Journal of Personality and Social Psychology Copyright 2000 by the American Psychological Association, Inc.

2000, Vol. 79, No. 4, 644-655 0022-351<M»/$5.00 DOI: I O.1037//O022-3 514.79.4.644

Emotional Experience in Everyday Life Across the Adult Life Span

Laura L. Carstensen Monisha Pasupathi

Stanford University University of Utah

Ulrich Mayr John R. Nesselroade

University of Oregon University of Virginia

Age differences in emotional experience over the adult life span were explored, focusing on the
frequency, intensity, complexity, and consistency of emotional experience in everyday life. One hundred
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

eighty-four people, age 18 to 94 years, participated in an experience-sampling procedure in which

This document is copyrighted by the American Psychological Association or one of its allied publishers.

emotions were recorded across a 1-week period. Age was unrelated to frequency of positive emotional
experience. A curvilinear relationship best characterized negative emotional experience. Negative emo-
tions declined in frequency until approximately age 60, at which point the decline ceased. Individual
factor analyses computed for each participant revealed thai age was associated with more differentiated
emotional experience. In addition, periods of highly positive emotional experience were more likely to
endure among older people and periods of highly negative emotional experience were less stable.
Findings are interpreted within the theoretical framework of socioemotional selectivity theory.

Emotions are central to human functioning, guiding thought and ing reasonably stable in midlife, and becoming dysregulated and
action from the earliest days of life (Frijda, 1988). In recent years, rigid in old age (cf. Bromley, 1990). Yet, evidence for emotional
research has revealed much about the astonishing developmental degradation in adulthood is hard to come by. On the contrary, a
gains made early in life concerning emotional differentiation (e.g., small but growing literature on the adulthood course of emotion
Camras, Sullivan, & Michel, 1993) and regulation (e.g., Rothbart paints a distinctly positive picture and suggests that improvements
& Ahadi, 1994). The social nature of emotion is evident through- in emotional functioning may continue well into middle-age and
out this literature. It appears that early on, regulation of emotion is perhaps old age (Carstensen & Charles, 1999). Older as compared
situated outside of the individual, with caregivers playing a pri- with younger adults, for example, display increasing complexity
mary role in soothing, exciting, comforting, and otherwise influ- in mental representations, infused by affect and subjectivity
encing infants' emotions. Gradually, however, regulatory pro- (Labouvie-Vief, DeVoe, & Bulka, 1989; see also Isaacowitz,
cesses are internalized; cognitive appraisals of the emotional Charles, & Carstensen, 1999); report better emotional regulation
significance of environmental stimuli begin to influence emotional (Gross et al., 1997; Labouvie-Vief, Hakim-Larson, DeVoe, &
responses (Lazarus, 1991). Although far less is known about the Schoeberlein, 1989; Lawton, Kleban, & Dean, 1993); display
developmental course of emotional experience and regulation in well-preserved expressive systems (Levenson, Carstensen, Friesen, &
adulthood, it is clear that successful regulation of emotion is Ekman, 1991; Tsai, Levenson, & Carstensen, in press; Malatesta &
central to functioning in interpersonal relationships, coping with Kalnok, 1984); and are relatively happy (Diener & Diener, 1996)
life's hardships, and optimizing mental health. and satisfied with life (Herzog & Rodgers, 1981). In a recent
We expect that part of the reason emotional development was survey, Mroczek and Kolarz (1998) found that age was associated
not studied in adulthood until relatively recently relates to long- with a self-reported increase in positive affect and a decrease in
held presumptions that emotional functioning in later life parallels negative affect. Studies also suggest that the salience of emotion
biological and cognitive functioning in adulthood and old age, may increase with age, such that emotional material is better
namely leveling in late adolescence and early adulthood, remain- remembered (Carstensen & Turk-Charles, 1994), is more central in
cognitive representations of other people (Carstensen & Fredrick-
son, 1998; Fredrickson & Carstensen, 1990), and is more centrally
involved in problem solving about interpersonal matters
Laura L. Carstensen, Department of Psychology, Stanford University; (Blanchard-Fields, 1997).
Monisha Pasupathi, Department of Psychology, University of Utah; Ulrich
Mayr, Department of Psychology, University of Potsdam, Potsdam, Ger- Two life span developmental theories, which have instigated
many; John R. Nesselroade, Department of Psychology, University of much of the empirical work described above, suggest that emo-
Virginia. tional development continues in adulthood. Both theories draw
This research was funded by National Institute on Aging Grant
heavily on the idea that because age and experience are inextrica-
RO1AG08816. We thank Susan T. Charles and Helene Fung for their
criticisms of earlier versions of this article.
bly intertwined and because experience and knowledge about
Correspondence concerning this article should be addressed to Laura L. emotions play an important role in emotion regulation, aging may
Carstensen, Department of Psychology, Jordan Hall, Stanford University, be associated with emotional maturation. Labouvie-Vief and her
Stanford, California 94305-2130. Electronic mail may be sent to colleagues have contended that cognitive functioning becomes
[email protected]. infused with affectivity in later adulthood such that peak intellec-

