known as operant conditioning. For example, Edward Thorndike found that animals were less likely to repeat behaviors that met with negativeconsequences than were animals given no punishment. This work convinced Watson that a few key conditioning principles would suffice to explain almost any human behavior. Personality, accordingto Watson, was “the end product of our habit systems. ”In other words, over the course of our lives we are conditioned to respond to certain stimuli in more or less predictable ways. You might have been conditioned by par-ents and teachers to respond to challenges with increased effort. Someone else might have learned to give up or try something new. Because each of us has a unique history of experiences that shaped our characteristic responses to stimuli, each adult has a slightly different personality. Watson had tremendous faith in the power of conditioning. His most out- rageous claim, which he admitted went “beyond my facts, ”was that given enough control over the environment, psychologists could mold a child into whatever kind of adult they wanted. “Give me a dozen healthy infants, well formed, and my own specified world to bring them up in, ”he wrote. “I’ll guarantee to take any one at random and train him to become any type of spe- cialist I might select —doctor, lawyer, artist, merchant —chief, and yes, even beggarman and thief ”(1924/1970, p. 104). This he promised regardless of the child ’s inherited abilities, intelligence, or ancestry. Although somewhat frightening in its implications for controlling human behavior, this type of thinking found a receptive audience among Americans who believed in the tra-dition of equal opportunity for all regardless of background or social class. Watson’ s legacy was extended by the career of another influential psy- chologist, B. F. Skinner. Skinner, who identified his particular brand of be-haviorism as radical behaviorism, took a small step away from the more extreme position Watson advocated. He did not deny the existence ofthoughts and inner experiences. Rather, Skinner challenged the extent towhich we are able to observe the inner causes of our own behavior. Suppose you are typically uncomfortable at social events. As you prepare for a party one evening, you begin to feel nervous. It ’s going to be a big party, and you don’t think you will know very many people. At the last minute, your anxiety becomes intense and you decide to stay home. Why did you skip the party?Most people would answer that they avoided the party because they felt anx-ious. But Skinner (1974) argued that behavior does not change because you feel anxious. Rather, in this example, the decision to skip the party and theanxiety are both conditioned reactions to the situation.Behaviorism 343

In other words, when we introduce an inner cause like anxiety to explain our actions, we may think we have identified the cause of the behavior, but we are mistaken. When you say you began eating because you were hungry, youhave only put a label on your behavior. You have not explained why you are eating. Similarly, saying that people behave the way they do because they are friendly or aggressive or introverted does not explain where these behaviorscome from. Although radically different in many ways, Skinner ’s view is much like Freud ’s in one respect. Both maintained that people simply do not know the reason for many of their behaviors, although we often think we do.B. F. Skinner 1904 –1990 When Burrhus Frederick Skinner was born inSusquehanna, Pennsylvania,his father, a lawyer,announced the birth in the local paper: “The town has a new law firm: Wm. A. Skinner & Son. ”But all of the father ’s efforts to shape his son into the legal profession failed. After growingup in a “warm and stable ”home, Skinner went to Hamilton College to study English. He planned acareer as a professional writer, not a lawyer. Thisambition was reinforced the summer before his senioryear when an instructor introduced Skinner to the poet Robert Frost. Frost asked to see some ofSkinner ’s work. Skinner sent three short stories, and several months later received a letter from Frostencouraging him to continue writing. Skinner devoted the 2 years after his graduation to writing, first at home and later in Greenwich Village inNew York. At the end of this time he realized he hadproduced nothing and was not likely to become a greatnovelist. “I was to remain interested in human behavior, but the literary method had failed me, ”he wrote. “I would turn to the scientific. The relevant science appeared to be psychology, though I had onlythe vaguest idea of what that meant ”(1967, p. 395). So Skinner went to Harvard to study psychology. He immersed himself in his studies, rising at 6 o ’clock each morning to hit the books. After teaching at theUniversity of Minnesota and Indiana University,Skinner returned to Harvard in 1948, where heremained the rest of his career. Literature ’s loss waspsychology’ s gain. A survey of psychology historians taken just about the time of his death ranked Skinner as the most influential of all contemporary psychologists(Korn, Davis, & Davis, 1991). Although his work in psychology earned him numerous professional awards and recognitions, Skinner never relinquished his interest in literature. Inthe 1940s, he returned to fiction, writing a novel,Walden Two, about a utopian society based on the principles of reinforcement he had found in hislaboratory experiments. “It was pretty obviously a venture in self-therapy, ”Skinner wrote, sounding more psychoanalytic than behavioristic. “I was struggling to reconcile two aspects of my own behavior representedby [the characters] Burris and Frazier ”(1967, p. 403). As this statement suggests, Skinner was not always as anti-Freudian as he is often described. Indeed, early inhis career, he developed what he described as aprojective test based on vague sounds emitted by a phonograph and once sought out an opportunity to go through psychoanalysis himself (Overskeid, 2007). Nonetheless, Skinner remained an adamant believer in the power of the environment and even-tually became an unwavering critic of those whointroduce nonobservable concepts to explain humanbehavior. “I do not believe that my life shows a type of personality à la Freud, an archetypal pattern à laJung, or a schedule of development à la Erikson, ” Skinner wrote nearly 8 decades after his birth. “There have been a few abiding themes, but they can betraced to environmental sources rather than to traitsof character. They became part of my life as I lived it;they were not there at the beginning to determine itscourse ”(1983, p. 401). “If I am right about human behavior, an individual is only theway in which a spe- cies and a culture produce more of aspecies and a culture. ” B. F. SkinnerBettmann/CORBIS344 CHAPTER 13 The Behavioral/Social Learning Approach

Naturally, Skinner’ s theory and some of the implications derived from it are highly controversial. Skinner described happiness as “a by-product of op- erant reinforcement. ”The things that bring happiness are the ones that rein- force us. In his most controversial work, Beyond Freedom and Dignity (1971), Skinner argued that it is time we moved beyond the illusion of per- sonal freedom and the so-called dignity we award ourselves for our actions. We don’ t freely choose to do something as the result of inner moral decisions. We simply respond to environmental demands. We attribute dignity to people for admirable behavior, but because behavior is under the control of external contingencies, dignity is also an illusion. If you rush into a burning building to save people, it is not because you are heroic or foolish but because youhave a history of reinforcements and contingencies in similar situations. BASIC PRINCIPLES OF CONDITIONING Traditional behaviorists explain the causes of behavior in terms of learningexperiences, or conditioning. They do not deny the influence of genetics butdownplay its importance relative to the power of conditioning. According to behaviorists, if we are to understand the processes that shape our personal- ities as well as develop procedures for changing problem behaviors, we must examine basic conditioning principles. It is convenient to divide conditioning into two categories: classical (or Pavlovian) conditioning and operant (or in- strumental) conditioning. Classical Conditioning Classical conditioning begins with an existing stimulus-response (S-R) associa- tion. For example, some people cringe (response) whenever they see a spider (stimulus). Although you may not be aware of them, your behavior repertoire contains a large number of S-R associations. You might feel faint when you see blood, want to eat whenever you smell chocolate, or become nervous when you find yourself more than a few feet off the ground. In his classic demonstration of conditioning, Pavlov used the S-R associa- tion of food and salivation. He presented hungry dogs in his laboratory withmeat powder (stimulus), to which they would always salivate (response). Because this S-R association existed without any conditioning from Pavlov,we call the meat powder the unconditioned stimulus (UCS) and the salivation theunconditioned response (UCR). Then Pavlov paired the old, unconditioned stimulus with a new, conditioned stimulus (CS). Whenever he presented the meat powder to the dogs, he also sounded a bell. After several trials of presentingthe meat powder and the bell together, Pavlov simply sounded the bell without the powder. What happened? As nearly every psychology student knows, the dogs began to salivate at the sound of the bell, even though no meat powder had been presented. The salivation had become the conditioned response (CR), part of a new S-R association (bell tone-salivation) in the dogs ’behavioral repertoire. The classical conditioning procedure is diagramed in Figure 13.1. Once the new S-R association is established, it can be used to condition still anotherBasic Principles of Conditioning 345

S-R association. If you were to pair a green light with Pavlov ’s bell tone, after a while the dogs would start to salivate when the green light came on. This process of building one conditioned S-R association on another is called second-order conditioning. Because the stimuli you experience are often inadvertently paired with other aspects of your environment, you are probably not aware of the manyS-R associations that influence your behavior. Research suggests that our pre-ferences in food, clothing, and even friends can be determined through this process. A friend of mine guesses that he enjoys country and western music because his father used to play it on Saturday, his favorite day of the week.Anxious participants in one study sat in a waiting room with a stranger (Riordan & Tedeschi, 1983). Although the two did not interact, participants reported unpleasant impressions of this other person. The researchers rea-soned that the incidental pairing of the anxiety with the stranger created a negative association with that person. However, researchers have also uncovered several limitations of classical conditioning. For a new S-R association to persist, the unconditioned and conditioned stimuli must be paired occasionally or otherwise reinforced. When Pavlov presented his conditioned dogs with just the bell tone, the dogs salivated less and less until finally the dogs failed to salivate to the tone at all. This gradual disappearance of the conditioned S-R association is called extinction. Moreover, two events presented together will not always produce an association (Rescorla, 1988). Certain stimuli are easily associable, but it may be impossible to create some S-R bonds through classical conditioning. Operant Conditioning At about the time Pavlov was demonstrating classical conditioning in Russia, American psychologists were investigating another type of learning through association. Edward Thorndike put some stray cats into “puzzle boxes. ”To es- cape from the box and thereby obtain a piece of fish, hungry cats had to engage in a particular combination of actions. Before long, the cats learned what they had to do to receive their reward. These observations helped Thorndike (1911)Unconditioned Stimulus (UCS) Conditioned Stimulus (CS)Unconditioned Response (UCR) Conditioned Response (CR) FIGURE 13.1 Classical Conditioning Diagram346 CHAPTER 13 The Behavioral/Social Learning Approach

formulate the law of effect: that behaviors are more likely to be repeated if they lead to satisfying consequences and less likely to be repeated if they lead to un- satisfying consequences. Thorndike ’s cats repeated the required behaviors be- cause they led to the satisfying consequences of escape and food. At first glance Thorndike ’s observations hardly seem insightful. Do you know any parents who don’ t occasionally try rewards and punishments to mold their children ’s behavior? Teachers, judges, and employers also regu- larly rely on the connection between actions and consequences to shape be-havior. But vague feelings that such a connection exists are not the same as understanding how this learning works or the most efficient and productive way to use it. Ask a group of parents the best way to deal with a problemchild, and you will soon understand how little agreement there is among non- scientists on how to use rewards and punishments. This poor understanding of basic learning principles is unfortunate, given the power of conditioning processes. It is especially tragic because several dec- ades of research have provided psychologists with a relatively good under- standing of how reinforcement and punishment shape and control behavior.Unlike classical conditioning, which begins with an existing S-R bond, operant conditioning begins with behaviors the organism (human or lower animal) emits spontaneously. We can observe these operant behaviors when a labora- tory rat is placed in a new cage. The animal moves about, scratches, sniffs, and claws in a haphazard manner, for none of these responses have been Much of what we know about the basic principles of conditioning was first demon- strated with laboratory animals. Here a researcher uses operant conditioning to teach arat to press a bar. The rat receives positive reinforcement (a pellet of food) whenever itpresses the bar.“Happiness is a … by-product of oper- ant reinforcement. The things whichmake us happy are the things which reinforce us. ” B. F. SkinnerBasic Principles of Conditioning 347

reinforced or punished. However, if one of these behaviors is always followed by a pellet of food, its frequency will increase. Operant conditioning concerns the effect certain kinds of consequences have on the frequency of behavior. A consequence that increases the fre- quency of a behavior that precedes it is called a reinforcement, one that de- creases the behavior is called a punishment. Whether a consequence is reinforcing or punishing varies according to the person and the situation. If you are hungry, strawberry ice cream is probably a reinforcement. But if youdon ’t like strawberry ice cream or if you are cold, the ice cream may serve as a punishment. Psychologists have discovered two basic reinforcement strategies for in- creasing the frequency of a behavior (Table 13.1). With positive reinforce- ment, the behavior increases because it is followed by the presentation of a reward. Hungry rats that receive a pellet of food every time they press a barwill begin to press the bar frequently. Students who receive an A after dedi- cated studying for a test are likely to study hard for subsequent tests. We can also increase the frequency of a behavior by using negative reinforcement, the removal or lessening of an unpleasant stimulus when the behavior occurs. Rats that can turn off an electric shock by pulling a string will quickly learn to pull the string. People whose headaches go away when they take a few minutes to relax will soon learn to relax. The other side of operant conditioning is the reduction of unwanted be- haviors. As with the task of increasing desired behaviors, operant condition-ing provides two methods for decreasing undesired behaviors. The most efficient method is to cease reinforcement and thereby allow the behavior to extinguish. Although this concept is simple enough, people often reinforce problem behaviors without realizing what they are doing. A teacher may re- act to a child who acts up in class by criticizing the child in front of the other students. The teacher may not realize that the attention the child gainsfrom other students in the form of laughter and classroom status has turned the intended punishment into a reinforcement. An observant teacher might take disruptive children out into the hall for discipline, thereby removing thereinforcer.TABLE 13.1 Operant Conditioning Procedures Procedure Purpose Application Positive reinforcement Increase behavior Give reward following behavior Negative reinforcement Increase behavior Remove aversive stimulus following behavior Extinction Decrease behavior Do not reward behavior Punishment Decrease behavior Give aversive stimulus following behavior or take away positive stimulus348 CHAPT ER 13 The Behavioral/Social Learning Approach

Alternatively, we can use punishment to eliminate unwanted behaviors. In theory, the frequency of a behavior is reduced when it is followed by an aversive stimulus, such as an electric shock, or the removal of a positive stimu- lus, such as taking away toys. The effects of punishment can be demonstrated in laboratory animals, and therapists have had some success applying this technique in special cases. But research shows the effectiveness of punishment is limited for several reasons. First, punishment does not teach appropriate behaviors, it can only de- crease the frequency of undesired ones. Rather than simply punish a child forhitting another student, it ’s better to help the child learn alternative ways to deal with frustrating situations. Second, to be effective, punishment must be delivered immediately and consistently. A parent needs to punish the problem behavior as soon as possible, not “when your father gets home. ”The punish- ment must also be fairly intense and should be administered after every in- stance of the undesired behavior. Parents who sometimes let their children use bad language but other times decide that to punish such talk will proba- bly have little success in changing their children ’s vocabulary. Third, punish- ment can have negative side effects. Although parents or therapists intend to suppress a certain response, a child might associate other behaviors with the punishment. A child who is punished for hitting a toy against a window may stop playing with toys altogether. In addition, through classical conditioning, aversive feelings that accompany the punishment may be associated with the person doing the punishing. Children who are spanked by their parents may associate the parent with the pain of the spanking. Another side effect is that undesirable behaviors may be learned through modeling. Children who are spanked may learn that physical aggression is okay as long as you are bigger and stronger. Punishment can also create negative emotions, such as fear and anxiety, strong enough to interfere with learning appropriate responses. Taken together, these factors make punishment one of the least desirablechoices for behavior therapists seeking to change problem behaviors. At most, punishment can temporarily suppress an undesirable response long en- ough for the therapist to begin reinforcing a desired behavior. Shaping Suppose you are hired to work with patients in a psychiatric hospital. Your job is to get reluctant patients more involved in some of the activities on the ward. You start with one patient who has never participated in any of the ward activ- ities. Your goal is to get him into daily art therapy sessions. Positive reinforce- ment seems the right tool. Every time the patient joins in one of the voluntary art sessions, you will reward him with coupons for free items in the hospital store. The patient skips art therapy the first day. So no reward. He skips arttherapy the rest of the week. Still no reward. You wait 2 months, and still the patient has not attended one of the sessions. By now, one of the problems en- countered when using operant conditioning is apparent to you: A behaviorcan be reinforced only after it is emitted. Does this mean operant conditioning is useless in this situation? Fortunately, the answer is “No.” A behavior therapist working with theBasic Principles of Conditioning 349

reluctant patient might use a technique known as shaping, in which successive approximations of the desired behavior are reinforced. For example, you might reward the withdrawn patient for getting out of bed and sitting among the other patients. Once this behavior is established, you might reinforce him only when he is near or in the art therapy room. From here, rewards might be limited to time spent in the room during the sessions and later to time spent attending to and participating in the sessions. Shaping is particularly useful when teaching complex behaviors. Children will learn to enjoy reading ifeach step along the way is reinforced. If learning the alphabet, letter sounds, and short words is difficult and unpleasant, it is unlikely the child will move on to reading sentences and stories. Generalization and Discrimination Operant conditioning would be rather limited if every situation required that we learn a new response. Fortunately, because of generalization , this is not the case. Pigeons trained to peck at large red circles to receive food will also peck at small orange circles, although not as frequently. This process, called stimulus generalization, helps explain why personality characteristics general- ize across situations. A child rewarded for acting politely around relatives will probably act politely around new acquaintances. The polite response has been generalized from the stimulus of the relative to the new stimulus, the stranger. When we observe polite behavior consistently across situations, we say thispattern is part of the child’ s personality. As long as the generalized response is met with reinforcement, the behavior is likely to continue. But if the pigeon is not rewarded for pecking at orangecircles, it will soon learn to discriminate between rewarded and nonrewarded stimuli and will peck only at the red ones. Similarly, the polite child may come in contact with adults who punish friendly behavior with harshness. Soon the child will learn to discriminate between people who are friendly and people who aren ’t. The difference between a good and a great tennis player or between a second-string baseball player and a star may be the ability to make fine discriminations between those actions that lead to a reinforcement (a win- ning shot or a home run) and those that do not. SOCIAL LEARNING THEORY It is difficult to overstate the impact traditional behaviorism had on psychology and subsequently the field of personality. Watson and his followers provided a scientific, easily testable account of human behavior that complemented thegrowing empirical flavor of psychology in American universities. The basic principles of learning were so universal they could be tested on lower animals. The image many people have of the lab-coated psychologist, pencil in hand,watching rats running through mazes comes from this era. But somewhere in the 1950s or 1960s the enthusiasm for traditional behaviorism began to wane. Psychologists questioned the assertion that all human learning is the result ofclassical or operant conditioning. “The prospects for survival would be slim in- deed if one could learn only from the consequences of trial and error, ”one350 CHAPTER 13 The Behavioral/Social Learning Approach

