Emotion

Most of the research on continuities and change in emotional patterns during the adult years shows that the fundamental properties of the emotions remain relatively stable. However, there are also subtle changes.

Physiological patterns. The emotions are linked to certain anatomical sites in the brain such as the amgydala and to neurological and neurochemical processes that govern the autonomic nervous system. The chief way that psychologists have studied the physiological aspects of emotion involves monitoring the autonomic nervous system during states of emotional arousal. Although there has been a good deal of work on the physiological aspects of emotion during the latter decades of the twentieth century, research on adult development and aging is particularly meager. In one study that is relevant (Levenson, Carstensen, Friesen, and Ekman, 1991), participants were asked to participate in two emotion evoking conditions while their skin conductance, heart rate, and other physiological measures were monitored: to recollect and relive salient emotional events, and to make facial expressions of emotion. The researchers found that older people showed the same emotion-specific pattern of response as younger individuals, though the magnitude of the response was less pronounced in older subjects. That is, older people showed smaller increases in heart rate for anger, fear, and sadness. Nevertheless, the older participants reported the same degree of subjective emotional experience. In another relevant study (Levenson, Carstensen, and Gottman, 1994), where younger and older couples were asked to talk about neutral, problem, and pleasant situations with their spouses, the older couples had smaller heart rate increases and greater lengthening of pulse transmission time to ear, indicating less physiological reactivity.

A much larger body of literature has examined whether emotion inhibition—the deliberate attempt to suppress emotional feelings or expressions—alters physiological reactivity in adults. The literature indicates that emotionally inexpressive individuals are more physiologically reactive than expressive persons, though this does not necessarily mean that inexpressive persons deliberately inhibit their emotions. There is also literature indicating that the chronic inhibition of some emotions, particularly anger or hostility, is linked to essential hypertension and coronary heart disease. However, these correlational studies do not necessarily mean that emotion inhibition causes disease. Emotion-expressive or inhibitory patterns and health may both be influenced by a third factor such as genetic background. Experimental work helps clarify the pattern. Indeed, laboratory studies in which participants are asked to deliberately suppress their emotions while physiological measures are monitored, have shown that inhibition is associated with a number of changes in physiological measures and that the changes are emotion specific. For example, the inhibition of the visible expression of disgust causes an increase in parasympathetic arousal, as indexed by increased skin conductance and decreased finger pulse amplitude; in contrast, suppression of sadness expressions leads to cardiovascular activation. These findings have been obtained with college-age research participants. There are as yet no experimental investigations examining the effects of emotion inhibition in older respondents. Nor is it known what the consequence might be of a lifetime pattern of emotion inhibition, though it has been argued that if inhibition requires a form of "work," then prolonged or repeated efforts at inhibition may cause wear and tear on bodily systems and result in disease.

Expressive patterns. There is a more substantial body of literature concerning expressive behavior, almost all of it having to do with facial expressions of emotion; this literature goes beyond the study of the association between facial expressions and physiological reactions. For the most part, the literature on facial expressions has centered on changes in the expression of emotion with development as well as changes in sensitivity to the perception of emotion in others.

Changes in emotion expression. In early development, facial expressions of emotion, which are innate and gradually unfold during the opening months of life, become more modulated and take on the characteristics of family and culture. In general, expressive behavior becomes more conventionalized, and this includes a dampening of expressive behavior; these changes are well in place by adolescence. But there are also changes during the adult years. For example, one study showed that when emotional memories were prompted and subjects asked to relate their experiences, older adults were more facially expressive in terms of the frequency of emotional expressions than younger individuals across a range of emotions, as detected by an objective facial affect coding system (Malalesta-Magai, Jonas, Shephard, and Culver, 1992). In another study, researchers also found that the expressions of older adults (women in this case) were more telegraphic in the sense that their expressive behaviors tended to involve fewer regions of the face, and yet more complex in that they used more blended or mixed expressions when recounting emotional events (Malatesta and Izard, 1984). These changes, in part, account for why the facial expressions of older adults are more difficult to read.