644
 EMOTION AND AGE 645

tual performance occurs in middle age. According to this theory, about their own maturation (McFarland, Ross, & Giltrow, 1992).
children must suppress idiosyncratic affective judgments about That is, if research participants believe that people "should" con-
collectively shared, symbol systems so that they acquire uniform, trol their emotions better as they get older, they are likely to say
culturally consistent representations of the world (Labouvie-Vief that they do so. Moreover, global evaluations of life are highly
& DeVoe, 1991; Labouvie-Vief, Hakim-Larson, et al. 1989). Adult cognitive and involve comparisons with the past and the present, as
intellectual development, in contrast, involves the reintegration of well as with idiosyncratic standards (Schwarz, Park, Knaueper, &
subjective information into existing knowledge structures. Theo- Sudman, 1999). Subsequently, such evaluations are susceptible to
retically, this increased complexity in cognitive operations is as- influence by cohort-specific experiences and mores (Elder, Odum,
sociated with increasingly more complex and adaptive emotional & Hareven, 1994), as well as by memory of past events (Levine &
responses and perhaps with greater flexibility in coping with new Bluck, 1997). A second important concern is that even though
life events (Diehl, Coyle, & Labouvie-Vief, 1996). older people may be able to respond well to emotional tasks in
A second theory relevant to emotional functioning in adulthood laboratory settings, emotional dysregulation may be evident in less
is socioemotional selectivity theory (Carstensen, 1993, 1995; structured settings.
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Carstensen, Gross, & Fung, 1997; Carstensen, Isaacowitz, & The purpose of the present study was to assess the frequency,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Charles, 1999). Although this theory also acknowledges experi- intensity, and the complexity of emotional experiences as they
ence as an important factor in emotional development, it focuses occur in everyday life. Our general hypothesis was that relative to
on perceived time left in life rather than past experience. The younger people, older people would show evidence of improved
theory contends that the distinctly human ability to consciously emotion functioning, including more differentiated emotional ex-
and subconsciously monitor time plays a fundamental role in periences and better regulation of their emotional states. By re-
motivation and emotion, providing the structure within which peatedly sampling the same participants over time, we were able to
goals are set, pursued, and evaluated. Because mortality places the examine the frequency and intensity with which positive and
ultimate constraint on time, chronological age is associated with negative emotions are experienced, as well as the stability of
changes in goals. Essentially, the theory contends that two primary negative and positive states over time and the complexity of
trajectories of social motives operate throughout life: the emotion emotional experience. Below we elaborate four experimental hy-
trajectory and the knowledge trajectory. The former is character- potheses about age differences in emotional experience.
ized by motives to achieve emotional satisfaction and meaning, the
latter by motives to acquire new information and to achieve in Hypothesis 1. Older people experience negative emotions less fre-
domains that are relevant to successful adaptation in the future quently than younger adults and experience positive emotions just as
(e.g., educational and occupational domains). The central change frequently as younger adults. According to socioemotional selectivity
in adulthood is a shift in the salience of social goals. Younger theory, increasing age is associated with greater appreciation of life
adults, having much to learn and relatively long futures for which and greater investment in emotionally meaningful social relationships.
to prepare, are motivated by the pursuit of knowledge—even when The theory predicts that this emphasis on emotionally meaningful
goals improves emotional experience in everyday life.
this requires that emotional well-being be suppressed. For older
adults, the reverse trend appears. Facing relatively shorter futures Hypothesis 2. The intensity of positive and negative emotional expe-
and having already accrued considerable knowledge about others, rience is comparable across age groups. Socioemotional selectivity
older adults prioritize emotional goals because they are realized in theory predicts that goal-directed behavior aimed at obtaining emo-
the moment of contact rather than banked for some nebulous future tionally meaningful goals results in less frequent negative emotions.
time. However, once negative emotions are elicited, the theory makes no
claim about the intensity of the experience.
The theory stresses that age does not entail the relentless pursuit
of happiness but rather the satisfaction of emotionally meaningful Hypothesis 3. Older as compared with younger adults show differen-
goals, which entails far more than simply feeling good. Finding tial stability of emotional experience such that positive states are
meaning in existing relationships, even conflictual ones, emerges maintained longer and negative states are terminated more quickly.
as a central task in later life. Emotional experience is subsequently Surveys that ask people how well they regulate their emotions suggest
expected to be more complex, and the experience of mixed emo- that where there are differences, older people report greater control. In
tions, more frequent. In short, socioemotional selectivity theory this study, we obviate global judgments about emotion control by
examining whether positive states last longer and negative states
suggests that constraints on time directly influence emotional
persist for shorter periods in older adults.
experience such that emotional states are increasingly mixed.
Whether pleasure or joy, sadness or pain, knowledge that an Hypothesis 4. Emotional experience is more complex in older as
experience will soon end changes the emotional experience itself. compared with younger adults. Because the pursuit of emotionally
Rather than simply prompting negative emotions related to antic- meaningful goals often entails mixed emotions, we anticipate a more
ipated loss, moments are savored, appreciated both for what they complex dimensional structure to the emotional experience of older
are and for their temporal fleetingness. adults.
Despite these relatively optimistic empirical and theoretical
pictures, well-documented declines in later life leave an uneasiness Method
that positive portrayals of emotion in later life may be overblown.
First, much of the literature on emotion, and to the best of our Sample
knowledge all of the literature on emotion regulation, has been One hundred eighty-four African American and European American
based on global self-reports, a practice that may be particularly research participants, ranging in age from 18 to 94 years of age, (M — 55,
precarious when questions elicit older adults* implicit theories SD = 20.4), were recruited by a survey research firm from the San
 646 CARSTENSEN, PASUPATHI, MAYR, AND NESSELROADE

Francisco Bay area to participate in an experience-sampling study of jectives representing all of the Big Five factors of personality are repre-
emotional experience. Ethnic composition of the sample was restricted to sented. Participants indicate whether a given statement describes them by
these two ethnic groups rather than sampling the ethnic diversity of the Bay placing a check next to it. Example items include "I am talkative" and "I
area so that sufficient numbers of participants in subsamples would allow can be somewhat careless." We computed summary scores for each of the
for statistically meaningful analyses. Thirty-one percent of the sample were Big Five factors: Neuroticism, Extraversion, Openness to Experience,
African American; the remaining 69% were European American. Forty- Agreeableness, and Conscientiousness.
one percent of the sample comprised blue-collar workers, and 59% were
white-collar workers; 54% of the participants were women, and 46% were
men. Education ranged from 5 to 22 years (M = 15.0, SD = 2.7). Gender, Procedure
blue- or white-collar status, and race were distributed evenly across age. As Following initial screening by the survey research firm to ensure that
shown in Table 1, the sample was diverse along many dimensions. participants met recruitment criteria for the project, participants were
scheduled at their convenience for an initial interview at Stanford Univer-
Measures sity or at the offices of the San Francisco-based survey research firm that
did the initial recruiting. Participants were informed that the purpose of the
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Although our principal aim was to sample emotional experiences in study was to examine feelings in everyday life. After obtaining informed
This document is copyrighted by the American Psychological Association or one of its allied publishers.

everyday life, we also assessed health, personality, and verbal fluency consent and obtaining background information, such as education level, the
because each of these factors may influence at least some features of following measures were administered: Category Instance Fluency, CMI,
emotional experience or performance on the sampling task (see, e.g., and the Adjective Checklist.
McCrae & Costa, 1991; Watson & Pennebaker, 1989). At this point, participants were provided with detailed instructions about
Emotion sampling booklet. On a 7-point scale that ranged from 1 (not the experimental procedures, familiarized with the operations of the elec-
at all) to 7 (extremely), participants indicated the degree to which they were tronic pager (e.g., how to set it for motion or sound, how to indicate that
feeling each of 19 emotions or feeling states. Ratings greater than 1, thus, they received the page by pushing a button, etc.), and instructed to
indicated that the emotion was present and, consequently, both frequency complete the emotion response sheets each time they were signaled. Next,
and intensity are captured in a single rating. The list of emotions included two practice trials were administered while participants were still in the
anger, guilt, pride, sadness, happiness, fear, accomplishment, shame, laboratory so that responses could be reviewed with the experimenter prior
amusement, anxiety/worry, joy, contentment, irritation, frustration, disgust, to beginning the study. The participant was left alone; the interviewer
interest, embarrassment, boredom, and excitement. An other blank was activated the pager from another room; the participant completed the
also provided on the response sheet to allow participants to record addi- questionnaire; and, on returning to die room, the interviewer reviewed the
tional emotions not included on the sampler. A week's supply of emotion participant's responses, clarified any apparent mistakes, and answered
response sheets were bound in a 5 in. by 5 in. pad for easy transport during questions.
the week of data collection. During the ensuing week, participants were paged five times each day.
Cornell Medical Index Health Questionnaire (CMI). The CMI (Brod- Paging times were determined by random selections from all possible
man, Erdmann, & Wolff, 1949) is a widely used 195-item index of physical 10-min intervals between 9 a.m. and 9 p.m. The only constraint on
and mental health problems that allows the computation of a general health sampling times was that participants were not sampled more than once
index as well as subscales that represent functioning in specific organ within a single 20-min period. At the end of each day, participants returned
subsystems and symptoms associated with specific psychological syn- the day's completed response sheets by mail in pre-addressed, stamped
dromes. Participants report whether they experience each of the 195 envelopes, allowing us to monitor responses during the data collection
symptoms. We computed two broad indexes from the CMI, one represent- period and assuring at least rough adherence to the experimental protocol.
ing the total number of recent symptoms of physical illness and the other Participants were encouraged to telephone the laboratory if procedural
representing the total number of recent symptoms of mental illness. An questions or problems arose and periodic calls were made to participants as
example of a physical illness symptom item is "Are you troubled by well to ensure that the highest quality data were obtained.
constant coughing?" A sample mental illness symptom item is "Do you After participants completed the week-long experience-sampling data
have to be on your guard even with your friends?" collection, they returned to the laboratory for a follow-up interview, at
Category instance fluency (Undenberger, Mayr, & Kliegl, 1993). As a which time they returned the pagers and were debriefed. Participants were
measure of verbal fluency, participants were asked to name as many paid $125 for their participation.
different kinds of animals as possible in 90 s. This test shows a strong
relationship to general intellectual ability and has been extensively used
with older adults. Results
Adjective checklist (John, Donahue, & Kentle, 1991). This is a list
We organize our results into four sections. The first section
of 54 adjectives presented in the form of self-descriptive sentences. Ad-
describes data reduction and preliminary analyses. The second
section of the results reports findings from analyses that examine
age differences in the frequency and intensity of emotional expe-
Table 1 rience, controlling for individual differences that may influence
Demographic Characteristics of the Sample emotional experience (Hypotheses 1 and 2). In this section we also
Characteristic Sample description examine the consistency of age differences across ethnic, gender,
and socioeconomic lines.
Age (in years) M = 55.0, SD = 20.4, range = 18.0-94.0 The third section of the results addresses emotion regulation
Education (in years) M = 15.0, SD = 2.7, range - 5.0-22.0 (Hypothesis 3), and the fourth section concerns the complexity of
Sex 54% women, 46% men
Ethnicity 31% African American, 69% European American emotional experience (Hypothesis 4). In these latter two sections,
Socioeconomic status 41% blue-collar, 59% white-collar hypotheses are tested on the basis of within-individual variability
Marital status 26% single, 43% married, 19% widowed, 13% that allowed us to examine emotional complexity and the temporal
divorced experience of emotional experiences. Here, too, we examined the
Number of children M = 1.6, SD = 1.6, range = O-9 consistency of effects across ethnicity, gender, and class. Because
 EMOTION AND AGE 647