psychologist wrote. “One does not teach children to swim, adolescents to drive automobiles, and novice medical students to perform surgery by having them discover the requisite behavior from the consequences of their successes and failures ”(Bandura, 1986, p. 20). Psychologists also began to question whether behaviorism was too limited in the scope of its subject matter. Why couldn ’t “internal” events like thoughts and attitudes be conditioned the same way as overt behaviors? For example, paranoid individuals who believe evil agents are out to get them might have been reinforced in the past for these beliefs.Thus began the transition from traditional behaviorism to a number of ap- proaches known collectively as social learning theory. One of the concepts introduced by social learning theorists is the notion ofbehavior-environment-behavior interactions (Staats, 1996). That is, not only does the environment influence our behavior, but that behavior then de- termines the kind of environment we find ourselves in, which can then influ-ence behavior, and so on. The way people treat you (environment) is partly the result of how you act (behavior). And, of course, how you act is partly a result of how people treat you. Other social learning theorists point out thatpeople often provide their own reinforcers. It is rewarding to live up to your internal standards or to reach a personal goal even if no one else knows about it. Social learning psychologists also helped to bridge traditional behavior- ism and cognitive approaches to personality (Chapter 15) by incorporatinginto their theories a number of concepts once deemed unscientific by John B. Watson. Among the most influential of these social learning theorists is Julian Rotter (1954, 1982; Rotter, Chance, & Phares, 1972). Rotter argues that the causes of human behaviors are far more complex than those of lower ani- mals. He introduced several “unobservable ”concepts to account for human behavior and personality. To get an idea of Rotter ’s approach, imagine someone has just insulted you at a party. How do you respond? You have several courses of action to choose from. You might attempt to top the remark with something clever and witty. You could calmly say the behavior was out of line and ask for anapology. You could get angry and hurl an equally rude insult at the offender, or you could simply leave the scene. The key to predicting your response lies in what Rotter refers to as the behavior potential for each option. The behav- ior potential is the likelihood that a given behavior will occur in a particular situation. Each possible response to the insult has a different behavior poten- tial. If you decide to scream out an insult, it means the behavior potential forthat response was stronger than for any of the other possible responses. But what determines the strength of the behavior potential? According to Rotter, two variables need to be considered: expectancy and reinforcement value(Figure 13.2). Before you decide to stay up all night studying for an exam, you probably ask yourself what the likelihood is that the all-nighter will help you do betteron the test. Similarly, when debating whether to join a softball game, you prob- ably try to figure out the probability that you will have a good time. Rotter re- ferred to these estimations as expectancies, and whether you study all night orSocial Learning Theory 351

play softball is determined by what you expect to happen. Of course, we base our expectancies largely on how things turned out other times we were in this situation. If you always do well after studying all night, you will develop an ex- pectancy of receiving the reward again. If you never enjoy yourself playing soft- ball, the expectancy of being rewarded for playing again is slim. Of course, traditional behaviorists would make these same predictions. People are more likely to engage in a behavior when it has been reinforced. But Rotter and the behaviorists disagree on how to explain the behavior. Behaviorists say that an operant conditioning association or habit has been strengthened by the earlier experience. However, Rotter argues that the more often people are reinforced for a certain behavior (for example, studying all night and receiving an A), the stronger their expectancy that the behavior willbe reinforced in the future. Of course, expectancies are not necessarily accu- rate. You might expect that studying for your SAT will result in a higher score, even if in reality the studying has little effect. In this case, your expectancy will probably predict your behavior better than the actual contingencies. But what about situations we encounter for the first time? In these situa- tions we rely on generalized expectancies —beliefs we hold about how often our actions typically lead to reinforcements and punishments. According to Rotter (1966), each of us can be placed along a continuum called locus of con- trol. At one end of this dimension we find individuals with an extreme internal orientation —those who believe that most of what happens to them is the re- sult of their own actions or attributes. On the other end we find people whohold an extreme external orientation —those who maintain that much of what happens to them is the result of forces outside their control, such as chance or powerful others. As we will see in the next chapter, where we fall on this di- mension has important implications in such areas as achievement and health. The second component in Rotter ’s model is reinforcement value, the de- gree to which we prefer one reinforcer over another. Naturally, the reinforce- ment value we assign a certain outcome can vary from situation to situation.When we are lonely, social contact holds a higher reinforcement value than when we aren ’t. Yet each of us also have reinforcers we almost always value more than others. Some people consistently work hard, placing their job ahead of family and recreation. We might call these people obsessive or driven. But using Rotter ’s model, their personalities can be explained in terms of the consistently high value they put on achievement.Behavior Potential (BP) = Expectancy (E) + Reinforcement Value (RV) FIGURE 13.2 Rotter ’s Basic Formula for Predicting Behavior352 CHAPTER 13 The Behavioral/Social Learning Approach

SOCIAL-COGNITIVE THEORY The evolution from traditional behavioral views of personality to more cogni- tive approaches is probably best illustrated by the work of Albert Bandura (1977a, 1986, 2001, 2006). Bandura rejects the behaviorist ’s depiction of human beings as passive recipients of whatever stimuli life throws their way. Certainly individuals respond to environmental events, and certainly they of- ten learn characteristic behaviors as the result of rewards and punishments. But people possess other capacities that are distinctly human. By reducingthe process through which people grow and change to the way a rat learns to press a bar, strict behaviorists overlook some of the most important causes of human behavior. Because these overlooked causes generally involve think-ing and symbolic processing of information, Bandura refers to his approach as a social-cognitive theory. Reciprocal Determinism Bandura adds a new twist to the question of whether behavior is determined by internal or by external forces. He argues that there are both internal andexternal determinants of behavior, but behavior is not determined exclusivelyJulian B. Rotter 1916 – Julian Rotter first learned about psychology in theAvenue J Library inBrooklyn, where he spent agreat deal of his grade school and high schoolyears. One day, after exhausting most of the books in other sections of thelibrary, he wandered over to the Philosophy andPsychology shelf. Among the first books he encoun-tered were Alfred Adler ’sUnderstanding Human Nature and Sigmund Freud ’sPsychopathology of Everyday Life. From that point on, he was hooked. But for a time his love of psychology took a backseat to the realities of the world. He decided to major inchemistry at Brooklyn College because “there was no profession of psychology that I knew of. And in 1933,in the depths of the Great Depression, one majored in a subject one could use to make a living ”(1982, p. 343). But circumstances soon intervened. One day during his junior year, Rotter discovered that Alfred Adler wasteaching at the Long Island School of Medicine. Rotter began attending the lectures. Eventually, Adler invitedRotter to attend the monthly meetings of the Society ofIndividual Psychology held in Adler’ s home. Unfortunately, Adler died the next year. Nonetheless, by then Rotter ’s enthusiasm for psychology dictated that he go to graduate school. He chose the University of Iowa so that he could studywith the famous Gestalt psychologist Kurt Lewin. Hewent to the University of Indiana for his PhD because itwas one of the few schools at the time to offer a degreein clinical psychology. He wanted an academicposition, but few were available when Rotter graduatedin 1941. After working in a hospital for a year, Rotterserved as a psychologist in the Army and later the AirForce during World War II. Circumstances altered Rotter ’s career path again following the war. The need for clinical psychologists was suddenly high, but their numbers were few. Rotter took a position at Ohio State University, finally fulfilling his ambition to be a professional academicpsychologist. He stayed there until 1963, when hemoved to the University of Connecticut. University of Connecticut, Department of PsychologySocial-Cognitive Theory 353

by either or by a simple combination. Bandura introduces instead the concept reciprocal determinism . That is, external determinants of behavior, such as re- wards and punishments, and internal determinants, such as beliefs, thoughts, and expectations, are part of a system of interacting influences that affect not only behavior but the various parts of the system as well. Put more simply, each part of the system —behaviors, external factors, and internal factors — influences each of the other parts. Some examples will help clarify the concept. Like Rotter, Bandura maintains that internal factors, such as our expectancies, affect our behavior. Suppose someone you don ’t like much asks you to play racquetball. You can just imagine what a dismal afternoon you would have with this person. Thus your internal ex- pectation will probably cause you to reject the invitation. But what would happen if this person offered to buy you that new, expensive racket you ’ve been eyeing if you play with him? Suddenly the external inducement is powerful enough to de- termine your behavior, and you say, “Let’sp l a y . ”Now imagine further that you have one of the most enjoyable sets of racquetball ever. You ’re evenly matched with this person, and he even cracks a few jokes to make the afternoon fun. You actually look forward to playing with him again. The behavior in this case has changed your expectations, which will affect future behavior and so on. The reciprocal determinism process is diagramed in Figure 13.3. You may notice that the arrows point in both directions, indicating that each ofthe three variables in the model is capable of influencing each of the other variables. This situation is very different from traditional behaviorism, which limits explanations of human behavior to a two-factor, one-way model in which external events cause behavior. How can we predict which of the three parts in the reciprocal determinism model is going to influence which other part? This depends on the strength of each of the variables. At times, environmental forces are most powerful; at other times, internal forces dominate. The example used in Chapter 7 of bothhigh and low self-esteem people fleeing a burning building illustrates how envir- onmental factors can override internal individual factors on occasion. Though at times we mold our environment to meet our needs, at other times we are Behavior External Factors (Rewards, Punishments)Internal Factors (Beliefs, Thoughts, Expectations) FIGURE 13.3 Bandura ’s Reciprocal Determinism Model354 CHAPTER 13 The Behavioral/Social Learning Approach

faced with environmental factors we cannot control. We often create our own opportunities and defeats, but they can also be created for us. Imagination and Self-Regulation Bandura identifies several features unique to humans that must be consideredto understand personality. Unlike lower animals, people use symbols and forethought as guides for future action. Instead of working our way through rewards and punishments in a trial-and-error fashion every time we face a new problem, we imagine possible outcomes, calculate probabilities, set goals, and develop strategies. We do all of this in our mind without engaging in ran- dom actions and waiting to see which will be rewarded or punished. Ofcourse, past experiences with reinforcements or punishments affect these judg- ments. But think about the way you prepare for a vacation. Most likely you think about several options of where and when to go, how to get there, who to go with, what to bring, what to do when you arrive, and so on. By imagin- ing what a vacation will be like at various locations and with various people,you don’ t have to literally try out each option to see if the experience will be reinforcing or punishing. Bandura also argues that most behavior is performed in the absence of external reinforcements and punishments. Our daily actions are largely con- trolled by what he calls self-regulation . Although we often strive to obtain external rewards, we also work toward self-imposed goals with internal Albert Bandura 1925 – Albert Bandura was born in Mundare, a small farmingcommunity located amongthe wheat fields of Alberta,Canada. His parents had emigrated from EasternEurope when they were teenagers. They had no formal schooling themselves,but communicated to their son the high value theyplaced on education. Bandura attended the only schoolin the area, a combined elementary and high schoolwith a total of about 20 students and two teachers.Summer jobs included filling in holes in the highwaysof the Yukon. He stayed in Canada for his undergraduate education, receiving a bachelor ’s degree from the University of British Columbia in 1949. Hehad intended to major in a biological science, but hebecame enamored with psychology after enrolling in anintroductory course one term simply because the early-morning class fit his schedule.Bandura chose the University of Iowa for graduate work, in part because of its strong tradition in learning theory. Among the Iowa faculty members whoinfluenced Bandura was the learning theorist KennethSpence. The faculty at Iowa also emphasized the need for empirical research. This training left Bandura withthe conviction that psychologists should “conceptualize clinical phenomena in ways that would make themamenable to experimental tests ”(as cited in Evans, 1976, p. 243). He received his PhD in 1952. After a year of clinical internship in Wichita, Bandura accepted a position at Stanford University in1953 and has remained there ever since. While atStanford, he has continued to build bridges betweentraditional learning theory and cognitive personalitytheories and between clinical psychology and empirically oriented approaches to understanding personality. Bandura has received numerousprofessional honors, including election to thepresidency of the American Psychological Associationin 1974. Linda A. Cicero/Stanford News ServiceSocial-Cognitive Theory 355

rewards. Amateur runners push themselves in races, even though few expect to win. The reward comes from the feelings of accomplishment and self-worth they get from setting a personal record or perhaps for just finishing the race. Of course, self-regulation also includes self-punishment. When we fail to maintain personal standards, we often degrade and feel bad about our-selves. You may have chastised yourself for being rude to a stranger or not sticking to your diet, even when no one else seemed to notice. Because much of our behavior is the result of self-regulation, Bandura challenges the radical behaviorist assertion that people will perform just aboutany action if the environmental contingencies are altered appropriately. “Anyone who attempted to change a pacifist into an aggressor or a devout reli- gionist into an atheist, ”Bandura wrote, “would quickly come to appreciate the existence of personal sources of behavioral control ”(1977a, pp. 128– 129). Observational Learning Perhaps social-cognitive theory ’s most important contribution to the understand- ing of human behavior and personality is the concept of vicarious orobserva- tional learning . In addition to classical and operant conditioning, we can learn by observing or reading or just hearing about other people ’s actions. Many behav- iors are too complex to be learned through the slow process of reinforcement andpunishment. We don ’t teach pilots to fly by putting them in the cockpit and rein- forcing correct behaviors and punishing incorrect ones. Bandura maintains that children would never learn to talk during their preschool years if they had to be reinforced for every correct utterance. Instead, the pilots and the toddlers watchothers fly and talk, noting which behaviors work and which don’ t. Bandura draws an important distinction between learning and perfor- mance. Behaviors learned through observation need not be performed. This idea again clashes with traditional behaviorists, who maintain that we cannot learn something until we have actually engaged in that behavior. But think for a moment of some of the behaviors you could perform if you wanted to,even though you never have. For example, although you have probably never picked up a pistol and shot another human being, you ’ve observed this be- havior in movies often enough for it to be part of your behavioral repertoire. You might even know to stand with your feet apart and to hold the weapon at eye level with both hands in front of you, just like the actors portraying police do. Fortunately, most of us will never perform this behavior, but it isone we have probably learned through observation. Why do we perform some of the behaviors we learn through observation but not others? The answer lies in our expectations about the consequences.That is, do you believe the action will be rewarded or punished? In the case of shooting another person, most of us expect this behavior will be punished —if not in a legal sense, then through self-punishment in the form of guilt and low- ered feelings of worth. But if we have never performed the behavior, where do we get our expec- tations about consequences? Again, from observing others. Specifically, wasyour model for the behavior rewarded or punished? For example, a high school boy may watch an older friend ask someone for a date. He pays close356 CHAPTER 13 The Behavioral/Social Learning Approach

attention to how the friend engages the potential date in conversation, what is said, and so on. If the friend ’s behavior is rewarded (a date is made), the boy may believe that he, too, will be rewarded if he acts just like his friend. Most likely, he ’ll soon get his courage up and ask out someone he ’s had his eye on for a while. And if the older friend is turned down? It ’s unlikely the boy will imitate the punished behavior. In both cases, the boy paid close enough atten- tion to learn how his friend went about asking for a date. But whether he will perform the behavior depends on what he thinks will happen. Bandura (1965) demonstrated this learning-performance distinction in a classic experiment with important social implications. Nursery school chil- dren watched a television program in which an adult model performed fournovel aggressive acts on an adult-size plastic Bobo doll: First, the model laid the Bobo doll on its side, sat on it, and punched it in the nose while remarking, “Pow, right in the nose, boom, boom. ”The model then raised the doll and pommeled it on the head with a mallet. Each response wasaccompanied by the verbalization, “Sockeroo …stay down. ”Following the mallet aggression, the model kicked the doll about the room, and these responses were interspersed with the comment, “Fly away. ”Finally, the model threw rubber balls at the Bobo doll, each strike punctuated with “Bang. ”(pp. 590 –591) The children saw one of three endings to the film. Some saw a second adult re- ward the aggressive model with soft drinks, candy, and lots of praise. Others sawthe model spanked with a rolled-up magazine and warned not to act aggressively again. A third group was given no information about the consequences of the ag- gressive behavior. Next, each child was left alone for 10 minutes of free playing time. Among the many toys in the room were a Bobo doll and all the materials needed to perform the aggressive acts they had seen. An experimenter watched through a one-way window to see how many of the four acts of aggression thechildren would perform spontaneously. Each child was then offered fruit juice and small toys for each of the four aggressive acts he or she could perform for the experimenter. This last step was included to see if the children could perform the behavior —that is, had they learned the responses from watching the model? The results are shown in Figure 13.4. Nearly all the children in all three groups could perform the behaviors when asked. However, whether theychose to perform the behavior when left alone depended on the consequences they expected. Although all of the children had learned how to act aggres- sively, the ones who had seen the model rewarded were significantly morelikely to perform the behaviors than those who had seen the model punished. APPLICATION: BEHAVIOR MODIFICATION AND SELF-EFFICACY THERAPY One of the appeals of traditional behaviorism is its presentation of a simple, rational model of human nature. Looking at the world through a behaviorist ’s eyes, everything makes sense. Employees work hard when they are reinforcedproperly. Children stop fighting when aggressive behavior is punished and working together is reinforced. But what about some of the seemingly irratio- nal behaviors we see in people suffering from psychological disorders? HowApplication: Behavior Modifi cation and Self-Efficacy Therapy 357

