Motivation and Behavioral Change
Getting people to develop and maintain healthier patterns of living involves the issue of motivation: How do you get people to perform and maintain behaviors that are in their own best interests but that can be bothersome or difficult to do, such as eating properly, exercising, moderating bad habits, and following through on doctors’ directives? More broadly, how do you motivate people to strive for goals that are realistic and adaptive as well as to modify those goals in response to new challenges and opportunities (e.g., for older adults, retirement, relocation, loss, illness)? The need for change is constant throughout life, as people seek new opportunities, try to improve their life-styles, enter new relationships, and control undesirable behaviors. Such behaviors carried out in early and midlife have profound effects on well-being throughout the life span. The committee considered the factors that promote or obstruct attempts at change, as well as the factors that support or interfere with the maintenance of change. At issue are the best ways to motivate efforts to change for the better at all stages of life, and, more specifically, to understand the special needs of older adults in motivating them to change for their own benefit.
Understanding the myriad factors that promote and maintain change is a daunting but crucial enterprise. Even when people say that they want to change, procrastination and acceptance of the status quo are commonplace (Anderson, 2003). More than two-thirds of older patients admitted to emergency rooms, for example, even those with known preexisting medical conditions, do not have advance directives, and a large percentage of those
surveyed say that procrastination is the cause (Llovera et al., 1999). Deferral of action has also been well documented in employer-sponsored savings plans and health insurance plans: default “options” are far more likely to be adopted than alternatives, for example (Choi, Laibson, Madrian, and Metrick, 2003). Thus, a great deal of inertia needs to be overcome before change can occur (Baumeister and Heatherton, 1996).
It is currently unknown whether older people are more or less likely than younger ones to initiate change, but it is clear that aging often entails the need to make changes and that the types of changes older people must consider are particularly pressing. For example, many older adults need to watch their diets not only for aesthetic reasons or general physical health, but because of immediate consequences to cardiovascular functioning, blood sugar regulation, or other health problems; failing to take medications may be imprudent and risky when one is in middle age, but downright disastrous in old age because of disease progression. Likewise, exercise can profoundly reduce the likelihood of falls in older people, the consequences of which are far more likely to result in death than in any other age group (Greenhouse, 1994). In short, poor health practices might interact with age to exacerbate negative health outcomes. Thus, behavior change can be an issue of life or death for older people. Understanding how change can be motivated and the ways in which older and younger people may differ in the initiation and maintenance of change is critically important. Large differences between individuals within groups of older and younger adults must also be investigated because broad variations due to chronological age, gender, or level of ability also influence the initiation and maintenance of behavior change.
Older people might have unique motives for change: for example, they might be especially and uniquely family oriented, and thus, wish to be less of a burden to their families, or they might be motivated to maintain an exercise program in order to retain physical functioning. Or they might be uniquely motivated by a behavior change that would promote global good. For instance, older adults might be willing to make a contribution to the needs of one generation in hopes that their contribution might flow through to other generations. Whether these unique age-specific motivators are sufficient to initiate and maintain change remains to be investigated.
The committee sees an important role for psychology in understanding the best strategies for motivating change. This includes a better scientific understanding of the factors that promote change and the factors that maintain change. In the fields of social and personality psychology, the committee viewed two distinct conceptual approaches as being relevant. First, many psychologists focus on self-regulation, which is concerned with personal efforts to initiate and maintain change. This approach is most concerned with internal sources of change, such as how people choose to
make changes, their sense of self-efficacy for change, the strategies they undertake to maintain change, and the factors that interfere with their abilities to remain changed. Internal sources of change include the implicit beliefs and values that are associated with different cultures, races, and ethnicity, which shape the lens with which one views the world and result in profound effects on behavior. Research is needed not only to understand the neural, cultural, and psychological processes involved in self-regulation, but also to understand how these processes may change during aging.
The second approach focuses on external sources of change, such as how information can be presented to change attitudes or persuade people to change. From this perspective, research examines social influence and the methods that can be used to change people’s behaviors. Research is needed to examine the specific social influence and attitude factors that are relevant to motivating change in older people. The rest of this chapter is divided into two parts, examining the internal and external approaches to self-regulation. Much of the experimental work discussed in this chapter has been carried out in younger or middle-aged people, which is fitting, given the consequences of earlier self-regulation (or its failure) to later life. But the studies must now be carried out in older people, as well.
MOTIVATION AND SELF-REGULATION
Self-regulation is the process by which people control or alter their thoughts, emotions, and behaviors. At its core, self-regulation involves overriding existing habits or contextually triggered impulses and sustaining efforts over time until a specified goal is reached. It involves the capacity to project oneself into the future, form adaptive attitudes, make plans, choose among alternatives, focus attention on pursuit goals, inhibit competing thoughts, and detect discrepancies between one’s current states and goal states (Bandura, 1997, 2001; Baumeister and Heatherton, 1996; Carver and Scheier, 1982; Gollwitzer, Fujita, and Oettingen, 2004; Leventhal, Leventhal, and Contrada, 1998; Mischel, 1996; Muraven and Baumeister, 2000). Although humans have the capacity to delay gratification, control appetites and impulses, and persevere in order to attain goals, people of all ages have difficulty with self-regulation. Indeed, failures of self-regulation are among the most important and perplexing problems facing modern society, and, for reasons discussed above, they may pose especially serious difficulties for older people.