our measure of differentiation was novel, we explored its relation- indicators of emotional experience: average frequency of negative
ship to other measures, such as personality, mental health, and emotions, average frequency of positive emotions, average inten-
intelligence. sity of negative emotions, and average intensity of positive emo-
tions. Thus, the average frequency of experiencing positive emo-
Data Reduction and Preliminary Analyses tions reflects the average of the proportion of times a person
experiences each of the positive emotions. Because there is some
Ratings of 19 emotions were obtained on 35 separate occasions, debate about the dimensionality of emotional experience and be-
generating a total of 665 experience sampler data points per cause the 38 scores were derived from only 19 ratings, this
participant. Data were reduced in the following way. Frequency aggregation was verified by factor analysis.
was represented as the proportion of times across the 35 sampling We conducted a descriptive factor analysis using varimax rota-
points that a participant acknowledged that he or she experienced tion with mean substitution for missing values. This revealed that
the emotion to some degree, namely, ratings were greater than 1. frequency and intensity scores for the emotions were reasonably
Intensity of emotional experience was calculated by computing the well-characterized by a four-factor solution (69% of the total
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

average rating for each felt emotion (see Schimmack & Diener, variance was accounted for by this solution). Although six factors
This document is copyrighted by the American Psychological Association or one of its allied publishers.

1997, for a discussion of this kind of decomposition). Table 2 with eigenvalues above 1 could be extracted, Factor 5 had high
presents the average frequencies and intensities for specific emo- loadings only for the intensity of Fear (.60) and of Shame (.80),
tions across the entire sample. As can be seen, people endorsed whereas only the intensity of Boredom loaded significantly on
negative emotions relatively infrequently and positive emotions Factor 6 (.80). Together Factors 5 and 6 accounted for only an
relatively frequently, a finding consistent with earlier experience additional 5% of variance in the data. Inspection of the scree plot
sampling studies (Diener & Diener, 1996). For each specific showed a clear dissociation of the latter two from the first four
negative emotion, some people indicated that they did not expe- factors. Thus, a four-factor solution was used (as per Tabachnik &
rience the emotion on any of the sampled occasions; however, Fidell, 1989). This solution clearly reflects the frequency and
only 2 individuals failed to endorse any negative emotions at all intensity with which positive and negative emotions were experi-
during the experience-sampling period.1 enced and was reliable both when cases with missing values were
The above procedure resulted in each participant having 38 deleted and when oblique rotation was used. Table 3 presents the
scores indicating the frequency and intensity with which he or she varimax rotated factor loadings of frequency and intensity vari-
reported 19 emotions over the sampling period. To reduce the ables on the four factors as well as eigenvalues and variance
number of statistical tests in our analyses and to increase the accounted for by each factor. As shown in Table 3, Factors 1 and 2
reliability of our measures, we collapsed these 38 scores into four reflect the frequency and intensity of negative affect, respectively.
Factors 3 and 4, respectively, represent the frequency and intensity
of positive affect.2

Table 2
Means and Standard Deviations of Frequency and intensity Hypotheses 1 and 2: Age Is Related to the Frequency but
of Experiencing Specific Emotions for the Entire Sample not the Intensity of Emotional Experience
Frequency" Intensity We hypothesized that the frequency of negative, but not posi-
tive, emotional experience decreases across age cohorts, and we
Emotion M SD M SD n hypothesized that intensity of emotion would not distinguish age,
Negative
Anger .20 .24 3.31 0.91 147 1
Findings remain essentially unchanged when very high and very low
Sadness .28 .30 3.31 1.05 159 scorers are eliminated.
Fear .20 .28 3.13 1.04 138 2
Disgust .21 .25 3.46 1.09 157 Oblique rotations suggested some relationships between the factors,
Guilt .17 .27 2.99 0.90 128 with the factors for intensity of positive and intensity of negative emotions
Embarrassment .17 .27 3.23 1.09 131 correlated at .34, and the factors for intensity and frequency of positive
Shame .14 .25 2.93 0.93 112 emotions also correlated (r = .30). Frequency of experiencing negative
Anxiety .44 .32 3.28 0.93 176 emotions and frequency of experiencing positive emotions were also
Irritation .39 .28 3.39 0.90 177 correlated, though less strongly (r = .21). No other interfactor correlations
Frustration .39 .28 3.40 0.86 176 were above .17. Relationships between the aggregate scores (not factor
Boredom .30 .32 3.38 0.95 149 scores) used in our analyses mirrored these oblique factor correlations, and
were somewhat stronger. Intensity of negative emotion and intensity of
Positive
positive emotion showed a moderate relationship (r = .40, p < .01).
Happiness .89 .17 4.35 0.97 184 Frequency and intensity of positive emotion were also correlated (r — .42,
Joy .78 .27 4.10 1.02 181 p < .01), as were frequency of positive and frequency of negative emotions
Contentment .90 .16 4.40 1.00 1S4 (r — .29, p < .01). All other correlations were much lower (maximum
Excitement .69 .32 3.98 0.94 181 absolute value r = .16). Both interfactor correlations and correlations
Pride .70 .34 4.09 1.10 180 between our variables imply the existence of individual differences in
Accomplish .74 .27 4.40 1.00 184
emotional intensity and in the general frequency with which emotions are
Interest .88 .19 4.55 0.99 184
Amusement .73 .27 3.95 0.86 182 experienced. These relationships (between the frequency and intensity of
positive and negative emotion) did not vary as a function of age at this, the
*n = 184. between-subjects, level.
 648 CARSTENSEN, PASUPATHI, MAYR, AND NESSELROADE