can basic conditioning principles explain a fear of stairs or a belief that people are out to get you? As you will see, not only can behaviorists account for these and other abnormal behaviors, but many psychotherapy techniques are basedon basic conditioning principles. Explaining Psychological Disorders John B. Watson was the first to demonstrate how seemingly “abnormal” be- haviors are created through normal conditioning procedures. Watson usedclassical conditioning to create a fear of white rats in an 11-month-old babyknown as Little Albert (Watson & Rayner, 1920). As shown in Figure 13.5, Watson began with the stimulus-response association between a loud noise and fear present in most infants. That is, whenever Watson would make theloud noise, Albert would cry and show other signs of fear. Next, Watson showed Albert a white rat, each time accompanied by the loud noise. Soon Albert was responding to the white rat with fear responses (crying, crawlingaway) similar to those he had made to the loud noise, even when the noise was not sounded. Watson demonstrated that what appeared to be an abnor- mal fear of white rats in an infant could be explained by knowing the pastconditioning of the child. Behaviorists argue that many of our seemingly irrational fears may have been developed in a similar manner. We may not recall when bridges or snakes were ever associated with an existing fear, but these associations could have taken place a long time ago or even without our awareness. However,1 0234Mean Number of Different Imitative Responses Reproduced Boys Girls Model RewardedBoys Girls Model PunishedBoys Girls No Consequences FIGURE 13.4 Mean Number of Aggressive Responses Performed Source: From “Influence of models ’reinforcement contingencies on the acquisition of imitative responses, ”by A. Bandura, Journal of Personality and Social Psychology , 1965, 1, 589 –595.358 CHAPTER 13 The Behavioral/Social Learning Approach

there is a problem with this explanation. As Pavlov discovered, new associa- tions formed through classical conditioning extinguish once the pairing is re-moved. Why, then, do phobias not just become extinct on their own without psychological intervention? One answer is that operant conditioning may take over. Imagine a 3-year-old girl who falls off a tall slide. The pain and fear she experiences are paired with the slide, and those feelings reemerge the next time she approaches the playground. Her anxiety increases as she gets closer and closer to the slide. Quite likely, she ’ll decide to turn away and try the slide some other time, thereby reducing the anxiety. What has happened in this situation is that the act of avoiding the slide has been reinforced through negative reinforcement. Running away was followed by a reduction in theaversive stimulus, the feelings of fear and anxiety. If this avoidance behavior is reinforced a few more times, the girl could develop a strong fear of slides. This fear might then be generalized to a fear of all high places, and years laterthe woman may seek therapy for this debilitating phobia. Psychologists also use conditioning principles to explain why some people develop fears after a traumatic experience whereas others do not (Mineka & Zinbarg, 2006). On one hand, previous learning can protect us from develop- ing phobias. A boy who has had many pleasant experiences with dogs is not likely to develop a fear of the animal when one untrained dog snarls at and bites him. One study found children who had many uneventful trips to the dentist were less likely than new patients to develop a fear of dentists after one painful visit (Kent, 1997). On the other hand, a learning history with sev- eral small fearful experiences can make a person more vulnerable to develop- ing a phobia when a very traumatic experience occurs (Mineka & Zinbarg, 2006). A girl who has been frightened several times by loud and splashing children at the local pool is especially vulnerable to developing a fear of water after a near-drowning experience. Behaviorists view other problem behaviors as the result of reinforcing the wrong behavior. A socially anxious girl may have found the only escape fromLoud Noise (Unconditioned Stimulus) White Rat (Conditioned Stimulus)Fear Responses (Unconditioned Response) Fear Responses (Conditioned Response) FIGURE 13.5 Diagram of Little Albert ’s Classical ConditioningApplication: Behavior Modifi cation and Self-Efficacy Therapy 359

the criticism and ridicule she received at home was to avoid family contact as much as possible (negative reinforcement), a behavior she then generalized toother people. A man suffering from paranoid delusions may believe he has thwarted a plan to kidnap him by staying in his house all day, thereby re- warding the behavior. Behavior theorists also explain a lack of appropriate behaviors as the result of too little reinforcement. For example, if the socially anxious woman ’s efforts to initiate conversations with others are never re- warded, she ’ll probably stop trying. Behavior Modification If we accept that problem behavior is the result of unusual conditioning experiences, then another principle of behaviorism should follow: Problem behaviors can be changed through appropriate conditioning experiences. Several therapy procedures, often grouped under the label behavior modifica- tion, have been developed from behaviorist theory and research. These proce- dures differ from more traditional therapies in several respects. The treatmentusually lasts for several weeks, as compared to perhaps years. The focus is onchanging a few well-defined behaviors rather than changing the entire person- ality of the client. And behavior therapists are often unconcerned with discov- ering where the problem behavior originated. Their goal is simply to removeit or replace it with a more appropriate set of responses. These features have made behavior modification techniques popular among therapists from a va- riety of theoretical orientations. Classical Conditioning Applications Pairing one stimulus with another is a powerful tool for creating newstimulus-response associations. Therapists often use classical conditioning to eliminate or replace stimulus-response associations that cause clients pro- blems. Although these techniques traditionally use physical pairing of objects and reactions, psychologists also find that mental images can be classically conditioned (Dadds, Bovbjerg, Redd, & Cutmore, 1997). Thus, in the safety of a behavior therapist ’s office, clients can imagine themselves facing the si- tuations they fear without actually visiting those places. One treatment for phobias pairs images of the feared object with a relaxa- tion response. Through systematic desensitization the old association between the feared stimulus and the fear response is replaced with a new association be-tween the stimulus and relaxation. Clients and therapists begin the treatmentby creating a list of imagined scenes ranging from mildly arousing to highly anxiety provoking. A man afraid of heights might begin his list with a scene of himself standing on a 2-foot-high footstool. The next scene might be walking up a flight of stairs, followed by a scene of him standing on an 8-foot ladder. Last on the hierarchy come the highly anxiety-provoking scenes, such as look- ing out from the top floor of a skyscraper or flying in a small airplane. After clients complete relaxation training, they imagine the scenes while they practice relaxing. One step at a time, they slowly move through the listuntil they can imagine the scene without feeling anxious. In theory, the fear response is being replaced with a new, incompatible response —relaxation.360 CHAPTER 13 The Behavioral/Social Learning Approach

If this therapy works, clients who used to be mildly anxious when thinking about standing on a 2-foot stool can imagine (and eventually perform) look-ing out over the city from the top of a tall building without experiencing fear. Aversion therapy is another example of classical conditioning used to al- ter problem behaviors. Here therapists try to rid clients of undesirable beha-viors by pairing aversive images with the behavior. For example, for a client trying to quit smoking, the image of a cigarette might be paired with images of becoming nauseated and vomiting. Operant Conditioning Applications Sometimes therapy can be as basic as reinforcing desired behaviors and pun-ishing undesirable ones. However, this is more difficult than it may sound. Behavior modification therapists begin this type of treatment by identifying the target behavior and defining it in specific operational terms. For example, what would you reinforce or punish when a child ’s problem is “acting imma- ture”? A behavior therapist would probably interview parents and teachers to determine which specific immature behaviors they wanted to reduce. Next, the therapist would want to determine a baseline of behavior frequency. How do you know if you ’re reducing the frequency of a behavior if you don’t know how often it occurs now? Through observation or interviews, the therapist might find that a child throws an average of two and a half tan-trums per week. Once we know how often the behavior occurs under the current system of rewards and punishments, we change the contingencies. If it ’s a desired be- havior, the environment is altered so the client is rewarded for it. If it ’s an un- desired behavior, punishment is introduced or reinforcement is reduced. Ideally, appropriate responses are reinforced at the same time undesired be- havior is extinguished or punished. In the case of the child throwing tantrums, parents might be told to stop rewarding the action with their attention and concern. In addition, punishments might be introduced, such as not allowing the child to watch television for 24 hours after a tantrum. At the same time, the child should be reinforced for handling frustrating situations in an appro- priate way, such as by seeking help. The frequency of the target behavior is monitored throughout the therapy. After a few weeks, the therapist can see if the treatment is working or if adjustments need to be made. If the child is down to one tantrum a week, the treatment will probably continue for a few more weeks until the tantrums disappear entirely. If they are still occurring two and a half times a week, a new therapy program may need to be developed. A therapist who wants to change a large number of behaviors for a large number of people at once might use another treatment system based on oper-ant conditioning called a token economy. People in a well-defined institutional unit, such as a psychiatric ward or a class, are given the opportunity to earn to- kens (for example, poker chips) worth a certain number of points. They can exchange these tokens for more tangible rewards, such as snack food or extra privileges. Psychiatric patients might be given 2 tokens for making their bedsin the morning, 5 for attending therapy sessions on time, 10 for doing theirApplication: Behavior Modifi cation and Self-Efficacy Therapy 361

assigned work on the ward, and so on. When clients show inappropriate behaviors, they might be punished by having tokens taken away. Biofeedback is another type of operant conditioning used to treat psycho- logical problems. Biofeedback requires special equipment that provides infor- mation about somatic processes. A woman suffering from anxiety might use a machine that tells her when she is tightening and relaxing certain facial and back muscles, a reaction she is otherwise not aware of. After several muscle relaxation sessions with the immediate feedback of the machine, she maylearn to reduce tension on her own and thereby overcome her anxiety. In op- erant conditioning terms, the woman was reinforced for producing the re- sponse that lowered her muscle tension, as indicated by the machine. As withother reinforced behaviors, she soon learned to make the relaxation response. One of the most common targets of biofeedback treatments is the tension headache. To reduce these headaches, psychologists use electromyographymonitors to inform clients about muscle constriction that they otherwise might not be aware of. A review of studies using this procedure found it was highly effective in reducing the frequency, intensity, and duration of tensionheadaches (Nestoriuc, Rief, & Martin, 2008). Moreover, the procedure was relatively quick and efficient. Participants averaged fewer than 11 biofeed- back sessions, and the improvements were seen in follow-up assessments 15 months later. Other bodily indicators that may be controlled through biofeed- back include blood pressure, heart rate, and brain waves. Self-Efficacy Every year millions of Americans seek professional help to stop smoking or lose weight. Although many of these people go several weeks without cigarettes or succeed in dropping a few pounds, only a small percentage permanently end their habit or keep the lost pounds off. What is it about these few successful cases that separate them from the others? The answer may lie in what Bandura calls self-efficacy . People stop smoking and lose weight when they convince them- selves they can do it. Many smokers complain that they have tried to quit many times but just can ’t. From a social-cognitive analysis, one reason these smokers are unable to kick their habit is precisely because they believe they cannot. According to Bandura (1977b, 1997), people aren ’t likely to alter their be- havior until they make a clear decision to expend the necessary effort. Bandura draws a distinction between outcome expectations and efficacy expectations. Anoutcome expectation is the extent to which people believe their actions will lead to a certain outcome. An efficacy expectation is the extent to which they believe they can bring about the particular outcome. Simply put, it is the differ- ence between believing that something can happen and believing that you can make it happen. You may hold the outcome expectation that if you devote sev- eral hours to studying each night and abandon social life on weekends, you will get good grades this term. However, you may also hold the efficacy expec- tation that you are incapable of such devoted work and sacrifice. Bandura argues that efficacy expectations are better predictors of behav- ior than outcome expectations. Students are unlikely to work hard for good362 CHAPTER 13 The Behavioral/Social Learning Approach

grades if they don ’t think it possible. Therapy clients are unlikely to stop smoking, lose weight, or overcome a fear of flying if they don’ t believe they are capable of doing so. Where do efficacy expectations come from, and how can therapists change these expectations in their clients? Bandura identifies four sources. The most important of these is enactive mastery experiences. These are suc- cessful attempts to achieve the outcome in the past. Sky divers suddenlystruck with fear before a jump may tell themselves that they ’ve done this many times before without incident and therefore can do it again. On the other hand, a history of failures can lead to low efficacy expectations. People with a fear of heights who have never been able to climb a ladder withoutcoming back down in a fit of anxiety will probably conclude they can ’t per- form this behavior. Although not as powerful as actual performances, vicarious experiences also alter efficacy expectations. Seeing other people perform a behavior with- out adverse effects can lead us to believe that we can do it too. People who are afraid to speak in front of an audience may change their efficacy expecta-tion from “I can’t do that ”to“maybe I can ”after seeing other members of a public speaking class give their speeches without disastrous results. When you tell yourself something like “If she can do it, so can I, ”you are changing your efficacy expectation through vicarious experience. A less effective way to alter efficacy expectations is through verbal per- suasion. Telling someone who is reluctant to stand up to the boss “you can do it”might convince the person to assert his or her rights. However, this ex- pectation will be easily crushed if the actual performance isn ’t met with the expected result. Physiological and affective states can also be a source of efficacy expecta- tions. A woman who has difficulty approaching men may find her heart beats rapidly and her palms perspire as she picks up the phone to call a man to askfor a date. If she interprets these physiological responses as signs of anxiety, she may decide she is too nervous to go through with it. However, if she no- tices how calm she is just before dialing, she may decide she is more coura-geous than she realized. The key to successful treatment programs is changing a client ’s efficacy expectation through one or more of these means. In one study, therapistshelped snake-phobic people overcome their fear of the reptiles by taking them through the process of touching and picking up snakes (enactive mas- tery experience) and/or watching someone else go through this procedure (vi-carious experience). In nearly every case, whether the people believed they could approach and touch the snakes was the best predictor of whether they would actually do it (Bandura, Adams, & Beyer, 1977). But if successful experiences are the most effective method for altering a client’s efficacy expectations, this creates a bit of a problem. How can a thera- pist provide the client with a mastery experience of overcoming a fear ofheights if the client is afraid to leave the first floor of a building? One answer is a procedure known as guided mastery (Bandura, 1997). Using this proce- dure, the therapist arranges the situation so that the client is almost guaranteedApplication: Behavior Modifi cation and Self-Efficacy Therapy 363

a successful experience. The treatment is broken down into small steps that can be accomplished with only a slight increase in the client ’s effort. A client with a fear of driving might begin by driving a short distance on a secluded street (Bandura, 1997). This step is followed with gradually longer drives on busier streets. With each successful experience, the client strengthens the belief that he or she is capable of driving an automobile. You may have noticed that this procedure sounds similar to systematic desensitization. Indeed, in many cases the distinction between the two therapy procedures may lie only in how theyare interpreted. Behavior modification therapists explain successful systematic desensitization in terms of replacing old stimulus-response bonds with new ones. Social-cognitive therapists maintain that mastery experiences change effi-cacy expectations, leading to the change in behavior. The other side of this process is that failure to instill a sense of efficacy in a client might very well doom therapeutic efforts. People battling alcohol anddrug abuse typically do not succeed in treatment programs when they doubt their ability to overcome the problem (Ilgen, McKellar, & Tiet, 2005). Similarly, investigators find ex-smokers who are not confident they can stopsmoking are the most likely to fall back into their habit, sometimes within a few weeks after quitting (Gwaltney, Shiffman, Balabanis, & Paty, 2005; Shiffman et al., 2000). But the power of self-efficacy goes far beyond eliminating fears and bad habits. Self-efficacy beliefs have been found to play a role in overcoming a widevariety of psychological problems, including childhood depression (Bandura, Pastorelli, Babaranelli, & Caprara, 1999), posttraumatic stress disorder (Solomon, Weisenberg, Schwarzwald, & Mikulincer, 1988), test anxiety (Smith, 1989), phobias (Williams, 1995), and bereavement (Bauer & Bonanno, 2001). Efficacy expectations also affect job performance (Stajkovic & Luthans, 1998), academic achievement (Bandura, Barbaranelli, Caprara, & Pastorelli, 1996), weight loss (Linde, Rothman, Baldwin, & Jeffery, 2006), and romanticrelationships (Lent & Lopez, 2002). Heart attack patients who believe they can effectively participate in their rehabilitation have better cardiovascular health and lower mortality rates than those with low self-efficacy expectations abouttheir health care (Burns & Evon, 2007; Sarkar, Ali, & Whooley, 2009). In short, believing that we are capable of making changes and moving forward is an important component for dealing with many of the challenges and problemslife tosses our way. ASSESSMENT: BEHAVIOR OBSERVATION METHODS Let’s begin this section by thinking about one of your bad habits. Unless you are quite different from the rest of us, you probably chew your nails, eat junk food, lose your temper, use harsh language, smoke, talk too much, or engage in some other behavior you probably would like to change. Now imagine that you seek out a behavior therapist for help with this problem. The thera- pist asks you a simple question: How often do you perform the behavior? If you have been keeping track, you may be able to say exactly how many cigar- ettes you smoke per day or how often you chew your nails each week. But364 CHAPTER 13 The Behavioral/Social Learning Approach

most likely your answer will be far from precise. Behavior therapists can’ t tell if a treatment program is effective unless they know how often the behavior occurs before treatment. Yet too often clients say they perform the unwanted behavior “every once in a while, ”“not too often,” or“all the time. ” Unlike those who practice other approaches to psychotherapy, behavior therapists typically do not spend much time trying to discover the initial cause of a client ’s problem. Instead, they focus on treating observable behaviors. Other therapists may see the behavior as a sign of some underlying conflict, but for behavior therapists, the behavior isthe problem. Therefore, objective and reliable assessment of behavior is critical. Behavior therapists use assess-ment procedures for a variety of purposes. Obviously, they want to determinehow often a problem behavior occurs. But they may also want to know about the events surrounding the behavior. Does the client smoke alone or with other people? Do the tantrums occur at a certain time of day or after a cer-tain kind of experience, such as a scolding? These data can be very helpful in designing treatment programs. Therapists probably also want to monitor the therapy ’s progress and make some judgment about its success. So how do behavior therapists obtain accurate information about the fre- quency of target behaviors? When one team of researchers surveyed members of a behavior therapist organization, they found the therapists relied on a va- riety of procedures (Elliott, Miltenberger, Kaster-Bundgaard, & Lumley, 1996). Some of the most common methods of assessment mentioned by the therapists are shown in Table 13.2. Let ’s look at a few of these in depth. Direct Observation The most obvious way to find out how often a behavior occurs is to observe the person directly. Although a therapist usually can ’t watch a client all day long, it is often possible to observe a representative sample of the client ’s behavior. If you want to know how much time a girl spends interacting with TABLE 13.2 Behavior Therapists ’Assessment Methods Method Percentage of Cases Using Interview with client 94.1 Direct observation 52.3 Client self-monitoring 44.1Behavioral rating scales 43.7Interview with client ’s significant others 42.0 Information from other professionals 38.0Role playing 19.4 Source: Cognitive and behavioral practice by Elliot, A. J., Miltenberger, R. G., Kaster-Bundgaard, J. Copyright 1996 by Elsevier Science & Technology Journals . Reproduced with permission of Elsevier Science & Technology Journals in the format Textbook via Copyright Clearance Center.Assessment: Behavior Observation Methods 365

children her own age, you can watch the child on the playground during several recesses. However, the therapist can ’t be everywhere. The therapist probably won ’t be around when a socially phobic person goes on a job inter- view or a married couple has an argument. In these cases, the psychologist might rely on analogue behavioral observation (Haynes, 2001). That is, the therapist creates a situation that resembles the real-world setting in which the problem behavior is likely to occur. For example, a therapist might stage a dance for clients suffering from acute shyness or ask a couple to enter into adiscussion that recently sparked a disagreement. Occasionally behavior thera- pists ask clients to role-play. A therapist helping a man to be more assertive might ask the client to imagine that someone just cut in front of him in line. The client then acts out what he would do in that situation. In this case, the way the client acts in the role-playing exercise is probably similar to the way he acts when confronting real-life situations. However, good behavioral assessment requires more than simply observ- ing a person. The behaviors to be observed must first be defined as precisely as possible. This is fairly simple when talking about the number of cigarettessmoked, but what if the target behavior is “appropriate classroom responses ”? In this case the therapist might define appropriate responses as those relevant to the topic being discussed or those in which the child waits for recognition before speaking. But even these definitions leave considerable room for ob- server interpretation. A good definition includes examples of behaviors to be counted and rules for dealing with borderline cases. One way to improve the accuracy of behavior observation is to have two or more observers independently code the same behaviors. For example, twojudges can watch the same child during the same set of recesses. If the two largely agree on how often they count the target behavior, we can be confident that the number is fairly accurate. However, if one coder sees few behaviors and a second coder sees many, we have no idea how often the target behavior actu-ally occurs. One solution is to videotape the behavior so that many different judges can be used and disagreements can be resolved by replaying the tape. Behavior therapists must also be concerned about bias. Sometimes observersunintentionally see what they want or what they expect to see. To guard against this problem, therapists should define behaviors in a manner that minimizes subjective judgment. If possible, they can use observers who don’ t know what the therapist expects to find. Self-Monitoring Although direct observation provides a relatively accurate assessment of behavior frequency, it is often too costly and time consuming to be useful. An alternative isself-monitoring —clients observing themselves. However, simply asking clients how often they engage in a certain act may be of little help. Clients frequently have a distorted idea about how often a behavior occurs. In addition, it is usually important to understand the circumstan ces surrounding the behavior. Are there places the client is particularly likely t o smoke, such as with certain friends or at a party? Is the smoking associated with a certain time of day, a certain type ofactivity, or a certain mood?366 CHAPTER 13 The Behavioral/Social Learning Approach