One of the other changes that comes with age that make older faces more difficult to read involves the wrinkling of the facial skin and the sag of facial musculature. What is particularly interesting is that the pattern of wrinkling is so different across individuals. Of course, part of this is due to biologically based aspects of aging, but individual differences also appear linked to personality process. One study (Malatesta, Fiore, and Messina, 1987) was able to show that facial patterns were related to dominant personality characteristics. For example, persons who tended to experience contempt frequently, had facial expressions that looked contempt-like, even when they were attempting to pose other kinds of emotion—that is, the contempt that was dominant in their personalities tended to leak through and contaminate the clarity of the posed expressions.

Other studies have looked at expressive behaviors in late life and under conditions of dementia. These studies, using observational methods and objective coding of facial and bodily expressions, indicate that for the most part the expressive behavior of dementia patients resembles that of nondemented older adults (Magai, Cohen, Gomberg, Malatesta, and Culver, 1996). The entire range of basic emotion expressions can be detected, and recognizable facial expressions of emotion are observed across the spectrum of cognitive deterioration, including the most severely impaired, end-stage patients, although there are changes with advancing intellectual impairment. Of note, there is a lower frequency and duration of happiness or pleasure, especially in the later stages of the disease. These findings indicate that the ability to express affect, as observed in interpersonal contexts, remains intact during the course of one form of debilitating disease, dementia; in fact, sensitivity to affect encoded in nonverbal communication may be even more acute in dementia patients since the neuronal fallout associated with dementia releases cortical inhibitory control of the centers that control emotion.

Emotion expressions seen in dementia patients appear linked to precipitating events in the environment rather than constituting random muscle movement patterns. For example, one study showed that as patients—all of whom were moderately to severely impaired—were observed during a family visit, their faces expressed sadness when their relatives prepared to leave (Malatesta et al., 1996).

Changes in sensitivity to the expressive behavior of others. A few studies have looked at potential age-related changes in the ability to read the expressive behavior of social partners. In one study (Malatesta, Izard, Culver, and Nicolich, 1987), where older and younger individuals were asked to identify negative and neutral facial expressions, the older adults were significantly less accurate than the younger participants. Similarly, in another study, where young, middle-aged, and older untrained "judges" attempted to label the facial expressions of other young, middle-aged, and older women, the older women did most poorly, but their performance was best for older faces. This suggests that contact with like-aged peers assists the process of understanding the emotional messages of interactants. Moreover, older individuals may have a greater facility in discounting the "noise" of facial wrinkling so as to discern the essential emotional messages of older social partners.

A handful of other studies have examined sensitivity to other nonverbal channels of emotion expression. One study found an overall decline in the ability of older adults to identify specific emotions from vocal indicators (Allen and Brosgole, 1993). In another case, where researchers examined age-related differences in the ability of adults to read emotion signals through body movements and gestures, the study found that while both younger and older adults made accurate emotion identifications well above chance levels, older adults made more errors overall. They were especially inaccurate with negative emotions, and were especially likely to misidentify emotional displays as being neutral in content (Montepare, Koff, Zaitchik, and Albert, 1999). It must be noted, however, that the actors who depicted the emotions were all young adults, and if age congruence between sender and receiver makes a difference, as one study suggested, this research will bear replication.

In summary, the data on the expression and perception of emotion across the life span indicates that there are developmental trends indicating increased complexity of facial expressions in terms of greater idiosyncratic responses, increased use of blended affect, and changes in facial musculature and skin elasticity, all of which make it more difficult to read the emotional expressions of adults as they age. However, this is compensated by familiarity with social peers. Moreover, since expressive behavior does not occur in a vacuum, but is typically accompanied by verbal expressions of emotion, the above findings do not suggest that older adults are at risk for misinterpretation of their feelings. Indeed, there is a small but substantive body of literature indicating that older adults up through middle age become more adept at describing their experiences than younger adults, and that their descriptions are more complex and nuanced; very old persons, however, seem to show a decline in this ability, though this may be a cohort effect, that is, one related to the historical epoch in which they were reared, rather than an aging effect per se. Persons born earlier in the twentieth century were born during less child-centered and permissive times and their parents were exposed to the exortations of John Watson, who remonstrated against the open expression of emotion and physical affection. In terms of the perception of emotion in others, the slim literature that exists suggests that there may be declines in the accuracy with which nonverbal emotion signals are understood, though age congruence between social partners may compensate.