Table 3 effect has both a linear and a nonlinear component, as shown by

Varimax Rotated Factor Loadings for Frequency and Intensity the joint effects of linear age, B = -.02, 0 = -1.47, z(157) =
of Positive and Negative Affect -3.42, p < .01, and age squared, B = .0001 £ = 1.41,
r(157) = 3.30, p < .01. Just to illustrate these effects in a more
NE PE PE NE intuitive way, simple correlations between age and frequency of
Emotion frequency intensity frequency intensity
negative emotion before and after age 60 were computed. These
Frequencies correlations reveal a decrease in the frequency of negative emo-
tions from 18 to 60 years (simple r = —.29, p < .01). After 60
Anger .86 -.04 .09 .03
Sadness .85 -.04 .11 -.03 years, the decrease ceases (r = .14, ns) and characterizes the
Fear .87 -.01 .08 .03 pattern from age 60 onward.3 This effect is shown in Figure 1.
Disgust .88 .02 .01 .03 Our next step was to explore the robustness of age as a predictor
Guilt .90 -.04 .11 .00 of emotional experience in conjunction with factors known to
Embarassment .88 .14 .09 -.08
influence emotions. We added the following predictors to the
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Shame .92 .06 .10 .01

regression equations described above: personality (Neuroticism,
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Irritation .86 -.09 .06 -.04

Frustration .85 -.14 .09 -.03 Extraversion, Openness to Experience, Agreeableness, and Con-
Anxiety .78 -.14 .13 -.05 scientiousness), health (self-reported mental health and self-
Boredom .65 -.17 .22 -.17 reported physical health), and demographic variables (ethnicity,
Happiness .06 .16 .85 -.12
Joy .13 .32 .79 -.02 gender, and socioeconomic status). In addition, because intellec-
Contentment .02 .02 .79 .00 tual functioning reliably declines with age, we included the mea-
Excitement .27 .30 .77 -.11 sure of general intellectual ability (i.e.. Category Instance Flu-
Pride .24 .22 .76 -.13 ency). All variables were entered simultaneously.
Accomplishment .14 .16 ,85 -.11
Amusement .19 .26 .80 -.18
The results, displayed in Table 4, show that age findings were
Interest .09 .07 .79 -.14 maintained even with these additional variables included. Con-
cerning the frequency of negative emotions, more mental health
Eigenvalues 9.34 8.35 5.60 2.00
% of Variance 25 22 15 5
symptoms and higher neuroticism were additionally associated
with higher frequencies of negative emotion. For the frequency of
Intensities positive emotions, African Americans and those who were more
Anger -.09 .18 -.05 .66 open and agreeable reported experiencing positive emotions more
Sadness .03 .20 -.02 .67 frequently.
Fear -.05 .07 -.13 .66 Concerning the intensity of emotions, negative affect was pre-
Disgust -.15 .08 -.08 .64 dicted by gender and socioeconomic status as well as by mental
Guilt -.01 .12 -.15 .56
Embarassment -.11 .12 -.02 .64 health symptoms, with women, blue-collar workers, and individ-
Shame -.07 .07 .01 .53 uals who were more extraverted endorsing more intense negative
Anxiety .17 .19 -.07 .72 affect. Intensity of positive affect was associated with ethnicity and
Irritation .06 .10 -.07 .83 extraversion, with African Americans and more extraverted indi-
Frustration .14 .03 -.03 .84
.15 -.08 .45
viduals endorsing more intense positive affect. We also explored
Boredom -.04
Happiness -.07 .84 .34 .18 the consistency of the pattern of age effects and noneffects in
Joy .03 .92 .04 .16 predicting affective experience across gender, race, and socioeco-
Contentment -.24 .76 .21 .15 nomic status by looking for interactions between age and these
Excitement -.01 .84 .07 .22 demographic factors in predicting our emotional experience vari-
Pride -.02 .75 .18 .16
Accomplishment -.07 .78 .33 .27 ables. There were no interactions for age and gender, socioeco-
Amusement -.05 .81 .14 .30 nomic status, or race in predicting any of these facets of emotional
Interest -.07 .72 .32 .22 experience.
Eigenvalues 9.34 8,35 5.60 2.00
% of Variance 25 22 15 5
Note. NE = negative emotions; PE - positive emotions. Boldface indi-
cates factor loadings above .40.
3
We chose 60 as an illustrative point because it represents approxi-
mately the bottom of the line depicted in Figure 1. Choosing other age
with the exception of excitement. Four regression analyses were cutoffs, such as 40,50, or 65, results in differences in the magnitude but not
computed to explore relationships between the frequency and the pattern of relationships. To be concrete, the correlations for the younger
portion of the sample are —.46, —.24, —.29, and —.24, respectively, for
intensity of experiencing positive and negative affect and age.
ages 40, 50, 60, and 65 years. The same correlations for the older portion
Both linear and quadratic age trends were explored. No significant
of the sample are .04, .17, .14, and .19. In no case is the age relationship
age effects were obtained for the frequency or the intensity of near zero for the younger portion of the sample—a decline is always seen.
positive affect, nor were there age effects for the intensity with What happens after the selected cutoff is somewhat more variable, ranging
which negative affect was experienced, all Fs(l, 181) < 1.0, ps > from zero relationships to small positive correlations with negative emo-
3. However, as predicted, age was associated with the frequency tion. However, again, these are meant only to illustrate the pattern detected
of experiencing negative affect F(2, 182) - 6.0, p < .01. This in the regression analyses.
 EMOTION AND AGE 649

Table 4
Emotional Experience, Age, and Other Variables

NA Frequency PA Frequency NA Intensity PA Intensity

Predictors r P B r B r IS 8 r B
Demographic variables
Race .07 .11 .06 .17* .21* .10 .16* .12 .19 .37** .32** .62
Sex -.04 -.01 -.01 -.13t -.13 -.05 .18* .18* .27 .07 .05 .09
Socio-economic status .03 -.07 -.03 .03 .02 .00 .25** .15 .23 .15* .12 .22
Health
Physical .11 .00 .0001 .01 -.06 -.001 .16* .02 .002 .04 -.04 -.003
Mental .28** .17t .01 .06 .06 .002 .22** .17f .02 .03 .04 .005
Personality
Neuroticism .20** -20t .17 -.11 .08 .70 .03 .04 .10 -.24** -.09 -.29
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

Extraversion .01 -.01 -.01 22** .11 .90 .11 .17* .50 .29** .20* .69
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Openness .03 -.07 .09 .12 •15t .18 .00 -.06 - .24 .14t .11 .51
Agreeableness -.11 .05 .05 .16* .17f .19 .01 .04 .15 .24** .11 .53
Conscientiousness -.18* -.09 -.10 .17* .05 .05 -.02 .02 .06 27** .09 .38
Verbal intelligence -.02 -.03 .0004 -.04 .04 .001 -.02 -.02 - .001 -.10 -.02 -.002
Linear age -.08 -1.30* -.01 .07 .39 .004 -.06 .37 .01 .06 .17 .007
Quadratic age -.04 1.25* .0001 .06 -.38 .00 -.06 -.47 - .0001 .05 -.21 .00
Overall K2 .20 .14 .17 .28
Regression, F(13, 129) 2.50** 1.60t 2.00* 4.00**

Note. NA = negative affect; PA = positive affect. Being female, being blue-collar, and being African American are represented by larger values.
tp < .10 (marginally significant). *p < .05. ** p < .01.

