Emotions can be defined as psychological states that comprise thoughts and feelings, physiological changes, expressive behaviors, and inclinations to act. The precise combination of these elements varies from emotion to emotion, and emotions may or may not be accompanied by overt behaviors. This complex of states and behaviors is triggered by an event that is either experienced or recalled. Someone insults you. Depending on the nature of the insult and your perception of the extent to which it was or was not intended to hurt you, you might feel angry or annoyed. If you feel angry, your face may redden, your heart may beat faster, your fists clench, and thoughts of retribution occur to you. In some cases you might take action against the person who was insulting. Days later, recalling the insult may re-evoke at least some features of the original emotional reaction. Similarly, clear-cut cases of emotion could be given for fear, joy, love, disgust, and sadness, among many others. However, there are also emotions that are less clear-cut, in that they do not always involve changes in physiological or motivational states and do not always result in behavioral change. Take the example of regret. Having made a decision or taken a course of action that turns out badly, one may well feel strong regret, but this subjective experience will typically not be accompanied by changes in physiology or behavior.
Further complications arise when considering psychological states that seem to be borderline cases of emotion: physical pain, generalized or free-floating anxiety, sexual arousal, boredom, depression, irritability, all of which can be seen as examples of affective states. Psychologists who study emotion tend to distinguish between affective states that have a clear object and those that do not, arguing that emotion is a term that should be reserved for psychological states that have an object. On this basis, chronic pain, general states of boredom, depression, or irritability would not be classed as emotions, whereas sexual arousal—to the extent that it has a clear object— would be treated as an emotional state. The distinction between affective states that have an object and those that do not is one that separates emotions, on one hand, from moods (e.g., irritability, boredom) and affective dispositions (depression, generalized anxiety), on the other.
Recognizing the difficulties inherent in trying to arrive at watertight definitions of what constitutes an emotion, theorists are generally agreed in regarding emotion as a set of states that has a fuzzy boundary with other psychological states, such as beliefs, attitudes, values, moods, and personality dispositions. What is not in dispute is that the set of states called emotion is defined by good examples, such as anger, fear, and passionate love. Where there is room for doubt, at or near the fuzzy boundary with neighboring states, psychologists are generally unconcerned with whether the state in question is an emotion. The difficulty of defining emotion is thereby finessed.
Emotions History and Background
Modern emotion theory is usually traced back to the writings of Charles Darwin or William James. Writing in the second half of the 19th century, these authors focused on issues that are still the subject of research and debate nearly 150 years later. Darwin’s focus was on the relation between subjective emotion and overt behavior. He argued that three principles explain the relation between emotions and expressive behavior. Of these, the first, the principle of serviceable associated habits, is the one most commonly linked to Darwinian explanations for expressive behavior. Here the argument is that movements of the face that originally served a purpose during emotional experiences have become automatic accompaniments of those emotions. Thus, the frowning that often accompanies anger might help to protect the eye socket by drawing the brows forward and together, or the eye widening that often accompanies surprise might help to take in more visual information when sudden, novel events occur. Surprisingly, given the general theory of evolution for which Darwin is better known, his writings on emotional expression did not treat this expression as the outcome of a process of natural selection. Rather, he saw the emotion-expression link as a learned habit that then gets passed on to one’s progeny. However, modern evolutionary theory can readily be applied to this issue, resulting in the view that it was the adaptive significance for the individual or the group that led to emotions being outwardly expressed. The notion that there is a close relation between emotional experience and bodily expression is certainly one that is echoed in modern emotion theory.
James focused on the fundamental question of the determinants of emotion. James advocated what has come to be called a peripheral theory of emotion, in which he argued that the perception of an arousing stimulus causes changes in peripheral organs, such as the viscera (heart, lungs, stomach, etc.) and the voluntary muscles, and that emotion is quite simply the perception of these bodily changes. To use James’s own example, it is not that people tremble and run because they are afraid; rather, they are afraid because they tremble and run. This raises the question of how the bodily changes come about. Here James argued for a direct link between perception and bodily change, using the analogy of a lock and a key. The fit between the perception of emotion-arousing stimuli and the human mind is, in James’s view, such that the stimuli automatically unlock physiological changes in the body, and it is the perception of these changes that is the emotion. The idea that there is a close link between perception and emotion, relatively unmediated by conscious cognition, is still found in modern emotion theory, as is the notion that changes in the peripheral activity of the body results in changes in emotion.
A third major plank in the theoretical analysis of emotion in psychology came with the rise of cognitivism (i.e., close study of mental processes) in the 1960s. The first proponent of a view that came to be known as appraisal theory was Magda Arnold. She argued that what makes people experience an emotion is not bodily change, but rather the cognitive process that makes one kind of stimulus emotionally arousing while another kind of stimulus leaves people cold. The difference, she argued, is that the emotionally arousing stimulus is personally meaningful and matters to people. Unless the stimulus matters to people, they will not become emotional. Clearly, what matters to one person may leave another person cold. This emphasis on subjective meaning in appraisal theory led researchers to shift their attention from the objective properties of emotional stimuli to the subjective processes (appraisal processes) by which perceivers attach significance and meaning to stimuli. Modern emotion theory is very much concerned with this process of meaning making.
Notice that these three key sources of influence on modern emotion theory map rather neatly onto three of the supposed components of emotion: expression, physiological activity, and cognitions. Before examining each of these three components in greater detail, consider the connection between emotion and social psychology.
Social Psychology and Emotions
Emotions are a topic studied within many subdisciplines of psychology, including clinical psychology, biological psychology, and developmental psychology. Yet if one reviews the history of psychological theory and research on emotion, it is noticeable that social psychologists have played a prominent role. In one sense this is surprising. There are certainly emotional reactions that have little or nothing to do with the social world that is the primary concern of social psychologists: Think of fear of heights, of snakes, or of grizzly bears. Yet these emotions are not typical of the range of emotions that people experience in everyday life. As noted earlier, emotions are always about something: They have an object. This object is very often social. It is a person (a rival for your loved one’s affection), a social group (an organization that does inspiring work in developing countries), a social event (your favorite sports team winning a trophy), or a social or cultural artifact (a piece of music). It turns out that these social objects are much more likely than nonsocial objects to be the source of our everyday emotions.
Furthermore, many emotions are either inherently or functionally social, in the sense that they either would not be experienced in the absence of others or seem to have no other function than to bind people to other people. Emotions such as compassion, sympathy, maternal love, affection, and admiration are ones that depend on other people being physically or psychologically present. Fear of rejection, loneliness, embarrassment, guilt, shame, jealousy, and sexual attraction are emotions that seem to have as their primary function the seeking out or cementing of social relationships.
A final point concerning the link between emotion and social life is that when people experience emotions, they have a strong tendency to share them with others. In research on what is called the social sharing of emotion, investigators have shown that the overwhelming majority of emotional experiences are shared with others, are shared with several others, and are shared soon after the triggering event. Moreover, this sharing of emotion with others elicits emotional reactions in the listeners, which is itself an interesting phenomenon, depending as it does on the listener’s tendency to empathize with the sharer. And the emotions experienced by the listeners tend to be shared with third parties, a phenomenon called secondary social sharing. There is an interesting paradox here. People tend to share their emotional experiences, some of which may be painful or shaming, with intimates because they trust them not to share their secrets with others. And yet these intimates are the very ones who are likely to empathize with other people and therefore to experience emotions themselves as a result of listening to what others divulge. This makes it likely that they will engage in secondary social sharing.
These points make it clear that emotions are invariably social in nature: They are about social objects, their function seems to be social, and they have social consequences. A parallel point is that the subject matter of social psychology is invariably emotional in nature: Topics such as close relationships, aggression and hostility, altruism and helping behavior, prejudice and stereotyping, and attitudes and persuasion entail concepts and processes that are often explicitly emotional. In short, there is an intimate connection between emotion and social psychology, which in turn helps to account for the prominent role that social psychologists have made to emotion theory and research.
