Opportunities Lost: The Impact of Stereotypes on Self and Others
There is ample evidence to suggest that negative expectations and stereotypes about the competence of older adults pervade Western culture (e.g., Hummert, 1999; Kite and Wagner, 2002). For example, older adults are characterized as more forgetful and less able to learn new information (e.g., Hummert, Garstka, Shaner, and Strahm, 1994). In addition, young and old people alike believe that there is general memory decline across the latter half of the life span (Lineweaver and Hertzog, 1998; Ryan, 1992; Ryan and Kwong See, 1993). Research corroborates these views: there is abundant evidence that older adults do perform more poorly than younger adults across a wide variety of cognitive tasks (for a review, see Zacks, Hasher, and Li, 2000).
Yet there is also evidence of older adults serving important roles in society. For example, nearly 40 percent of the nation’s 1,200 working federal judges have reached senior status and could retire. But, these senior judges are crucial to the justice system and, handling reduced caseloads, carry out nearly 20 percent of the federal judiciary’s work (Markon, 2001). This fact is also consistent with the literature on cognitive aging, which shows that reasoning about complex matters relevant to daily life—what some call wisdom–shows no deterioration with age (Baltes and Kunzmann, 2003). Yet pervasive beliefs about age-related decline tend to outweigh beliefs about positive aging in our culture. Most people expect that losses will outnumber gains as they get older (Heckhausen, Dixon, and Baltes, 1989); most people expect their abilities to decline with age (Staudinger, Bluck, and Herzberg, 2003).
Most of the work on stereotyping and aging documents this phenomenon. Far less examines the degree to which negative and positive stereotypes have an effect on the quality of life for older adults. Do negative expectations of older people and ageist beliefs lead people in general, as well as older adults themselves, to underestimate the capacities of older adults? Do positive expectations have the opposite effect?
Negative stereotypes can have harmful consequences for the quality of life of older adults and can also result in a major loss to society. With increases in life expectancy as well as reduced infirmity, many adults are aging well, but negative stereotypes of aging may put society at risk for losing the contributions of these vital and knowledgeable people. The potential individual and social effects underscore the need to understand the content of aging stereotypes in terms of their accuracy and applications. It is especially important to understand how negative stereotypes exacerbate poor performance in areas in which decline is real. That is, beliefs that memory is bad in old age can reduce motivation when increased motivation is needed instead. A framework for predicting and interpreting individuals’ behavior is imperative to understand how aging stereotypes drive behavior in both positive and negative ways.
Social psychologists have a long history of studying stereotypes and their effects on judgment and behavior. As outlined in more detail below, stereotypes people have about others can influence how those others are treated and in turn elicit particular behaviors from the others that are consistent with those stereotypes (e.g., Snyder, 1992). In addition, stereotypes can exert a direct influence on the stereotype holder. In particular, activation of a stereotype can cause people to act in a manner consistent with the stereotype (Dijksterhuis and Bargh, 2001), regardless of whether they are members of the stereotyped group or not (Wheeler and Petty, 2001).
The current literature suggests that both positive and negative stereotypes influence judgments made about older adults in everyday life. There are countless ways in which negative stereotypes can have serious personal consequences on the way older adults are perceived and treated (Pasupathi and Löckenhoff, 2002). For example, Erber and colleagues (2001) find that memory failures are seen as more serious for older adults than younger adults and support the perceiver’s negative expectations of aging. Older adults are repeatedly reminded of negative stereotypes associated with aging in a variety of settings, such as media advertising of products and services that focus on such aspects of aging as memory loss, frailty, incontinence, and loss of mobility. Other examples include ageist views of older
workers on the job and its harmful effects on employee satisfaction (Gordon et al., 2000; McMullin and Marshall, 2001) and ageist views of patients in a medical setting, who receive less aggressive medical treatment because their physical complaints are dismissed simply as signs of aging. In many settings, patronizing forms of communication are used with older adults despite the fact that it is viewed as debasing and disrespectful (see Hummert, 1999; Kemper, 1994; Ryan, Meredith, and Shantz, 1994). As noted by Richeson and Shelton (in this volume), negative stereotypes of age-related cognitive deficits are far more severe than the actual deficits. Those stereotypes may inhibit older people from attempting and actively participating in new activities or exercising their full potential.