Finally, we looked specifically at age differences in the intensity feels more positive or less negative affect than usual) and the
of excitement. Contrary to our hypothesis of reduced intensity of cessation of undesirable emotional states (defined as states where
excitement given the literature reviewed above, once again we the person is feeling less positive or more negative than usual). The
failed to reveal a significant relationship between age and the former reflect adaptive aspects of emotional stability, and the latter
intensity with which excitement was experienced, B = — .001 J3 = reflect adaptive aspects of emotional lability. To test Hypothesis 3,
.08,1(182) = 1.1, p > .25. we computed four scores that reflect these four aspects of emotion
regulation.
Hypothesis 3: Compared With Younger People, Older For each sampling occasion, participants were classified as high
People Better Regulate Emotional Experience on positive affect relative to their own idiosyncratic ally calculated
We operationalized emotion regulation as the maintenance of mean across all sampled situations or as below or equivalent to
desirable emotional states (defined as those where the individual their own idiosyncratically calculated mean across situations. A

.341

£ .32
ft
M .30

S-28
"8
g .26

u.

I
18-34 35-64 65-94
Age in Years

Figure 1, Frequency of negative affect across the life span.

 650 CARSTENSEN, PASUPATHI, MAYR, AND NESSELROADE

similar split was made for negative emotional states. We then emotion and for negative emotion. This is equivalent to a cross-
computed four conditional probabilities: (a) maintaining high pos- lagged correlation but applied to the categorical states (above or
itive states: the probability that, given participants were more below one's idiosyncratic mean) that we defined for this analysis.
positive than average on Occasion 1, they would be more positive These correlations arc instructive. For positive emotions, small
than average on the subsequent sampling occasion; (b) maintain- age-associated increases in stability (whether of low or high pos-
ing the absence of highly negative states: the probability that, itive states) are evident (r = .17, p < .02). This is not the case for
given participants were less negative than average on Occasion 1, negative emotions (r = .11, p > .15).
they would be less negative than average on the subsequent occa- Finally, the curvilinear pattern of results obtained for lability in
sion; (c) moving from low positive states to high positive states: the highly negative states with age raises the possibility that this kind
probability that, given participants were less positive than average of lability accounts for the age differences observed in the fre-
at Occasion 1, they would be more positive than average at quency of negative emotions. Note that this is not necessarily the
Occasion 2; and (d) moving from highly negative states to low case. Older and younger adults could have different frequencies of
negative states: the probability that, given participants were more negative emotions without those emotions occurring in temporally
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

negative than average at Occasion 1, they would be less negative linked ways. However, if older adults are moving out of negative
This document is copyrighted by the American Psychological Association or one of its allied publishers.

than average on the following occasion. Note that these scores are states more quickly than young adults (at least up to some point in
not necessarily related to the overall frequencies of experiencing
adulthood), this might mean that older adults' better emotion
positive or negative emotions, because these scores reflect some-
regulation, as assessed here, accounts for age differences in the
thing about the temporal distribution of positive and negative
frequency of negative emotions. We examined this by computing
states. Also, note that these scores do not simply reflect stability of
a hierarchical regression predicting the frequency of negative
positive or negative states, but rather, adaptive features of stability
emotion (from Hypothesis 1 above). The probability of moving
and adaptive features of lability.
from highly negative to low-negative states was entered as the first
Age-related patterns were reasonably positive. Older men and predictor, after which we examined whether age (linear and qua-
women showed greater stability of highly positive states (r = .17, dratic) made any additional significant contribution to the equa-
p < .05), and this pattern was consistent across all gender, eth- tion. This analysis showed that once lability of highly negative
nicity, and socioeconoraic groups. Age was also correlated with states was entered, age (both linear and quadratic) contributed an
stability of low negative states (r — .20, p < .001). Age was additional 2% of the variance to predicting the frequency of
negatively, though not significantly, correlated with moving from negative emotion F(2, 179) = 2.3, p = .10. Thus, changes in the
a low positive to a high positive state (r = —.13, p < .11). There frequency of negative emotion with age may be interpreted as
were no quadratic trends for age and these effects were constant stemming from changes in the lability of highly negative states.
across genders, ethnicities, and socioeconomic status groups. Fi-
nally, age was uncorrelated with the likelihood of moving from a
highly negative state to a low negative state (r = .04, ns). How- Hypothesis 4: Age Is Associated With the Complexity of
ever, here a quadratic trend was present. Together, linear age (B = Emotional Experience
.014, 0 = 1.47, p < .002) and quadratic age (B - -.0001, /3 =
—1.44, p < .002) accounted for 6% of the variance in the likeli- To test Hypothesis 4, we computed the eigenvalues of each
hood of moving from high negative states to low negative states. individual's 19 X 19 emotion ratings correlation matrix on the
This pattern was consistent across both ethnicities and socioeco- basis of his or her 35 occasions of measurement. We took as an
nomic status groups, but did appear to vary by gender (Linear index of differentiation the number of eigenvalues greater than 1.
Age X Gender interaction: A* 2 = .03, &F = 6.7, p < .02). Across the whole sample, the average number of eigenvalues
Examining the regression equation separately for men and women exceeding 1.0 was 5.8 (SD = 1.1, range = 2 to 9). The corre-
revealed that the pattern of age effects (linear and quadratic) was sponding principal components accounted for, on average, 77% of
identiqal for both genders, but the magnitude was lower for women the total variance in emotional ratings across time (SD ~ 3.9). The
(overall R2 = .0(5) than for men (overall J?2 = .13). In both cases, amount of variance accounted for by the principal components
however, the age regression attained statistical significance. To get with eigenvalues larger than unity was uncorrelated with age (r =
a feel for what these results suggest, we again computed correla- .09). The correlation between age and the number of eigenvalues
tions between age and the lability of highly negative states sepa- larger than 1.0 was, as predicted, positive and significant (r = .28,
rately for those under 60 (r = .30, p < .002) and those over 60 p < .01) and is shown in Figure 2. This evidence of age-related
(r = — .15, p > .19). Thus, older adults in our sample, up to some differentiation held across all levels of ethnicity, gender, and
age, were less likely to remain in a highly negative state over socioeconomic class. There was no quadratic trend for age. The
occasions than were younger adults. At some point in adulthood, relationship between differentiation and age was not accounted for
this trend reverses, although perhaps not significantly,4 by individual difference variables of personality, health, or verbal
fluency.
To summarize, then, older adults were more likely to maintain
highly positive states and were more likely to maintain the absence
of negative emotional states. Thus, the stability indicators suggest 4
Again, if these correlations are computed for cutoff points of 40,
greater stability of emotional experience in older adults for the 50, 60, and 65, the respective correlations in the younger portion of the
adaptive portions of emotional stability (i.e., one could also be sample are .23, .31, .30, and .26. The respective correlations in the older
very stable by being continuously very angry). This raises the sample are —.14, —.15, —.15, and —.17. Once again, the pattern is clear:
question of whether older adults are simply more stable in general. Highly negative states are increasingly labile across adulthood, but at some
This can be examined by computing a phi correlation for positive point, this increase levels off or becomes negative.
 EMOTION AND AGE 651
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

18-34 35-64 65-94

Age in Years

Figure 2. Differentiation of emotional experience over time.

