How positively or negatively someone feels toward an object refers to which property of an attitude?

Attitudes are a tendency to evaluate a target object with some degree of favor or disfavor, and are based on emotional reactions, behaviors toward, and cognitive evaluations of the attitude object (Eagly & Chaiken, 1993). Although these three components are often interrelated, they are distinct (Abelson, Kinder, Peters, & Fiske, 1982; Breckler & Wiggins, 1989).

The affective component is related to how people feel about the attitude object (Eagly & Chaiken, 1993). It can reflect a general liking or disliking, or more specific affective reactions toward the object. With respect to attitudes toward emotion, the affective component is likely related to the extent to which one likes or dislikes the target emotion. The behavioral component is related to how people behave toward the attitude object (Eagly & Chaiken, 1993). Approaching the object is typically associated with more positive attitudes toward it, whereas avoiding the object is typically associated with more negative attitudes. With respect to attitudes toward emotion, the behavioral component is likely related to whether people approach or avoid the target emotion. The cognitive component is related to how people think about the attitude object (Eagly & Chaiken, 1993). This component reflects beliefs that people hold about the object and the attributes they associate with it. The more people associate an object with positive attributes, the more positive their attitudes toward it. With respect to attitudes toward emotion, the cognitive component is likely related to the extent to which one thinks positive or negative thoughts about the emotion, or associates it with positive or negative attributes.

Emotions are evaluative states. However, emotions can also be the object of evaluation. Various constructs related to evaluations of emotions have been proposed and assessed in the literature. Emotion norms (Eid & Diener, 2001), desired emotions (e.g., Tamir, Bigman, Rhodes, Salerno, & Schreier, 2015), and ideal affect (Tsai, Knutson, & Fung, 2006) refer to personally and culturally relevant evaluations about the desirability of emotions and affective states. These motivational constructs are likely informed by evaluations of emotions, but they are not entirely equivalent to them.

Harmon-Jones et al. (2011) were the first to focus on evaluations of emotions per se, by directly assessing attitudes toward emotions. They introduced the Attitudes toward Emotion (ATE) scales to measure people's attitudes toward five discrete emotions (i.e., joy, sadness, anger, fear, and disgust). They found that people differ in their attitudes toward discrete emotions, and that these differences are linked to emotion-related behavior. In a series of studies, they showed that more positive (or more negative) attitudes toward an emotion were related to attempts to increase (or decrease) experiences of that emotion. For example, individuals with more negative attitudes toward fear were more motivated to avoid fearful stimuli after viewing a fear-inducing film clip.

The research by Harmon-Jones et al. (2011), using the ATE, was critical in highlighting the importance of studying attitudes toward emotions. However, as reviewed above, such attitudes are not homogeneous constructs. Instead, they are likely comprised of three distinct components. We argue that the ATE taps primarily the affective and potentially the behavioral components of attitudes toward emotions. The affective component is targeted by items that refer to how much people like or enjoy the emotional experience (e.g., I like how it feels when I am furious), or stimuli that elicit the emotion (e.g., I like conversations that make me feel happy). The behavioral component is targeted by items that describe behaviors that regulate the emotional experience (e.g., I do things just because they scare me). Nonetheless, we argue that the ATE does not contain items tapping the cognitive component of attitudes toward emotions.

As the different components of attitudes might carry different theoretical and pragmatic implications (e.g., Millar & Tesser, 1986; Trafimow & Sheeran, 1998), it is important to also assess the cognitive component of attitudes toward emotions. This, therefore, was the goal of the current investigation. We sought not only to develop and validate a measure of attitudes toward emotions that taps the cognitive component of such attitudes, but also to demonstrate that this component is conceptually distinct from the other components, and may underline distinct motivational outcomes.

The cognitive component of an attitude is based, in part, on associations between the attitude object and valence attributes. Such associations are best captured by the semantic differential scale, which has often been used to measure attitudes (Himmelfarb, 1993). The scale was developed by Osgood, Suci and Tanenbaum (1957) to measure the connotative meaning of a concept. For each attitude object, the scale introduces a series of bipolar adjectives (e.g., bad-good). Each of the adjective pairs ranges on a 7-point scale from the negative attribute (“1”, e.g., bad) to the positive attribute (“7”, e.g., good). The adjectives used in semantic differential scales are general and abstract, rather than tailored to fit a specific attitude object.

Semantic differential scales can be used to assess affective components of attitudes, if they include adjectives that refer to hedonic experiences (e.g., pleasant-unpleasant). However, they can also be used to assess cognitive components of attitudes, to the extent that they include adjectives that are evaluative, but not necessarily hedonic (e.g., good-bad; useful-harmful). Therefore, to assess the cognitive component of attitudes toward emotions, we constructed a measure based on semantic differential scales, using adjectives that capture positive or negative attributes. We expected our scale to capture the cognitive component of attitudes toward emotions, and we further expected it to be related, yet distinct, to other components of attitudes, as captured by the ATE.

The components of attitudes may be differentially linked to perceptions and behaviors toward the attitude objects. The affective component of attitudes is more closely associated with pleasure related aspects, whereas the cognitive component is more closely related to utilitarian behaviors and appraisals (e.g., Millar & Tesser, 1986; Trafimow & Sheeran, 1998). Such distinction should apply to attitudes toward emotions. Accordingly, the affective component of attitudes toward emotions may be associated with hedonic judgments. For instance, the affective component may be linked to judgments of how pleasant or unpleasant the target emotion is. In contrast, the cognitive component of attitudes toward emotions may be linked to utilitarian judgments. For instance, it may be linked to judgments of how useful or harmful the target emotion is (see Chow & Berenbaum, 2012).

This differentiation could ultimately lead to different emotion-related behaviors. People may be motivated to experience emotions for hedonic or instrumental reasons (Tamir, 2016). When people regulate emotions for hedonic reasons, they are guided by how pleasant or unpleasant an emotion is. In contrast, when people regulate emotions for instrumental reasons, they are guided by how useful or harmful an emotion is. To the extent that the cognitive component of attitudes toward emotions is linked to utilitarian judgments, it may underlie instrumental motivation in emotion regulation.

In the current investigation, we sought to show that attitudes toward emotions involve more than one component. We hypothesized that the cognitive component of such attitudes is distinct from the other components. Furthermore, we tested whether affect-based attitudes toward emotion are more strongly related to the perceived pleasantness of emotions, whereas cognition-based attitudes toward emotions are more strongly related to the perceived utility of emotions. Finally, we tested the predictive validity of the cognitive component of attitudes toward emotions, by assessing links to instrumental motivation in emotion regulation. To accomplish these aims, we developed and validated a scale designated to capture the cognitive component of attitudes toward emotions (i.e., the EVE). In Study 1, we tested whether scores on the new scale is psychometrically distinct from the existing scale (i.e., the ATE), using an exploratory factor analysis. In Study 2, we verified this distinction using a confirmatory factor analysis. In Studies 1–3, we tested our hypothesis that the ATE is more strongly and consistently linked to perceived pleasantness of the target emotion, whereas the EVE is more strongly and consistently linked to perceived utility of the target emotion. In Study 4, we tested whether the EVE, but not the ATE, would be related to instrumental motivation in emotion regulation.