| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
a Department of Epidemiology and Public Health, Yale University School of Medicine, New Haven, Connecticut.
b Beth Israel Deaconess Medical Center and Harvard Medical School, Boston, Massachusetts
c Harvard Graduate School of Education, Cambridge, Massachusetts
Becca R. Levy, Department of Epidemiology and Public Health, Yale University School of Medicine, 60 College St., P.O. Box 208034, New Haven, CT 06520-8034 E-mail: becca.levy{at}yale.edu.
Decision Editor: Toni C. Antonucci, PhD
 |
Abstract |
|---|
 TOP Abstract The Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
THE majority of individuals older than 65 have some form of cardiovascular disease; it is the leading cause of disability and death within this age group (Wenger 1997
). A growing body of research suggests that cardiovascular response to stress contributes to the development of cardiovascular disease (Barnett, Spence, Manuck, and Jennings 1997; Gullette, et al. 1997; Manuck, Adams, McCaffrey and Kaplan 1997). In the current study, we examined a previously unexplored social psychological mechanism that may contribute to how older individuals experience stress. Specifically, the negative stereotypes of aging that exist in our society may increase cardiovascular stress in this population.
|
The Power of Age Stereotypes |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
; Palmore 1998; Perdue and Gurtman 1990). Yet, it is the negative stereotypes that predominate (Butler 1980; Kite and Johnson 1988; Stone and Stone 1997). These stereotypes tend to marginalize perceptions of elderly people within our younger-adult-centered society by shifting them in opposite directions along the lifespan continuum. One set of negative age stereotypes categorizes elderly people as a group that has regressed to childhood. Older individuals encounter the outcomes of these stereotypes in a variety of situations. For example, younger adults sometimes address older people in a form of patronizing speech that resembles baby talk (Caporeal 1981; Williams and Giles 1998). A restaurant menu one of the authors recently encountered included a section for "Patrons under 12 and over 61."
Another set of marginalizing stereotypes depicts old age as a time of debilitation and imminent death. Older individuals encounter the outcomes of these stereotypes when they are among the first to be laid off and among the last to be rehired or when physicians discourage them from receiving life-prolonging medical procedures (Finkelstein, Burke, and Raju 1995; Guigliano et al. 1998).
To date, research on gender, racial, and age stereotypes has focused on how individuals use stereotypes to judge others (Banaji and Hardin 1996; Palmore 1998). Considerably less is known about how stereotypes influence individuals when the stereotypes concern themselves, particularly in the case of aging self-stereotypes, or older individuals' internalized beliefs about elderly people.
In a series of studies, we demonstrated the effects of the automatic activation of negative and positive aging self-stereotypes, designed to resemble those that arise in everyday life. We found that the negative self-stereotypes of aging reduced memory performance, self-efficacy, and the will to live of older individuals, whereas the positive self-stereotypes of aging improved these factors (Levy 1996; Levy, Ashman, and Dror in press). In contrast, these same stereotypes exerted no effect on the young participants, for whom the stereotypes were not personally relevant (Levy 1996; Levy et al. in press; Levy and Langer 1994).
|
Evidence That Response to Stress Contributes to Illness |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
; McEwen 1998). The autonomic nervous system can be adaptive in mobilizing an individual to respond to an acute stressor. But when an individual perceives chronic stressors, so that the autonomic system is frequently activated and does not get a chance to sufficiently relax, a maladaptive process can occur. Chronic activation of the autonomic nervous system may lead to the body's overreacting and producing too many stress hormones. Over time, this process may cause the body to damage itself (McEwen 1998). In the current study, we examined whether one such chronic stressor could be negative aging stereotypes. We focused on cardiovascular response to stress. Thus, we included heart rate, systolic blood pressure, and diastolic blood pressure as the primary outcomes. We also included a measure of electrodermal activity to determine whether age stereotypes influence other physiological measures that are influenced by the autonomic nervous system.
Most of the studies that have found a relationship between the autonomic response to stress and illness have focused on the cardiovascular system. These studies have found support for the hypothesis that repeated elevations of heart rate and blood pressure in response to stressors can cause and accelerate heart disease (Barnett et al. 1997; Uchino, Cacioppo, and Keicolt-Glaser 1996).