Unfortunately, few clients can provide accurate information about these variables from memory. Therefore, therapists often ask clients to keep records of when and where they engage in certain behaviors. Clients are sometimes sur- prised by what they find. For example, people trying to watch their weight may discover that they eat more when they ’re alone, when watching television, or after they ’ve had a drink. An interesting benefit of the self-monitoring method is that watching your own behavior can be therapeutic in itself. Clients forced to pay attention to their eating or smoking sometimes show improvement even before the treatment is started (Mahoney & Arnkoff, 1979). Self-monitoring is also used to assess progress throughout the treatment period. One problem thatsometimes surfaces when relying on self-monitoring is the client ’s honesty. Clients may not want to admit that they increased their smoking or lost theirtemper several times in one week. Therapists who suspect a problem may wantto use other assessment methods, such as the one discussed next. Observation by Others Some clients are unwilling or simply unable to provide accurate information about themselves. For example, self-monitoring is probably inappropriate with children or those with severe psychological disorders. In these cases, itmay be possible to rely on other people to make the observations. Parents and teachers can often record the frequency of a child ’s problem behaviors. A teacher might be asked to record each time she punishes a child for acting Psychologists working with children often use direct observation. This procedure allows them to assess how a child plays alone, how parents interact with their child, or howwell a child interacts with other children. Many psychologists have also discovered thevalue of videotaping behavior samples for more extensive observation and coding later.Photo by Marlene SomsakAssessment: Behavior Observation Methods 367

aggressively. Nurses and aides can record the occurrence of patients ’behaviors. Although this process can introduce bias, it often provides the most accurate assessment of a client ’s behavior. Many psychologists use these reports to complement data obtained through other methods. For example, children sometimes act differently inthe presence of a therapist than they do at home. A client may be able to role-play the appropriate behaviors when confronting a make-believe man- ager but may become timid when facing the real boss at work. Getting familymembers involved in the process can have other advantages, such as making them aware of the client ’s problem and of how their reactions might affect his or her behavior. STRENGTHS AND CRITICISMS OF THE BEHAVIORAL/ SOCIAL LEARNING APPROACH Behaviorism roared onto the psychology scene in the 1920s and put a grip on the discipline that didn ’t loosen for several decades. Although not as influential as it once was, behaviorism in various forms remains alive and well today. Explanations of behavior that evolved from behaviorism, such as social-cognitive theory, also remain popular. Obviously, the behavioral/social learning approach to personality could not have withstood this test of time without having some unique strengths. Of course, no theory as influential as this one can hope to es- cape criticism either. Strengths One reason for the endurance of the behavioral/social learning approach is itssolid foundation in empirical research. This contrasts with other approaches to personality, which are sometimes based on intuition or on data gathered from biased samples. Most of the theorists covered in this chapter relied on empirical data when developing and refining their theories. Critics often chal- lenge the existence of Freud ’s Oedipus complex, but it would be difficult to deny the mountain of evidence demonstrating that behaviors sometimes change through operant and classical conditioning. Another strength of the behavioral/social learning approach lies in the development of some useful therapeutic procedures. Studies find these proce-dures to be effective in treating a number of psychological problems, especially when combined with elements from cognitive therapies (Christensen, Atkins, Yi, Baucom, & George, 2006; Mitte, 2005; Shadish & Baldwin, 2005).Moreover, behavioral treatments are popular. One survey asked marriage and family therapists about their primary treatment approach when working with clients (Northey, 2002). By a large margin, “cognitive-behavioral ”was the most commonly cited approach. Behavior modification procedures have several advantages over other therapy approaches. One advantage is their use of baseline data and objective criteria for determining success or failure. Other approaches often begin treat- ment without first determining the level of the problem; therapy is declared a success when the therapist or the client decides there has been some368 CHAPTER 13 The Behavioral/Social Learning Approach

ASSESSING YOUR OWN PERSONALITY Assertiveness Indicate how characteristic or descriptive each of the following statements is of you by using this code: þ3¼very characteristic; þ2¼rather charac- teristic; þ1¼somewhat characteristic; –1¼somewhat uncharacteristic; –2¼rather uncharacteristic; –3¼very uncharacteristic. 1. Most people seem to be more aggressive and assertive than I am.* 2. I have hesitated to make or accept dates because of “shyness. ”* 3. When the food served at a restaurant is not done to my satisfaction, I complain about it to the waiter or waitress. 4. I am careful to avoid hurting other people ’s feelings, even when I feel that I have been injured.* 5. If a salesman has gone to considerable trouble to show me merchandise that is not quite suitable, I have a difficult time saying “No.” * 6. When I am asked to do something, I insist upon knowing why. 7. There are times when I look for a good, vigorous argument. 8. I strive to get ahead as well as most people in my position. 9. To be honest, people often take advantage of me.* 10. I enjoy starting conversations with new acquaintances and strangers. 11. I often don’ t know what to say to attractive persons of the opposite sex.* 12. I will hesitate to make phone calls to business establishments and institutions.* 13. I would rather apply for a job or for admission to a college by writing letters than by going through with personal interviews.* 14. I find it embarrassing to return merchandise.* 15. If a close and respected relative were annoying me, I would smother my feelings rather than express my annoyance.* 16. I have avoided asking questions for fear of sounding stupid.* 17. During an argument I am sometimes afraid that I will get so upset that I will shake all over.* 18. If a famed and respected lecturer makes a statement that I think is incorrect, I will have the audience hear my point of view as well. 19. I avoid arguing over prices with clerks and salesmen.* (continues)Strengths and Criticisms of the Behavioral/Social Learning Approach 369

improvement. In addition, behavior modification may be the most useful ap- proach when working with certain populations, such as children or severely emotionally disturbed patients. These individuals might have a difficult time discussing abstract psychoanalytic concepts or dealing with some of the exis- tential questions posed by humanistic therapists. Behavior modification is also relatively quick and easy to administer. Treatment often lasts a matter of weeks, compared with months or years with other approaches. The basic20. When I have done something important or worthwhile, I manage to let others know about it. 21. I am open and frank about my feelings. 22. If someone has been spreading false and bad stories about me, I see him or her as soon as possible to “have a talk ”about it. 23. I often have a hard time saying “No.” * 24. I tend to bottle up my emotions rather than make a scene.* 25. I complain about poor service in a restaurant and elsewhere. 26. When I am given a compliment, I sometimes just don ’t know what to say.* 27. If a couple near me in a theater or at a lecture were convers- ing rather loudly, I would ask them to be quiet or take their conversation elsewhere. 28. Anyone attempting to push ahead of me in a line is in for a good battle. 29. I am quick to express an opinion. 30. There are times when I just can ’t say anything.* Many people have difficulty asserting their rights. In behavioral terms, these people need to increase the frequency of their assertive behaviors in appropriate situations. A behavior modification treatment called assertive- ness training allows participants to watch models asserting themselves ap- propriately, role-play their own assertive responses, and receive immediate reinforcement for appropriate assertive actions. To obtain your assertive- ness score, first reverse the sign for your answer on each of the items with an asterisk (that is, a plus becomes a minus, and vice versa). Then add all30 answer values together. A positive score indicates high assertiveness; a negative score reflects low assertiveness. The average score is around 8 for college women and 10 for college men (Nevid & Spencer, 1978). Two-thirds of all college females score between 31 and –17, and two-thirds of college males have scores that fall between 33 and –11. Scale: Rathus Assertiveness Inventory Source: Behavior Therapy by Rathus, S. A. Copyright 1973 by Elsevier Science & Technology Journals . Reproduced with permission of Elsevier Science & Technology Journals in the format Textbook via Copyright Clearance Center.370 CHAPTER 13 The Behavioral/Social Learning Approach

methods can be taught to parents, teachers, and hospital personnel, who can carry out the therapy without the therapist present. This means that morepeople can benefit from therapy procedures at a lower cost than is possible with most other types of psychotherapy. The social learning theories and Bandura ’s social-cognitive theory added cognitive variables to the behavioral approach and thereby expanded the range of phenomena explained by this perspective. These theories have helped to fill in the gaps many psychologists see in traditional behaviorism. Sociallearning models of personality allow us to understand thoughts, expectancies, and values along with basic behavior conditioning principles within one theo- retical framework. These models have also helped to bridge traditional behav-iorism with cognitive approaches to personality. Criticisms A persistent criticism of the behavioral/social learning approach is that it is too narrow in its description of human personality. Although the approach touches on several crucial aspects of human experience such as thinking, emo-tion, and levels of consciousness, many psychologists feel that it does so in a limited way. Critics are particularly concerned with the Skinnerian brand of behaviorism, which rejects the usefulness of examining inner feelings and intu-ition. Others criticize the behavioral/social learning approach for giving inade- quate attention to the role of heredity. In addition, research points to some limits on conditioning procedures. For example, it is more difficult to create afear of food in animals by pairing the food with electric shocks than by pairing it with nausea (Garcia & Koelling, 1966; Seligman & Hager, 1972). Another criticism, directed primarily at traditional behaviorism, is that human beings are more complex than the laboratory animals used in behav-ioral research. As Bandura and some of the social learning theorists recognize, people are capable of considering alternative courses of action, of weighing the probabilities and values of different reinforcers, of looking at long-term goals, and so forth. These critics do not deny that we often respond to stimuli in an automatic fashion or that some of our behaviors are conditioned. But they maintain that these are the least important and least interesting human behaviors. An example of the difficulty in generalizing from animal data to human behavior is seen in research on the effects of extrinsic reinforcers on intrinsically motivated behavior. Although this remains an area of contro- versy (Eisenberger & Cameron, 1996; Sansone & Harackiewicz, 1998), re- searchers often find that paying people to engage in a behavior they already enjoy results in a reduction in the frequency of the behavior. People seem to redefine the behavior as work instead of play ( “I play the piano because I am paid ”) and therefore lose interest unless rewarded. Despite the success of behavior therapists in dealing with many problem behaviors, some critics argue that these therapists sometimes distort the real therapy issues when they reduce everything to observable behaviors. For ex-ample, a man who complains that he has no meaning in his life might be asked to define this abstract issue in terms of measurable behaviors. A behav- ior therapist might count the number of times the man engages in pleasantStrengths and Criticisms of the Behavioral/Social Learning Approach 371

activities and set up a treatment program that rewards him for going to par- ties, talking with friends, reading good books, and so on. These activitiesmight make the man feel better. However, critics might argue that the therapy has not addressed the real problem but, instead, has temporarily diverted the client’s attention from the concerns that caused him to seek out therapy. SUMMARY 1. Behaviorism was introduced by John B. Watson in the 1920s. In its most extreme form, behaviorism limits psychology to the study of observable behaviors. Classical conditioning and operant conditioning are used bybehaviorists to explain the development and maintenance of behaviors. Personality is described as the end result of one’ s history of conditioning. B. F. Skinner later became the spokesperson for what he called radical behaviorism. He rejected the use of inner states, such as anxiety, as explanations of behavior in favor of observable external events. 2. Traditional behaviorism identifies two basic kinds of conditioning. Classical conditioning occurs when a new stimulus is paired with an existing stimulus-response bond. Operant conditioning results when a behavior is followed by either reinforcement or punishment. 3. Later social learning theorists expanded on the basic behaviorist position. Rotter argues that the probability of engaging in a behavior changes after rewards and punishments because our expectancies change. He uses these expectancies and the values given to particular reinforcers to predict which of many behavior options will be enacted. 4. Bandura proposes that internal states, the environment, and behavior all affect one another. He maintains that people often regulate their ownbehavior and that we engage in purposeful, future-oriented thinking. Bandura has added to classical and operant conditioning the notion that we learn through observing others, although whether we perform the behaviors we learn depends on our expectancies for rewards or punishments. 5. Behavior modification therapists apply basic conditioning principles when working with their clients. Some of these therapeutic procedures, such assystematic desensitization, are based on classical conditioning. Others, such as token economies, are based on operant conditioning. Bandura has identified clients ’self-efficacy beliefs as crucial for psychotherapy progress. Whether clients expect to succeed is an important determinant of therapy success. These expectancies come from a variety of sources, including past performance accomplishments and vicarious learning. 6. Behavioral assessment includes a variety of techniques, including direct observation, self-monitoring, and observation by others. Each of thesetechniques can provide useful data for determining baseline frequencies, the conditions under which the target behavior occurs, and the success ofthe treatment procedure. 7. The behavioral/social learning approach has its strengths and its criticisms. Among the strengths are its empirical base and the useful therapeutic372 CHAPTER 13 The Behavioral/Social Learning Approach

procedures it has generated. The criticisms include an inappropriate atten- tion to important causes of behavior such as heredity. People have alsocriticized the way behavior therapists interpret problems into observable behaviors. KEY TERMS behavior modification (p. 360) classical conditioning (p. 345)discriminate (p. 350) generalization (p. 350) observational learning (p. 356)operant conditioning (p. 348) reciprocal determinism (p. 354)self-efficacy (p. 362) self-regulation (p. 355) MEDIA RESOURCES Visit our website. Go to www.cengage.com/ psychology/burger to find online resources directly linked to your book, including quizzing, glossary, flashcards, and more.Summary 373

CHAPTER14 The Behavioral/Social Learning Approach Relevant Research Individual Differences in Gender-Role Behavior Observational Learning of Aggression Learned Helplessness Locus of ControlSummary 374

Behaviorists are sometimes portrayed as aloof, data-oriented scientists more concerned with how many times a rat presses a bar than with the people intheir lives. Although it ’s true that these researchers often attend to minute ex- perimental details and theoretical issues that seem overly esoteric to an outside observer, it is unfair to say they have lost sight of the human element or their goal of improving the human condition. Even B. F. Skinner, who conducted most of his research on rats, wrote extensively on how we can use the informa- tion coming out of animal laboratories to overcome many of the problems fac-ing society today. This concern can be seen in each of the four research topics reviewed in this chapter. Each has something to say about pressing social pro- blems or personal lifestyle issues. First, society ’s expectations for men and women have undergone a great deal of reevaluation and change in the past few decades. In increasing numbers,women are abandoning traditional gender roles to take important positions inbusiness and government. Some men are experimenting with nontraditional male roles, such as assuming child-rearing responsibilities. Understanding why we make some of the gender-related choices we do requires an examina-tion of how operant conditioning and observational learning shape those choices. We ’ll look at these processes and individual differences in masculinity and femininity. Second, in response to the ever-present issue of violence in our society, many psychologists have focused their attention on the impact aggressive mod- els have on aggressive behavior. Bandura ’s observational learning model helps explain some of this process. We ’ll look at relevant research and the question of how mass media violence affects the behavior of those who consume it. Third, applying animal research findings to human beings is a standard feature of the behavioral approach to personality. A particularly fruitful ex- ample of this application is the work on learned helplessness. From some surprising observations of dogs in a classical conditioning experiment, re-searchers have developed a theory with implications for depression and adjust- ment among the elderly. Fourth, we ’ll look at one aspect of Rotter ’s social learning theory. Individual differences in locus of control have been the focus of an enormous amount ofpersonality research. Some of these findings provide important information about how our expectancies are related to our well-being and our physicalhealth. INDIVIDUAL DIFFERENCES IN GENDER-ROLE BEHAVIOR I would like to describe two friends of mine. The first is a very caring and loving person. This friend never forgets my birthday, is sensitive to my needs and moods, and is the person I seek first when I need someone to talkto. This friend also confides in me and is not afraid to share intimate feelings. My other friend is on the way to becoming a leader in the business world. This friend knows how to be assertive when necessary, how to express opinionsdirectly, and how to get others to do what is needed for the company.Individual Differences in Gender-Role Behavior 375

