In 1884, William James asked the fundamental question about the nature of emotions in his famous article “What Is an Emotion?” More than 120 years later, psychologists still wrestle with this question, and a single, precise definition has proven elusive. Definitional precision has been difficult both because emotion is a word lifted from common language—it is not a scientific term—and because scientists studying emotion approach it from many different perspectives. Psychologists from evolutionary, cognitive, and physiological traditions each focus on different antecedents, components, and outcomes of emotions. Further, an emotion is not one thing; it is a cluster of responses.

Defining Emotion

In light of these obstacles to a precise definition, however, most researchers agree that emotions have the following characteristics. First, they include a subjective, experiential feeling state. This is the prototypical idea of an emotion: It is what we commonly refer to as feelings and what psychologists call affect. Second, emotions include a physiological component. Anger, for example, is associated with autonomic changes in areas such as heart rate and galvanic skin response. Third, emotions have a behavioral component. This includes expressive behavior as seen in facial and postural changes, and action tendencies (e.g., the tendency to recoil when experiencing fear). Finally, most definitions of emotion also include an evaluative component connecting the emotion to a specific person, object, or event. That is, emotions have a focus: We are angry at someone or sad about something.

It is this final component that is useful for distinguishing emotion from the closely related concept of mood. Moods are affective states, similar to the subjective, experiential feeling state of emotions. Moods also typically are of a longer duration, and are less intense, than emotions. But the primary feature that distinguishes emotions from moods is that unlike emotions, moods lack a specific focus. Moods are broad and diffuse, whereas emotions are associated with a person, object, or event that has been evaluated as significant for the individual. In distinguishing emotions from related concepts, it is useful to think in terms of a hierarchy, with affect, the subjective feeling state, as a broader, higher order category characteristic of both emotions and moods. If the affective state is accompanied by distinct physiological changes, behavioral tendencies, and a referent person, object, or event, it is most appropriate to classify it as an emotion.

Because emotion is a term lifted from common language, its use, even in the scientific literature, does not always match the definition provided here. Concepts such as emotion regulation, emotional contagion, and emotional labor often focus more on general affective states than on emotions per se. Thus, more technically correct terms might be affective regulation, mood contagion, or affective labor. For most uses of these concepts, however, the distinction between affect and emotion is relatively unimportant, and because these terms are well established in the literature, their use will continue.

Discrete Emotions versus Dimensions

There are two traditions in psychology for how to best conceptualize emotions. In one camp are those who argue for the existence of a small set (5 to 9) of discrete primary emotions. These typically include happiness, interest, surprise, fear, anger, sadness, and disgust. Evidence for the existence of basic emotions is mostly drawn from studies showing the existence of distinct, specific, and universal (across cultures) emotional antecedents and responses. Particularly influential in this area is the work of Paul Ekman, whose research has demonstrated the existence of a small number of distinct, universal facial expressions of emotion. Across numerous cultures (Western and Eastern, modern and premodern), basic emotions such as fear are expressed in the face in the same manner and are universally recognized in pictures of facial expression. There have, however, been several influential critiques of the entire notion of basic emotions in recent years.

In contrast to the basic emotions view is the view held by researchers who assert that the domain of affective experience (mood and emotion) can be described by two underlying dimensions. Typically, one dimension is described as valence or hedonic tone (from pleasant to unpleasant), and the other dimension is intensity or arousal (from high to low). The structure of affective experience can then be described by a model called the affective circumplex. The affective circumplex is represented as a circle created by two bisecting dimensions: valence and arousal. All emotions can be represented as falling in a circle around these dimensions, as each can be described by different levels of valence and arousal. Anger is a combination of negative valence and high arousal; sadness is negative valence at moderate levels of arousal; boredom is negative valence at low levels of arousal. There is a vigorous debate surrounding how to best use the circum-plex to represent affective experience. One group of researchers argues that affective experience is best represented by the orthogonal dimensions of positive and negative affect (each dimension representing a combination of valence and arousal), whereas others argue that the dimensions of valence and arousal best represent affective experience (see the May 1999 issue of the Journal of Personality and Social Psychology for a special section devoted to this debate). Despite this ongoing debate, virtually all researchers studying emotions from a dimensional perspective use the circumplex model in some form.