Self-regulation is becoming increasingly well understood in social psychology, with an ever-increasing emphasis on how motivational factors may be important for successful self-regulation. For instance, recent models view self-regulation as a limited resource that can be depleted with situational demands (Gross and Levenson, 1997; Muraven and Baumeister,
2000). In contrast is the theory that repeated acts of self-regulation lead to greater self-regulatory capacity, and this may be especially important among older adults. For example, Park et al. (1999) reported that older adults were more adherent in taking medication than middle-aged adults, in part because of lower levels of environmental demands. The increased regularity and predictability of everyday life for older adults may promote habit and self-regulation with age.
The issue of self-regulation must be addressed at multiple levels of analysis: from neurochemistry, neuroanatomy, brain systems, cognition, and emotion; to such familial and societal contexts as socioeconomic status and including gender, race, culture, and ethnicity. At each level, attention is needed to whatever encourages people to inhibit or exhibit certain behaviors, thoughts, or emotions. Research is needed to understand how neural and psychological processes involved in self-regulation may change with age and influence effective change.
A lack of cross-disciplinary interaction and communication among scientists may be one reason that there is still limited understanding of why and when self-regulation succeeds or fails. Social and developmental psychologists have developed complex theories and models of self-regulation, but these theories often fail to consider underlying neurobiological functions or structures. At the same time, neuroscientists have made tremendous gains in understanding the link between brain and behavior, but too often these findings are not informed by the social and cultural contexts that shape and guide human lives. Thus, research is needed that crosses levels of analysis, from molecular genetics to social psychology and sociology.
Moreover, further attention needs to be given to self-regulation outside of the laboratory, such as attempted changes in people’s lives (Heatherton and Nichols, 1994). There is consistent evidence that people are able to make important changes in their lives, and many people who do change do so without any form of professional assistance (Prochaska, Velicer, Guadagnoli, Rossi, and DiClemente, 1991). Indeed, Stanley Schachter (1982) observed that clinical studies provide an especially pessimistic assessment of change, in that he found that most individuals who had lost substantial amounts of weight and kept it off or who had quit smoking had done so without formal treatment. Similarly, there is evidence that those who curb problematic drinking do so on their own, without formal treatment of any kind (Peele, 1989). Even those with serious drug dependencies, including alcoholism and heroin addiction, are able to change or alter their problematic behaviors (Klingemann, 1991; Sobell, 1991). Still, there is a surprising paucity of information about the factors that predict change in natural environments, and this may be especially relevant as people become older. Schachter (1982) observed that a greater number of attempts to lose
weight or quit smoking predicted greater success, perhaps in part because people develop more successful strategies over time. It is also possible that people develop greater self-regulatory strength through repeated efforts (Muraven and Baumeister, 2000). Researchers must move beyond laboratory and clinical settings to study the way people go about trying to make important changes in their daily lives.
Age Differences in Self-Regulation
Self-regulation is clearly important for people of all ages, although certain aspects of self-regulation may be especially relevant for older adults, such as adherence to medical regimens, following a specified diet to control health problems, and maintaining physical activity (Balkrishnan, 1998; Brown and Park, 2003; Christmas and Andersen, 2000; Diehl, Coyle, and Labouvie-Vief, 1996; Kahana and Kahana, 1975; Schneider, Friend, Whitaker, and Wadhwa, 1991). Social psychologists have long been interested in examining and promoting methods that can be used to encourage people to engage in healthful behaviors (Salovey, Rothman, and Rodin, 1998). This interest is driven both by researchers wanting to test their theories in applied settings outside the laboratory and by a growing awareness in the medical community that people’s personal beliefs and actions can have a profound effect on their physical and mental health. Individuals’ beliefs and attitudes about aging are also related to outcomes, as are perceptions about the controllability of life. In fact, older people’s subjective perceptions of their health status predict mortality better than physicianrated health (Idler et al., 2004; Mossey and Shapiro, 1982).
Given that self-regulation is crucial for successful living across the life span, it is surprising that it has received only modest attention in the research on developmental aging. Developmental changes are also scientifically intriguing because successful self-regulation draws on areas in which there are both age-related gains, such as in emotion regulation, and age-related declines, such as in working memory and attention. Because models developed in social psychology allow for the breakdown of these factors, they are likely to be particularly useful in understanding age differences in self-regulation.
Some evidence warrants optimism regarding self-regulatory capacities among older adults (Blanchard-Fields and Chen, 1996; Hess, 1994). Research has found older adults to be better at emotional control and emotional stability (Gross et al., 1997; Lawton, Kleban, Rajagopal, and Dean, 1992; Thayer, Newman, and McClain, 1994) and that they may also be able to delay gratification longer (e.g., Green, Fry, and Myerson, 1994) than younger adults. Similarly, Diehl, Coyle, and Labouvie-Vief (1996) demonstrated that older adults have greater impulse control than children
and young adults, primarily because of the use of more efficacious coping strategies (see also Labouvie-Vief, Hakim-Larson, DeVoe, and Schoeberlein, 1989). The evidence indicates that self-regulatory capacity increases over the life course, with children being the least capable and older people being the most capable.