Unlike the first person I described, this one sometimes has difficulty being in- timate with others or sharing feelings. I ’ve never seen this friend cry. Unless you’ ve already caught on to my point here, you probably imag- ined that the first person is a woman and the second is a man, even though I never identified the gender of either. If so, it doesn ’t mean you’ re gullible or sexist. Rather, it suggests you are aware of the gender-role stereotypes that af- fect the way men and women behave in this culture. Traditional stereotypes portray men as aggressive, independent, and unemotional, whereas womenare depicted as passive, dependent, and affectionate. Much has been written recently about changes in these gender roles, with men being told it is all right to show emotion and women being encouraged to be assertive and business-like. However, although some gender restrictions may have loosened in recent years, gender roles remain a part of our culture. Why do women tend to behave in certain ways and men in others? Although biological differences between the sexes certainly play a role, beha- viorists and social learning theorists point to a lifelong process of gender-role socialization. Children and adults acquire and maintain gender-appropriatebehaviors largely through operant conditioning and observational learning. You can see the effects of operant conditioning whenever young children act in gender-inappropriate ways. Boys often tease one another for crying, play- ing with dolls, or showing an interest in cooking or sewing. Similarly, play- mates make fun of girls when they act like tomboys. Boys are rewarded with camaraderie and parental nods for playing football and standing up to those Most little girls occasionally play “dress up.” Girls put on their mother ’s clothes, jewelry, and makeup after identifying that this is something females, but not males, do. We would not expect to find little boys imitating this behavior.376 CHAPTER 14 The Behavioral/Social Learning Approach

who try to push them around. Girls win approval for showing an interest in caring for babies and dressing up in pretty outfits. You can appreciate the difficulty in changing these behavior patterns when you realize how early this operant conditioning starts. Even before the child can talk, parents speak to and play with their daughters differently than they do with their sons (Clearfield & Nelson, 2006). One team of re- searchers interviewed parents of sons and daughters within 24 hours after the birth of their first child (Rubin, Provenzano, & Luria, 1974). Parents rateddaughters as softer, finer featured, smaller, and less attentive than sons. The daughters were also described as beautiful, pretty, and cute. In reality, the new- borns did not differ in terms of weight, length, or measures of general health. One team of investigators examined the choices parents made for boys and girls when the child was fewer than 25 months old (Pomerleau, Bolduc,Malcuit, & Cossette, 1990). Girls were more likely to receive dolls and toy fur-niture, whereas boys were more often given sports equipment, toy tools, and toy cars and trucks. Perhaps predictably, the girls were more likely to wear pink clothing, and the boys blue. In another study, prior to Christmas research-ers asked elementary school children what they wanted and after Christmas asked what they had received (Etaugh & Liss, 1992). Most children asked for toys traditionally associated with their gender, and most of the time that ’sj u s t what they received. However, a few children asked their parents for toys atypi- cal for their gender. These boys and girls were much less likely to get what they wanted. Many parents simply won ’t buy their daughter a football, no matter how much she wants one. Messages about the different ways boys and girls are supposed to act are communicated to the child within the first few years of life. By the time children enter kindergarten, they are well aware of gender-role expectations (O ’Brien et al., 2000; Vogel, Lake, Evans, & Karraker, 1991). Preschool boys and girlsin one study were given a choice between traditional “boy”toys (tools) and “girl”toys (dishes) during a free-play period (Raag & Rackliff, 1998). Not only did most children select the toys traditionally associated with their gender, butmost of the boys explained that their fathers would not approve of them playingwith girls ’toys. Preschoolers in another study were told that the other gender “really likes ”a certain toy (Martin, Eisenbud, & Rose, 1995). After receiving this news, both boys and girls said they liked the toy less. In short, children aresurrounded by adults and peers ready to reward gender-appropriate behaviors and punish inappropriate ones. Gender-role behaviors are also acquired through observational learning. Children learn which behaviors are expected of males and which are expectedof females by watching parents, neighbors, siblings, playmates, and television characters. When children are very young, parents are probably the most in-fluential models, which may explain why people’ s gender-role behavior tends to resemble that of their mother or father (Jackson, Ialongo, & Stollak,1986). Later, children are more likely to take their cues about appropriateand inappropriate behavior from their friends. However, it is not the case that boys imitate only male models and girls only female models. Instead, the child must first notice that a certain behaviorIndividual Differences in Gender-Role Behavior 377

is performed more often by one gender than the other (Bandura & Bussey, 2004; Bussey & Bandura, 1999). Children may recognize that men, but rarelywomen, work on mechanical things. When an appliance needs fixing, father takes care of it. All the garage mechanics seem to be men, and if someone on television uses a screwdriver or a wrench, it is almost always a male. The chil- dren are likely to conclude that men are rewarded for mechanical behavior but women are not. Thus boys are likely to get involved with mechanical things, anticipating rewards, whereas girls seek out other activities. At this point, oper-ant conditioning may also come into play, such as when a father rewards his son for showing an interest in cars and laughing when his daughter asks to help with an oil change. Masculinity –Femininity After a lifetime of socialization through operant conditioning and observa-tional learning, we should not be surprised that most adult men and womenact in gender-appropriate ways. But even casual observation of the people you meet in the next few hours will confirm that there are large individual differences in the extent to which people act in a masculine or feminine man- ner. Although men are generally more aggressive and independent than women, there are many exceptions. Similarly, finding women who do not fit the stereotypic affectionate, emotional, and sensitive pattern is not difficult. As with other individual differences, personality psychologists are inter- ested in identifying, measuring, and describing the way people typically act interms of their gender-role behavior. Unfortunately, researchers do not alwaysagree on what to call these traits. Originally, psychologists used the terms masculinity and femininity . People who act in ways that fit the traditional role expectations for men were classified as masculine, and those who acted like the traditional role expectations for women were feminine. Other psychologists ar- gue that we should replace these terms with more specific and less emotionally loaded labels. In particular, many researchers prefer the terms agency andcom- munion (Helgeson, 1994; Spence, 1993). Agency refers to independence, asser- tiveness, and control and is roughly similar to masculinity. Communion refers to attachment, cooperation, and interpersonal connection and is similar to fem-ininity. Nonetheless, because many researchers continue to rely on the mascu- linity and femininity labels, we will also use these terms here. Early scales developed to measure individual differences in gender-role behavior were based on two assumptions. First, masculinity and femininitywere assumed to represent two extreme positions on a continuum. As shown in Figure 14.1, masculinity and femininity were considered opposites. The Masculine Feminine FIGURE 14.1 Traditional Masculinity –Femininity Model378 CHAPTER 14 The Behavioral/Social Learning Approach

more a person was of one, the less he or she was of the other. Each of us can be placed on this continuum, with very masculine and very feminine peopleon the extremes and those who are both, but not much of either, toward the middle. The second assumption was that the more people ’s gender-role behavior matched the stereotype for their gender, the more psychologically healthy they were. Masculine men and feminine women were considered well adjusted. But a man who acted too much the way society said a woman was supposedto act or a woman who behaved too much like a stereotypic man was said to have adjustment problems. One of the original scales on the Minnesota Multiphasic Personality Inventory (MMPI) is the Mf (Masculinity-Femininity)Scale. Initially researchers maintained that scoring too far on the wrong side of this scale for one ’s gender was indicative of psychological disturbances. Androgyny Investigators soon uncovered several problems with the masculinity –femininity model (Constantinople, 1973). In response, psychologists developed a new approach for measuring and identifying gender-role behaviors. The mostinfluential of these new approaches is called the androgyny model (Bem, 1974, 1977). Coupled with society ’s rising concern for women ’s issues during these years, the androgyny model triggered an explosion in research on genderroles. The model begins by rejecting the notion that masculinity and femininity are opposites on a single continuum. Instead, masculinity and femininity are seen as independent traits. People can be high on both traits, on only onetrait, or on neither. Further, because these traits are independent, knowing that someone is high in masculinity tells us nothing about how feminine that person is. A great deal of research supports the notion that masculinity and femi- ninity are best thought of as independent concepts. For example, researchers find that women tend to increase in both masculinity and femininity as they move through their middle adult years (Kasen, Chen, Sneed, Crawford, & Cohen, 2006; Strough, Leszczynski, Neely, Flinn, & Margrett, 2007). This pat- tern would not be possible if masculinity and femininity were opposite poles on a single dimension. The androgyny model also challenges the assumption that a person ’s gen- der should match his or her gender type. Advocates of the model maintainthat the most well-adjusted person is both masculine and feminine; that is, an- drogynous. According to this perspective, people who are only masculine or only feminine often find they lack the ability to engage in adaptive behavior.Masculine people do well as long as the situation calls for a masculine re- sponse, such as asserting one ’s rights or taking over the leadership of a group. But when masculine individuals are called on to act in a traditionally feminine manner, such as showing compassion or sensitivity, they falter. A well- adjusted person must have the flexibility to engage in masculine behaviors when the situation demands and in feminine behaviors when those are mostappropriate. Psychologists soon developed scales to measure the traits of masculinity and femininity separately (Bem, 1974; Lenney, 1991; Spence, Helmreich,Individual Differences in Gender-Role Behavior 379

& Stapp, 1974). By using the median score on each scale as a cutoff point, re- searchers can place people into one of four sex-type categories, as shown inFigure 14.2. Those who score high in both masculinity and femininity are classified as androgynous. People scoring high on one scale but not the other fall into either the masculine orfeminine category. Those who score low on both scales are classified as undifferentiated. Gender Type and Psychological Well-Being The androgyny model was born out of a belief that possessing both masculine and feminine characteristics was important for psychological well-being. But how has that assumption held up under the scrutiny of research? In addition to the androgyny model, two other explanations for the relation between gen- der type and well-being have been proposed and tested. The first explanation is the traditional congruence model. According to this account, masculine men and feminine women are the most well adjusted. Although this approach reflects old-fashioned attitudes and may even border on sexism, a case can be made. Think about all the pressure society puts on men and women to act in gender-appropriate ways. What can we concludeabout people who emerge from this socialization without developing the gen- der type dictated by society? Perhaps they are merely liberated from the re- straints society attempts to place on them. But remember rewards andpunishments for gender-appropriate behavior begin in childhood and con- tinue into adulthood. Society is geared to give masculine men and feminine women most of the rewards in life, whereas masculine women and feminine men face social rejection and possibly ridicule. We might therefore expect gender-congruent individuals to be the happiest and most content. Although this reasoning makes some sense, researchers rarely find support for the con- gruence model (Taylor & Hall, 1982; Whitley, 1983). The second explanation, the masculinity model, maintains that being masculine is the key to mental health. Before rejecting this view as masculinepropaganda, consider that in many ways our society is still geared toward admiring and rewarding the traits traditionally associated with men andAndrogynous Feminine Masculine UndifferentiatedHigh High LowLowMasculinity Femininity FIGURE 14.2 Androgyny Model380 CHAPTER 14 The Behavioral/Social Learning Approach

masculinity. Stereotypically, men are independent, and women are dependent. Men are achieving and powerful; women are unassertive and conforming.Men are leaders, whereas women are followers. Given these descriptions, it makes sense that those who fit the masculine role might accomplish more and feel better about themselves than those who do not. Women do not have to abandon their femininity to get ahead in the traditionally male business world, but they may need some traditionally masculine attributes to be successful. Several investigations find support for the masculinity model (Cheng, 1999; Marsh, Antill, & Cunningham, 1987; O’ Heron & Orlofsky, 1990; Orlofsky & O’ Heron, 1987; Roos & Cohen, 1987). Because masculine peo- ple are more likely to use direct, problem-focused strategies, they are better able to deal with stressors than those low in masculinity (Helgeson & Lepore, 1997). Masculine men in one study coped better with the loss of theirspouse and the subsequent changes in their life than did other widowers (Bowers, 1999). Women who possess masculine traits are good at influencing others and getting what they want and are, therefore, less likely than mostwomen to suffer from helplessness and depression (Sayers, Baucom, & Tierney, 1993). Support for the masculinity model is particularly strong when looking at the relationship between gender type and self-esteem (Whitley, 1983). People who possess traditionally masculine attributes (such as achieving, athletic, pow- erful) also feel good about themselves. The third explanation is the androgyny model. According to this view, people whose behavioral repertoires lack either masculine or feminine beha-viors are ill-prepared to respond to many situations they encounter. Without masculine characteristics such as decisiveness and assertiveness, both men and women are likely to falter in achievement situations. At the same time, people unable to express emotions have difficulty establishing good interper- sonal relationships. Only androgynous people are capable of getting ahead onthe job while relating well with friends and loved ones in their leisure time. Several investigations find support for the androgyny model (Bem & Lenny, 1976; Bem, Martyna, & Watson, 1976; Cheng, 2005; Lefkowitz &Zeldow, 2006; Shaw, 1982; Stake, 2000). When taking care of a baby, feminine and androgynous —but not masculine —individuals show appropriate nurturant behavior. Feminine people are easily swayed by the opinions of others, whereas masculine and androgynous people better resist conformity pressures. Individuals who possess only masculine or feminine characteristics often do poorly when faced with stressful situations where taking action and seeking comfort from others is beneficial. However, overall support for the androgyny model is mixed. Whereas many studies show the superior adaptability of androgynous individuals,others do not (Taylor & Hall, 1982; Woodhill & Samuels, 2004). Although androgynous people may be well-prepared to deal with many situations, this does not always translate into a sense of well-being or high self-esteem. So what are we to conclude? First, very little research supports the con- gruence model. Second, some of the inconsistent findings may reflect the waymasculinity and femininity are measured. For example, the most widely usedIndividual Differences in Gender-Role Behavior 381

measure in this research, the Bem Sex Role Inventory, asks people the extent to which 20 masculine and 20 feminine terms describe them. Unfortunately,the masculine terms in the scale tend to be more desirable than the feminine terms (Pedhazur & Tetenbaum, 1979). It makes sense that people who describe themselves with the more flattering and positive masculine terms (for example, self-reliant, ambitious) have higher self-esteem than those who describe them- selves with feminine terms (for example, gullible, shy). Third, it seems quite possible that some aspects of a healthy personality, such as dealing with stressand personal achievement, are related to masculinity, whereas other aspects, such as developing good interpersonal relationships, are not (Marsh & Byrne, 1991). Individual differences in gender-role behavior are clearly tied to well-being in some way. However, just how these two are related remains the fuel for continued research. Gender Type and Interpersonal Relationships Who would you turn to if you needed to talk to someone about a personal problem —a masculine, feminine, androgynous, or undifferentiated person? Who would you prefer for a friend? For a romantic partner? Advertisements and TV shows often portray masculine men and feminine women as the most desirable partners for romantic encounters. And Americans spend a con- siderable amount of money on makeup, body-building equipment, and the like to make themselves appear more feminine or masculine. But is this the road to a perfect relationship? Some research suggests it may not be. A simple way to examine how people react to different gender types is to ask participants about hypothetical character sketches of masculine, feminine,androgynous, and undifferentiated individuals. In general, researchers usingthis procedure find the androgynous character is liked more than the other three (Brooks-Gunn & Fisch, 1980; Gilbert, Deutsch, & Strahan, 1978; Jackson, 1983; Korabik, 1982; Kulick & Harackiewicz, 1979; Slavkin &Stright, 2000). Undergraduates in one study said the androgynous person was more popular, more interesting, better adjusted, more competent, more intelli- gent, and more successful than people described in masculine, feminine, orundifferentiated terms (Major, Carnevale, & Deaux, 1981). And when research- ers asked college students to estimate the desirability of hypothetical romantic partners, both men and women showed a preference for the androgynous person (Green & Kenrick, 1994). But do these impressions of hypothetical people translate into actual be- haviors? To answer this question, one team of researchers created four typesof male-female pairs: a masculine man and a feminine woman, an androgy- nous woman and a masculine man, a feminine woman and an androgynous man, and two androgynous people (Ickes & Barnes, 1978). The couples, who did not know each other before the study, were left alone in a room for 5 minutes. The participants were free to carry on a conversation or simply sit quietly and wait. Their behavior was recorded with a hidden video camerafor later evaluation. Participants also were asked to rate afterward how much they had enjoyed the interaction. As shown in Figure 14.3, members of the masculine man –feminine woman dyads enjoyed their interactions the382 CHAPTER 14 The Behavioral/Social Learning Approach

least. Analyses of the videotapes revealed that these couples talked to each other less, looked at each other less, used fewer expressive gestures, and smiled and laughed less than people in the other combinations. These results argue against the masculine man –feminine woman combina- tion as the ideal couple. When we examine the different ways masculine and feminine people approach an interpersonal encounter, some of the reasons for this finding become clear. The masculine style emphasizes control, self-monitoring, and self-restraint, whereas femininity is associated with inter- personal warmth and actively expressing one ’s feelings. Little wonder, then, that this combination didn ’t work out well in this or other experiments (Ickes, 1993; Ickes, Schermer, & Steeno, 1979; Lamke & Bell, 1982). But what about long-term relationships? After the initial awkwardness, it’s possible a masculine man and a feminine woman will get along well once they get to know one another. However, this notion is also not supported bythe evidence. Researchers find the highest level of relationship satisfaction in people married to someone with feminine characteristics (Antill, 1983). That is, people with either a feminine or an androgynous spouse tend to be satis- fied with their relationship. Being married to a partner who lacks feminine characteristics (that is, a masculine or undifferentiated partner) is indicative of an unhappy marriage. This same pattern has been found in cohabitating heterosexuals, gay couples, and lesbian couples (Kurdek & Schmitt, 1986).1020304050Liking Rating Masculine Male– Feminine FemaleMasculine Male– Androgynous FemaleAndrogynous Male– Feminine FemaleAndrogynous Male– Androgynous Female FIGURE 14.3 Mean Liking Rating Between Couples During 5-Minute Interaction Source: From “Boys and girls together —and alienated: On enacting stereotyped sex roles in mixed-sex dyads, ” by W. Ickes and R. D. Barnes, Journal of Personality and Social Psychology , 1978, 36, 669 –683.Individual Differences in Gender-Role Behavior 383