A critical issue that emerges from these findings is the extent to which negative stereotypes affect the behavior of older adults in an everyday context. For example, negative stereotypes may not only affect the attributions of medical personnel regarding an older adult’s symptoms (i.e., viewing them as normal aging instead of as treatable conditions), but may also affect the older person’s understanding of what normal aging is. Thus, the older adult does not receive enough medical care or doesn’t want more medical care because of his or her own stereotypes about normal aging. Do older adults themselves overlook symptoms of disease because they view them as part of normal aging, when they should be taking these symptoms more seriously? Older adults’ perceived choices also need to be taken into account. Research should examine knowledge and individual choice on the part of older adults in making medical decisions (see Chapter 4).
Fortunately, positive stereotypes and attitudes toward aging can also affect how older adults are treated. For example, Erber and Szuchman (2002) found that a forgetful older adult is seen as having more desirable traits than a forgetful young adult. Similarly, in legal settings older witnesses are believed to be just as credible as younger witnesses despite older adults’ memory failures (Brimacombe, Jung, Garrioch, and Allison, 2003). Thus, despite perceptions of declining memory capacity on the part of older adults, they can still be viewed as credible or desirable. There is even a recent emergence in the mass media of positive stereotypes of aging, with older characters described as powerful, active, and healthy (Pasupathi and Löckenhoff, 2002).
What can be abstracted from these few studies is that the social context moderates perceptions and treatment of older adults. Research is needed to determine the degree to which age-differentiated perceptions of behavior are ageist, where they are prominent, and the extent to which behaviors distance and exclude older adults and the extent to which behaviors are beneficial and protective of older adults. For example, are ageist attitudes less prominent in interpersonal settings? Research is also needed to identify the conditions under which positive or negative stereotypes affect decisions
made about older people in everyday life—such as whether an older person should continue to drive or requires assisted living or in communications between older people and health care providers.
IMPLICIT STEREOTYPE ACTIVATION
From a sociocultural perspective, negative age stereotypes are socialized early in life (Kwong and Heller, 2005; Montepare and Zebrowitz, 2002) and become so well ingrained that they may be automatically activated in the mere presence of an older person (Hummert, Gartska, O’Brien, Greenwald, and Mellot, 2002; Perdue and Gurtman, 1990). A social psychological perspective further suggests that stereotypes can be viewed as person perception schemas. By examining the cognitive representation of stereotypes important questions can be addressed: Under what conditions are stereotypes activated? Under what conditions do stereotypes guide social judgments and behavior? Why do behaviors reflect negative stereotypes more so than positive ones?
As reviewed in the paper by Richeson and Shelton (in this volume), there is a wealth of evidence describing positive and negative stereotypes of older adults and a growing literature indicating the conditions under which stereotypes are activated. With respect to stereotype activation in perceivers, the literature identifies physical cues, the subcategory of “old-old” adults, contexts involving physically or mentally demanding tasks, and off-target verbosity as elicitors of negative stereotypes (Bieman-Copland and Ryan, 2001; Hummert, Garstka, and Shaner, 1997). More “youthful” behavior and young-old status are elicitors of positive stereotypes (Hummert et al., 1994). Finally, although both younger and older adults hold negative views associated with aging (Hummert et al., 2002; Nosek et al., 2002; Parr and Siegert, 1993; Ryan, 1992), there is conflicting evidence as to whether older adults themselves have more positive or negative views than younger adults.
Other questions with respect to stereotype activation are the degree to which people are or are not aware of having such evaluative attitudes about the elderly, and the effect that such attitudes have on people’s thoughts and actions regarding the elderly. In other words, there is a need to distinguish implicit from explicit activation of stereotypes. For example, with respect to older adults who belong to multiple identity categories (age/gender/race), are the same attitudes activated both consciously (explicitly) and unconsciously (implicitly)?