In contrast to the evidence just described, research in cognitive neuroscience suggests that some aspects of self-regulation may degrade with aging. Given declines in executive functions that frequently accompany aging, it is important to examine how these declines affect the capacity and motivation for self-regulation. Self-regulation requires a number of executive functions, such as working memory, allocating attention, inhibiting prepotent behavioral responses, and initiating novel strategies. A variety of evidence indicates that various neural circuits in the prefrontal cortex are involved in executive functions and self-regulation (Banfield et al., 2004; Bechara, 2003). There is currently considerable debate about whether different brain structures are involved in different types of executive functions or whether some structures enable all executive functions and direct other structures to perform subsidiary tasks (e.g., Breiter and Rosen, 1999; Cohen, Botvinick, and Carter, 2000; Gehring and Knight, 2000, 2002; Grafman, 1999; Kerns et al., 2004; Posner and Rothbart, 1998; Raichle, 2000; Richeson et al., 2003; Smith and Jonides, 1999). Understanding the coordination and integration of brain structures that support behavioral and mental self-regulation is one important goal of neuroscience research, especially to the extent that there is reduction of prefrontal cortex functioning with aging (Salat et al., 2004; Tisserand and Jolles, 2003). Basic research must be conducted to examine the neural mechanisms that support self-regulation with a particular emphasis on which, if any, of the mechanisms change with age.
Initiating or Maintaining Change
Among young people, the maintenance of change programs appears to be more difficult than initiation. Young people are eager to start exercise programs, quit smoking, or lose weight, but failure rates in maintenance are very high (Baumeister, Heatherton, and Tice, 1994). Indeed, a sense of hope that permeates people’s lives when they decide to make changes is a strong motivator of initial efforts to change, even if it does not itself predict long-term change (Polivy and Herman, 2002). Inherent in this idea is that notion that some degree of negative affect is crucial for initiating change; people do not attempt to fix what they don’t believe is broken. Thus, people attempt to diet when they are dissatisfied with how they look, they quit drinking when it causes problems in their lives, and they save money because they are worried about not having enough when they need it. To the
extent that older adults experience fewer negative emotional states, it may be that they are less likely to spontaneously initiate life change on their own. At the same time, there is a greater likelihood that they will be encouraged by physicians or family members to make healthy changes to prolong their lives. Thus, research is urgently needed to understand more thoroughly the factors that initially prompt change, especially for older people.
Rothman (in this volume) notes that one factor that prevents initiation of change is that many people are hesitant to accept information that calls their basic health practices into question. People often respond to negative health diagnoses by minimizing the seriousness of the condition or emphasizing its prevalence (Ditto and Croyle, 1995). This reaction is especially a concern during the aging process when people might misinterpret their symptoms as resulting from aging rather than from disease processes.
Social and health psychologists have developed a number of methods that encourage efforts to change in people of all ages. For instance, message tailoring involves presenting information that is personally relevant for specific individuals, and this method is generally more successful than generic messages that are aimed at a broad audience (see Rothman, in this volume). The manner in which the message is framed also matters a great deal. It is generally believed that if the goal is to promote the use of detection measures (such as screening for cancer), then messages that emphasize losses have stronger effects than those that emphasize gains. In contrast, the promotion of preventive behaviors (such as eating a healthy diet) is best accomplished through messages that focus on gains. However, there is evidence that with older adults emphasizing gains may be more effective in both cases (Löckenhoff and Carstensen, 2004).
One hypothesis is that with age it becomes harder to make behavioral changes, but once those changes are initiated, older adults find them easier to maintain. There is reason to infer from the existing aging literature that change in older adults may entail special challenges in the area of initiation. Passivity on the part of institutionalized older adults has been widely documented (Baltes, 1995, 1996), perhaps because of reinforcement processes (Baltes and Wahl, 1991). And some memory decline in normal aging directly affects self-initiated cognitive strategies (Cabeza et al., 2004; Logan et al., 2002). For example, in healthy older adults, self-initiated memory strategies show reliable deficits with age, while strategies that are externally primed show no age decline (Einstein et al., 1995; Wingfield and Kahana, 2002). It is also possible that older adults see fewer opportunities for change or believe that their efforts to change will have less positive outcomes (Leventhal, Leventhal, and Contrada, 1998).
Yet older people are able to change their behavior. Medication adherence is better among older than middle-aged adults, attributed in part to
more stable daily routines (Park et al., 1999) and psychotherapy efficacy is as good with adults between the ages of 61 and 90 as with their younger counterparts ages 21 to 59 (e.g., Reynolds et al., 1996). The difficulty of initiating change, as well as the ease of maintenance, may be related to the stability of contextual cues in late adulthood. There is sound evidence that older adults lead more routine and less varied lives than young adults (Martin and Park, 2003). This stability may make initiating change difficult; at the same time, the presence of stable environmental cues may help maintain behavior. Thus, context plays an important role in maintaining behavior.