The evidence that exaggerated response to stress, or reactivity, can contribute to illness comes from two lines of research. The first series of studies suggests that reactivity can influence the development of cardiovascular disease in healthy animals and humans. For example, several studies with monkeys have demonstrated that response to psychosocial stress may contribute to atherosclerosis. Researchers found that cardiovascular reactivity measured during a capture by humans predicted the presence of coronary atherosclerosis 2 years later (Manuck et al. 1997). In a complementary study, monkeys that frequently had to defend their dominant social status from other monkeys in an unstable environment were at increased risk of atherosclerosis (Kaplan and Manuck 1997).
Studies with human participants also suggest that response to stress predicts development of cardiovascular disease. For instance, one study of healthy individuals found that their blood pressure response during a frustrating cognitive task predicted development of atherosclerosis measured 2 years later (Barnett et al. 1997). Another study of healthy individuals found that those who experienced a greater blood pressure response to a mathematical task were more likely to develop hypertension 5 years later (Falkner, Onesti, and Hamstra 1981). In a third study, young men with higher heart rate responses to stress had higher heart rate and blood pressure levels than peers with a lower heart rate response 10–15 years later (Light, Dolan, Davis, and Sherwood 1992).
The second line of research supporting the reactivity hypothesis comes from studies demonstrating that stress can trigger cardiovascular events in individuals with cardiovascular disease. For example, studies show that when individuals with coronary artery disease experience mental stress, both in laboratories and in daily life, it can lead to myocardial ischemia, defined as reduced blood supply to the heart (Gullette et al. 1997). Mental-stress-induced myocardial ischemia is associated with higher rates of nonfatal and fatal cardiac events (Jiang et al. 1996).
|
Present Study |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
The subliminal presentation of stereotypes may have several advantages over a more explicit presentation. Participants are less likely to give socially desirable responses if they are unaware of the nature of the primes. Also, priming is designed to mimic the way stereotypes are activated in everyday life. That is, stereotypes are frequently activated without awareness (Banaji, Blair, and Glaser 1997; Bargh and Chartand 1999).
In summary, we predicted that those exposed to the negative age stereotypes would demonstrate a heightened response in the cardiovascular measures (systolic blood pressure, diastolic blood pressure, and heart rate) as well as skin conductance, compared with those exposed to the positive age stereotypes. Specifically, we hypothesized that:
|
Methods |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
). To more closely resemble the repeated exposure to stereotypes and stressors that tend to occur over a person's lifetime, the protocol consisted of two sets of positive or negative aging stereotype primes and two sets of mathematical and verbal challenges. Thus, our primary analyses examined how the physiological indices of those in the two aging stereotype groups responded to the total intervention, which was measured by change from baseline to following the two sets of primes and two sets of mathematical and verbal challenges.
Participants
The 54 participants (aged 62–82 years) consisted of 29 women and 25 men, living independently in the Boston area, who tended to describe their health as very good. We recruited individuals who (a) were able to read and write; (b) were native English speakers; (c) had no major psychiatric illness; and (d) were older than 60.
In addition, participants were removed from the sample if they were able to identify any of the prime words. Three participants who accurately identified several of the prime words were excluded from the analysis. These participants did not differ from the included participants in self-reported health, age, or years of education.
All the participants and one of the two experimenters were blind to the priming conditions. The two experimenters tested equal numbers of individuals in both priming groups. Participants were randomly assigned to experimenters. As described at the end of the Methods section, the findings of the participants tested by the two experimenters were not significantly different.
Members of the two groups did not significantly differ on any of the cardiovascular risk factors at baseline (Table 1 ), including (a) demographic features, such as age and educational level; (b) psychosocial factors, such as self-rated health, level of social support as measured by the Inventory of Socially Supportive Behaviors (Barrera and Ainlay 1983), depression as measured by the Geriatric Depression Scale (Garrard et al. 1998), and anxiety level as measured by the short form of the Spielberg State-Trait Anxiety Inventory (Marteau and Bekker 1992); and (c) behavioral factors, such as current smoking, the mean number of alcoholic drinks consumed in a week, and scores on a Type A behavior scale (Bortner 1969). In addition, the two groups did not differ in number of individuals reporting a history of various cardiovascular illnesses (Table 1 ).