Feeling states. Research in this area has covered the frequency and intensity of emotion states, as well as the quality of emotional feelings; the latter have included the salience of emotional stimuli and their ability to elicit emotional reactions, hedonic tone (pleasure/unpleasure), discrete emotions (anger, fear, sadness, etc.), and emotional complexity (emotional/cognitive interaction).

Intensity. Although there are conflicting data on changes in affective intensity with age, in general, studies have indicated that there is no substantive decline in affective intensity or the intensity with which a person experiences emotions. This pattern holds true when adults of different ages are asked to gauge whether the intensity of their emotions has changed over the years, when they report the intensity of their emotional experiences as sampled randomly throughout the day, or when they recount emotional experiences in the laboratory.

Hedonic tone. In terms of general hedonic tone, the bulk of the literature indicates that the frequency or level of positive affect either remains the same or increases over the adult years until very late in life, when there is a decline. In terms of negative affect, the most sophisticated studies seem to indicate that there is a gradual lessening of negative affect over the adult years up to approximately age sixty, after which there is a nonsignificant increase. The subtle changes in positive and negative affect during late life are likely linked to release from childrearing and work roles—in the case of gains in happiness— and chronic debilitating illnesses of old age as well as the deaths of important intimates such as spouses and friends—in the case of negative affect.

Discrete emotions. Investigators have also looked at changes in discrete emotions—that is, the various emotions that comprise positive affect and negative affect. This is the area of greatest disparity in findings, largely due to the use of different methodologies for assessing change and stability in emotion, and in the range and kinds of emotions sampled. However, one fairly consistent finding stands out. Across a range of studies, both cross-sectional and longitudinal in nature, results indicate that there is a decrease in the level and frequency of anger across the adult years. The reason for this may relate to the functional role of anger, which is to overcome obstacles and barriers to goals. To the extent that age brings with it changes in a reduction of blocked goals, its prevalence should decline. Research has also suggested that older adults shift from an external to internal strategy to control feelings associated with success and failure of goals. Rather than acting on the environment—expressing anger openly—they may use more internally controlled responses such as positive reframing or repression. Finally, as social networks narrow and older adults depend on a fewer, more intimate partners for their socioemotional needs, the expression of anger may pose a threat to these relationships.

In terms of emotional complexity, in memory tasks, narrative analyses of recollected emotional experiences, and in experience-sampling methodologies, research indicates that emotional experience becomes more complex across the adult years. In the experience-sampling work, older adults report more mixed and bittersweet emotions and more poignancy within the same sampled moment. Older adults also orient to the emotional significance of events more so than younger adults. For example, when older and younger participants were asked to recall narrative material that they had read, older individuals recalled more of the emotional versus the neutral material. In recounting emotional experiences, young adults rarely refer to inner subjective feelings and are likely to describe their experiences in terms of normative proscriptions; middle-aged adults, in contrast, are able to acknowledge complex feelings, are less influenced by conventional standards, and can sustain feelings of ambivalence and tension without needing to resolve them immediately. Moreover, research indicates that emotional complexity—the interaction between emotion and cognition and their integration—appears to improve with age, although the data also suggest that this trend peaks during middle age, after which there is a decline.

In summary, emotions become more important to individuals as they mature, they are better at tolerating mixed feelings, and there is a deepening appreciation of the emotional complexity of the world both within and around the self. The fact that these trends do not continue into old age, at least on the basis of this research, may be related to cohort effects and/or to changes in emotional regulatory strategies, as discussed in the next section.

Emotion regulation. Emotion regulation is the ability to modulate feeling states and emotional responses in reaction to emotional elicitors. People can moderate emotion by avoiding emotionally charged situations, by cognitive strategies such as denial and intellectualization, or by engaging in processes that permit the cognitive elaboration and integration of experience. The literature suggests that there is an age trend toward affective optimization or the tendency of persons to actively create environments that permit them to achieve a better mix of emotionally stimulating versus insulating features. Older adults also report greater abilities to control their emotions and to moderate affective responses. However, there are also distinct individual differences in terms of the tendency to dampen versus elaborate emotional experiences. One style, which is responsible for maintaining an affectively positive self core involves equilibrium-regulating strategy; negative affect is dampened, lessening the deviation from positive affect. In contrast, the other style, an affect-elaborating strategy, permits movement into a zone that is more chaotic or unstable; it tolerates and even seeks out a range of experience and is therefore more flexible and results in the ability to integrate positive and negative affect. There may be a preference for affect dampening strategies in old age, given greater physical frailty and reduced energy reserves. If so, this would explain why older persons seem oriented toward actively avoiding conflict. Another means by which older adults accomplish affective regulation involves the selective narrowing of social networks. Both cross-sectional and longitudinal studies show that adults shed more peripheral relationships in later life, while retaining those that are most intimate and emotionally gratifying.