This section will now return to the three components of emotion identified earlier, namely, physiological changes, cognitions, and expressive behaviors, and review modern developments in research on each component.
Emotions and Physiological Change
The theory of emotion proposed by James, already referred to, is one that places physiological change at the center of emotion. As James put it, if people could imagine themselves perceiving an emotional stimulus without any accompanying bodily changes, the result would be a pale and colorless imitation of the real emotion. This seems correct in the case of emotions such as fear and anger: What would such experiences be if they were stripped of all the accompanying physiological changes? For James, this was evidence of the necessary role played by such bodily activity. However, there are several possible problems with James’s approach, one of them being the fact that the large variety of emotion terms found in English and many other languages is not matched by an equally large variety of distinguishable patterns of physiological activity during emotion. This is one of five problems noted by Walter Cannon who focused his attention on that aspect of James’s theory concerning visceral changes in organs such as the heart, as opposed to changes in voluntary muscles such as those in the face or the limbs. Another of the problems noted by Cannon is that visceral changes tend to be rather slow, whereas emotional reactions can be, and often are, rather fast. If this is the case, how can the experience of an emotion be the perception of the bodily changes that occur on exposure to the right emotional stimulus?
These were some of the considerations that Stanley Schachter took into account in developing his two-factor theory of emotion. Schachter had previously conducted research on the way in which people who were made to feel anxious and uncertain liked to be in the company of other people who were in the same situation, so that they could compare their own emotional reactions with those of other people. This suggests that social context may play an important role in emotion, by helping people to interpret their stirred-up internal state. Rather than there being a particular pattern of bodily change associated with each subjectively distinguishable emotion, Schachter suggested that the key role played by bodily change was to energize emotion; without a state of physiological arousal, no emotion would ensue. Bodily change, on this account, is a necessary but not sufficient condition for emotion. For emotion to be experienced, the second of the two factors is also needed. This factor is cognition, and the role it plays in Schachter’s model is that of labeling the general state of arousal. In theory, then, exactly the same physiological state of arousal could be interpreted in quite different ways and therefore experienced as quite different emotions.
Despite its elegance, this bold prediction made by two-factor theory did not attract enough experimental support for the theory to be able to remain as influential as it was in the 1970s and early 1980s. What has survived relatively unscathed is the proposition that people can misattribute the cause of any felt arousal, with the result that they will tolerate more pain if they are led to believe that at least part of the arousal they experience when exposed to a painful stimulus is due to a drug (in fact a placebo, which means a drug that has no genuine effect, like fake pills) that they have swallowed. Equally, if they are led to think that they have ingested a tranquilizer (again, in fact a placebo, so it has no tranquilizing effect) then any arousal they experience will tend to be overattributed to the most plausible source of arousal. If the most plausible source of the arousal is the fact that they have just written an essay advocating a position that runs contrary to their beliefs, and they are concerned about the effect this essay may have on others, they change their attitudes even more than they would in a no-tranquilizer control condition, apparently because they believe themselves to be experiencing a lot of cognitive dissonance as a result of writing the essay; and the most effective way to reduce the dissonance is to change one’s attitude to bring it more into line with the position taken in the essay.
The focus both of Cannon’s critique of James and of Schachter’s attempt to build a theory that took account of this critique was the state of arousal of the individual’s autonomic nervous system. Yet bodily change clearly can involve more than how fast one’s heart is beating, how dry one’s mouth is, and how much tremor there is in one’s hands—all of which are perceptible signs of autonomic nervous system arousal. James’s theory was as much concerned with the activity of the voluntary muscles as with visceral changes, as is evident from his assertion that we feel angry because we strike (rather than striking because we feel angry). This aspect of James’s theorizing was taken up by researchers interested in the effects of manipulating the feedback individuals receive from their facial or postural musculature. A series of studies has shown, for example, that people tend to find humorous stimuli more amusing if their faces are induced to smile during exposure to these stimuli, that they find painful stimuli more noxious if they are led to adopt more negative facial expressions while exposed to these stimuli, and that they feel more dejected if they are induced to adopt a stooping posture. Thus modern research has provided support for one feature of James’s theory, even if that evidence is more consistent with the view that the perception of bodily change moderates (rather than mediates) the experience of emotion.