|
On an at-home health questionnaire, completed 1 week before participants came to the hospital for the experiment, individuals in the two priming groups did not differ in the average number of prescription medications they mentioned using 
. Most (35) participants reported that they did not use cardiovascular medications. Among those who used these medications, no differences emerged between the two groups in use of ACE inhibitors, beta blockers, calcium blockers, digitalis, diuretics, or nitrates (Table 2 ).
|
). The priming paradigm, including the number of words and the frequency of their appearance, followed the models of presenting stimuli developed by Devine 1989 and Bargh and Pietromonaco 1982. The experimenter asked participants to indicate whether a flash appeared above or below a bull's-eye, located at the center of the computer screen, by pressing a corresponding arrow key. Participants were told to try to keep their focus on the bull's-eye and to respond as quickly, but accurately, as possible. Participants were first exposed to a practice trial at slower speeds than the actual primes that allowed perception with awareness. To prevent the participants from expecting to see words in the actual trials, the practice trials consisted of 20 stimuli made up of scrambled letters. Following the practice trial, they were exposed to 100 words (five blocks of 20 words) that were presented in a random order within the blocks. The 20 words in each block consisted of 14 stereotype of aging primes, two category words (old or senior), and four neutral words.
The stereotype primes in the positive condition were accomplished, advise, alert, astute, creative, enlightened, guidance, improving, insightful, learned, sage, and wise. The stereotype primes in the negative condition were alzheimer's, confused, decline, decrepit, dementia, dependent, diseases, dying, forgets, incompetent, misplaces, and senile. To increase the potency of the primes, two of the stereotype primes were repeated in each block (Devine 1989). The two positive stereotypes that repeated were sage and wise, whereas the two negative stereotype words that repeated were alzheimer's and senile. The neutral words in both groups were another, between, sentence, and together.
Previously, Levy 1996 generated the stereotype of aging primes by asking one set of individuals to write down as many positive and negative descriptions of aging as they could think of and asking another set to rate these descriptions according to how positive and negative they appeared and how characteristic of aging they seemed. Both panels consisted of old and young participants. The words that were considered characteristic of aging and either positive or negative were selected.
Because of the variability in visual-processing speeds of elderly participants, we presented the first set of primes at either 55 or 66 ms. A mask covered each prime word for 2–7 s. All participants began at 55 ms. If they could not see any flashes after a presentation of 20 stimuli, we slowed down the presentation rate to 66 ms. We repeated this procedure with the second set of primes, again starting at 55 ms. At the end of the Results section, we present evidence that the paradigm allowed perception without awareness.
Physiological outcomes
We examined physiological response to stress with four measures: systolic blood pressure, diastolic blood pressure, heart rate, and skin conductance. Systolic blood pressure corresponds to the force of the blood against the arterial walls when the heart contracts. Diastolic blood pressure corresponds to the force of the blood against the arterial walls when the heart relaxes. Heart rate is a measure of the number of times the heart contracts in a minute. Skin conductance is a measure of electrodermal activity, which is associated with the activation of the sweat glands in the hands when an individual faces a challenge.
Systolic blood pressure, diastolic blood pressure, and heart rate were assessed automatically with the Critikon Dinamap Vital Signs Monitor (Tampa, FL). A blood pressure cuff was attached to the participant's left arm soon after arrival. When a button was pressed by an experimenter, the monitor measured blood pressure and heart rate and printed out the numbers. Skin conductance was assessed with a NeuroDyne Monitor (Cambridge, MA) that includes two gold-plated galvanic skin response sensors that attach to the middle and ring finger of the nondominant hand with Velcro-sealed strips. For every skin conductance data point, three consecutive values, 1 s apart, were averaged to provide a more robust estimate of arousal. The screen for both monitors faced away from the participants. To prevent inquiries during the session, participants were told at the beginning that the experimenter would give a summary of their cardiovascular measures at the end of the session.
At the end of the session, in addition to learning about their cardiovascular measures, participants were told about the true nature of the priming intervention. The participants were also told to which priming group they had been assigned. For those assigned to the negative priming group, we read the list of positive aging stereotypes. We further informed the participants that we believed everyone has internalized both images of aging and that the positive stereotypes could have been just as easily activated as the negative ones. The participants were then given a National Institute on Aging publication about the elderly cardiovascular system that includes recommendations for health-promoting behaviors.
The experimenters measured the four autonomic nervous system variables at six points: (a) baseline; (b) after the first set of primes; (c) after the first set of verbal and mathematical stressors; (d) after the second set of primes; (e) after the second set of verbal and mathematical stressors; and (f) following a final questionnaire that participants completed between 15 and 43 min after the second prime, with an average duration of 26 min. Priming group did not have an effect on this duration.
Mathematical and verbal challenges.
In the mathematical challenge, the researcher asked participants to count backward from 956 in decrements of 7, as quickly, but accurately, as possible within 1 min (Uchino, Keicolt-Glaser, and Cacioppo 1992). In the second mathematical challenge, participants were asked to count backward from 375 by 7. The verbal challenge consisted of a request to discuss for 3 min a particularly stressful event that the participant had experienced within the last 5 years. The second time participants engaged in this task we asked them to talk about a different stressful event from the last 5 years. To make the tasks more stressful, the experimenter kept track of time with a stopwatch, a metronome ticked in the background, and a tape recorder placed in front of the participants recorded their performance.
To better recreate the kind of everyday general stress an older person might encounter, and to simplify analyses, we averaged the responses to the combined mathematical and verbal challenges for each of the cardiovascular measures. Combining the two challenges was supported by the exploratory analysis described in the next section and the finding that a high intertask consistency exists in cardiovascular response to a variety of psychological challenges designed to induce stress (Turner, Sherwood, and Light 1994).
To measure the impact of the challenges, we asked participants to rate how stressful they found each of the mathematical and verbal challenges on a scale from 1 (not at all stressful) to 7 (extremely stressful).
Self-Efficacy
We calculated a self-efficacy score by asking participants, before they tried each of the mathematical challenges, to estimate the number of times they could correctly subtract from a three-digit number by 7 in a minute. We averaged the two responses.
Mathematical Performance
We created the mathematical performance measure by scoring the number of times participants correctly subtracted by 7 from 956 in 1 min and correctly subtracted by 7 from 375 in 1 min. We averaged the two scores.
Exploration of Potential Confounds and Data Reduction
To explore potential confounds and simplify the analyses, prior to examining our hypotheses we conducted an analysis designed to discern whether gender, reported use of cardiovascular medications, experimenter, type of challenge (mathematical or verbal), or age led to any significant main effects or interactions. We conducted a 2 (priming group) x 2 (gender) x 2 (cardiovascular medication) x 2 (experimenter) multivariate analysis of covariance (MANCOVA) with repeated measures on eight time points and four outcome measures. The eight time points consisted of baseline, after the first set of primes, after the first verbal challenge, after the first mathematical challenge, after the second set of primes, after the second verbal challenge, after the second mathematical challenge, and at follow-up. The outcome measures were systolic blood pressure, diastolic blood pressure, heart rate, and skin conductance. Age served as a covariate.
In this analysis, we found that gender, reported use of cardiovascular medications, experimenter, and type of challenge produced no significant main effects or interactions. Using 8 time points and 54 participants provided more statistical power than 1 time point with 
, enough power to reveal meaningful relationships between these variables and the change in cardiovascular scores over time, including the difference between priming groups across time. Thus, we excluded gender, reported use of cardiovascular medications, experimenter, and type of challenge from the primary model examining our hypotheses. Although age did not produce a main effect or interactions, we kept age as a covariate in the model examining our hypotheses because previous studies have found a relationship between age and cardiovascular response to stress (Steptoe and Tavazzi 1996).
|
Results |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
Overall Prediction
Results of the MANCOVA supported our overall prediction that those exposed to the negative stereotypes would demonstrate a heightened cardiovascular response compared with those exposed to the positive stereotypes. A significant interaction emerged between priming group and time, 
. (Each of the 
2 that we report are standardized.) In addition, main effects emerged for autonomic measures, 
, and interactions emerged between priming group and autonomic measures, 
, and priming group, cardiovascular measures, and time, 
. The 2 for the overall model was .40.