In industrial/organizational (I/O) psychology, most researchers have adopted a dimensional view of emotions. This is seen in, for example, the wide use of the Positive and Negative Affect Schedule (PANAS) measure, which is based on the affective circumplex. The prevalence of the dimensional view in I/O psychology research likely stems from the fact that it allows researchers to move seamlessly between discussing general affect, mood, and emotion. The dimensional perspective also meshes well with research on personality: Research shows links between positive and negative affect and, respectively, the Big Five dimensions of extraversion and neuroticism. Further, many outcomes of interest in I/O psychology— for example, prosocial organizational behavior—are sufficiently broad that they are likely better predicted by general categories of emotion than by specific, discrete emotions. Critics of the dimensional view of emotions have argued, however, that conceptualizing emotions in terms of two general dimensions masks important differences between discrete emotions. Anger and fear, for example, are both emotions characterized by high arousal and negative valence. The dimensional view treats these emotions as similar, yet they have unique antecedents and outcomes. As such, in certain areas of study (for example, abusive supervision, in which employees may be expected to experience anger, fear, or both), theoretical precision may warrant the study of discrete emotions.

Individual Differences in the Experience and Expression of Emotion

Individuals differ in the quantity and quality of emotions they experience and express. How these differences emerge can be conceptualized by using an input-output model of an individual’s emotional experience. The experience of emotion begins with an emotional stimulus, such as an interaction with an angry customer. The input triggers emotion-response tendencies, the behavioral, physiological, and experiential changes that represent the initial experience of emotion for the individual. Certain individuals (e.g., those high in the personality trait of neuroticism) will be more sensitive to these negative stimuli than others, producing stronger response tendencies in some individuals. Response tendencies do not, however, necessarily translate into visible emotional responses. An individual’s emotion regulation determines whether the generated emotion experience translates into a visible emotional response. Emotion regulation is how individuals influence the emotions they have and how they experience and express their emotions.

Research by James Gross suggests that emotion regulation can occur at two points: (a) antecedent to the activation of emotional response tendencies and (b) after the emotion has been generated. Antecedent-focused emotion regulation techniques act on the stimuli coming into the system before emotional response tendencies are activated. Techniques include selecting or modifying situations to alter the emotional stimuli (e.g., avoiding an angry customer); thinking about something different to avoid the stimuli; or cognitive change, which involves attaching a different meaning to the situation. Response-focused emotion regulation works only on changing the outward expression of emotion, not the emotion itself. Suppressing the emotion is a common response-focused strategy. Although antecedent- and response-focused emotion regulation may look similar in terms of outward expression of emotion, their effects are actually quite different. Specifically, when presented with a negative emotional stimulus, suppression changes the outward expression of emotion but not the subjective experience. Suppression also produces increased sympathetic nervous system activation and decrements in cognitive functioning, whereas antecedent-focused coping strategies do not.

Distinctions between types of emotion regulation strategies highlight the importance of differentiating between the experience and expression of emotion. We typically make judgments of people’s emotional state based on their expressive behavior: facial expressions, vocalizations, and body posture. What we can see, however, is only emotional expression. Individuals differ in terms of emotional expressivity, the extent to which they express outwardly the emotions they are experiencing. Differences in expressivity are a key to understanding the common stereotype that women are more emotional than men. The stereotype is not true: research demonstrates that males and females do not differ reliably in terms of emotional experience. What is true is that women are more emotionally expressive than men. Although men and women experience similar types of emotions, women are more likely to manifest those emotions behaviorally than are men.


  1. Gross, J. J. (2001). Emotion and emotion regulation. In L. A. Pervin & O. P. John (Eds.), Handbook of personality (2nd ed.). New York: Guilford Press.
  2. James, W. (1884). What is an emotion? Mind, 9,188-205.
  3. Larsen, R. J., Diener, E., & Lucas, R. E. (2002). Emotion: Moods, measures, and individual differences. In R. G. Lord, R. J. Klimoski, & R. Kanfer (Eds.), Emotions in the workplace. San Francisco: Jossey-Bass.
  4. Pugh, S. D. (2002). Emotion regulation in individuals and dyads: Causes, costs, and consequences. In R. G. Lord, R. J. Klimoski, & R. Kanfer (Eds.), Emotions in the workplace. San Francisco: Jossey-Bass.
  5. Russell, J. A. (2003). Core affect and the psychological construction of emotion. Psychological Review, 110, 145-172.
  6. Weiss, H. M. (2002). Conceptual and empirical foundations for the study of affect at work. In R. G. Lord, R. J. Klimoski, & R. Kanfer (Eds.), Emotions in the workplace. San Francisco: Jossey-Bass.

See also: