At the broadest level, an emotion is a highly valenced experiential state. More precisely, emotions comprise coordinated neural, neuromuscular/expressive, and experiential responses to meaningful stimuli or events. In the affective sciences, the term emotion (or basic or discrete
Basic emotion assessments are appropriate in situations where a pharmacological agent is intended to directly regulate or reduce the occurrence of a basic emotion such as fear or anger, as might occur during treatment of a phobia or borderline personality disorder. Basic emotions may also warrant assessment when administering pharmacological agents with extremely fast pharmoki-netics, since such agents may directly induce a discrete emotion, such as fear or euphoria.
Because basic emotions are most frequently seen in response to characteristic situations or stimuli, it is essential to assess these emotions in relation to the occurrence of such triggering events. This can be accomplished either through retrospective self-report ratings or through laboratory-based exposure paradigms in which the individual is exposed to triggering stimuli or situations. Retrospective ratings have the advantage of potential aggregation over multiple time periods and triggering events in the person's everyday life, but provide limited, if any, ability to standardize the frequency, qualitative nature, or intensity of triggering events across conditions or individuals. Furthermore, the verification and coding of such events can be highly subjective and difficult to reliably code. Additionally, unless the person is fitted with an ambulatory psychophysiological and videorecording system, it is not possible to collect any additional objective measures of the emotion.
In contrast, laboratory exposure techniques can be highly standardized, and allow the collection of data time-locked to the triggering event. By time-locking self-report ratings, an individual does not have to evaluate already long-past experiences to determine ratings. More importantly, time-locking allows the use of objective measures, which can compliment, or in some cases replace, self-report ratings. These objective measures consist of coding or direct measurement of facial muscle activity and psychophysiological measurements of correlates of sympathetic and para-sympathetic autonomic activity. Because basic emotions are accompanied by specific patterns of facial muscle movement (indeed, this is a common criteria for a basic emotion), the coding of facial expression are frequently used as markers of emotion. Detailed coding systems of facial expressions are well validated, and can be applied to video recordings of participants (Ekman et al. 2002). An alternative approach is to measure facial muscle activity with electromyography (EMG). EMG can detect weak activity of muscles even in cases in which a full expression is not completed. However, it is important in using this technique to record from a number ofmuscle groups simultaneously, as single muscle groups are often multidetermined. For instance, the corrugator supercilii muscles above the orbits are sensitive to anger, but also show increased activity during exposure to ambiguous or difficult stimuli (Pope and Smith 1994). ► Psychophysio-logical measures, such as electrodermal response, blood pressure, blood volume, heart rate, detailed analysis of electrocardiograms, respiration, and skin temperature are all useful as objective measures of the presence of an emotional experience. By combining psychophysiological measurements, it is possible to observe some degree of autonomic patterning specific to basic emotions. However, true discriminant validity is difficult to achieve with these measures, which often show properties characteristic of dimensional models of valence (or approach-avoidance) and arousal (Levenson 1988).
Although laboratory exposure techniques have a number of strengths for the assessment of basic emotions, they suffer from two major shortcomings. First, the laboratory events may be weak and somewhat unnatural approximations of real-life triggering events. Second, it is often difficult to aggregate over multiple exposures in the laboratory setting due to habituation. In such cases, the reliability of the assessment is often compromised.
In most psychopharmacological contexts, the assessment of more enduring mood states is more relevant than assessing basic emotions. For instance, the depressive disorders are diagnosed based on sustained enduring mood states, rather than individual discrete periods of emotion. Similarly, the goal of treatment for depression is to alter these enduring mood states as opposed to targetting discrete emotions. Such mood states maybe less intense than the basic emotions, but they do not require assessments that are time-locked to triggering events.
Self-report remains the only viable option in most studies of mood, because there are few objective measures of mood. A first question arises regarding which mood terms a subject should rate. Factor analytic studies indicate that mood data are marked by two higher-order factors (Watson and Tellegen 1985), which are respectively labeled positive affect (PA) and negative affect (NA). (Note: The use of the term "affect" here refers to the subjective mood factor, rather than to an individual's observable expressed emotion, as the term "affect" is frequently used in psychiatric settings.) PA reflects a person's level of pleasurable engagement with the environment. High states of PA are characterized by terms such as interested, excited, and determined, which denote positive behavioral engagement. NA comprises a general factor of subjective distress, with high states of NA marked by descriptors such as distressed, nervous, and hostile. Taken together, PA and NA account for 50-75% of the common variance of mood.