To examine such implicit constructs and processes, social psychologists have developed a battery of implicit measures that do not call for conscious self-reports of the construct or process. The earliest such measures were in essence disguised self-reports (e.g., thematic apperception tests) or behav-
ioral observations (e.g., how close one sat next to a stranger) from which researchers inferred an underlying attitude or motive. Recently, implicit measures based on reaction times have demonstrated considerable utility in predicting behaviors that could not be predicted by direct self-reports (e.g., Dovidio, Kawakami, and Gaertner, 2002). Furthermore, even when direct self-reports were useful in predicting behavior, implicit measures have been shown to account for additional variance (e.g., Vargas, von Hippel, and Petty, 2004).
Two measures have captured the bulk of recent research attention. One measure is based on priming procedures developed initially by cognitive psychologists. With this measure, participants are presented with different stimuli (e.g., elderly or young faces—the primes) and then asked to indicate the evaluative meaning (i.e., good/bad) of various words (e.g., dirt or flower) as quickly as possible by pressing an appropriate response key on a computer. Reaction times for the classification of the words are assessed. To the extent that elderly faces facilitate responses to negative words or inhibit responses to positive words in comparison to young faces, one can infer that a negative attitude toward the elderly is automatically activated when the face appears (e.g., Fazio, Sanbonmatsu, Powell, and Kardes, 1986).
The second measure is the implicit association test (Greenwald, McGhee, and Schwartz, 1998). It assesses the strength of association between a target concept (e.g., the elderly) and an attribute dimension (e.g., good/bad) by examining the speed with which participants can use two response keys to categorize words (or pictures) when each key is assigned a dual meaning (e.g., elderly/good versus young/bad). If people are classifying young (e.g., spring break) and old (e.g., retirement) terms or positive (e.g., flower) and negative (e.g., dirt) words, the question is whether it is easier to do so when elderly is paired with good or with bad on the response keys. The relative pattern of reaction times to the categorization task is informative with respect to whether the category of elderly is more closely associated with good or bad. Both the priming measure and the implicit association test have been used successfully in research on prejudice toward a wide variety of social groups (see recent reviews by Blair, 2002; Fazio and Olson, 2003).
With respect to aging, Hummert and colleagues (2002), using the implicit association test, found that people of all ages were faster to respond to young-pleasant and old-unpleasant trials than to old-pleasant and young-unpleasant trials. Furthermore, all individuals had implicit age attitudes that strongly favored the young over the old. Again, these experiments and others (Levy, 2003) demonstrate that the activation of negative stereotypes about aging affects people’s automatic evaluations without their necessarily being aware of it.
Interestingly, just as activating stereotypes about the elderly can cause
elderly individuals to act in a more elderly fashion, so too can activating stereotypes about the elderly cause young people to act in a more elderly manner. Thus, after activating the stereotype of the elderly, young college students walked more slowly down the hall (Bargh et al., 1996) and endorsed more politically conservative attitudes (Kawakami, Dovidio, and Dijksterhuis, 2003). Understanding the mechanisms behind the effect of stereotypes is an area ripe for research. Different explanations have been favored—self-stereotype activation versus other stereotype activation—although the behavioral and judgmental effects of activating these stereotypes are quite similar (Wheeler and Petty, 2002).
MULTIPLE CATEGORIES OF IDENTITY
A few studies have examined categories in addition to age as moderators of age-related stereotypes. Older men are perceived more positively than older women (Kite and Wagner, 2002). A gender-based double standard is applied to typical, but not optimal aging (Canetto, Kaminski, and Felicio, 1995). Simulated juries are more likely to vote for conviction when the victim is an older statesman than an older grandfather (Nunez et al., 1999). Few studies have considered race or ethnicity, with only a handful of studies examining cultural differences.