These findings are in agreement with research demonstrating that imagining a context in which a future behavior is to be performed greatly enhances the probability that the behavior will be completed (Gollwitzer and Schaal, 2001; Gollwitzer et al., 2004). Chasteen, Park, and Schwarz (2001) reported that older adults were much more likely to remember to perform a simple action of writing the data on a paper at the end of a laboratory study if they imagined completing the action, in comparison with simply repeating the planned action. In a recent study, Liu and Park (2004) demonstrated that older adults were much more likely to maintain use of a glucose monitor over a 4-week period if they spent only a few minutes imagining the context in which they would perform the action at the time they received the monitor, in comparison with people who rehearsed trying to remember performing blood glucose monitoring. These researchers hypothesize that it is the stability of older adults’ lives and the regularity in which they encounter the same contextual cues (e.g., they eat meals at the same time every day) that resulted in the maintenance of complex behavior for a month, based on only a few minutes of rehearsal. Further investigation of mechanisms for initiating and maintaining changes in health behavior is needed, with an emphasis on the role of contextual stability. The findings with older adults will also provide valuable strategies for younger adults who are trying to maintain change and also have the potential to inform the understanding of self-regulation at a very broad level.
The Avoidance of Novelty
Generally speaking, change involves novelty. It involves learning new information, modifying routines, and very often making new social contacts, whether change agents or simply new people. There is good evidence that older people have less motivation to explore new ways of living. They tend to be lower in the personality trait of sensation seeking (Lawton et al., 1992) and, as shown in cross-cultural studies, also less open to new experiences (McCrae et al., 2000). Instead, they tend to focus attention away
from novelty toward familiar, predictable aspects of life. One prominent conceptual model in life-span psychology noted above, selective optimization with compensation, postulates explicitly that a narrowing of breadth coupled with heightened effort and expertise in selected areas of life reflects successful adaptation to old age (Baltes and Baltes, 1990).
For somewhat different reasons, socioemotional selectivity theory makes similar predictions. According to selectivity theory, perceived constraints on time result in motivational changes that favor goals related to regulating emotional states over goals associated with gaining knowledge or otherwise expanding one’s horizons. In studies that use social choice paradigms, for example, older participants are reliably more likely to opt for well-known, emotionally significant partners over novel social partners (Fredrickson and Carstensen, 1990; Fung and Carstensen, 2004; Fung, Carstensen, and Lutz, 1999; Fung, Lai, and Ng, 2001). Even basic cognitive processes, like memory and attention, appear to be affected by motivational changes. Empirical tests of selectivity theory demonstrate that memory for advertisements is better when slogans promise emotional rewards than when they promise informational rewards (Fung and Carstensen, 2003). Thus, there is considerable evidence that, with age, people grow more interested in emotional satisfaction and less interested in seeking novelty. This resistance to trying new things has important implications for initiating change, and it is an important priority for research to illuminate how the motivation to avoid change influences self-regulation, as well as basic cognitive and emotional processes.
Emotional Processes and Self-Regulation
It has been widely noted that negative emotional states lead to relapse for a number of addictive behaviors, such as alcoholism and smoking. Distress increases craving for alcohol among those trying to control alcohol intake (Litt et al., 1990), which, in turn, leads to drinking (Hull and Young, 1983; Miller et al., 1974). Similarly, the single most important trigger for smoking urges is a negative emotional state (Marlatt, 1985; Shiffman, 1982): people smoke in order to control stress (Kassel, Stroud, and Paronis, 2003). Emotional distress has also been identified as a major determinant of diet failure and binge eating (Greeno and Wing, 1994; Heatherton and Baumeister, 1996). According to this research, when a person’s emotional state conveys negative implications about the self, people are especially motivated to shut out painful self-awareness, either by external means (such as consuming alcohol) or by restricting attentional focus to potent stimuli through cognitive narrowing. The resulting mental state may indeed be less distressing, but it is also likely to lead to disinhibition, and the long-term consequences of these escapist strategies might exacerbate future dis-
tress (see Heatherton and Baumeister, 1996). Thus, it is people who feel personally deficient or who experience threats to the self who are most likely to break their diets, spend excessively, or binge drink.
The literature on self-concept and aging suggests that despite objective threats to self, from physical disabilities to ageism, older people are surprisingly satisfied with their self-views (Greve and Wentura, 2003). Discrepancies between ideal and actual selves, for example, are smaller than those observed in younger people, when groups with mean ages of 19.3, 46.0, and 73.4 are compared (Ryff, 1991). Studies of well-being also suggest that aging is associated with less interest in personal growth (Ryff, 1995). In other words, aging is associated with greater satisfaction with the status quo and less interest in improving the self. One experience sampling study also revealed less variability in self-descriptions in everyday life (Charles and Pasupathi, 2003). Thus, because negative affect is an important trigger of self-regulation failure, it may be that older adults succeed at self-regulation more often than younger adults because they do not experience as much dissatisfaction. This hypothesis further reinforces the idea that although initiating change may be more difficult for older adults, maintaining change may be easier.
It is also possible that the relation between negative emotions and self-regulation is due to inadequate coping and the possibility that people use alcohol, drugs, or food as a coping strategy. Accordingly, it may be that older adults are less prone to self-regulation failure because they use strategies that minimize negative affect in their daily lives (Mather, in this volume). It is unclear whether putative self-regulatory differences observed as a function of age are due to changes in cognitive skills or capacities or to changes in the degree of experienced negative affect. Considered together, it is apparent that the more positive emotional experiences of older adults may discourage initial change but serve to enhance any changes that do occur. Research should examine the general role of affect in self-regulation, with an emphasis on how emotional changes that occur during aging may be associated with initiation or maintenance. Moreover, additional research should examine how emotional processes are involved in change that adds new behaviors (such as exercise) as opposed to change that eliminates them (such as poor health habits).