An alternative dimensional schema for understanding mood states has been proposed by Russell (1980), who identifies a bipolar valence dimension (unpleasant-pleasant), and a second dimension of arousal. This schema has some advantages, particularly when examining immediate responses to stimuli, in which one often sees a clear inverse relationship between positive and negative experiences. However, when aggregating over multiple time points, or measuring longer retrospective periods, the inverse relationship between positive and negative experience weakens. This independence allows for the study of separable influences on PA and NA (which is not feasible with a single bipolar valence dimension in which pleasant and negative states are measured in opposition to each other). An important distinction can also be made between the Russell model and the PA/NA model of Watson and Tellegen in that the experience of pleasantness or unpleasantness may be seen as a consummatory response to stimuli or events. In contrast, PA and NA are more motivational in nature, with clear links to approach and avoidance. This consummatory vs. motivational separation parallels Berridge's (1996) distinction between wanting and liking as applied to mesolimbic dopamine and opioid functions. This distinction between wanting and liking also appears useful in monitoring drug-induced mood changes. For instance, the Drug Effects Questionnaire (Fischman and Foltin 1991) asks subjects to separately rate drug wanting and drug liking, and these two ratings often show distinct patterns of correlations, and differential time courses following drug exposure.
PA and NA show different temporal patterns. Individuals typically show at least a moderate level of PA, with variability occurring around their own traitwise mean level. In contrast, NA is typically very low, with spikes occurring in response to specific negative or potentially negative events. This has implications for experimental designs, because the ability to observe a pharmacological agent's impact on NA may be limited in the laboratory environment if the person is not exposed to potentially unpleasant experiences.
The positive and negative affect schedule (PANAS), developed by Watson et al. (1988), provides a widely used measure of mood that taps the relatively pure factor structure of PA and NA. The scale includes 20 mood labels marking high (activated) PA and NA states, and has been repeatedly found to be sensitive to individual differences in both current and long-term mood. There are, however, a few important limitations of this type of scale. First, as originally designed, the PANAS does not measure mood states associated with reduced PA or NA. Specifically, terms such as fatigue (which appear to reflect an absence of PA) or calm (which appears to reflect an absence of NA) are not assessed. Subsequent measures, such as the expanded version of the PANAS (the PANAS-X), have attempted to capture these "low activation" markers. Yet, low activation markers do not possess as pure a factor structure as high activation markers of PA and NA. This occurs because terms such as boredom or tired not only reflect an absence of PA, but are also experienced as unpleasant, while states such as serenity or calm not only reflect an absence of NA, but are also experienced as pleasant. Because of this, low activation states require direct assessment, and should not be inferred by simply reverse scoring high NA or high PA terms. This issue takes on importance when we consider that many medications and street drugs are taken to alleviate feelings such as fatigue or to induce calm. Treatment studies provide further evidence of the need to directly assess low activation states, as antidepressants can produce differential effects on low activation vs. high activation PA states (Tomarken et al. 2004). Interestingly, alterations in low activation states figure prominently in the subjective effects of certain drugs. For instance, alcohol researchers frequently utilize the Biphasic Alcohol Effects Scales (Martin et al. 1993), which is a 14-item scale containing two factors: (1) a stimulant factor that corresponds to high PA, and increases during the rising limb of blood alcohol levels; and (2) a sedative factor that corresponds to an absence of PA (i.e., low activation), which increases during the descending limb of blood alcohol levels.
A second issue with the pure factor approach implemented by the PANAS arises because some important mood states reflect combinations of different levels of PA and NA. For instance, sadness can be viewed as a combination of reduced PA and heightened NA. Given the importance of sadness to the affective disorders, it is often essential to capture subjective states such as sadness, and the continued use of measures, such as the Profile of Mood States (McNair et al. 1981), which includes a Depression-Dejection scale, attest to this. In the NA domain, there is also sometimes utility in examining lower-order factors, such as anxiety or hostility. In doing so, it is important to determine to what extent observed associations are specific to the lower order factor vs. reflecting the higher order factor of NA more generally. For instance, even a scale such as the State Trait Anxiety Inventory (Spielberger et al. 1983), which is often treated as a specific measure of anxiety, captures elements of general distress, and thus cannot be used to draw conclusion about a specific subfactor of NA.
Studies examining the psychological effects of drugs of abuse require attention to the specific subjective effects of the agent. In particular, because of the intensity of basic emotion or mood states induced by drugs of abuse, typical mood scales may fail to appropriately capture such experiences, or may demonstrate ceiling effects (in which too many subjects give maximal ratings). For instance, ratings of joy may fail to capture the intensity of euphoric states. Several scales are in circulation that attempt to rectify this problem, by asking at least one question related to euphoria or related experiences. For instance, Van Kammen and Murphy (1975) developed the Amphetamine Interview Schedule to capture subjective responses to amphetamine, and include an item for euphoria, as well as related experiences of closeness to others, confidence, and overall feeling good. The Drug Effects Scale (Fischman and Foltin 1991) includes a rating of "feeling high," that may tap euphoric effects. However, the term "feeling high" is ambiguous, as it can also refer to a broad range of subjective experiences including perception of altered reality.
Was this article helpful?
Are You Depressed? Heard the horror stories about anti-depressants and how they can just make things worse? Are you sick of being over medicated, glazed over and too fat from taking too many happy pills? Do you hate the dry mouth, the mania and mood swings and sleep disturbances that can come with taking a prescribed mood elevator?