Researchers in social psychology have recognized the importance of examining the degree to which social context and shifting standards moderate automatically activated stereotypes, such as race and gender (Blair, 2002). They argue that studying a single status category (such as age) from any physical context may exaggerate the importance of global stereotypes and attitudes and obscure the importance of contextual variation. Social judgments typically result from multiple categorizations of the same individual, such as age and role, age and race, or age and gender. There is recent evidence that shows that the automatic evaluations that result from multiple simultaneous categorizations reflect emergent properties of combined categories (e.g., Barden, Maddux, Petty, and Brewer, 2004). For example, whites generally have more negative automatic evaluations of blacks than they do of whites (e.g., Greenwald et al., 1998), but automatic evaluations of “black lawyers” are more positive than automatic evaluation of “white lawyers” (Barden et al., 2004). That is, this particular race and role combination changes the automatic response. This research relies on a social cognitive approach that examines how individuals extract information from multiple sources and combine them in complex ways to produce both controlled and automatic patterns of bias. Research has not yet addressed how automatic evaluations of the elderly are affected by other variables, such as occupational or other roles, gender, or race.
Given the importance of automatic evaluations and stereotypes in af-
fecting behavior, there is a need to further examine the nature of age stereotypes and multiple categories embedded in a social context. As the population of older workers is growing, an important topic is stereotyping in the workforce—the interfaces among employment status and age and occupational roles and stereotypic beliefs. This topic is particularly important in that older adults’ alleged incompetence often lies in the eye of the beholder. Since relatively little relationship has been found between age and job performance (Salthouse and Maurer, 1996), it is important to identify social context effects that moderate such perceptions. Other status variables such as health, gender, and ethnicity may interact with age to produce combined categories. How various stereotype categories become activated has implications for hiring practices, training, and retirement.
Another important process that needs to be considered is the context in which the elderly person is being evaluated. According to the shifting standards framework, people make judgments about individuals who belong to stereotyped groups on the basis of within-category judgments (Biernat, 2003; Biernat and Manis, 1994). For instance, because men are assumed to be stronger than women, when a women is described as strong there is an implied assumption that she is strong for a woman. Thus, the description depends on the context and how the meaning of the word “strong” differs for men and women. Similarly, if one says an elderly person is healthy, it likely reflects a judgment made in comparison to other older adults. And even if there are objective differences in memory performance, an elderly individual may be judged to have a good memory because of the implicit comparison against other older adults.
AGE IDENTITY AND SELF-CONCEPT
Recent research suggests that older adults do not necessarily internalize negative aging stereotypes (Zebrowitz, 2003). It is possible for people to know certain information about themselves, but have contradictory feelings about that information. For instance, although people may be accurate in reporting their objective age and even use terms that are appropriate for their age group, they may subjectively view themselves as being much younger than that age. That is, they may believe certain stereotypes about older adults, but not believe that those stereotypes apply to them because they are not, subjectively, old. On implicit measures of age identity, older adults identify with the category “young” as strongly as do younger adults (Levy and Banaji, 2002). What accounts for individual differences in age identity and what are the effects of those differences on the quality of life for older adults?
It may be possible, paradoxically, that a lifetime of discrimination is protective in old age. There is considerable evidence in the social psychol-
ogy literature that being the target of discrimination is stressful (Crocker, Major, and Steele, 1998). This stress may be associated with numerous physical health problems, including shortened life expectancy. However, the evidence on mental health is more ambiguous (see Richeson and Shelton, in this volume). For example, decades of research comparing the self-esteem of African Americans to European Americans have found small differences between the groups, with slightly higher self-esteem among African Americans (Porter and Washington, 1979; Rosenberg, 1965). Crocker and Major (1989) argued that recognizing that negative treatment and outcomes are the result of prejudice is protective for self-esteem. That is, African Americans can attribute negative outcomes, such as doing poorly on a task or being treated poorly, to prejudice rather than anything personal. One possibility is that stigmatized group members learn coping strategies for dealing with discrimination (Miller and Major, 2000), perhaps by comparing their outcomes to those who share their stigmatizing condition rather than to the population as a whole. Thus, they might value domains that favor their own group and devalue those that favor other groups (Steele, Spencer, and Aronson, 2002). For those who survive and manage to cope with life-long prejudice, the experience of aging might be quite different than it is for those who have not faced discrimination. Their coping methods may allow them to deal with ageism in much the same way that was adaptive in the past. Research could examine how the experience of being discriminated against over the life course may prepare elderly members of stigmatized groups to cope with age discrimination.