Cognition and Emotions
Although cognition was given a central role in Schachter’s two-factor theory, that role was distinct from the role it has in appraisal theories of emotion. In Schachter’s theory, the role of cognition was to label arousal that was already present. In appraisal theory, the role of cognition is to interpret the significance and meaning of the unfolding emotional event. Imagine that you hear a strange noise coming from your ground-floor kitchen in the middle of the night. The sense you make of this event through a process of appraisal is regarded as determining whether and how you will react emotionally. Interpreting the noise as caused by a human intruder will give rise to a very different set of emotions than will interpreting the noise as caused by your cat or by the wind blowing some-thing off the window sill. Another important factor, in the view of appraisal theorists, is your sense that you will be able to cope with any possible threat to your well-being. A young, physically able person will experience less threat under these circumstances than will an elderly or disabled person. The essence of the appraisal theory view of the role played by appraisal is nicely summarized in Nico Frijda’s law of situational meaning, which states that emotions arise from the meanings people ascribe to situations and that if the meaning changes (such that your initial thought that the noise from your kitchen was made by an intruder now changes as you remember that your cat had been outside when you went to bed, and the noise is the sound of her entering the house), so too will be emotion (in this case from fear to relief).
The singular term appraisal theory makes it seem as though there is one theory to which all appraisal theorists subscribe. In fact there are several appraisal theories, all of which share the view that emotions arise from cognitive appraisals. Where they differ is with respect to the details of how this common assumption is worked into a full-fledged theoretical position. Some theorists, like Richard Lazarus, emphasize the importance of a relatively small set of core relational themes. These are, in effect, clusters of configured appraisals that capture the essence of an emotion. Thus, the core relational theme of irrevocable loss is one that holistically defines sadness, whereas the core relational theme of other blame is one that holistically defines anger. The advantage of these core relational themes is that they capture the key relational meaning of a situation and are therefore likely to be predictive of physiological and behavioral changes. Other appraisal theorists, like Frijda, emphasize the readiness for action that appraisals entail. Even if one does not act on these so-called action tendencies, the felt tendency to aggress, to retreat, to freeze, to cry, to laugh, and so on, represents an important element of the experienced emotion. Still other theorists, such as Klaus Scherer, emphasize the importance of the temporal dimension of appraisal. On this view appraisal is a sequential business, starting with rudimentary checks, such as whether the stimulus is novel, pleasant, and expected (in that order), and ending with more complex assessments, such as whether the stimulus conforms to personal or social norms.
There is a wealth of evidence showing that individual emotions are associated with distinct appraisals or patterns of appraisal. There is no doubt, then, that people are able to make connections between emotions and appraisals in much the same way that appraisal theorists propose. Less plentiful is good evidence showing that appraisals are causally linked to emotions. This leaves open the question of whether appraisals are causes, constituents, or even consequences of emotion. This turns out to be a critical issue, because the most sustained attack on appraisal theory, initiated by Robert Zajonc, has argued that affective reactions (in the sense of like, approach, dislike, avoid) often precede cognitive reactions (such as beneficial or detrimental to one’s goals), and therefore cannot be caused by them. The sheer speed of emotional reactions is an important component of Zajonc’s critique, raising doubts about the potential for relatively time-consuming cognitive processes to mediate these reactions. Also important for Zajonc’s argument is evidence that people are able to arrive at evaluations of stimuli without being aware of having been exposed to them, as in the mere exposure effect, which again raises questions about the necessity of appraising a stimulus before having an emotional response to it.