Individual Beliefs and Attitudes
A vast literature demonstrates the role that efficacious beliefs play in instigating and maintaining change in young people (Bandura, 2001), and there is additional evidence that they play a similar role in older adults. Self-efficacy and personal beliefs have long been known to play critical roles in self-regulation (Bandura, 1989, 1997; Mischel, Cantor, and Feldman, 1996;
Mischel, Shoda, and Rodriguez, 1989) and research demonstrates that feelings of control over the environment are important for self-motivation across the life span (Brandtstädter and Rothermund, 1994; Seeman, McAvay, Merrill, Albert, and Rodin, 1996). Racial, cultural, and ethnic group membership confer an identity that is part of the conception of the self and connectedness of the self to others that directs choices and decision behaviors. Such group membership plays a crucial role in the development of self-efficacy and personal beliefs, especially about control over the environment. A comparison among different groups could highlight the mechanisms of most importance and suggest interventions for members of all groups. Studies of individual variability within these groups could provide further information for designing the most useful interventions.
There is considerable evidence that self-efficacy beliefs are related to motivation and health. In terms of the former, self-efficacy beliefs are associated with the selection of goals, effort, and persistence in the face of frustration, emotion (e.g., anxiety and depression), and coping with stress and disappointment. In relation to the latter, self-efficacy beliefs are associated with healthy behaviors, enhanced immunological system functioning, and the avoidance of relapse (Bandura, 1997). Self-efficacy is especially important for encouraging patient compliance with medical regimens (Schneider et al., 1991). Tying the two together, it is clear that motivating people to change involves helping them to set specific health-related goals and influencing their self-efficacy beliefs in relation to these goals and other life tasks. But far more needs to be known about the mechanisms involved, especially in specific individuals, and how to activate them.
Social Facilitation and Barriers to Change
A variety of evidence suggests that social support is an important component of successful life change (Clifford, Tan, and Gorsuch, 1991). For instance, social support has been found to be valuable for achieving and maintaining weight loss in younger and middle-aged people (Perri et al., 1988). Social support may enhance change because it contributes to self-efficacy (Major et al., 1990) and increases general well-being, and it also may buffer against the strain caused by high-stress life events (Cohen and Wills, 1985; Gentry and Kobasa, 1984). In the context of racial, cultural, and ethnic identity, it is also important to recognize the shielding mechanism and social support that may be provided by membership in a specific racial or ethnic group. As discussed above, relapse is commonly associated with perceived emotional distress (Brownell, Marlatt, Lichtenstein, and Wilson, 1986), but significant others may be able to assuage negative emotional experiences by helping people cope with high-risk situations (Marlatt, 1985). Family, friends, coworkers, and health care professionals can pro-
vide the emotional and esteem support, feedback, information, reinforcement, and direct assistance that a person involved in change frequently needs (Clifford et al., 1991; Marlatt, 1985). Moreover, making public commitments and having a “buddy” who shares attempted change might also be a useful strategy for change. The social networks of older people typically involve a small circle of close relatives and friends, suggesting that those in their social networks might be potent motivators of change (Lang and Carstensen, 1994).
However, some research has also shown that social support may interfere with motivation or behavioral change (Kelly, Zyzanski, and Alemagno, 1991). This finding may reflect the fact that people tend to associate with others who have similar ideas, personalities, and backgrounds (Caspi and Herbener, 1990), which would tend to promote stability. But when people change, their ties with other people change as well. It is possible that others sometimes will actively hinder change because they feel threatened by the implications of potential changes. For example, spouses who wish to continue smoking may be unlikely to offer support and encouragement to a partner who is contemplating quitting smoking because such advice would be dissonant with their own behavior. Social support is likely to be effective only when the supporter already holds the attitudes and identity that the person is trying to adopt. Again, because the social networks of older people are small and restricted to significant others, it is likely that any negative social effects will be larger for older than for younger adults. It is crucial that researchers examine the role of other people in efforts to initiate and maintain change.
In general terms, it might be that older adults are less likely to be influenced by social forces. Most theories of social influence do not consider adult development. Theoretical and empirical work in life-span developmental psychology, however, suggests that age may reduce susceptibility to social influence. Under certain conditions and relative to their younger counterparts ages 18-35, people between the ages 63 and 85 display lower rates of social conformity (Pasupathi, 1999).
PERSUASION AND ATTITUDE CHANGE
In addition to self-regulation processes, social psychologists have developed a rich literature on the psychological processes underlying attitude formation and change. In fact, the topic of attitudes and persuasion is one of the most well developed in social psychology, with numerous established theories and findings (for reviews, see Eagly and Chaiken, 1993; Petty and Wegener, 1998). Attitudes refer to individual’s evaluations of people (including themselves), objects (e.g., one’s medication), issues (e.g., changing Social Security), and actions (e.g., moving to a retirement home). Attitudes
can be based on a person’s values, specific beliefs, emotions, and behaviors (both actual and anticipated). As noted above, forming attitudes is typically necessary before decisions can be made or behaviors enacted. Thus, if older people are going to change their diets or exercise more, they first need to develop favorable attitudes toward these behaviors and toward making these changes.