Attention should be focused on the strategies that older adults use for preserving well-being in the face of age discrimination. For example, they might compare themselves only to people of a similar age and value only those domains in which positive outcomes are associated with aging. Some research suggests that older adults may disidentify with their age group in the face of age discrimination (Zebrowitz and Montepare, 2000) to avoid a stigmatized identity and its harmful effects on well-being. Other studies suggest that identifying with one’s older group provides a positive identity in spite of age-related stigmatization (Branscombe et al., 1999; Schmitt et al., 2002). Group identification may enable older adults to avoid negative effects of age discrimination as a form of secondary control (Garstka et al., 2004). In this way older adults could adaptively respond to negative and uncontrollable consequences of age discrimination. It would be important to tease apart and expand the repertoire of control strategies in the context of age discrimination. Furthermore, learning about this repertoire of control strategies may also suggest new ways for people to cope with other stereotypes, including those surrounding the cognitive or physical limitations that frequently occur with aging.
EFFECTS OF INTERNALIZED STEREOTYPES
As noted above, there is emerging evidence that negative stereotypes about aging may affect not only others’ perceptions about the abilities of older adults, but also the actual behavior of older adults. In recent years, experimental approaches to social cognition have demonstrated that at least part of the documented decline in cognitive functioning can be attributed to beliefs about aging and the social context of the testing. Because there are widespread beliefs in the culture that memory declines with age, tests that explicitly feature memory may invoke performance deficits in older people. In one study, Tammy Rahhal and her colleagues compared memory performance for adults of different ages (17 to 24 and 60 to 75) under two conditions (Rahhal, Hasher, and Colcombe, 2001). In one, the experimental instructions stressed that memory was being tested, with the experimenter repeatedly stating that participants should “remember” as many statements from a list as they could. In the other condition, the experimental instructions were identical except that emphasis was placed instead on learning, that is, participants were instructed to “learn” as many statements as they could. This study showed rather dramatic effects: there were age differences when memory was emphasized, but there were no age differences when learning was emphasized.
One possible explanation for this finding is “stereotype threat” (Steele, 1997). In a series of studies, Steele and colleagues have shown that when people fear confirming a negative stereotype they may perform badly, possibly due to the anxiety involved (e.g., Steele and Aronson, 1995). Thus, when the task was labeled as a memory task, it might have invoked the forgetful stereotype of the elderly. Alternatively, considerable research also suggests that merely priming stereotypes or specific traits can cause people to act in accordance with the primed concepts even if the stereotypes and traits are not self-relevant. Thus, priming the elderly causes young people to walk more slowly (Bargh et al., 1996), and priming the trait of hostility causes people to cooperate less in a game (Carver et al., 1983). Research is needed to pin down the mechanism by which the elderly and other individuals respond to activated stereotypes and primes.
Thomas Hess and his colleagues have also documented depressed performance when aging decline is emphasized to participants (Hess, Auman, Colcombe, and Rahhal, 2003). In their study, older participants read one of three simulated newspaper articles prior to completing a memory task. One article reaffirmed memory decline and stated that older people should rely on others to help them. The second article described research findings that suggested that memory may improve in some ways with age. The third article was neutral about memory. In this study, younger people outper-formed older people in each condition, but the age difference was signifi-
cantly reduced in participants who read the positive account of memory. Most important, Hess’s team identified a potential mediator of these performance differences. Participants had been required to write down as many words as they could remember. Those who had read the positive account about memory were more likely to use an effective memory strategy, called semantic clustering, in which similar words are grouped together. Thus, it appeared that strategic efforts were not recruited as skillfully in those participants who were reminded of age deficits.
Indeed, one recent study by Levy and colleagues (2002) suggests that beliefs about aging, as assessed more than 20 years earlier, predicted longevity: people with strong, positive attitudes lived years longer than those with negative attitudes. Of course, it is possible that those with most negative attitudes were already experiencing more serious declines in health, but this finding raises the important alternative that what people believe about the aging process influences biological aging. Much greater attention needs to be given to how personal beliefs about aging influence psychological abilities (such as memory), as well as physical health.