When we examined each of the cardiovascular measures separately within the MANCOVA with contrast analyses, all the measures demonstrated the predicted linear increase in differences between the change scores of the positive and negative priming groups and time: systolic blood pressure, 
; diastolic blood pressure, 
; heart rate, 
; and skin conductance, 
.
Hypothesis 1
The findings also supported our first hypothesis that the negative age stereotype intervention would lead to a significant increase in autonomic measures compared with baseline. To examine this, we conducted contrast analyses for each of the measures separately within the negative priming group, with weights representing the prediction that among those exposed to the negative stereotype primes, a significant increase should occur between baseline and following the intervention (after the second set of mathematical and verbal stressors). These contrasts were significant for three of the four measures: systolic blood pressure, 
; diastolic blood pressure, 
; and skin conductance, 
(1,25) 
(see Fig. 1). The contrast analysis did not reach significance for heart rate.
|
p 
; diastolic blood pressure, 
; and skin conductance, 
(see Fig. 2).
|
; diastolic blood pressure, 
; and skin conductance, 
.
The second set of contrasts examined the relative contributions of each time point (after the first set of negative primes, after the first set of mathematical and verbal challenges, after the second set of negative primes, and after the second set of mathematical and verbal challenges) to the overall increase from baseline to after the intervention (after the second set of mathematical and verbal challenges) for the four outcome measures. In these analyses, we found that most of the increase in outcome measures was due to the first set of negative stereotype primes. The only significant increase occurred from baseline to after the first set of negative primes: systolic blood pressure, 
; diastolic blood pressure, 
; and skin conductance, 
. The next three changes (from the first set of negative primes to after the first set of mathematical and verbal challenges, from the first set of mathematical and verbal challenges to after the second set of negative primes, and from the second set of negative primes to after the second set of mathematical and verbal stressors) demonstrated nonsignificant increases.
In support of the large contribution of the first set of negative stereotype primes to the overall increase from baseline to after the intervention for systolic blood pressure, diastolic blood pressure, and skin conductance, we found the amount of variability accounted for by the first set of primes (when averaged across the three measures) was at least six times 
the contribution of the second set of primes 
, or either the first set of mathematical and verbal challenges 
or the second set of mathematical and verbal challenges .
Hypothesis 2
The findings also supported our second hypothesis that the positive age stereotype intervention would protect individuals from showing a change from baseline, despite the exposure to multiple challenges. To examine this, we conducted a series of contrast analyses for each autonomic measure. First, we specified contrast weights that compared baseline and postintervention scores (immediately after the second set of mathematical and verbal challenges) for those exposed to the positive stereotypes. As predicted, none of these contrasts showed significant differences between baseline and the postintervention scores (see Fig. 1).
To further examine whether the positive stereotype primes exerted a protective effect, we conducted contrast analyses to examine whether any significant changes occurred between the six time points. We found two significant changes. Significant increases occurred in three of the four measures with the first set of mathematical and verbal challenges: systolic blood pressure, 
; diastolic blood pressure, 
; and skin conductance, 
. A significant decline occurred following exposure to the second set of positive aging stereotypes for these same measures: systolic blood pressure, 
; diastolic blood pressure, 
; and skin conductance, 
. Heart rate, in contrast, showed no significant increase with either the first or second set of mathematical and verbal challenges.
It appears that the positive stereotype primes exerted a protective effect the second time participants were exposed to them. The second set of positive aging primes not only brought down the autonomic response of systolic blood pressure, diastolic blood pressure, and skin conductance close to baseline, but also appeared to help protect participants from experiencing further stress. No significant increase occurred with the second set of mathematical and verbal challenges. Thus, by the end of the positive stereotype intervention, the participants' autonomic indices did not differ significantly from baseline.
Effect of Aging Stereotypes on Cognition and Self-Efficacy
Those exposed to positive aging stereotypes had significantly higher mathematical self-efficacy scores than those exposed to negative aging stereotypes, 
(Fig. 3).
|
(Fig. 3). Furthermore, the mathematical self-efficacy measure significantly correlated with mathematical performance, 
.