Indeed, a major premise of research in social psychology is that attitudes are important because they guide people’s decisions and actions (e.g., Fishbein and Ajzen, 1975; Sanbonmatsu and Fazio, 1990). Stated simply, people will choose an option they like over one they dislike. When two options are both evaluated positively (or negatively), the option evaluated more positively (or less negatively) typically will be selected. Thus, if a person likes Coke more than Pepsi, Coke is more likely to be purchased. If two options are equivalent in their desirability, then the option toward which one holds the more confident or accessible attitude will be chosen (Petty and Krosnick, 1995). The attitudes of others are also important in influencing actions. For example, if people believe that you are forgetful, they will tend to act in a manner toward you that is consistent with this belief, and this action may in turn elicit the forgetful behavior that is expected (see Snyder, 1992). Thus, both people’s own attitudes and the attitudes that others have about them can influence behavior.
Because of the critical role that attitudes have in guiding behavior, understanding how evaluations are formed and changed and whether these judgments are strong or weak (e.g., how confident or accessible they are) is an important undertaking. Considerable research has examined attitude change processes in younger adults and much can be learned from this work. In contrast, relatively little effort has been aimed at understanding how older adults form or change their attitudes.
There are several areas of research that seem ripe for exploration with respect to older people. One area focuses on the underlying mechanisms by which attitudes are formed or changed and what techniques are most effective in persuading older people. These may or may not be dependent on cultural values and racial or minority group membership. Understanding of persuasion can be useful from the standpoint of producing desirable changes (e.g., a physician persuading older patients to take their medications) and protecting people from undesirable agents of influence (e.g., avoiding telephone scams). A second area concerns the unique role that emotional factors might play in the choices that older adults make. This is of particular interest because of evidence that emotional regulation processes become better and more important with age. A third area involves investigating the factors in older people that produce an initial change in possible contrast with those that are responsible for maintaining a change once it has oc-
curred. A fourth area of research concerns the distinction between implicit and explicit attitudes (e.g., Greenwald and Banaji, 1995). Explicit attitudes are those evaluations that a person consciously endorses; implicit attitudes are evaluative tendencies that may be automatically activated without a person’s awareness (Fazio and Olson, 2003). Although much attention has been paid to how to change explicit attitudes, researchers are only beginning to examine how to modify implicit attitudes, and very little work in this area has been undertaken with older adults.
Attitude Change Processes: Thoughtful or Automatic
Researchers are just beginning to consider the unique aspects of communicating with and modifying the attitudes and behaviors of older people (e.g., Spotts and Schewe, 1994). Factors that affect their compliance with medical regimens have received particular attention (e.g., Brown and Park, 2003), with some clear findings, especially with respect to such factors as the cost of medication, insurance coverage, complexity of the medical regimen, and certain demographic variables (e.g., Balkrishnan, 1998; Salzman, 1995). Far less attention has been paid to persuasion processes in older people and how they might differ from those of younger adults.
One reason persuasion processes might differ in older and younger adults relates to the effects of aging on the frontal regions of the brain. For example, a reduced working memory capacity may lead older adults to seek less information when making a decision, and older people ages 65-75 may thus form attitudes based on less information and with less thought than adults ages 28-55 (e.g., Streufert et al., 1990). Much research in social psychology suggests that the underlying processes of attitude formation and change can be placed along a continuum, ranging from extensive thought about the merits of objects to evaluations based on relatively simple (even automatic) processes that require little thinking (Petty and Cacioppo, 1986). Specifically, when people are not thinking very much, they rely on their intuitions, gut feelings, and simple heuristics to form attitudes and make choices (e.g., if an expert said it, it must be true), but when people are thinking carefully, their idiosyncratic thoughts in response to the message and their thoughts about their thoughts (e.g., was my thought valid?) become important determinants of attitudes (for reviews, see Chen and Chaiken, 1999; Petty and Wegener, 1999).
Many individual and situational variables have been identified that determine where a person is along the thinking continuum and thus what type of influence technique might be most effective. For example, people engage in greater thinking when the message is perceived as personally relevant or when the person feels accountable for a decision. In terms of
personality, some people are prone to rely on deliberative processing (Cacioppo and Petty, 1982) while others tend to be more reliant on heuristics and intuitions (see Epstein, 2003).
The thinking continuum is particularly interesting to examine in light of aging because, as noted above, people tend to rely less on deliberative forms of thinking and more on intuitive modes as they grow older. It is important to note that extensive thinking does not necessarily produce better decisions than does relying on gut feelings or intuition. This is because a person’s extensive thinking can be biased in various ways, and intuition can be a proxy for a person’s expertise. Nevertheless, if older adults are less motivated or less able to seek and process information than younger adults, attitude modification processes that rely more on heuristics and intuitions might be more effective than strategies that rely on high amounts of information processing and thinking. Thus, older people may be more likely to base evaluations on the first information presented (primacy effect), or on a smaller number of information dimensions, or on simple inferences, shortcuts, and associations (e.g., if it makes me feel good, I like it; Schwarz and Clore, 1983). The strategies that are most effective in modifying the attitudes of older adults should be investigated.