Because our operational measure of differentiation represented a ative affect across the 35 sampling occasions. On average, this
novel way of thinking about qualitative aspects of emotional correlation was - . 3 5 (SD = .33), suggesting that positive and
experience, it was unclear how greater differentiation was associ- negative affect tended not to be present on the same occasion. A
ated with mental health. For this reason, we conducted exploratory linear-age effect emerged, with older age associated with the
correlation analyses about the relation of our complexity measures greater potential for co-occurrence of positive and negative emo-
with other variables. We focused on three questions. First, is tions, (r = .26; p < .01). This effect is shown in Figure 3. For
differentiation associated with better mental health? In other people under age 60, the average correlation between positive and
words, is it meaningfully related to indicators aside from the negative emotion within occasions was -.42 (SD = .28), whereas
emotion sampling data on which complexity scores were derived? for those above age 60, that correlation was - .25 (SD = .36). Just
Second, how does complexity relate to the emotional experience as in the above analyses, these age differences remained after
measures (frequency and intensity of negative and positive affect)? controlling for personality, health, and verbal fluency. Age differ-
Third, is complexity meaningfully related to other conceptually ences in poignancy also held within race, gender, and socioeco-
relevant measures, specifically, verbal fluency and neuroticism? nomic classes. Poignancy was unrelated to the frequency of expe-
Verbal fluency especially may be related to individuals' capacity riencing positive or negative affect, suggesting that individuals
to represent situations in complex ways. Neuroticism, on the other who frequently endorse all emotions are not more likely to expe-
hand, is a diffuse tendency to perceive life negatively. rience negative and positive emotions in the same moment. Fi-
Differentiation was uncorrelated with overall mental health (r = nally, differentiation and poignancy were correlated, (r = .23, p <
-.07). Differentiation was negatively associated with the fre- .01), but differentiation did not account for the age and poignancy
quency of experiencing both negative affect (r = —.30, p < .01) relationship, nor did poignancy account for the age and differen-
and positive affect (r — —.22, p < .01). Differentiation was not tiation relationship, suggesting these are separate aspects of emo-
associated with the intensity of positive affect (r = —.06) but was tional complexity.
negatively correlated with the intensity of negative affect (r =
-.16, p < .05). It was unrelated to verbal fluency (r = -.03) but
Discussion
was negatively associated with neuroticism (r = — .19, p < .05).
In sum, differentiation appears to be a positive feature of emotional To the best of our knowledge, this project represents the first
experience, as it is associated with greater emotional control and experience-sampling study of emotion based on a cross-sectional
less intense negative affect and with less neuroticism. sample that spans most of the adult age range. Across a 76-year
In addition to this factor-based indicator of emotional complex- span, encompassing participants between the ages of 18 and 94, a
ity, we computed a second analysis to examine the degree to which pattern of age differences emerged that is notably consistent with
individuals experienced both positive and negative emotions on positive reports based on survey data. Older people experience
the same sampling occasion. We refer to this feature of emotional positive emotions just as often as their younger counterparts and—
experience as "poignancy." Poignancy was computed by calculat- until the age of roughly 60 years—experience fewer negative
ing, for each participant, a correlation between positive and neg- emotions in their everyday lives. Of considerable importance is the
 652 CARSTENSEN, PASUPATHI, MAYR, AND NESSELROADE

00

•a -.3"
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

13

4>

SP
I -.5)
18-34 35-64 65-94
Age in Years
Figure 3. Poignancy with age: The correlation between positive and negative affect across the life span.

finding that age does not diminish the subjective intensity of meaningful goals in the face of limited time (Fredrickson &
positive or negative emotions in everyday life. In other words, Carstensen, 1990; Fung, Carstensen, & Lutz, 1999). The theoret-
when older people experience emotions, they are felt just as ical assertion that the pursuit of such goals results in a richer, more
intensely as in younger people, addressing an important piece of complex, and overall deeply satisfying emotional experience has
the descriptive puzzle representing emotion in the second half of not been tested previously.
life. The first two hypotheses we tested concerned the intensity and
Findings from this study are especially compelling in that emo- frequency of positive and negative emotional experiences in sub-
tions were sampled over an entire week while participants went jects of different ages. We postulated that the intensity of felt
about their daily lives. Whereas most previous reports about age emotions would be comparable for young and old adults but that
differences in emotion—including one's own—were based on older adults would experience negative emotions less frequently
global judgments about emotional functioning or, alternatively, than their younger counterparts. Our prediction about the fre-
were obtained under tighly controlled laboratory conditions, the quency of positive emotional experience was more tentative.
experience sampling procedure represents emotional experience in Changes in the frequency of positive emotions are not necessarily
everyday life. Furthermore, our sample was diverse, and with few predicted by our theoretical model and the literature is equivocal
exceptions, the pattern of results held for both African and Euro- on this point. Survey studies have found both greater (Mroczek &
pean Americans, men and women, and blue- and white-collar Kolarz, 1998) and lesser (Rossi & Rossi, 1990) happiness among
workers. The consistency of findings obtained in this project thus older adults. Our experience-sampling findings reveal no age dif-
relieves the persistent concern that the rosy picture of emotional ferences in positive emotional experience. We point out, however,
aging is simply the product of cognitive distortions or implicit that the nonsignificant correlation between positive affect and age
theories about emotional aging held by a select group of middle- in our data set (r = .07) is not meaningfully different from the
class adults. significant correlation (r = .10), based on a much larger sample,
The primary theoretical basis for the research was socioemo- reported by Mroczek and Kolarz (1998). Both studies find very
tional selectivity theory, the fundamental tenet of which is that slight increases in positive emotion in older people. Thus, our
boundaries on time lead to the prioritizing of emotional goals findings weigh in on the side of very small but reliable increase in
(Carstensen, 1993, 1995, 1998). Because chronological age is the frequency of positive emotional experience among older
related to time left in life, as people move through adulthood, they adults. Moreover, our findings certainly speak against a diminish-
grow keenly aware of the fragility of the human circumstance. ment in positive emotional experience found in some studies.
When they do, the pursuit of emotionally satisfying goals takes Measurement instruments may account for discrepancies in the
precedence over the pursuit of other goals. Most of the research literature. Age-related reductions in positive affect tend to be
testing postulates from socioemotional selectivity theory has ex- found when positive emotions are operationalized in terms of
amined goal selection. Over the years, this line of research has surgency and excitability. Four out of the five questions tapping
generated considerable evidence that people prefer emotionally positive emotions included on the Bradburn Affect Balance Scale
 EMOTION AND AGE 653

(Bradburn, 1969), for example, ask about high energy emotions ingly smaller at older ages. Although very preliminary, such find-
(e.g., feeling "on top of the world" or excited). None of the items ings support the utility of applying intraindividual approaches to
tap happiness, contentment, or joy. Because of this, we predicted the study of emotion.
excitement would be less frequent in older adults. Although we did Regarding the differentiation findings, there are many possible
not find support for this hypothesis, that is, frequency of excite- interpretations of increased factorial complexity in individuals'
ment did not distinguish older and younger adults, we suspect that emotional experiences. Among Ihem is the idea that older adults'
it may have been due to the relatively infrequent endorsement of emotional experience is less coherent, thereby producing more
excitement in the sample. Thus, interesting questions about the factors that are essentially meaningless. There are several reasons
frequency and the character of positive emotional experience in that we believe the findings reflect greater differentiation rather
old age remain. than incoherence. First, the solutions accounted for approximately
The hypothesis that negative emotional experience is less fre- the same reasonably high amount of variance across all ages.
quent among older people was strongly supported. Our findings, Second, the variability in older adults' emotions was, in general,
along with previous reports in the literature, clearly show that early lesser than younger people, making it doubtful that the amount of
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

adulthood is the life stage in which negative affect is most fre- error in the reports somehow increased over the week. For exam-
This document is copyrighted by the American Psychological Association or one of its allied publishers.