Perception Without Awareness Check
To check the participants' awareness of the primes, after the two sets of primes the experimenter asked participants to report all words they saw. As mentioned, the three participants who correctly identified any of the words were removed from the analyses.
As further assurance that this method allowed perception without awareness, we conducted two control conditions (Bargh and Pietromonaco 1982; Devine 1989). Ten participants who were of similar age to the study sample were randomly assigned to the positive or negative priming conditions. Participants were asked to guess each word that flashed. To encourage guessing, participants were told it was better to free associate than not guess. Their overall hit rate was 0.8%. The hit rate was 1.67% for the neutral words and 0.7% for the prime words.
The control participants were exposed to the two sets of primes at 55 ms. Then they were given a recognition check that consisted of 36 words: 18 prime words to which they had been exposed and 18 new words that appeared in the other priming condition. Participants were asked to circle all the words that they recognized, or that looked at all familiar, from the computer task. Their overall recognition rate was at chance level: 46%. These hit and recognition rates meet the standard for perception without awareness used by other researchers studying implicit stereotyping (Devine 1989).
|
Discussion |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
Impact of Negative Age Stereotypes
In partial support of our first hypothesis, the negative age stereotype intervention led to a significant increase, compared with baseline, in three of the four measures (systolic blood pressure, diastolic blood pressure, and skin conductance). These increases still appeared at the follow-up measurement that occurred an average of 26 min after the intervention.
The nonsignificant heart rate response to the negative stereotypes, as compared with blood pressure, is consistent with research that shows blood pressure response tends to increase with advanced age, whereas heart rate response tends to diminish (Steptoe and Tavazzi 1996; Wei 1992). Thus, in older individuals, heart rate may not be as effective a marker of the cardiovascular system's response to stress as blood pressure.
The first set of negative aging stereotypes led to a significant increase from baseline in three of the four measures and made a greater contribution to the overall effect of the intervention than was made by the two sets of mathematical and verbal challenges. These findings suggest that the negative stereotypes had immediate effect. The relatively weaker impact of the sets of challenges and the second set of negative aging stereotypes may be due, in part, to the first negative primes' causing a ceiling effect.
Impact of Positive Age Stereotypes
The data support our second hypothesis that the positive age stereotype intervention would protect individuals from showing a significant increase in their autonomic measures, despite the exposure to multiple challenges. It appears, however, that the first set of positive aging stereotypes did not immediately protect participants. Instead, a significant increase occurred with the first set of mathematical and verbal challenges and a significant decline occurred with the second set of positive aging stereotypes. No significant increase from baseline occurred in the autonomic measures after the second set of stressors. This suggests that the positive age stereotypes not only protected the participants from the later challenges, they also helped them recover from the earlier ones.
It therefore seems that the positive stereotypes had a cumulative effect. The positive age stereotypes took longer to exert a protective effect than the negative stereotypes took to elevate stress. This cumulative protective finding suggests that multiple exposures to positive aging stereotypes could help older individuals both reduce the amount of stress they experience and recover from stressful events.
Self-Efficacy as a Potential Mediator
Our results suggest that self-efficacy may help mediate the relationship between self-stereotypes of aging and experiencing stressors. Those who were in the negative prime group expressed lower self-efficacy (as measured by predicted mathematical performance), which was associated with lower mathematical performance, whereas those who were in the positive prime groups expressed higher self-efficacy, which was associated with higher performance. Apparently, the negative stereotype primes activated negative self-stereotypes, which, in turn, lowered the expectation of performance. This expectation may have acted as a self-fulfilling prophecy because the participants' performance was in fact hampered. This inability to perform successfully may have caused frustration and, consequently, a heightened degree of stress.
The concomitant process would be that the positive primes activated positive self-stereotypes, which led to the expectation of successful performance. This expectation may have facilitated a successful performance that led to reassurance and, therefore, lowered stress.
Influence of Implicit Aging Self-Stereotypes on Health
This study demonstrates that aging self-stereotypes can activate the autonomic nervous system of older individuals without their awareness. Not only were the participants in our study unable to name the stereotype primes that flashed on the screen, the positive and negative aging stereotype groups rated the mathematical and verbal challenges as equally stressful. Thus, the influence of the self-stereotypes on physiology was not due to how the older participants consciously perceived the stressors. Rather, it appears the priming unconsciously influenced the participants' self-efficacy, cognitive performance, and autonomic response to stress.
