Appraisal components and emotion traits: Examining the appraisal basis of trait curiosity

Individual differences related to emotions are typically represented as emotion traits. Although important, these descriptive models often do not address the psychological dynamics that underlie the trait. Appraisal theories of emotion assume that individual differences in emotions can be traced to differences in patterns of appraisal, but this hypothesis has largely gone untested. The present research explored whether individual differences in the emotion of interest, known as trait curiosity, consist of patterns of appraisal. After completing several measures of trait curiosity, participants read complex poems (Experiment 1) or viewed simple and complex pictures (Experiment 2) and then gave ratings of interest and interest's appraisal components. The effect of trait curiosity on interest was fully mediated by appraisals. Multilevel analyses suggested that curious people differ in the amount of appraisal rather than in the kinds of appraisals relevant to interest. Appraisal theories can offer a process-oriented explanation of emotion traits that bridges state and trait emotional experience. Article: One of the oldest issues in the study of emotion is the relationship between emotion and personality. Emotions and personality intersect in many ways (Arnold, 1960; Haviland-Jones & Kahlbaugh, 2000; Lewis, 2001; Magai & Haviland-Jones, 2002; Silvia & Warburton, 2006; Tomkins, 1979; van Reekum & Scherer, 1997; Vansteelandt & Van Mechelen, 2006). A popular intersection is the study of emotion traits. In this approach, research identifies individual differences in the intensity or frequency of experiencing an emotion, such as anger, anxiety, shame, happiness, or positive and negative moods (see Watson, 2000; Watson & Clark, 1997). Identifying and assessing emotion traits is central to understanding stable patterns of emotionality, but it is only the first step. The study of individual differences is most powerful when it adopts a process-oriented approach (Cronbach, 1957; Underwood, 1975). If the dynamics that create and sustain the individual differences are known, then state and trait approaches can be integrated, thus enriching the study of both states and traits. Appraisal theories have been successful in explaining some of the central problems of emotion psychology (Ellsworth & Scherer, 2003; Roseman & Smith, 2001), and they have much to offer the study of emotion traits. The present research thus uses appraisal theories of emotions to examine the psychological processes that underlie individual differences in trait curiosity, the emotion trait associated with feelings of interest (Kashdan, Rose, & Fincham, 2004; Silvia, 2006b, chap. 4). Trait curiosity has been widely studied, and many reliable scales assess it (see Litman & Silvia, 2006). Like many emotion traits, however, little is known about why curious people experience interest in response to specific situations. The present research examines the appraisal basis of trait curiosity, and, in doing so, addresses the broader theoretical problem of how appraisal theories inform the process-oriented study of emotion traits. Appraisal theories and emotion traits One of the central questions appraisal theories were developed to handle, according to Roseman and Smith (2001), is the problem of individual differences in emotional experience. People respond differently to similar situations, and they vary in their chronic patterns of emotional experience. Appraisal theories explain this variability by referring to covarying patterns of appraisal (Lazarus, 1991; Roseman, 2001; Scherer, 2001a; Smith & Ellsworth, 1985). The role of appraisals in individual differences has been recognised in past work. Scherer (2001b, p. 383), for example, suggests that individual differences probably “massively contribute to the variance in phenomena studied by appraisal theorists”. Nevertheless, individual differences have not received much attention in appraisal research. Lewis (2001, p. 211) notes that it is “somewhat of a mystery why appraisal theorists have spent so little time examining individual differences explicitly”. One intersection between personality and appraisal that has received some attention is how individual differences affect appraisals (e.g., Smith & Pope, 1992). In an extensive treatment, van Reekum and Scherer (1997) outline ways in which individual differences affect emotions by influencing appraisal processes, such as individual differences related to levels and complexity of information processing. A second intersection of personality and appraisal—and the concern of the present research—is the appraisal basis of individual differences themselves (Lewis, 2001). Some emotion traits might be constituted by appraisals. In this case, people are typically angry, sad, or afraid because they typically appraise situations in a manner that