What is it that makes feminine and androgynous people preferable part- ners? Research suggests at least three reasons. First, look at the characteristics that make up the feminine trait. People scoring high on this scale are affec- tionate, compassionate, and sensitive to others ’needs. Feminine people are better able to express their feelings and understand the feelings of others. It only makes sense that we turn to them when we want to talk. Second, an- drogynous people are more aware of and better able to express romantic feel- ings (Coleman & Ganong, 1985). They have both the sensitivity and theunderstanding needed for intimacy as well as the assertiveness and willingness to take the risks required to make things happen. People married to spouses high in both expressiveness and sensitivity report the highest level of satis-faction with their relationships (Bradbury, Campbell, & Fincham, 1995; Zammichieli, Gilroy, & Sherman, 1988). Consequently, androgynous people may make the best romantic partners. Third, because they communicatewell, feminine and androgynous individuals are better able to resolve prob- lems and avoid unnecessary disputes (Voelz, 1985). They are more sensitive to their partners ’feelings and needs, are better able to express their own feel- ings, and are thus more likely to live harmoniously than people who lack these qualities (Aube, Norcliffe, Craig, & Koestner, 1995). Unmitigated Communion Feminine individuals clearly have an advantage when interacting with others in a nurturing and loving way. But what if we take this characteristic to an extreme? Imagine a woman who is not simply sensitive to others ’problems but subjugates her own needs to the needs of the people around her. Is it possible to be so fo- cused on taking care of other people that you fail to take care of yourself? These questions lead some research ers to draw a distinction between commu- nion and unmitigated communion (Helgesen, 1994). Communion —interacting with others in a compassionate and caring manner —is a positive attribute. But people high in unmitigated communion are so concerned with taking care of others that they tend to sacrifice their own needs and interests. A woman high in unmitigated communion may put her own educational and career ambitions aside to concentrate on the needs of her spouse, children and friends. Unmitigated com-munion is also associated with difficulty asserting oneself, a fear of expressing feel- ings that might lead to conflict, and a vuln erability to being exploited by others. Failure to take care of one’ s own needs can exact a heavy toll. People high in unmitigated communion tend to score low on measures of well-being and self-esteem (Aube, 2008; Fritz & Helgeson, 1998; Helgeson & Fritz, 1999; Saragovi, Koestner, Di Dio, & Aube, 1997). These individuals tend to view their personal value in terms of what others think of them (Fritz & Helgeson, 1998). Thus, as discussed in Chapter 12, their sense of self-worth is fragile and highly vulnerable to events outside their control. Not surpris- ingly, researchers find unmitigated communion is related to high scores on measures of depression (Helgeson & Fritz, 1999). One investigator collected measures of unmitigated communion from a group of men and women when they were 31 years old (Aube, 2008). The participants were contacted again at age 41 and assessed for their level of depression. For both men and women,384 CHAPTER 14 The Behavioral/Social Learning Approach

high scores on unmitigated communion predicted higher levels of depression 10 years later. Because women are more likely than men to suffer from unmiti-gated communion, these findings suggest one reason for the higher rates of depression found in women. The problems associated with unmitigated communion are especially evident when facing health issues (Helgeson & Fritz, 2000). People high in unmitigated communion may neglect their own needs at a time when self- attention is called for. Resting, eating well, attending rehabilitation sessionsand other steps necessary to improve their health may take a back seat to taking care of the family. In one study, unmitigated communion was associ- ated with poor psychological and physical health among women diagnosedwith breast cancer (Helgeson, 2003). In another investigation, women high in unmitigated communion who were diagnosed with rheumatoid arthritis ASSESSING YOUR OWN PERSONALITY Unmitigated Communion Indicate the extent to which you agree or disagree with each statement using the following scale: 1 ¼Strongly disagree, 2 ¼Slightly disagree, 3¼Neither agree nor disagree, 4 ¼Slightly agree, 5 ¼Strongly agree. 1. I always place the needs of others above my own. 2. I never find myself getting overly involved in others ’ problems. 3. For me to be happy, I need others to be happy. 4. I worry about how other people get along without me when I am not there. 5. I have great difficulty getting to sleep at night when other people are upset. 6. It is impossible for me to satisfy my own needs when they interfere with the needs of others. 7. I can ’t say no when someone asks me for help. 8. Even when exhausted, I will always help other people. 9. I often worry about others ’problems. To determine your score, reverse the answer value for item # 2 (i.e., for thisitem only, 1 ¼5, 2¼4, 3¼3, 4¼2, 5¼1). Then add all nine answer values together. The test creators found mean scores of 29.97 ( sd¼6.14) for undergraduate women and 28.55 ( sd¼5.53) for undergraduate men. The higher your score, the more you tend to subjugate your own needs tothe needs of other people. Scale: Unmitigated Communion Scale Source: Copyright © 1998 by the American Psychological Association. Reproduced with permis- sion. Fritz, H. L., & Helgeson, V. S. (1998). Distinctions of unmitigated communion from commu- nion: Self-neglect and overinvolvement with others. Journal of Personality and Social Psychology , 75, 121 –140. doi: 10.1037/0022-3514.75.1.12. No further reproduction or distribution is permitted without written permission from the American Psychological Association.Individual Differences in Gender-Role Behavior 385

became more psychologically distressed than other patients as the disease progressed (Danoff-Burg, Revenson, Trudeau, & Paget, 2004). One team of re-searchers looked at how teenagers responded to being diagnosed with diabetes, a disease that requires a good deal of personal monitoring and treatment (Helgeson, Escobar, Siminerio, & Becker, 2007). High levels of unmitigated communion predicted more depression and higher anxiety among these adoles- cents one year after the diagnosis. In short, being compassionate and nurturing is a good thing, but on occasion people need to devote some of that nurturing tothemselves. OBSERVATIONAL LEARN ING OF AGGRESSION On May 25, 2009, a 17-year-old set off a bomb outside a ManhattanStarbucks. When arrested weeks later, the boy confessed to plans for a seriesof similar attacks. He explained he was just imitating Brad Pitt ’s character in his favorite movie, Fight Club . The incident was just the latest in a long string of violent crimes linked to violence in a motion picture. The list goesback many years, and many of the examples on that list have ended in trag- edy. In December 1997 a 14-year-old boy entered his Kentucky high school carrying five guns. The boy opened fire on classmates who had gathered fora prayer meeting, and three students were killed. Later the boy said he was acting out a scene from the movie The Basketball Diaries. In July 1991 the motion picture Boyz ’nt h eH o o d opened around the country. Although calm was the norm at most of the theaters, some became the setting forreal-life violence, including severa l shootings. Thirty-five people were reported wounded or injured the first night the movie was shown. A man in Chicago was killed. In May 1981 John Hinckley tried to assassinate President Ronald Reagan. Investigator s soon discovered that Hinckley had viewed the motion picture Taxi Driver several times before the shooting. The film portrays a man who falls in love with a young prostitute, playedby Jodie Foster, and who later attempts to shoot a presidential candidate. The subsequent investigation uncover ed that Hinckley also had a strong at- traction to Jodie Foster. These incidents are painful examples of one of the most widely re- searched aspects of Bandura ’s social-cognitive theory, the relationship between modeled aggression and performed aggression. Studies not only demonstrate how people often learn behaviors through observing models, but the findings also raise some impo rtant questions about the portrayal of violence in the mass media. Bandura ’s Four-Step Model As you will see, decades of research demonstrate that people exposed to ag- gressive models sometimes imitate the aggressive behavior. Before reviewingsome of that research, an observation is in order. Anyone who watches televi- sion or goes to an occasional movie (which is just about all of us) undoubtedly has seen some murders, beatings, shootings, and the like. Yet rarely do we step386 CHAPTER 14 The Behavioral/Social Learning Approach

away from our TV set or leave the theater in search of victims. Obviously, sim- ple exposure to an aggressive model is not enough to turn us into violent peo-ple. Why, then, do individuals sometimes imitate aggression when most of the time they do not? Bandura (1973, 1986) has an answer to this question. He explains that observational learning and performance consist of four interrelated pro- cesses. People must go through each o f four steps before exposure to agg- ression leads them to act aggressively. They must attend to the aggressive action, remember the information, enact what they have seen, and expect that rewards will be forthcoming. Fortunately, most of the time circum- stances prevent people from moving through the entire process. Unfortuna-tely, sometimes they do. Let ’s look at each of the four steps in the process more closely. For observational learning to take place, people must first attend to the significant features of the model ’s behavior. We can sit in front of violent TV programs all day long, but the aggressive models will have little or no impact unless we pay attention to them. Children who watch a lot of television haveprobably seen so many TV characters punched in the face or shot that only the most graphic and spectacular action grabs their attention. Children in one study imitated aggressive models only when the acts were carried out quite vigorously (Parton & Geshuri, 1971); less intense action apparently failed to hold the children ’s attention. A viewer ’s mental state can also make him or her more attentive to the aggression. Children in one study were more likely to pay attention to an aggressive model when they were frustrated (Parker & Rogers, 1981). This finding ties back to the connection between frustration and aggression described in Chapter 6. But attending to an aggressive act is only the first step in the observational learning process. People must also remember information about the model ’sb e - havior. You are unlikely to recall any one aggressive behavior you saw on tele- vision a few weeks ago unless the behavior was quite gripping. And if you can’t recall the action, you are not likely to imitate the model. Unfortunately, although most aggressive acts we witness soon fade from our memories, not all do. Practice and mental rehearsal can keep the action fresh in our minds. Children who play with toy guns and plastic combat equipment may embed the actions of their aggressive heroes permanently into their memories. The importance of this recall was demonstrated in a study with first- and second-grade children (Slife & Rychlak, 1982). The researchers asked thechildren how much they liked each of the aggressive acts they saw on a video-tape. They also determined which toys used by the aggressive model each child liked. Then, as in Bandura ’s classic study, the children were watched for 5 minutes while they played in a room containing all the equipment neces- sary to imitate the aggressive acts they had just seen. As shown in Figure 14.4, the children were most likely to imitate the aggression when it was an act they liked and when it was performed with a toy they liked. The researchers arguethat these are the acts the children remember. This interpretation helps to ex- plain why the boys in the study were more aggressive than the girls: They liked and recalled the aggressive behavior more.Observational Learning of Aggression 387

Research indicates that children learn aggression by imitating aggressive models. Rehearsing aggression, as when children play with toy guns, is one step in this process. 123456Number of Acts Imitated Liked Toy and Liked ActDisliked Toy but Liked ActLiked Toy but Disliked ActDisliked Toy and Disliked Act FIGURE 14.4 Mean Number of Aggressive Acts Imitated Source: From “Role of affective assessment in modeling aggressive behavior, ”by B. D. Slife and J. F. Rychlak, Journal of Personality and Social Psychology , 1982, 43, 861 –868.388 CHAPTER 14 The Behavioral/Social Learning Approach

The third step in the observational learning process is that people must enact what they have seen. Remember that Bandura draws a distinction be- tween learning and performance. One reason we don ’t carry out every aggres- sive act we notice and recall is that we may lack the ability to do so. Few of us can imitate the behavior of a martial arts champion, even after watching a dozen Jackie Chan movies. We must also have the opportunity to carry out the act. I may remember from repeated exposure in movies how to hold and fire a gun. But because I don’ t have access to a gun and because I hope I am never in a situation where a gun would be useful, shooting someone with a handgun is one learned behavior I will probably never enact. The final step in the process requires individuals to expect that the aggressive act will lead to rewards and not to punishment. One study of ele-mentary school children found that aggressive boys were particularly at- tracted to what they saw as the positive consequences of aggression, such ascontrolling other children (Boldizar, Perry, & Perry, 1989). These same boys were not very concerned about potentially negative consequences, such as causing suffering or being rejected by their classmates. Where do aggressive children develop these expectancies? As described in the previous chapter, we not only learn what to do from models, we also learn what is likely to happen to us as a result of imitating them. If an aggres- sive model is declared a hero and praised, we may expect that we, too, would be rewarded for the same behavior. If the model is arrested or hurt by some- one even more aggressive, we will probably anticipate punishment. Aggressive children typically learn what consequences to expect by watching children their age or slightly older (Huesmann, 1988; Huesmann & Guerra, 1997). If an older child who pushes and punches gets his choice of toys or gets to bat first, there is a good chance the behavior will be imitated. In the same way, par- ents who physically punish children for fighting may communicate that bigger and stronger people can do what they want, which may be why corporal punish-ment is related to more aggression in children, not less (Gershoff, 2002). People are also more likely to imitate aggressive behavior that is por- trayed as justified (Paik & Comstock, 1994). Children are more likely to imi-tate a superhero who smacks around a bad guy for the good of society than a supervillain who acts violently only for his own good. Children see that the villain is punished, but they also see the good guy ’s aggressive behavior re- warded. Unfortunately, in most cases people believe their side in a conflict is the correct and just one. Therefore, like the superhero, aggression may seeman appropriate solution to their problems (Smith & Donnerstein, 1998). Thisobservation leads us to the next issue —the impact of mass media violence. Mass Media Aggression and Aggressive Behavior If you watch even a small amount of television, you are surely aware that the average American receives a heavy dose of modeled aggression almost daily. For several decades, psychologists and other professionals have been con-cerned about how this constant exposure to stabbings, shootings, beatings, and so on affects viewers, especially children. Although today the action may consist of a space monster being killed by a superhero ’s laser beam instead ofObservational Learning of Aggression 389

a bank robber felled by a bullet from a sheriff ’s gun, one estimate claimed the average American child will view about 8,000 murders and more than 100,000 other acts of violence on television before leaving elementary school (Smith & Donnerstein, 1998). And the situation isn ’t getting any better. Studies find there is more violence on prime-time television today than everbefore (Bauder, 2005). As the examples at the beginning of this section suggest, there are some very convincing instances of people witnessing and then imitating media vio-lence. However, we can ’t determine conclusively from these examples that viewing the aggressive act actually caused the person to behave violently. Itis possible that John Hinckley would have committed some other violent act IN THE NEWS Television Violence In 1951, when television was in its infancy, Senator Estes Kefauverraised questions about the impactthe new medium might have onimpressionable children. The debate over appropriate subject matter fortelevision programming has conti-nued ever since. In the middle of thisdebate is the question of violence.Psychologists have produced strong evidence suggesting exposure to violent scenes and themes increasesthe likelihood of violence byviewers. Because of its prevalence in American homes and its easy accessby children, violence on televisionmay pose a particular problem. Although objections to television violence have been around fordecades, the issue has drawnincreased attention in recent years.A 3-year study commissioned bythe National Cable TelevisionAssociation confirmed the concernabout television violence (Brown,1998; Murray, 1998). The $3.5million study found that 61% of alltelevision programs contained someform of violence. Psychologists werealso disturbed by the finding that nearly 40% of the violence wasperformed by “good”characters and that more than 70% of theaggressors showed no remorse for their actions. Thus viewers are oftenexposed to the kind of violencemost likely to be imitated —violence performed by desirable role modelswith apparently positive consequences. Professional organizations, the television industry, and political leaders have entered into the ongoing discussion. The AmericanAcademy of Pediatrics, among otherprofessional organizations, hascalled on the entertainment industryto reduce the amount of violencechildren are exposed to. Congresspassed a law requiring all newtelevision sets to come equippedwith an electronic blocking deviceknown as a V-chip (the V stands forviolence). The device allows parentsto block their children ’s access to programs identified as too violentfor young viewers. A rating systemwas also put into place so thatparents would be warned before- hand of inappropriate material in aparticular program. And the U.S.Surgeon General released a reportdeclaring that violent television programs were an important causeof aggressive behavior in children.Nonetheless, the amount of violenceshown on television has increased inrecent years (Bauder, 2005). Despite widespread agreement that there is a lot of violence ontelevision, there is little consensuson what to do about it. Even the harshest critics are often reluctantto interfere with the broadcasters’freedom of speech. Televisionrepresentatives argue that parentsshould take the lead in regulatingwhat their children watch. Othersfear that the rating system willlead to even more graphic violencein programs identified as unfitfor children, similar to what hashappened in motion pictures.What seems certain is that thedebate over televised violence willcontinue.390 CHAPTER 14 The Behavioral/Social Learning Approach

if he hadn ’t watched Taxi Driver. After all, millions of people saw the movie without reacting aggressively. Although most of us find it difficult not to see a link between viewing aggression and performing aggression in these exam- ples, these incidents supply only weak evidence for this connection. Fortunately, we don ’t have to rely on this circumstantial evidence. Researchers have provided a wealth of experimental data concerning the impact of viewing aggression on performing aggression. The vast majority of these stud- ies find the causal link irrefutable: Viewing aggression increases the likelihood ofacting aggressively, especially over a short time span (Anderson & Bushman, 2002b; Bushman & Huesmann, 2001; Friedrich-Cofer & Huston, 1986; Geen, 1998; Paik & Comstock, 1994; Smith & Donnerstein, 1998; Wood,Wong, & Chachere, 1991). Most of the data come from controlled laboratory experiments. Typically, participants watch a segment from either a violent or an arousing but nonviolent program or movie. Then they are given the opportunityto act aggressively against another person, often by administering electric shocks or loud noise they believe will hurt the other person. In almost all cases, partici- pants who watch the violence act more aggressively than those who see thenonviolent clip. Although impressive, this research also contains some limitations. The effects typically are short-lived, and the opportunity to hurt another personprovided by the experimenter is unique. Therefore, it is reasonable to wonder how much these studies tell us about the impact of aggressive movies and television shows in real-life situations. In response to this question, several investigators have conducted long- term field studies to gauge the impact of exposure to violence and aggressivebehavior outside the laboratory (Eron, 1987; McCarthy, Langner, Gersten, Eisenberg, & Orzeck, 1975; Singer & Singer, 1981). In each case the re- searchers use the amount and kind of television children watched at one point in their lives to predict how aggressive the children would be at a later date.Like the laboratory studies, these investigations find significant evidence indicating that watching a lot of aggressive television leads to more aggres- sion in children and adults. One team of researchers began their investigation by determining how much television a group of 8-year-old children watched (Eron, 1987;Huesmann, Eron, Dubow, & Seebauer, 1987; Lefkowitz, Eron, Walder, &Huesmann, 1977). They then waited 22 years before measuring aggressive behavior in these same individuals at age 30. The researchers found a signi- ficant relationship between the am ount of television the participants watched as children and the likelihood that they would have been convicted for criminal behavior by age 30. As shown in Figure 14.5, the seriousness of the criminal act was directly related to the amount of televisionwatched. The more TV the 8-year-old had watched, the more serious the adult crime. Another investigation measured the amount of television boys and girls watched at age 14 and incidences of aggression over the next 8 years(Johnson, Cohen, Smailes, Kasen, & Brook, 2002). As shown in Figure 14.6, the percentages of men and women who engaged in some act of aggressionObservational Learning of Aggression 391