In sum, the degree to which older adults reach their performance potential may be hindered as a function of stereotype threat. Research can identify the conditions under which stereotype threat effects operate in many everyday situations, especially those that involve relations with others who may hold negative stereotyped beliefs about aging (e.g., work settings). Research can also investigate ways to eliminate stereotype threat, such as exploring socially based remediation strategies for aging-related reductions in basic skills in contrast to more traditional cognitive interventions, such as strategy training or providing a more supportive cognitive environment.
INTERVENTIONS TO CHANGE AGEIST ATTITUDES
There are very few studies that examine ways to change negative beliefs about aging. Changing negative stereotypes is difficult because people encode social information in a stereotype-maintaining way: stereotypes reinforce themselves, and thus are resistant to change (Wigboldus, Dijksterhuis, and Van Knippenberg, 2003). However, social psychological research has investigated ways of changing racist and sexist attitudes, and other forms of in-group and out-group behavior, particularly among youth (Aronson and Bridgeman, 1979; Greenfield, Davis, Suzuki, and Boutakidis, 2002; Jackson et al., 2002; Pettigrew, 1998). This literature has identified some important principles that may have relevance to dealing with ageism in the society. More research connecting the two literatures might lead to more effective means of combating ageism. In turn, such research might benefit social psychological knowledge and theory about preventing and
remediating discrimination because of ways that age discrimination may differ from other forms of discrimination.
What has been learned in past research is that contact by itself is generally not enough to change stereotyping and negative attitudes. Yet, those working with ageism often focus on exposing young people to older adults. For example, techniques to change ageist attitudes have included frequent exposure to the elderly, heightening sensitivity to the stereotyping of older people, increasing perspective taking, increasing intergenerational cooperation opportunities, among others (Braithwaite, 2002). However, there is little research that examines the effectiveness of such techniques. Changing communication patterns for younger and older people has been generally ineffective because of the very significant differences in communication style and the content of communication between these groups (Gould and Shaleen, 1999; Harwood et al., 1995; Kemper et al., 1995; Williams, 1996). In fact, these differences result in even more negative perceptions and lessened interest in interaction (Giles et al., 1994; Giles and Coupland, 1991).
Work on challenging automatic biases may prove to be a productive direction for future research. For example, Dasgupta and Greenwald (2001) and Duval et al. (2000) have been more successful in changing negative stereotypes about the elderly by using older exemplars who are associated with fewer automatic biases. However, the mechanisms by which exemplars of admired older adults influence negative stereotypes associated with aging have not been identified. Overall, research on changing negative stereotypes about older age has only begun to illuminate the problem. Research can examine the degree to which paradigms used to change racist and sexist beliefs may apply to ageism and negative stereotypes. Similarly, research can better clarify whether and how greater contact between older and younger people can lead to fewer negative stereotypes.
A vast literature documents both positive and negative stereotyping about older people, but little is known about the effects of these stereotypes on their behavior, self-concept, and motivation. Unlike stereotypes that have no basis in fact, aging stereotypes tend to hold a kernel of truth. Yet, since negative stereotypes can undermine optimal performance, it is particularly important to consider the social context that moderates age-related differences in cognitive functioning and others’ perception of the elderly. It is important to determine the extent to which the observed declines in older adults’ cognitive functioning reflects a cognitive deficiency and the extent to which cognitive impairment is influenced by mechanisms associated with a negative social context.
From a practical perspective, research on stereotypes also suggests that healthy older adults are capable of more effective cognitive functioning when operating within a facilitative social context. Thus, another important topic is identifying those conditions in which positive stereotypes serve a beneficial and protective function that may enhance both the performance of older adults and others’ perceptions of the elderly. The complicated roles played by gender; socioeconomic status; and racial, ethnic, and cultural identities in the development, internalization, and rejection of ageist stereotypes also require further investigation. The overall challenge is to develop a more nuanced understanding of the mechanisms underlying both explicit and implicit stereotypes and their effects on society and the functioning of older people.