A compromise position on the role of cognitive appraisal in emotion is one that recognizes that there is more than one route to an emotional response. Take fear as an example. Modern animal and clinical research shows that there are two distinct ways in which fear can be triggered in the brain, one of which is cortically mediated (thereby implying a role for appraisal), the other of which is mediated by the amygdala (implying a fast response that would be adaptive in predator-prey situations). Note that this subcortical route harks back to one of James’s central assumptions, namely, that there is an automatic link between perception and bodily change. The two-routes argument works best for emotions such as fear, which have clear implications for the survival of the individual. It is less plausible to argue for two routes in the case of an emotion such as guilt, for example. Yet here, too, there is debate about the extent to which the appraisal that is assumed by many theorists to be a necessary condition for guilt, namely, perceived responsibility for harm to another, is in fact necessary. Roy Baumeister, for example, has argued that the root cause of guilt is loss of love in a valued relationship and that people who experience this loss of love feel guilty and, as a component or consequence of these guilt feelings, feel responsible for the hurt experienced by the other party to the relationship.
Expression and Emotions
It is obvious that emotions can be expressed in the face and other parts of the body. If someone is intensely angry, sad, afraid, or surprised, others are likely to be able to see signs of the emotion in question in that person’s face. The outward expression of inner emotional experience is, of course, of particular interest to social psychologists because it affords others the opportunity to understand how someone is feeling without this person needing to explain the feeling in words. It is sometimes argued that the nonverbal means of communicating emotion are more important and effective than the verbal means. Clearly, this is true of interpersonal communication whereby one or both persons are unable to communicate verbally, because they are prelingual (as in the case of infants) or deaf or simply unable either to speak or to hear each other because of the context (as in a silent Trappist Order or in noisy working environments). It is known that babies are especially interested in faces and that there is a tendency for humans to mimic each other’s nonverbal behaviors. People tend to like other people more when others mimic them in this way. To the extent that people do what others do, facially and posturally, and to the extent that feedback from the face and from body posture moderates their emotional experience, it is likely that people come to feel what others are feeling, thereby strengthening understanding of and bonding with those others. However, all of this depends to an important degree on the extent to which subjective experience of emotion translates into overt expression. It may well be obvious that there are conditions under which this does happen, but it is also obvious that people are capable of appearing to feel one thing when they really feel something else. To avoid giving offense, people pretend to like things that they do not; actors pretend to feel things that they do not so as to produce a convincing portrayal of a character in a particular setting. To what extent is bodily behavior a reliable reflection of someone’s emotional state?
Paul Ekman and his colleagues have tested the notion that there is a close relation between emotion and facial behavior, which is what Ekman’s notion of a facial affect program, a hardwired system linking experienced emotion to facial behavior, would predict. Their research program employs two kinds of methods. The first is based on Darwin’s idea that the ways in which emotions are expressed are universal and therefore independent of culture. To provide more scientific support for this idea than Darwin had been able to muster, Ekman and colleagues took photographs of faces that were recognized by Westerners as clearly expressing certain emotions, and they showed these to persons in a variety of other cultures. The most telling studies are those conducted in preliterate cultures, such as the highlands of Papua New Guinea. What the researchers found was that members of tribes living in these remote cultures, who had had little exposure to Westerners or to Western media images, could match the photographs to short stories of an emotional nature in ways that showed that they broadly understood the emotional meaning of the faces. This is taken as evidence that emotions are expressed facially in the same way across the world: How else could researchers account for the ability of those living in isolated cultures to attribute the same meaning to faces as Westerners do? However, it is important to recognize that these findings relate to a limited set of emotional expressions—happy, sad, angry, afraid, disgusted, and surprised—and that the stimuli used in this type of research are still photos taken at the apex of an extreme, iconic version of an expression. It is also worth noting that although members of remote tribes could match the photos to emotional stories with above-chance accuracy, their performance on average tends to be worse than that of their Western counterparts. Bearing in mind that the expressions they are asked to judge in these studies are of Western faces, this raises the possibility that people may be better at recognizing emotions in own ethnicity faces than in other ethnicity faces. Recent research suggests that this is the case, pointing to the existence of emotional dialects that are easier for persons who are familiar with the dialect to decode.