(assault, physical fights resulting in injury, robbery, crime committed with a weapon) increased dramatically with an increase in television viewing. One potential difficulty in interpreting this research concerns the possibil- ity that the children watched television because they were aggressive, not the other way around. Not surprisingly, research shows that aggressive people prefer aggressive television programs (Bushman, 1995). However, when re-searchers control for the child ’s initial aggressiveness level statistically, the find- ings still suggest that watching television causes the later aggressive behavior.Moreover, when researchers account for other possible influences on aggres-sive behavior, such as neighborhood violence, childhood neglect, and family income, they still find the association between television viewing and aggres- sion (Johnson et al., 2002). In short, frequent exposure to aggressive models on television appears to increase aggressive behavior over the short run and many years later. Some of this relationship can be explained through Bandura ’s observational learning model. However, closer examination of the model suggests imita- tion may be only part of the picture. In many studies the aggressive acts dis- played by participants are different f rom the acts modeled in the films the participants are shown (Geen & Thomas, 1986). That is, exposure to an ag- gressive motion picture increases aggressive acts, but not necessarily the acts shown in the movie. Why might this be so? As we will see in Chapter 16, some concepts borrowed from the cognitive approach to personality help5 01015202530354045Seriousness of Criminal Act by Age 30 Low Medium Frequency of TV Viewing at Age 8High Low MediumFemales Males High FIGURE 14.5 Seriousness of Criminal Act at Age 30 as a Function of Frequency of TV Viewing at Age 8 Source: From Eron (1987).392 CHAPTER 14 The Behavioral/Social Learning Approach

us complete the picture. That research provides a good example of how psychologists often blend aspects of the behavioral approach with featuresfrom the cognitive approach to obtain a better understanding of complex phenomena. Violent Video Games Over the past three decades, the video game business has grown from a few quarter-eating machines in restaurants and arcades to a multibillion dollar in-dustry. Along the way, the nature of many of these games also changed: they became violent. In some cases, very violent. Players of violent video games are rewarded for killing police, prostitutes, and innocent bystanders. Weapons in- clude cars, guns, flame-throwers and chain saws. In some games, the player takes on the role of a criminal on a violent crime spree. Given what we know about the effects of violent movies and television, psychologists and other professionals began to worry about the impact of allthis exposure to simulated violence in video games. A couple of features about video games are cause for particular concern. First, players do not merely watchthe action. They pay close attention and actively engage in practicing violent acts. Second, virtually all violent video games are designed to reward violence. The more thugs, monsters, prostitutes and police you kill, the higher your score.Thus, some of the necessary elements for imitating aggression identified by Percent Committing Violent Act Less than 1 hour/dayMalesFemales 1 to 3 hours/dayMore than 3 hours/day50 4540353025201510 5 Television Viewing at Age 14 FIGURE 14.6 Violence as a Function of Television Viewing Source: Adapted from Johnson et al. (2002).Observational Learning of Aggression 393

Bandura —attending to the behavior, enhanced recall through rehearsal, seeing the behavior rewarded —are built into most violent video games. Numerous studies on the effects of violent video games have now been conducted. As with research on media violence, the findings consistently point to the same conclusion —playing violent video games increases the likelihood that the player will act aggressively (Anderson, 2004; Anderson & Bushman, 2001; Anderson, Carnagey, Flanagan, Benjamin, Eubanks, & Valentine, 2004). Participants who play violent video games are more aggressive immediatelyafterward than participants who play nonviolent video games (Anderson & Dill, 2000; Bartholow, Bushman, & Sestir, 2006). Undergraduate men and women in one study played a car-driving video game (Carnagey & Anderson, 2005). In one condition players were rewardedfor killing pedestrians and opponents. In a second condition players lost points when they hit other cars or pedestrians. In a third condition all pedes-trians were removed from the screen and players earned points simply by passing checkpoints. A little later, participants were given an opportunity to hurt someone who had insulted them by administering what they thoughtwere blasts of loud noise to their insulter. As shown in Figure 14.7, partici- pants who had been rewarded for video game violence gave louder and lon- ger blasts of noise than participants who either had been punished for violence or who played the nonviolent game. This increased aggressiveness from playing violent video games typically lasts for several minutes after people stop playing (Barlett, Branch, Rodeheffer, &Harris, 2009). Moreover, the more realistic the violence in the game, the stronger the effect (Barlett & Rodeheffer, 2009). And, consistent with Bandura’ s theory, the fact that players are actively engaged in the simulated violence may make the games especially worrisome. Participants who play violent video games act more aggressively afterward than participants who simply watch someone else play (Polman, de Castro, & van Aken, 2008). Aggression Score Nonviolent Punishment Reward180 160 140120100 FIGURE 14.7 Aggression as a Function of Video Game Violence Source: Adapted from Carnagey and Anderson (2005).394 CHAPTER 14 The Behavioral/Social Learning Approach

Psychologists have also looked at the long-term effects of playing violent video games. One study found that adolescents who play a lot of violent video games are more likely to argue with teachers and get into physical fights (Gentile, Lynch, Linder, & Walsh, 2004). Similarly, college students who frequently play these games were more likely to have engaged in violent acts (destroying property, hitting, threatening to hurt someone) during the past year than students who rarely played such games (Anderson & Dill, 2000). One team of researchers measured how often young adolescents (aver-age age 13 years) played violent video games and how often they engaged in violent behavior 30 months later (Moller & Krahe, 2009). The investigators found a significant increase in violence (hitting, threatening to hit, pullinghair) during the two and a half year period among those who had played a lot of violent video games. In short, consistent with Bandura ’s theory and research on the effects of mass media violence, it appears violent video games also contribute to aggressive behavior. LEARNED HELPLESSNESS Consider the following three cases. A woman is fired from her job becauseher employer believes the position is too demanding for her abilities. After afew frustrating weeks of job-hunting, she decides to just stay home. She stops going out with friends and shuts down other parts of her life she once enjoyed —dancing, movies, jogging. She becomes more and more depressed, develops lower and lower self-esteem, and has little faith in her ability to get another job. An elderly man is moved to a senior residential community and is told the staff will take care of all the chores he used to do. He no longerhas to cook for himself or clean his room or even do the shopping. Shortly after the move, he becomes less active. He is less talkative and less cheerful. His health begins to fail. A fourth-grade boy does poorly on a math test. Hebecomes frustrated and distressed on his next few math assignments and eventually refuses to even try. He begins to struggle in other subjects and soon loses interest in school altogether. What these three hypothetical people have in common is that they are all examples of what researchers refer to as learned helplessness . Psychology ’s interest in learned helplessness began with the curious behavior of some dogs in a classical conditioning study but soon evolved into a widely applied phenomenon. Learning to Be Helpless Like so many of the topics to come out of the behaviorist tradition, research on learned helplessness began with studies on laboratory animals. In the original learned helplessness experiments, harnessed dogs were subjected to aseries of electric shocks from which they could not escape (Overmier & Seligman, 1967; Seligman & Maier, 1967). After several trials of inescapable electric shock, the animals were placed in an avoidance learning situation.Whenever a signal sounded, the dogs could avoid electric shocks by jumping over a small partition to the other side of a shuttle-box (Figure 14.8). Naturally, dogs that had not gone through the earlier shock experience scurried aboutLearned Helplessness 395

frantically when the electric shock came on and quickly learned to leap over the barrier to safety whenever they heard the signal. But the researchers were totally surprised by the response of the dogs that had first gone through theinescapable shock experience. These dogs also ran around for a few seconds after the shock came on, but then the dogs stopped moving. “To our surprise, it lay down and quietly whined, ”a researcher explained. “After one minute of this we turned the shock off; the dog had failed to cross the barrier and had not escaped from the shock ”(Seligman, 1975, p. 22). What had happened to these dogs? According to the researchers, the ani- mals had learned that they were helpless. During the inescapable shock trials, the dogs tried various moves to avoid the shock and found that none were re- warded. The animals eventually learned there was nothing they could do toturn off the shock, and they became resigned to their helplessness. Of course, this reaction was no surprise. It ’s probably the most reasonable response to inescapable shock. The problem became apparent when the dogs experienced shock in the shuttle-box situation —shock from which they could escape. In behavioral terms, the dogs inappropriately generalized what they had learned in the first situation to the second situation. Although the dogs could easily have escaped the shock in the shuttle-box, they responded with the helplessness they had learned earlier. In fact, before the animals could learn the simple response, researchers had to physically move the dogs to the other side of the shuttle-box to show them the shock was escapable. Learned Helplessness in Humans Soon after the first demonstrations of learned helplessness in animals, psy- chologists wondered if learned helplessness could also be found in people. Ethically, we can ’t put human volunteers in a harness and subject them to inescapable shock. But with a few modifications in the basic procedure, re- searchers figured out a way to test whether humans were also susceptible toShock grid turned on on one sideShock grid turned off on other side28 in.43 in.8 ft. FIGURE 14.8 Shuttle-Box for Learned Helplessness Experiments396 CHAPTER 14 The Behavioral/Social Learning Approach

this effect (Hiroto, 1974; Hiroto & Seligman, 1975). Instead of inescapable shock, irritating (but not painful) loud noise was used. Participants were told they could turn off the noise by solving a problem (for example, pressing some buttons in the correct sequence). Some participants quickly worked through dozens of these problems, turning off each noise blast by figuring out the answer. However, other participants were given problems for which there were no solutions. Like the dogs in the earlier studies, these peoplesoon learned there was no way to escape the aversive stimulus. Would these people generalize their feelings of helplessness to other situa- tions? Participants were taken out of the noise situation and given a different Residents of old-age homes often have many of their daily tasks, such as cooking and cleaning, taken care of by the staff. However, a learned helplessness analysis suggests thatthis reduction in control may lead to problems with adjustment and health for the elderly. Photo by Marlene SomsakLearned Helplessness 397

kind of problem to work on. The ones who had found the earlier problems solvable had little difficulty with the new problems. In fact, they did no worsethan a comparison group of participants who received no noise. However, participants who had felt helpless to turn off the noise performed significantly worse on the second set of problems. Like the dogs in the shuttle-box, they appeared to have inappropriately generalized their perception of helplessness in one situation to a new, controllable situation. Numerous replications of this experiment confirm that humans are as susceptible as laboratory animals to learned helplessness (Overmier, 2002;Peterson, Maier, & Seligman, 1993). People learn they are helpless in the initial uncontrollable setting and can ’t break out of that association in subsequent situations. On occasion, people can learn to be helpless by simply observing other people who are helpless (Brown & Inouye, 1978; DeVellis, DeVellis, & McCauley, 1978). Imagine your reaction when you see several people withability similar to yours trying yet failing to pass an important test. You might conclude that you also can ’t pass the test, even though you have yet to try (“There ’s no use in trying; nobody ever passes ”). These feelings of helplessness might then be generalized to a new situation, and you could suffer fromlearned helplessness without ever experiencing failure yourself. Some Applications of Learned Helplessness Since it was first demonstrated in humans, learned helplessness has been stud- ied in hundreds of investigations and used to explain a wide variety of human problems. We ’ll look at two of those problems here —well-being among older individuals and psychological disorders. Learned Helplessness in the Elderly We commonly assume in Western society that elderly people deserve to rest after a lifetime of hard work. Retirement is structured to relieve older indivi- duals of their daily concerns and responsibilities. Retirement communities areoften designed to take care of the cooking and cleaning and structuring of daily activities. But is this approach really in the best interests of the retired person? If we apply a learned helplessness analysis to the situation, we seethat these living situations may be taking away the older persons’ control over their daily experiences. For formerly active people used to exercising agreat deal of control, living under such conditions may be similar to the expe-rience of research participants presented with uncontrollable noise. And, like the research participants, the elderly may generalize this perception of uncon- trollability to other areas of their lives. In short, the lack of motivation andactivity seen in many retired individuals may be a form of highly generalized learned helplessness. One team of investigators tested this possibility in a classic study that in- volved residents on two floors of a retirement residence (Langer & Rodin, 1976). With the administrators ’cooperation, they altered the usual treatment given to one of these groups. The researchers increased the amount of398 CHAPTER 14 The Behavioral/Social Learning Approach

responsibility and control typically exercised by these residents in several ways. Administrators gave a presentation urging the residents to take controlof their lives. Here is an excerpt from that talk: You have the responsibility of caring for yourselves, of deciding whether or not you want to make this a home you can be proud of and happy in. You should bedeciding how you want your rooms to be arranged —whether you want the staff to help you rearrange the furniture. You should be deciding how you want tospend your time, for example, whether you want to be visiting friends or whether you want to be watching television, listening to the radio, writing, reading, orplanning social events. In other words, it ’s your life and you can make of it whatever you want. (Langer & Rodin, 1976, p. 194) In addition, participants were offered a small plant as a gift. The residents de- cided whether they wanted a plant and which plant they wanted, and weretold they were responsible for taking care of it. Residents on the other floor served as the comparison group. They listened to a talk about allowing the staff to take care of things for them. They were given a plant (chosen by the staff) and were told the staff would take care of the plant for them. The differences between the two floors were soon evident. Within a few weeks, the residents in the responsibility-induced condition reported feelinghappier. Staff members noted they were visiting more and sitting around less. Nurses, who did not know a study was going on, reported 93% of these residents showed improved adjustment. Only 21% of the residents in the com-parison group showed improvement. But the effects of the treatment did not stop there. The researchers returned to the home 18 months later to find that many of these differences in happiness and activity level remained (Rodin &Langer, 1977). Most dramatically, only 15% of the responsibility-induced resi- dents had died during the 18-month period, compared to 30% of the compari- son group. Several subsequent investigations have discovered similar advantages when elderly people retain a sense of control over their lives (Chipperfield & Perry, 2006; Schulz & Heckhausen, 1999; Wrosch, Schulz, & Heckhausen, 2002). One team of researchers looked at mortality rates from all causes over a 5-year period for 20,323 middle-aged and older adults living in the com- munity of Norfolk, England (Surtees, Wainwright, Luben, Khaw, & Day, 2006). Citizens who had indicated on earlier psychological tests that they felt a sense of mastery over their lives were significantly more likely to survive than those who expressed a general sense of helplessness. These findings do not mean we should abandon those who genuinely need assistance, but some- times letting people take care of themselves is in everyone ’s best interest. Learned Helplessness and Psychological Disorders Soon after the demonstrations of learned helplessness in humans, psycholo- gists noticed some striking parallels between research participants and people suffering from depression (Seligman, 1976). Clinical psychologists have long observed that depressed patients often act as if they are helpless to controlLearned Helplessness 399

what happens to them (Beck, 1972). Severely depressed people sometimes lack the motivation even to get out of bed in the morning. Little intereststhem, and they often believe that nothing they do will turn out well. Like the dogs that lie whimpering in the shuttle-box, they seem to have given up on their ability to do anything about their problems. These observations led psychologists to suggest that depression some- times develops in a manner similar to the way research participants acquire learned helplessness (Seligman, 1975). That is, people perceive a lack ofcontrol over one important part of their lives and inappropriately generalize that perception to other situations. A college student might have difficulty in a particular class. No matter how hard she tries, she can ’ti m p r o v eh e rt e s t scores. At first she studies harder and gets advice from other students, but it doesn ’t seem to help. If it ’s important to her to do well in school, she may continue her efforts to change her grade. However, at some point she may decide that no matter what she does, she can ’t avoid the bad grade that is bound to come at the end of the term. In other words, she has learned she is helpless in this class. As a result, she may become mildlydepressed. Unless other information is forthcoming to counteract these feelings, this student may soon conclude there is no sense trying in other classes or in otherparts of her life, such as sports, friendships or keeping up her appearance. She may decide she lacks control over most of life ’s outcomes and may eventually lose the motivation to try. In learned helplessness terms, she has inappropri- ately generalized her feelings of helplessness in one situation she can ’t control to others that she might be able to control. Consistent with this interpretation of depression, people who find they cannot control relatively simple laboratory tasks, such as escaping the irritating noise, show significant increases in depressed feelings (Bodner & Mikulincer, 1998; Burger & Arkin, 1980). Other support for a learnedhelplessness –depression connection comes from research with animals. Investiga- tors find changes in neurotransmitters and receptors in animals exposed to ines-capable shock are similar to what we see in the neurotransmitters and receptorsof depressed individuals (Dwivedi, Mondal, Payappagoudar, & Rizavi, 2005; Ferguson, Brodkin, Lloyd, & Menzaghi, 2000; Joca, Zanelati, & Guimaraes, 2006; Kram, Kramer, Steciuk, Ronan, & Petty, 2000; Maier & Watkins, 2005).In particular, the neurotransmitter serotonin appears to play a role in the develop- ment of both learned helplessness and depression. Data from many different sources suggest exposure to uncontrollable events can be a cause of depression. But one difference between laboratory- induced learned helplessness and genuine depression needs to be addressed. Learned helplessness in laboratory animals is short, typically lasting no morethan a few days in rats and dogs (Maier, 2001). But clinical depression often lasts considerably longer, in some cases for years. One explanation for this discrepancy is that there are many different causes of depression, only one ofwhich is learned helplessness. Another possibility is that, in a sense, people suffering from depression continually relive the initial helplessness induction. Depressed patients typically ruminate about the causes of their depression400 CHAPTER 14 The Behavioral/Social Learning Approach