creates the emotion of anger, sadness, or fear. To use trait curiosity as an example, the disposition of curiosity may be composed of the stable pattern of appraisals that create the emotion of interest: curious people are more often interested because they tend to make the appraisals that cause interest. Trait curiosity and interest The study of trait curiosity dates to the 1960s, inspired by the Berlyne (1960) tradition of curiosity research (see Day, 1971; Litman, 2005; Silvia, 2006b, chap. 4; Spielberger & Starr, 1994, for reviews). This early generation of research has been criticised in several reviews for poor psychometric practices (Boyle, 1983; Langevin, 1976; Loewenstein, 1994). Recently, a new generation of models has emerged (Kashdan et al., 2004; Litman & Jimerson, 2004). Given the youth of these models, not much is known about the processes that underlie individual differences in curiosity. To date, research has primarily correlated self-report curiosity scales with other self-report instruments; the psychological processes that constitute trait curiosity are not well understood.1 The emotion of interest is the emotion associated with curiosity, exploration, intrinsic motivation, and information seeking (Fredrickson, 1998; Izard & Ackerman, 2000; Sansone & Smith, 2000; Silvia, 2005c, 2006b; Tomkins, 1962). Interest is thus the emotion most closely tied to trait curiosity. Appraisals of interest seem like a promising way of explaining trait curiosity. The appraisal structure of interest, according to recent research (Silvia, 2005c), involves two dimensions: an appraisal of novelty-complexity, and an appraisal of coping potential. As understood within the multilevel sequential-check model of appraisal (Scherer, 1997, 1999, 2001a), people first appraise an event's novelty, viewed broadly as appraisals of incongruity, complexity, unexpectedness, obscurity, and uncertainty (cf. Berlyne, 1960, chap. 2). Following this appraisal, an appraisal of coping potential assesses the person's ability to comprehend the new, complex event. Events appraised as new and complex yet potentially comprehensible are experienced as interesting. This appraisal structure is congruent with past research (see Silvia, 2005b, 2005c, 2006b, chap. 2). One literature shows that the family of novelty-complexity variables affects interest (see Berlyne, 1960, 1971, 1974; Walker, 1981); a different literature shows that appraisals of coping potential affect interest (Millis, 2001; Russell, 2003; Russell & Milne, 1997). More critically, several direct tests demonstrate that novelty and coping potential predict the experience of interest (Silvia, 2005a, 2005c, 2006a; Turner & Silvia, 2006). These effects replicated for measured and manipulated appraisals, for self-report and behavioural measures of interest, and for interest in random polygons, abstract visual art, classical paintings, and poetry. Moreover, this appraisal structure is specific to interest (Turner & Silvia, 2006): it discriminates interest from enjoyment, a related positive emotion (Ellsworth & Smith, 1988; see Silvia, 2006b, chap. 1, for a review). The appraisal basis of trait curiosity If these two appraisal components comprise the appraisal structure of interest, then the appraisal basis of trait curiosity can be examined. According to an appraisal analysis, trait curiosity should predict interest because it predicts appraisals. This hypothesis breaks down into two variants. First, trait curiosity might be mediated by both appraisals—it predicts interest by predicting both appraised novelty-complexity and appraised coping potential. Second, trait curiosity might be mediated by only one of these appraisals. It is difficult to predict whether one or both appraisals will mediate the effects of trait curiosity. In either case, this possibility would manifest as indirect, mediated effects of trait curiosity on interest. A second possibility is that trait curiosity will predict interest, but not by predicting appraisals. No modern theory of trait curiosity is rooted in appraisal theories or in emotion psychology more generally (Kashdan, 2004; Litman, 2005; Spielberger & Starr, 1994). To the extent that they have offered mechanisms that connect curiosity to emotional experience, these models have not proposed appraisals as an explanation. Furthermore, research on related constructs (e.g., sensation seeking, openness to experience) has traditionally preferred psychobiological mechanisms (Bergeman et al., 1993; Zuckerman, 1994). Thus, it isn't necessarily obvious that curiosity would predict interest because of appraisals. This second possibility would manifest as direct, unmediated effects of curiosity on interest. The Present Research Two experiments examined whether appraisal processes explain why trait curiosity predicts the experience of interest. In each experiment, interest and appraisals were measured in response to real