One reason the amount of thinking behind one’s attitudes is important is because thoughtful attitudes tend to be more persistent over time, more resistant to change, and more influential on one’s behavior. Also, people tend to have more confidence in judgments that are based on large amounts of information, and judgments based on considerable thinking tend to be more accessible (see Petty and Krosnick, 1995). Thus, if the newly developed attitudes of older adults are based on less information processing, these attitudes may be held with little confidence, or one’s new attitudes may be low in their accessibility, causing decision making to be more stressful (see Fazio, 1995). If decision making is somewhat stressful because of holding attitudes with low confidence, decision making will be avoided. The committee believes that this is an important topic for exploration.
How to maximize thoughtful decision making and how to tailor or frame information to be maximally effective for older adults are among the many other research issues that could be examined with respect to persuasion and the concerns of the older population (see Rothman, in this volume). From the perspective of contemporary models of persuasion, there are several mechanisms by which framing or tailoring might work. For example, targeting a message for older people (rather than a younger audience) might be effective because tailoring makes the message seem intuitively appealing (e.g., “the message resonates with me, so it must be good”) or because tailored messages get people to think carefully about the information because of the heightened perceived self-relevance. Which of these mechanisms is operating is of importance because the latter process is likely
to lead to stronger attitudes than is the former. Also of interest is the possibility that for some older message recipients, tailoring may not have the expected effects. For example, if some older adults hold negative implicit or explicit attitudes toward old people generally (Richeson and Shelton, in this volume), the normal positive effects of tailoring would not be expected. This, too, should be explored.
Emotional Factors in Attitudes and Decision Making
Older adults show increased focus on emotional goals, such as avoiding regret and maximizing satisfaction (Mather, in this volume). Considerable research in social psychology has focused on how emotions, in comparison with cognition, affect individuals’ attitudes and ultimately their actions. Common models of attitude formation assume that people consider the cognitive (utilitarian) and affective (emotional) consequences of adopting an advocacy position or making a choice (e.g., Abelson et al., 1982). Although some attitudes are based largely on cognition, others are based largely on affect, and still others are based on both (Crites, Fabrigar, and Petty, 1994; Eagly, Mladinic, and Otto, 1994).
As noted in Chapter 4, psychological models hold that two aspects of beliefs are of particular importance for people—the likelihood of the consequences considered and the desirability of those consequences (e.g., Fishbein and Ajzen, 1975). Thus, if a person is deciding whether to move into a new retirement community, she might consider whether the move would be too expensive and whether the move would make her happy. For each of these perceived consequences, the person would then consider how likely it was that the move would be expensive and how likely the move would be to produce happiness. The person would also consider how desirable it would be to spend the anticipated amount of money and how desirable it would be to attain happiness from the move. Furthermore, social psychological research suggests that being in an emotional state can influence the perceived likelihood and desirability of consequences: for example, being in a happy state can make positive consequences seem even more likely than when not happy (DeSteno et al., 2000; Lerner and Keltner, 2001). These likelihood and desirability “forecasts” (Gilbert and Wilson, 2000) about the cognitive and affective consequences of a considered action are important because they can determine the person’s attitudes, which would in turn influence the particular choices made and behaviors implemented.
If one wants to modify people’s attitudes in a thoughtful way, persuaders can introduce new consequences that people have not considered previously, or they can try to influence the perceived likelihood or desirability of consequences that are already known. Thus, if a person thinks that it is somewhat likely that moving will produce a little happiness, a persuader
can attempt to convince a person that moving is very likely to produce a lot of happiness. To the extent that the perceived likelihood or desirability of the consequences change, the overall attitude toward the move will also change.
Within this framework, research might focus on whether older decision makers weigh emotional consequences more than cognitive ones in attitude formation and decision making. In addition, the role that emotional factors or anticipated emotions have in influencing likelihood and desirability forecasts for older individuals could be investigated. Other topics of interest would include understanding why older adults show greater susceptibility to some forms of persuasion (e.g., phone scams). Is this due to cognitive deficits (e.g., inability to counterargue a sales pitch) or enhanced weighting of emotional benefits (e.g., enjoyment of talking to someone)?
As noted above, research suggests that emotion regulation improves with age (Gross et al., 1997). One strategy people use to maintain their positive moods is to avoid negative information (Wegener and Petty, 1994), including avoidance of processing of messages with negative overtones or consequences (Wegener, Petty, and Smith, 1995). Thus, it could be that older adults are more likely than younger ones to avoid threatening or fearful messages as a mood regulation strategy. Because of emotional regulation and lack of ready attitudes (or attitudes lacking in confidence), they may avoid decision making in order to avoid stress. The notion of avoiding distress also suggests that older adults may be more susceptible to the effects of cognitive dissonance (Festinger, 1957) than younger adults. For example, research has shown that decision making can produce dissonance (unpleasant tension) when people must accept some undesirable features of a chosen alternative and forgo some attractive features of a rejected one (Brehm, 1956). As a result, people tend to exaggerate the positive aspects of the chosen alternative and minimize the positive features of the rejected alternative after making their choice. To the extent that older adults have a greater aversion to negative states, they could be more susceptible to this dissonance effect (see Mather, in this volume).