The second line of research pursued by Ekman and colleagues has directly examined the extent to which different emotions are accompanied by measurable differences in facial behavior. In this line of work, researchers have made use of the Facial Action Coding System (FACS), a measurement system for coding all visible movement in the human face, which was developed by Ekman in collaboration with Wallace Friesen. This sort of research has shown, for example, that happiness and disgust, as induced by film clips, are associated with different facial actions. This seemingly uncontroversial finding has been the subject of some debate in the literature. Various researchers have used films and other types of stimulus to induce emotional states, surprise being a recent example, and have failed to find that these states are accompanied by the distinct facial actions (brow-raising, eye-widening, jaw-dropping) that would be expected if the notion of a facial affect program were correct. Other researchers have questioned the assumption that there is a close relation between emotion and facial action, arguing instead that facial actions evolved to communicate intentions or motives, not emotions, to conspecifics. This line of argument leads one to predict that facial behavior should vary as a function of how social a situation is, rather than how emotional it is. The debate concerning the closeness and robustness of the relation between emotion and overt behavior is far from settled, but it is evident to most commentators that the strength of the relation is variable. The challenge for future researchers is to identify the factors that moderate this relation.
The Social Life of Emotions
Although much of the research on emotion has a distinctly social psychological flavor, emotion researchers have started to address more explicitly social psychological issues, and social psychologists have started to incorporate emotional concepts and measures into their study of mainstream social psychological issues. Thus, on the one hand, there are emotion researchers who study social or self-conscious emotions, such as shame, guilt, embarrassment, envy, and jealousy—emotions that depend on a real or imagined social context. The importance of this work is that it treats emotion as embedded in a social context and thereby helps to counterbalance the tacit assumption in much theorizing that emotion is essentially a private experience that arises from socially isolated individuals’ assessments of the implications of events for their personal well-being. Also noteworthy in this connection are emotion researchers who study the impact of culture on emotional experience and expression. What this type of research makes clear is that the ways in which cultures promote certain kinds of values (e.g., honor) or self-construals (e.g., the self as an autonomous agent) have an impact on the conditions under which emotions are experienced and communicated.
On the other hand, there are social psychologists who study phenomena such as interpersonal, group, or intergroup relationships and who have found that by taking emotional processes into account they gain a richer understanding of these phenomena. Harmony and discord in close personal relationships can be better understood by examining the quality and quantity of emotional communication in those relationships. Variations in productivity in work groups can be better understood by examining the emotional climates that prevail in these groups. Acceptance or rejection of other social groups can be better understood by taking account of the emotions that are felt toward members of those groups.
Although social psychologists have played a central role in emotion research, it is only relatively recently that emotion has become a central topic of research for social psychologists, but there is every indication that the relationship between emotion and social psychology will be mutually beneficial.
- Ekman, P., & Rosenberg, E. (Eds.). (1997). What the face reveals: Basic and applied studies of spontaneous expression using the facial action coding system (FACS). New York: Oxford University Press.
- Kitayama, S., & Markus, H. R. (Eds.). (1994). Emotion and culture: Empirical studies of mutual influence. Washington, DC: American Psychological Association.
- Leach, C. W., & Tiedens, L. Z. (Eds.). (2004). The social life of emotions. Cambridge, UK: Cambridge University Press.
- Parkinson, B., Fischer, A., & Manstead, A. S. R. (2005). Emotion in social relations: Cultural, group and interpersonal perspectives. New York: Psychology Press.
- Parrott, W. G. (Ed.). (2000). Emotions in social psychology: Essential readings. Philadelphia: Psychology Press.
- Philippot, P., Friedman, R. S., & Coats, E. J. (Eds.). (1999). The social context of nonverbal behavior. New York: Cambridge University Press.
- Russell, J. A., & Fernandez-Dols, J. M. (Eds.). (1997). The psychology of facial expression. New York: Cambridge University Press.
- Scherer, K. R., Schorr, A., & Johnstone, T. (Eds.). (2001). Appraisal processes in emotion: Theory, methods, research. London: Oxford University Press.