This finding adds another dimension to the risk for older individuals of the widespread occurrence of negative aging stereotypes. If the activation of negative self-stereotypes causes deleterious health consequences, the affected individuals will be unaware of the causation. The health outcome may be attributed to the inevitability of aging, which could reinforce the negative stereotype and prevent successful coping. If, as a consequence, there is suppressed anxiety, it may compound the problem.
The implicit stereotype intervention had a powerful impact on the cardiovascular and skin conductance measures. The overall model accounts for 40% of the total variability in the outcome measures, and the stereotype priming effects account for nearly half of the model variance. The effect size of the priming intervention is especially impressive in light of the unexpected degree of variability in the responses on the different cardiovascular measures, which accounts for a substantial amount of the remaining variance. The sample used in this study included men and women with diverse medical histories. Gender and the use of cardiovascular medications did not contribute to the screening MANCOVA, but this does not indicate that these factors had no influence at all on individual responses. It is likely, in fact, that the effects of these factors account for a substantial portion of the error variance, and that a study that eliminated those with a history of cardiovascular problems would demonstrate an even stronger effect of the priming intervention. The power of the priming intervention in this study to overcome such individual differences may be an indicator of the impact of positive and negative stereotypes on the general population.
Conclusion
We found that the negative aging stereotypes continued to exert an effect on three of the four physiological measures at follow-up, which occurred about half an hour after the intervention. In future research, it would be interesting to continue monitoring the participants so that the length of the stereotype effect could be determined. Even if the effect lasts no longer than half an hour, it is ominous. Older individuals are likely to encounter numerous negative aging stereotypes in their daily lives, which could result in frequent activation of the negative self-stereotypes. Chronic stressors that lead to repeated elevations of the sympathetic nervous system have been associated with the development of heart disease (Barnett et al. 1997; Uchino et al. 1996).
This research presents for the first time the effect of self-stereotypes of aging on a physiological process: the cardiovascular response to stress. Our data suggest that interventions designed to improve cardiovascular health and cognitive functioning in the elderly should include the reduction of negative aging self-stereotypes and the promotion of positive ones. The reinforcement of positive stereotypes offers the prospect of reducing infirmity and enhancing independence for our graying world.
|
Acknowledgments |
|---|
Received for publication November 17, 1998. Accepted for publication January 11, 2000.
|
References |
|---|
|
TOPAbstractThe Power of Age...Evidence That Response to...Present StudyMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  B. R. Levy Mind Matters: Cognitive and Physical Effects of Aging Self-Stereotypes J. Gerontol. B. Psychol. Sci. Soc. Sci., July 1, 2003; 58(4): P203 - 211. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Levy Conscious Versus Unconscious Levels of Aging Self-Stereotypes: Author's Reply J. Gerontol. B. Psychol. Sci. Soc. Sci., July 1, 2003; 58(4): P215 - 216. [Full Text] [PDF]
|
|
|
|
|
|
V. Igier and E. Mullet Application of the Five-Factor Model of Personality to Intergenerational Perception J. Gerontol. B. Psychol. Sci. Soc. Sci., May 1, 2003; 58(3): P177 - 186. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Hess, C. Auman, S. J. Colcombe, and T. A. Rahhal The Impact of Stereotype Threat on Age Differences in Memory Performance J. Gerontol. B. Psychol. Sci. Soc. Sci., January 1, 2003; 58(1): P3 - 11. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. R. Levy, M. D. Slade, and S. V. Kasl Longitudinal Benefit of Positive Self-Perceptions of Aging on Functional Health J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2002; 57(5): P409 - 417. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Hooker, S. R. Bowman, D. P. Coehlo, S. R. Lim, J. Kaye, R. Guariglia, and F. Li Behavioral Change in Persons With Dementia: Relationships With Mental and Physical Health of Caregivers J. Gerontol. B. Psychol. Sci. Soc. Sci., September 1, 2002; 57(5): P453 - 460. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. R. Levy Eradication of Ageism Requires Addressing the Enemy Within Gerontologist, October 1, 2001; 41(5): 578 - 579. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