Initiation and Maintenance of Change
Just as research on self-regulation suggests that the processes involved in the initiation of change may differ from those involved in the maintenance of change, so too does the literature on externally initiated change. In particular, as noted above, when people are considering change, they are largely influenced by their expectations, both emotional and cognitive, regarding the outcomes of changing (e.g., Will I be happier? Will it be easier to walk to the store if I move?). However, once change has occurred, research suggests that people are influenced by their assessment of their
post-change experiences. Research on post-change processes is especially prominent in the work on consumer behavior. Researchers have studied expectancies about product performance as a determinant of initially select-ing a brand, but post-trial satisfaction with the product as a determinant of whether the brand will continue to be purchased (e.g., Oliver, 1993). That is, once a person has purchased a new product or engaged in some new behavior (e.g., taking high blood pressure medicine), the experience can either match, exceed, or fall short of one’s expectations. The higher one’s initial expectations for a behavior change, the more likely one is to initiate the change because more positive attitudes will be formed when expectancies of desirable outcomes are high. However, the higher one’s positive expectancies for a change, the less likely it is that the experience of the change will actually match the expectancies. The disappointment following failure to meet expectations can lead to behavioral termination (e.g., Westbrook, 1987). One interesting possibility is that if older people are more focused on positive rather than negative emotions (Ybarra and Park, 2002), any disappointments following behavior change are less likely to undermine the new behavior than they are for younger people.
In addition to post-change dissatisfaction, other factors may work against maintenance of change, and these factors with their implications for change in later life are only beginning to receive serious research attention. For example, once a new attitude is formed, the old attitude and the associated behaviors do not just disappear. Rather, long-standing behaviors can re-emerge, and old attitudes can still be potent when people are not thinking about how to behave (e.g., Ouellette and Wood, 1998; Petty, Baker, and Gleicher, 1991; Wilson, Lindsey, and Schooler, 2002), or if antithetical cultural practices engulf the person. Research has yet to explore the consequences of conflict between old and new attitudes and behaviors and how it might be possible to both decrease the effects of old attitudes and associated behaviors as well as reinforce new ones at all stages of life. Since the consequences of attitudes and behaviors formed in early and midlife are so profound in later life, this and related research topics are on the committee’s suggested research agenda.
Changing Implicit and Explicit Attitudes
As noted above, social psychologists have recently begun to explore the possibility that individuals may hold not only explicit attitudes, those that can be reported on direct self-assessments, but also implicit attitudes, automatic evaluative tendencies of which people may not be aware and that can sometimes conflict with the explicit attitude (see Wilson et al., 2002). Implicit attitudes can stem from a variety of sources, such as media portrayals, and they tend to govern behavior when people are not deliberately consid-
ering their actions (Dovidio et al., 1997). Social psychologists have developed various procedures to assess people’s automatic attitudes, such as the priming measure (Fazio et al., 1995) and the implicit association test (Greenwald, McGhee, and Schwartz, 1998).
Although decades of research studies have examined how to change people’s more conscious explicit attitudes, attention has only recently begun to turn to modifying implicit attitudes. Yet, such understanding is of potentially great importance for older people to the extent that automatic processes do not decline with aging as much as deliberative processes. That is, if older adults are more prone to rely on automatically activated evaluations rather than deliberately considered ones, understanding how to produce desirable automatic attitudes in older people can reap considerable benefits.
The need for change is constant throughout life, as people strive to live healthful, productive lives. Aging often entails the need to make significant life-style modifications, from taking new medications to relocation to developing new social networks. Unlike younger adults, who may not experience any negative effects from their life-styles for many years, the consequences for older adults of failing to commence or sustain health behaviors are often immediate and potentially life threatening.
The literature on self-regulation among older adults is sparse, focusing primarily on emotional regulation or adherence to specific medical directives. Interestingly, the literature suggests that older adults may be less likely to initiate behavioral changes, but more likely to maintain any changes that do occur. A major factor that prompts efforts to change, but also sabotages those efforts, is emotional distress. Given evidence that older adults are generally satisfied with their lives and that they avoid negative information, it is possible that the factors that prompt and support change are very different for older adults than for younger ones. However, it is as yet unclear whether self-regulatory differences observed as a function of age are due to changes in cognitive skills, functional neurological capacities, degree of experienced negative affect, or some other factors. Understanding self-regulation among older people is not only important for bettering the lives of older adults, but also provides a unique opportunity for psychologists to examine theoretical models.
People live longer in part because of advances in medicine, but also because they have quit smoking, watched their diets, and generally are motivated to look after themselves. Many of the most common causes of mortality are related to behaviors that people should be doing more often (e.g., eating healthfully, exercising) or avoiding (e.g., excess alcohol con-
sumption, smoking). Motivating change often entails the communication of persuasive messages to change people’s attitudes about engaging in these behaviors. Very little is known about persuasion processes in older people—and how they might differ from those of younger adults—or about the roles of racial, cultural, or ethnic preferences in those processes over the life course. Given that older adults are motivated to avoid processing negative information and perhaps are more likely to use heuristic processing than younger adults, it is possible that framing or tailoring messages to older audiences might be an especially efficacious means of encouraging long-term change. It is apparent that research that examines the role of socioemotional processes in self-regulation and persuasion holds great promise for developing methods to motivate older adults to make needed changes in their lives.