In summary, the literature on emotion regulation indicates that there is increasing facility at emotion regulation with age. Some of this is achieved through an ability to tolerate and sustain ambiguity and mixedness of experience and to integrate positive and negative affect, while some is due to active avoidance of emotionally charged circumstance and the narrowing of social relationships.

Emotion in personality. As people mature their personalities become more stable or crystallized. That is, a young woman who is extroverted is likely to be extroverted in later life as well; a man who shows neurotic behavior in his thirties, is likely to show these same features in his fifties and sixties.

One personality dimension, sensitization/repression, relates to the way in which persons habitually attend to or avoid emotionally disturbing or threatening stimuli. Repressors tend to avoid or disattend to threatening stimuli or to route them from consciousness, whereas sensitizers monitor the environment for the presence of such stimuli. The literature indicates that these tendencies are relatively stable over short periods of time; less is known about the stability of these traits over longer stretches of the adult life course, however, sensitization is related to trait anxiety and thus stability for this characteristic may be quite strong, as indicated in the literature on trait emotion.

Many dimensions of personality have an affective aspect to them. Some theorists have even argued that personalities are organized around distinctive affective patterns or emotional traits. Indeed, on an informal basis, it is widely observed that some individuals are more hostile than others, some more somber, some more ebullient, and so forth. The empirical research on emotion traits has generated substantial support for this observation. These affective organizations create emotionally biased ways of processing the world that affect attention, perception, memory, judgment, and other cognitive processes, and also affect behavior and interpersonal process. Emotion traits for the broad dimensions of positive and negative affect, as well as the specific mood states of anxiety, sadness, and anger, among others, have shown substantial stability over months and even years. Such data suggest that emotional moods may represent enduring dispositions. However, these stability figures mask individual differences in change. Sophisticated modeling procedures indicate that there are substantial inter-individual differences in longitudinal change over a number of years, at least when it comes to broad personality traits such as neuroticism and extroversion; some people's trajectory remains level, others show increases, and still others show decline. Inter-individual differences in the pattern of personality change may as well hold for more circumscribed emotion traits or aspects of emotional organization, as suggested by a slim but growing literature.

Given the evidence of both stability and change, some commentators have suggested that one of the more challenging jobs for today's personality researchers is to account for the processes underlying personality change. Examination of the factors that support personality change have only begun, but two factors have emerged. There is some preliminary evidence that severe illness can precipitate personality change. Another major precipitant appears to involve emotionally charged events that involve interpersonal process. Events associated with change include psychotherapy, and participation in groups, workshops, or retreats, where intense affiliative relationships are experienced. One study (Magai, 1999) tracked changes in adults' emotion traits and their perspectives, goals, personality, feelings, and ways of relating to others over an eight-year period, using personality scales, as well as self-reports and observer reports. There was evidence of both stability and change. Emotion traits tended to remain stable, although individual differences were also in evidence. An aggregate measure of personality change in terms of ratings on feelings, attitudes, and goals, indicated that change was associated with positive and negative interpersonal life events including marriage, divorce, and death of loved ones. Change was unassociated with other emotionally charged high and low points in people's lives such as job losses, career advances, changes in residence, and more distant social relationships.

In summary, there is substantial support for the stability of emotion traits in personality when mean level or aggregate data are examined. When individual trajectories are examined, we find more evidence of inter-individual variation, with some persons evidencing no change, other showing increments or decrements. It is now clear that structural change in personality, including aspects related to emotional organization, is possible under certain circumstances. Experiences with acute or chronic illness is one such precipitant, and emotionally charged inter-personal encounters is another. However, a full description of the causes, the process, and the types of affective change have yet to be charted.