quent. In this adulthood sample, both linear and quadratic func- ple, if one calculates standard deviations across time for each
tions were revealed. There was a steady decrease in the frequency emotion, within individuals (since this is the thing at stake for the
of negative emotion from 18 to roughly 60 years, at which point within-person factor analyses), and then averages to obtain the
the linear decline stopped and a very slight nonsignificant upward average variability for negative and positive emotions, the result-
trend appeared. Although, on the one hand, we do not want to ing variabilities for negative emotion correlated at - . 2 6 with age,
overinterpret a nonsignificant trend, on the other hand, our rela- and positive emotions correlated at —.29 with age (ps < .01).
tively small sample size may not have allowed detection of a These age decreases in variability hold even when controlling for
significant trend. In any event, the correlation of negative emotions the average level of endorsement of positive and negative emo-
and age in people over 60 was slight (r = .12). Not even at the tions. Third, the negative correlations between the number of
oldest ages did the frequency of negative emotions approach that factors obtained and the frequency of endorsing positive emotions
observed in younger adults. A quadratic function observed in a or negative emotions imply a more differentiated and specific
sample spanning adulthood may help to reconcile differences in endorsement of emotions on any given occasion.
published reports. The sample studied in Mroczek and Kolarz's The overall profile of findings generated in this study are
(1998) report, for example, included people only up to 74 years of consistent with the most classic of developmental hypotheses:
age. Smith and Baltes (1997), in contrast, found increasingly Development brings increasing differentiation. We found greater
negative affective status with age, but their sample consisted differentiation in emotional experience in older as compared with
entirely of older people and encompassed participants between the younger people, and we found that emotional differentiation is
ages of 64 and 103 years. By focusing on different periods in related to a positive profile of characteristics, including less neu-
adulthood, researchers may be tapping different parts of a non- roticism and better emotional control. Such findings about emotion
linear course (Baltes, 1998). are especially important given widely documented decrements
The pattern of findings about the temporal stability of emotion in cognitive and biological aging and conclusions that de-
is consistent with reports of improved emotion regulation with age, differentiation better characterizes development in adulthood
and interestingly these findings are linear, suggesting increasingly (Salthouse, Hancock, Meinz, & Hambrick, 1996). Findings from
better emotion regulation into the oldest ages we studied. To the this study stand in stark contrast. At the same time in life when
best of our knowledge, these data are the first to address emotion cognitive speed and biological hardiness are on the decline, emo-
regulation without asking participants for judgments about their tional functioning may continue to improve.
ability to control their emotions. Thus, circumventing the potential To recapitulate, socioemotional selectivity theory maintains that
subjective bias in global reports, the experience sampling data boundaries on time imposed by human mortality elicit complex
provide support for self-reported improvement in emotion regula- emotional reactions in later life that are better characterized by
tion with age. Older adults were not simply more emotionally poignancy than happiness. People realize not only what they have
stable in day-to-day life. Rather, highly positive emotions were but also that what they have cannot last forever. A good-bye kiss
stable, whereas highly negative emotions were unstable. Although to a spouse at the age of 85, for example, may elicit far more
these data cannot address questions about the mechanisms used to differentiated and complex emotional responses than a similar kiss
influence stability, they strongly suggest that investigation into to a spouse at the age of 20. The theory maintains that emotions are
potential age differences in regulatory processes is an important deeper and more complex as the end of life nears because life's
direction for future research. fragility comes fully into awareness. Some forms of positive
To the best of our knowledge, our tests of emotional complex- emotions may decline, yet emotional satisfaction does not, nor
ity—namely, differentiation and poignancy—are entirely new and, does negative affect become predominant. If, as socioemotional
as such, must be interpreted conservatively. In some ways, how- selectivity theory asserts, such effects are related to approaching
ever, they provide the most original contribution of the research the end of life more than chronological age per se, the theory offers
and the theoretical model. Two findings emerge: (a) More dimen- an explanation for the apparent paradox that people suffering from
sions are required to reflect the structure of older as compared with terminal diseases often describe life as better than ever before
younger individuals' emotions, and (b) although the average cor- (Taylor & Brown, 1988; Taylor et al., 1992).
relation between positive and negative affect within a sampled Although any and all claims about age change based on cross-
moment is negative at all ages, the negative correlation is increas- sectional designs must be tempered accordingly, our findings
 654 CARSTENSEN, PASUPATHI, MAYR, AND NESSELROADE

contribute to a remarkably reliable pattern of findings in the How old age and anticipated endings make us more selective. Psychol-
literature that suggest that emotional functioning remains vital in ogy and Aging, 5, 335-347.
the second half of life. The entire array of emotions are experi- Frijda, N. H. (1988). The emotions. Cambridge, England: Cambridge
enced in the later years, but negative emotions are experienced less University Press.
frequently and are better controlled, and emotional experience is Fung, H., Carstensen, L. L., & Lutz, A. (1999). The influence of time on
social preferences: Implications for life-span development. Psychology
more complex. More so than younger adults, older adults also
and Aging, 14, 595-604.
appear to experience complex mixes of emotions, and positive and
Gross, J. J., Carstensen, L. L., Pasupathi, M., Tsai, J., GotesUm-Skorpen,
negative experiences are less independent in older adults.
C , & Hsu, A. (1997). Emotion and aging: Experience, expression, and
control. Psychology and Aging, 12, 590-599.
Herzog, A. R., & Rodgers, W. L. (1981). Age and satisfaction: Data from
References
several large surveys. Research on Aging, 3, 142-165.
Baltes, M. M. (1998). The psychology of the oldest old: The fourth age. Isaacowitz, D., Charles, S., & Carstensen, L. L. (1999). Emotion and
Current Opinion in Psychiatry, 11, 411-415. cognition. In G. Craik & T. Salthouse (Eds.), Handbook of aging and
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.

cognition (2nd ed., pp. 593-631). Mahwah, NJ: Erlbaum.

This document is copyrighted by the American Psychological Association or one of its allied publishers.