(Nolen-Hoeksema, 2000). By frequently thinking about the circumstances leading up to their depression, people may re-experience the helplessness-inducing events. Recall that even imagining oneself in an uncontrollable situation sometimes is sufficient to generate learned helplessness. Occasional reminders about the initial helplessness event can also reactivate depression. When rats in one study were exposed periodically to the location in which their initial learned helplessness experience had occurred, researchers found no decline in learned helplessness over time (Maier, 2001). In short, learned helplessness has become an important model for under- standing some kinds of depression. Experiences with uncontrollable aversiveevents can be the first step into a downward spiral of helplessness. Fortunately,research also suggests a treatment. People who experience success at control- ling outcomes soon overcome feelings of helplessness (Klein & Seligman, 1976). Thus, all the failing student may need is a good grade in another classto appreciate that she still has the ability to succeed in school, make friends and generally control her life. Moreover, as you ’ll discover in Chapter 16, whether people fall into learned helplessness may depend on how they explain their lack of control. LOCUS OF CONTROL If you’ re in good health, is it because you take care of yourself or because you’re lucky? Are lonely people without friends because they don ’t try to meet other people or because they don ’t have many opportunities? When you win in a sporting contest, is it because you did your best or because of some lucky breaks? These are the kinds of questions researchers ask when they investigate individual differences in locus of control . The key is not whether your health habits actually contribute to your good health, but whether youbelieve this to be true. Research on locus of control developed out of Julian Rotter ’s concept of generalized expectancies, described in Chapter 13. In a new situation, we have no information upon which to draw an expectancy of what might hap-pen. In these cases, Rotter argued, we rely on general beliefs about our ability to influence events. If you answered that good health comes from taking care of yourself, that loneliness is caused by not trying, and that winning a sport-ing contest is the result of effort, you probably maintain an internal locus of control orientation. Your generalized expectancy is that people can affect what happens to them and that good and bad experiences are generally ofour own making. However, if you feel that health is a matter of luck, that people are lonely because of the circumstances they find themselves in, and that winning means you got some lucky breaks, you probably fall on the external end of the locus of control dimension. More than most people, you believe that what happens to you and others is outside of your control. Because locus of control represents a generalized set of beliefs, it can potentially play a role in many areas of your life. In this chapter, we ’ll look at how indi- vidual differences in locus of control affect your psychological well-being and your physical health.Locus of Control 401

Locus of Control and Well-Being Who is happier —internals who believe they can control most things or exter- nals who recognize the limits that outside forces place on them? Which person is more productive, better liked, and better adjusted? A case can be made for either position. On one hand, we could argue that internals probably work harder and thus achieve more because they feel they control outcomes. Externals, who give up in the face of setbacks and who quickly conclude there is nothing they can do to correct a problem, are unlikely to get far in a world filled with obstacles and challenges. On the other hand, just because peopleASSESSING YOUR OWN PERSONALITY Locus of Control Indicate the extent to which each of the following statements applies to you.Use the following scale: 1 ¼Disagree strongly, 2 ¼Disagree, 3 ¼Disagree slightly, 4 ¼Neither agree nor disagree, 5 ¼Agree slightly, 6 ¼Agree, 7¼Agree strongly. 1. When I get what I want, it ’s usually because I worked hard for it. 2. When I make plans, I am almost certain to make them work. 3. I prefer games involving some luck over games requiring pure skill. 4. I can learn almost anything if I set my mind to it. 5. My major accomplishments are entirely due to my hard work and ability. 6. I usually don ’t set goals because I have a hard time following through on them. 7. Competition discourages excellence. 8. Often people get ahead just by being lucky. 9. On any sort of exam or competition I like to know how well I do relative to everyone else. 10. It ’s pointless to keep working on something that ’s too difficult for me. To determine your score, reverse the point values for items 3, 6, 7, 8, and 10 (1¼7; 2¼6; 3¼5; 5¼3; 6¼2; 7¼1). Then add the point values for each of the 10 items together. A recent sample of college students found a mean of 51.8 for males and 52.2 for females, with a standard deviation of about 6 for each. The higher your score, the more you tend to believethat you are generally responsible for what happens to you in personal achievement situations. Scale: Personal Efficacy Scale Source: Copyright © 1998 by the American Psychological Association. Reproduced with permission. Paulhus, Delroy (1983) Sphere-specific measures of perceived control. Journal of Personality and Social Psychology . 44, 1253 –1265. doi 10.1037/0022-3514.44.6.1253. No further reproduction or distribution is permitted without written permission from the American Psychological Association.402 CHAPTER 14 The Behavioral/Social Learning Approach

believe they are in control does not mean they actually exercise control. Highly internal people may invest their efforts inefficiently chasing rainbows or mak- ing plans at odds with reality. Perhaps externals understand their limits andwork to achieve only what is reasonably attainable. Of course, happiness is determined by many factors, and we can point to happy and unhappy people at any point of the locus of control spectrum.Nonetheless, researchers find that, with a few exceptions, internals tend to be happier than externals (DeNeve & Cooper, 1998; Ng, Sorensen, & Eby, 2006). To better understand this conclusion, let ’s look at the connection between locus of control and a few markers of psychological well-being: psychological disorders, achievement, and psychotherapy results. Psychological Disorders People suffering from psychological disorders tend to be more external than internal (Lefcourt, 1982; Phares, 1976; Strickland, 1978). Researchers areparticularly interested in the relationship between locus of control and depres- sion. The reasons for this connection tie back to the research on learned help- lessness. It may be that externals often find themselves in situations similar to that of learned helplessness participants who cannot control important out- comes. Consider the findings of a study in which recently diagnosed cancer Of the millions who try to lose weight each year, only a small number of people succeed in taking it off and keeping it off. One variable that may affect a diet ’s success or failure is the extent to which the dieter believes he or she is capable of losing the weight.Photo by Kathryn MacLeanLocus of Control 403

patients were tested for level of depression (Marks, Richardson, Graham, & Levine, 1986). For external patients, the more severe the diagnosis, the moredepressed they became. However, the severity of the disease had no impact on the depression experienced by internal patients. These patients believed they could still control the course of the disease, and this belief shielded them from giving up and becoming depressed about their situation. One review found an average correlation of .31 between locus of control scores and measures of depression, with external scores associated with higher levels ofdepression (Benassi, Sweeney, & Dufour, 1988). As discussed in Chapter 7, this is an impressively high correlation. A dramatic example of how locus of control is related to depression was demonstrated in a study of suicidal patients (Melges & Weisz, 1971). Patientswho had recently attempted suicide were asked to relive the events that took place immediately before the attempt. Patients were left alone with a tape re-corder and asked to describe in the present tense what had happened to them during this time. Analysis of the recordings revealed that patients described themselves in more external terms as they became more suicidal. Other stud-ies find suicide attempters often experience an increase in events outside their personal control just prior to the attempt (Slater & Depue, 1981) and that external adolescents and college students report more suicidal thoughts than internals (Burger, 1984; Evans, Owens, & Marsh, 2005). Although studies suggest locus of control is related to depression, we need to add a few notes of caution when interpreting the findings. First, thevast majority of people scoring on the external end of locus of control scales are happy and well-adjusted. Second, because the relationship is correla- tional, it is difficult to make strong statements about external locus of control causing the disorder. It may be that externals are susceptible to depres- sion, but it is also possible that depressed people become more external. Achievement One indicator of well-being in Western society is how much we achieve in school and in our careers. Although high achievers are by no means shieldedfrom psychological problems, we often point to a deteriorating job perfor- mance as a reason for concern. Similarly, improved performance in school or work is often seen as evidence that a therapy client is getting better. When re- searchers use locus of control scores to predict achievement, they consistently find that internal students receive higher grades and better teacher evaluations than externals (Cappella & Weinstein, 2001; Findley & Cooper, 1983;Kalechstein & Nowicki, 1997). This finding is true of elementary, high school, and college students, but the relationship is especially strong among adolescents. Why do internals do better in school? One reason is that they see them- selves as being responsible for their achievements. Internal students believestudying for tests pays off, whereas externals are less likely to feel that theirefforts affect their grades. Internals and externals also respond differently to feedback (Martinez, 1994). Internal students are likely to attribute high test scores to their abilities or to studying hard, whereas externals who do well404 CHAPTER 14 The Behavioral/Social Learning Approach

might say they were lucky or that the test was easy. Internals also are better at adjusting their expectancies following feedback, which means they have abetter idea of how to prepare for the next exam. Externals are more likely to make excuses following a poor performance (Basgall & Snyder, 1988). An external student who decides the teacher is an unfair grader probably will not prepare much for the next test. Because they believe academic success is up to them, internal students also pay attention to information that will help them reach their goals. One investigator found internal undergraduates weremore likely than externals to know about test dates, grading policies, and other relevant information that would help them do well in their classes (Dollinger, 2000). Internal students also are more ambitious than external stu-dents. Because they see outcomes as under their control, they are more likely to work to reach their goals. Internal college students are more likely than exter- nals to complete their degrees in a timely fashion (Hall, Smith, & Chia, 2008).They also are more likely to apply to graduate school (Nordstrom & Segrist, 2009). Higher achievement by internals is not limited to the classroom. Studies in career settings also find higher levels of performance for internal workersthan externals (Judge & Bono, 2001; Ng et al., 2006). People who believe making a sale, inspiring employees, or completing a job on time is largely up to them are more likely to reach those goals than workers who fail to see their role in achieving work objectives (Judge, Erez, & Bono, 1998). Internal workers also tend to seek out positions that provide the opportunities and re- wards they prefer (Spector, 1982). Not surprisingly, researchers find internals score higher than externals on measures of job satisfaction (Judge & Bono, 2001, Ng et al., 2006). The relationship between perceived control and achievement may be universal. One team of investigators examined locus of control and job satis- faction among managers in large corporations in 24 countries (Spector et al.,2001). Across different cultures, managers who experienced a great deal of control over their work environment reported consistently higher levels of satisfaction with their jobs than managers who felt they had little control. Psychotherapy As a general rule, clients tend to become more internal as they pass through successful psychotherapy (Strickland, 1978). Consider the case of Israeli sol- diers suffering from posttraumatic stress disorder following their experiences with intense combat (Solomon, Mikulincer, & Avitzur, 1988). These men suf- fered from a variety of symptoms often found after a profoundly stressful ex- perience. When tested shortly after combat, the soldiers scored fairly external on locus of control measures. However, as they recovered from their traumaover the next 3 years, they became increasingly internal. As the soldiers came to appreciate the control they could exercise over many parts of their lives, they took an important step toward recovery. Does this mean that therapists should focus on giving clients more control over therapy? Not necessarily. Although internals respond well when givencontrol over their treatment, externals sometimes do better when treatmentLocus of Control 405

remains in the therapist ’s hands (Schwartz & Higgins, 1979). One team of in- vestigators looked at depression levels in patients with rheumatoid arthritis (Reich & Zautra, 1997). The external patients became less depressed when their spouses provided them with a lot of support and assistance. However, internal patients showed an increase in depression when their spouse gave this same amount of assistance. The researchers speculate that the spouse ’s care was seen as helpful by the externals but as an indication of dependenceby the internals. Locus of Control and Health One of the most frustrating problems health care professionals face is lack ofpatient cooperation. Many patients discontinue their therapy programs or simply stop taking their medicines. On the other hand, other patients do an excellent job of watching their diets, taking medication, attending therapy, and keeping appointments. Observations like these lead some psychologists to suggest that locus of control might play a role in health behaviors (Strickland, 1989; Wallston, 2005). Those who believe their health is largelyin their own hands will do what they can to get better. Those who attribute poor health to bad genes or fate may see little reason to make the effort. Research tends to support this analysis. One study followed 5,114 middle-aged men and women for an eight and half year period (Sturmer &Hasselbach, 2006). Locus of control scores taken at the beginning of the study were strong predictors of which participants would suffer from heartattacks and cancer by the end of the study. As predicted, internals were less likely than externals to experience these health problems. Another investiga- tion measured locus of control in 10 year olds (Gale, Batty, & Deary, 2008). When the researchers contacted these individuals again 20 years later, they found the participants identified as internals two decades earlier were in sig- nificantly better health than those identified as externals. Psychologists explain these effects in terms of the way internals and exter- nals approach their physical health. People who take an external orientationtoward their health believe there is little they can do to improve their physical condition or avoid disease. Whether they become ill is out of their control, and when they become ill they depend on health professionals to make them well again. On the other hand, people with an internal locus of control be- lieve they have a significant role in maintaining good health. Because they see a relationship between what they do and how they feel, internals are more likely than externals to eat well and participate in health-maintaining exercise programs, such as aerobics or jogging. Consistent with these descriptions, several studies find that internals prac- tice better health habits and are generally healthier than externals (Johanssonet al., 2001; Klonowicz, 2001; Ng et al., 2006; O ’Hea, Grothe, Bodenlos, Boudreaux, White, & Brantley, 2005; Perrig-Chiello, Perrig, & Staehelin,1999; Simoni & Ng, 2002). One investigation found college students whoheld an external locus of control toward their health were more likely to smoke, drink alcohol, skip breakfast, eat fatty foods, and consume less fruit and fiber than internals (Steptoe & Wardle, 2001). Internals also have more406 CHAPTER 14 The Behavioral/Social Learning Approach

confidence in their ability to control stressful situations and, therefore, are less likely to suffer the health-harming consequences of stress (Weinstein &Quigley, 2006). A study of business executives in high-stress positions found that internals were less likely to become ill than externals (Kobasa, 1979). Internals also are more likely than externals to seek out information about health problems (Wallston, Maides, & Wallston, 1976). However, as with psy- chotherapy success, a match between locus of control and the health message may be the most effective approach. One study looked at the effectiveness of cam-paigns encouraging middle-aged women to obtain a mammogram (Williams- Piehota, Schneider, Pizarro, Mowad, & Salovey, 2004). Half the women received a brochure and phone call that was targeted to internals. The brochure was titled“The Best Thing YouCan Do for Your Health, ”and contained messages like, “You hold the key to your health. ”The other half received messages targeted to externals. The brochure was titled, “The Best Thing Medical Science Has to Offer for Your Health, ”with messages like, “Health care providers hold the key to your health. ”As shown in Figure 14.9, internal women were more likely to get a mammogram within the next 6 months when they received the internally worded messages. In contrast, external women were more likely to get a mammo- gram when receiving the externally oriented message. Although researchers often find a connection between locus of control and health, this is not always the case. Some investigations fail to find health dif-ferences between internals and externals or find only weak effects (Bettencourt, Talley, Molix, Schlegel, & Westgate, 2008; Norman & Bennett, 1996). Why might this be so? To answer this question, we need to return to Rotter ’st h e o r y . Recall that Rotter said behavior was a function of both expectancy andrein- forcement value. That is, I might expect that studying for a test will result in agood grade. However, if I don ’t value that grade, I am still unlikely to make the effort. Psychologists make similar predictions when they apply Rotter ’s model to health behaviors (Wallston, 1992; Wallston & Smith, 1994). That is, believ- ing that your actions can affect your health is not enough. You also need to 55 50 45 3540 30Percentage Obtaining MammogramsInternals Externals Internal MessageExternal Message FIGURE 14.9 Percentage of Women Obtaining Mammograms Source: Adapted from Williams-Piehota et al. (2004).Locus of Control 407

place great value on good health. Of course, everyone wants good health. But if you think about some of the people you know, you probably can identifythose who place health at the top of their concerns and those who don ’t. According to the theory, people who place a high value on their health and who believe there is something they can do to control their health are theones who watch what they eat, exercise, and get regular checkups. You might believe that daily exercise leaves you feeling fit and full of energy. But if you don’t particularly value these effects (especially if you value less-exhausting activities more), it ’s unlikely you ’ll enroll in a fitness program. Several investigations find evidence for this reasoning (Norman & Bennett, 1996; Wallston & Smith, 1994). Internal participants in one studywho placed a high value on health were found to eat more fruits and vegeta- bles and fewer fatty foods and snacks than either external participants or in- ternals who did not value physical health (Bennett, Moore, Smith, Murphy, &Smith, 1994). Similar results have been found in studies looking at breast self- examination (Lau, Hartman, & Ware, 1986) and efforts to stop smoking (Kaplan & Cowles, 1978). In short, health professionals face two tasks whentrying to get patients to take better care of themselves. Patients must place their health high on their list of things they value, and they must believe that they can influence the extent to which they are healthy. SUMMARY 1. From the day we are born, most of us face tremendous socialization pressures to take on the gender roles deemed appropriate by society. Through a combination of operant conditioning and observational learn-ing, boys tend to act like other boys, and girls like other girls. Research on individual differences in gender-role behavior was originally stifled by a model that viewed masculinity and femininity as polar opposites. Theandrogyny model sees these as two independent traits and argues that the most well-adjusted people are those who are androgynous —that is, high in both masculinity and femininity. 2. Researchers agree that exposure to aggressive models increases a person ’s likelihood of acting aggressively. Bandura ’s four-step model helps explain why people sometimes imitate aggressive acts they see and sometimes do not. Before people imitate aggression, they must attend to the act, recall it, have the opportunity to engage in the behav ior, and believe the aggression will lead to rewards. Research from laboratory and long-term field studies indi- cates that exposure to mass media violence increases aggressive behavior. 3. Like much behavioral research, work on learned helplessness began with experiments on laboratory animals. Researchers observed that dogs thatlearned they were helpless to escape shock in one situation inappropri- ately generalized this perception of helplessness to a new situation. Subsequent research found that humans are also susceptible to this effect. Research suggests elderly people may adjust better to retirement commu- nities when they are allowed to retain some control over their situation. Depression may develop when people perceive a lack of control over an408 CHAPTER 14 The Behavioral/Social Learning Approach

important event and inappropriately generalize that perception to other aspects of their lives. 4. The most widely researched aspect of Rotter ’s social learning theory is the notion of individual differences in generalized expectancies, or locus of control. At one end of this dimension we find internals, who generally believe they control what happens to them. On the other end are exter- nals, who generally hold that what happens to them is under the control of outside forces. Internals generally do better than externals on measuresof well-being and health. KEY TERMS androgyny (p. 379) learned helplessness (p. 395) locus of control (p. 401)masculinity-femininity (p. 378) unmitigated communion (p. 384) MEDIA RESOURCES Visit our website. Go to www.cengage.com/ psychology/burger to find online resources directly linked to your book, including quizzing, glossary, flashcards, and more.Summary 409

CHAPTER15 The Cognitive Approach Theory, Application, and Assessment Personal Construct Theory Cognitive Personality VariablesCognitive Representations of the Self Application: Cognitive (Behavior) Psychotherapy Assessment: The Repertory Grid Technique Strengths and Criticisms of the Cognitive Approach Summary 410