Blanchard-Fields, F. (1997). The role of emotion in social cognition across

the adult life span. In K. W. Schaie & M. P. Lawton (Eds.), Annual John, O. P., Donahue, E. M., & Kentle, R. L. (1991). The Big Five
review of gerontology and geriatrics (Vol. 17, pp. 238-265). New York: Inventory: Versions 4a and54 [Technical Report]. Berkeley: University
Springer. of California, Institute of Personality and Social Research.
Bradburn, N. M. (1969). The structure of psychological well-being. Chi- Labouvie-Vief, G., & DeVoe, M. (1991). Emotional regulation in adult-
cago: Aldine. hood and later life: A developmental view. In K. W. Schaie (Ed.),
Brodman, K., Erdmann, A. J., Jr., & Wolff, H. G. (1949). Cornell Medical Annual review of gerontology and geriatrics (pp. 172-194). New York:
Index Health Questionnaire Manual. Ithaca, NY: Cornell University Springer.
Medical College. Labouvie-Vief, G., DeVoe, M., & Bulka, D. (1989). Speaking about
Bromley, D. B. (1990). Behavioral gerontology: Central issues in the feelings: Conceptions of emotion across the life span. Psychology and
psychology of aging. New York: Wiley. Aging, 4, 425-437.
Camras, L., Sullivan, J., & Michel, G. (1993). Do infants express discrete Labouvie-Vief, G., Hakim-Larson, J., DeVoe, M., & Schoeberlein, S.
emotions? Adult judgments of facial, vocal, and body actions. Journal of (1989). Emotions and self-regulation: A life span view. Human Devel-
Nonverbal Behavior, 17, 171-186. opment, 32, 279-299.
Carstensen, L. L. (1993). Motivation for social contact across the life span: Lawton, M. P., Kleban, M. H., & Dean, J. (1993). Affect and age:
A theory of socioemotionai selectivity. In J. E. Jacobs (Ed.), Nebraska Cross-sectional comparisons of structure and prevalence. Psychology
Symposium on Motivation (pp. 209-254). Lincoln: University of Ne- and Aging, 8, 165-175.
braska Press. Lazarus, R. S. (1991). Emotion and adaptation. New York: Oxford Uni-
Carstensen, L. L. (1995). Evidence for a life-span theory of socioemo- versity Press.
tionai selectivity. Current Directions in Psychological Science, 4, Levenson, R. W., Carstensen, L. L., Friesen, W. V., & Ekman, P. (1991).
151-156. Emotion, physiology, and expression in old age. Psychology and Ag-
Carstensen, L. L. (1998). A life-span approach to social motivation. In ing, 6, 28-35.
J. Heckhausen & C. Dweck (Eds.), Motivation and self-regulation
Levine, L., & Bluck, S. (1997). Experienced and remembered emotional
across the life span (pp. 341-364). New York: Cambridge University
intensity in older adults. Psychology and Aging, 12, 514-523.
Press.
Lindenberger, U., Mayr, U., &. Kliegl, R. (1993). Speed and intelligence in
Carstensen, L. L., & Charles, S. T. (1999). Emotion in the second half of
old age. Psychology and Aging, 8, 207-220.
life. Current Directions in Psychological Science, 7, 144-149.
Malatesta, C. Z., & Kalnok, M. (1984). Emotional experience in younger
Carstensen, L. L., & Fredrickson, B. F. (1998). Socioemotionai selectivity
and older adults. Journal of Gerontology, 39, 301-308.
in healthy older people and younger people living with the Human
McCrae, R. R., & Costa, P. T. (1991). Adding Liebe and Arbeit: The full
Immunodeficiency Virus: The centrality of emotion when the future is
constrained. Health Psychology, 17, 1-10. five-factor model and well-being. Personality and Social Psychology
Bulletin, 17, 227-232.
Carstensen, L. L., Gross, J., & Fung, H. (1997). The social context of
emotion. In M. P. Lawton & K. W, Schaie (Eds.), Annual review of McFarland, C , Ross, M., & Giltrow, M. (1992). Biased recollections in
geriatrics and gerontology (pp. 325-352). New York: Springer. older adults: The role of implicit theories of aging. Journal of Person-
Carstensen, L. L., Isaacowitz, D., & Charles, S. T. (1999). Taking time ality and Social Psychology, 62, 837-850.
seriously: A theory of socioemotionai selectivity. American Psycholo- Mroczek, D., & Kolarz, C. M. (1998). The effect of age on positive and
gist, 54, 165-181. negative affect: A developmental perspective on happiness. Journal of
Carstensen, L. L., & Turk-Charles, S. (1994). The salience of emotion Personality and Social Psychology, 75, 1333-1349.
across the adult life course. Psychology and Aging, 9, 259-264. Rossi, A. S., & Rossi, P. H. (1990). Of human bonding: Parent-child
Diehl, M., Coyle, N., Labovie-Vief, G. (1996). Age and sex differences in relations across the life course. New York: Aldine de Gniyter.
strategies of coping and defense across the life-span. Psychology and Rothbart, M., & Ahadi, S. A. (1994). Temperament and the development
Aging, 11, 127-139. of personality. Journal of Abnormal Psychology, 103, 55-66.
Diener, E., & Diener, C. (1996). Most people are happy. Psychological Salthouse, T. A., Hancock, H. E., Meinz, E. J., & Hambrick, D. Z. (1996).
Science, 7, 181-185. Interrelations of age, visual acuity, and cognitive functioning. Journal of
Elder, G., Odum, H., & Hareven, T. (1994). Rising above life's disad- Gerontology: Psychological Sciences, 51, 317-330.
vantage: From the Great Depression to war. In G. H. Elder Jr., J. Schimmack, U., & Diener, E. (1997). Affect intensity: Separating intensity
Modell, & R. E. Parke (Eds.), Children in time and place: Develop- and frequency in repeatedly measured affect. Journal of Personality and
mental and historical insights (pp. 47-72). New York: Cambridge Social Psychology, 73, 1313-1329.
University Press. Schwarz, R , Park, D., Knaueper, B., & Sudman, S. (Eds.). (1999). Cog-
Fredrickson, B. F., & Carstensen, L. L. (1990). Choosing social partners: nition, aging, and self-reports. New York: Erlbaum, Taylor & Francis.
 EMOTION AND AGE 655

Smith, J., & Baltes, P. B. (1997). Profiles of psychological functioning in Tsai, J., Levenson, R. W., & Carstensen, L. L. (in press). Autonomic,
the old and oldest old. Psychology and Aging, 12, 458-472. expressive, and subjective responses to emotional films in younger and
Tabachnik, B. G-, & Fidell, L. S. (1989). Using muttivariate statistics (2nd older adults of European American and Chinese descent. Psychology
ed.)- New York: Harper Collins. and Aging.
Taylor, S., & Brown, J. (1988). Elusion and well-being: A social psycho- Watson, D., & Pennebaker, J. W. (1989). Health complaints, stress, and
logical perspective on mental health. Psychological Bulletin, 103, 193— distress; Exploring the central role of negative affectivity. Psychological
210. Review, 96, 234-254.
Taylor, S., Kcmeny, M. E., Aspinwall, L. G., Schneider, S. G., Rodriguez,
R., & Herbert, M. (1992). Optimism, coping, psychological distress, and
high-risk sexual behavior among men at risk for acquired immunodefi- Received March 15, 1999
ciency syndrome (AIDS). Journal of Personality and Social Psychol- Revision received February 2, 2000
ogy, 63, 460-473. Accepted February 18, 2000
This article is intended solely for the personal use of the individual user and is not to be disseminated broadly.
This document is copyrighted by the American Psychological Association or one of its allied publishers.

Call for Nominations

The Publications and Communications Board has opened nominations for the
editorships of Journal of Applied Psychology, Journal of Consulting and Clinical
Psychology, Journal of Educational Psychology, Psychological Bulletin, and Journal
of Personality and Social Psychology: Interpersonal Relations and Group Processes
for the years 2003-2008. Kevin R. Murphy, PhD, Philip C. Kendall, PhD, Michael
Pressley, PhD, Nancy Eisenberg, PhD, and Chester A. Insko, PhD, respectively, are the
incumbent editors.

Candidates should be members of APA and should be available to start receiving

manuscripts in early 2002 to prepare for issues published in 2003. Please note that the
P&C Board encourages participation by members of underrepresented groups in the
publication process and would particularly welcome such nominees. Self-nominations
are also encouraged.

To nominate candidates, prepare a statement of one page or less in support of

each candidate and send to

• Margaret B. Spencer, PhD, for the Journal of Applied Psychology

• Donna M. Gelfand, PhD, and Lucia Albino Gilbert, PhD, for the Journal of Con-
sulting and Clinical Psychology
• Lauren B. Resnick, PhD, for the Journal of Educational Psychology
• Janet Shibley Hyde, PhD, and Randi C. Martin, PhD, for Psychological Bulletin
• Sara B. Kiesler, PhD, for JPSP: Interpersonal Relations and Group Processes

Address all nominations to the appropriate search committee at the following

address:

[Name of journal] Search Committee

c/o Karen Sellman, P&C Board Search Liaison
Room 2004
American Psychological Association
750 First Street, NE
Washington, DC 20002-4242

The first review of nominations will begin December 11,2000.