events. Much appraisal research has used responses to hypothetical scenarios or retrospective reports of memorable emotions (see Roseman & Evdokas, 2004). Stronger inferences about the appraisal basis of emotion traits can be made by placing people in potentially emotional situations and then assessing momentary appraisals and momentary emotional experience. In Experiment 1, people read a series of complex poems and rated their experience of interest and their appraisals of coping potential. In Experiment 2, people viewed pictures and gave ratings of interest and of appraisals; novelty-complexity appraisals were manipulated by presenting simple versus complex pictures. By replicating the effects across type of interesting object (poetry vs. visual art) and across five measures of trait curiosity, the two experiments can provide strong evidence for convergent validity. Study 2 explored an additional intersection between trait curiosity, appraisals, and interest: do curious people differ in the kinds of appraisals relevant to interest? Kuppens and his colleagues have recently suggested that people can vary in an emotion's appraisal structure (Kuppens, Van Mechelen, Smits, & De Boeck, 2003). In the case of anger, for example, people vary in whether an appraisal of high intentionality is necessary to become angry (Kuppens, Van Mechelen, Smits, De Boeck, & Ceulemans, in press). Perhaps curious and incurious people differ in kind, not just in amount, of appraisal. To explore this, Study 2 assessed whether trait curiosity affected the within-person relationships between appraisals and interest. Experiment 1 Experiment 1 was an initial test of whether appraisals accounted for the effects of trait curiosity on interest. This experiment focused on appraisals of coping potential. Because several experiments have found that appraisals of coping potential predict interest only when novelty-complexity is high (Silvia, 2005c), Experiment 1 held the dimension of novelty-complexity constant at a high level to simplify the analyses. People read complex poems and rated each poem for interest and for appraised coping potential. An appraisal model predicts that coping potential will at least partially mediate between trait curiosity and the experience of interest. Method Participants A total of 83 students—60 women, 23 men—enrolled in general psychology at the University of North Carolina at Greensboro (UNCG) participated and received credit toward a research option. Procedure People participated in groups of two to eight. The experimenter explained that the study was about personality and impressions of poetry. The participants expected to complete some measures of personality, read some poems, and provide their “impressions and reactions to the poems”. Measures of trait curiosity Before reading the poems, people completed three measures of trait curiosity. Multiple measures were used to avoid idiosyncrasies associated with any particular scale. All items were answered on 5-point Likert scales (endpoints: strongly disagree, strongly agree). The scales and their psychometric properties are described in detail elsewhere (Litman & Silvia, 2006). The Curiosity/Interest in the World subscale of the Values in Action Inventory is a 10-item measure of trait curiosity (Peterson & Seligman, 2004). The items, which are fairly general, have few references to positive emotional experience or to specific activities (e.g., “I find the world a very interesting place”; “I have many interests”). The Perceptual Curiosity Scale (Collins, Litman, & Spielberger, 2003) is a 10-item measure of curiosity associated with perceptual and sensory experience (e.g., “I like exploring my surroundings”). The third scale was the 20-item measure of Openness to Experience from the International Personality Item Pool (Goldberg et al., 2006). Openness to experience involves curiosity as a central component (see McCrae, 1996, 2007; Silvia, 2006b, chap. 4). Complex poems Eleven poems were taken from books and journals of experimental language art. Participants read the poems in the same random order. These poems were selected by pretesting a large set of poems; the 11 poems that received the highest ratings on a cluster of novelty-complexity variables were selected for the experiment. For example, one poem (Ingersoll, 1999) begins with: Library free book night in the outside of the woman whose house photograph apology, little black camera; my layered noodle hanging below sun's whereof a sliding stair, uniforms like a fast, a liked spot in the angry confrontation. Ratings of appraisals and of interest After reading a poem, people rated their impressions on 7-point semantic-differential scales. Appraised ability to understand the poem was measured with three scales: comprehensible-incomprehensible, coherentincoherent, and easy to understand-hard to understand. Interest was measured with two scales: interestinguninteresting and boring-exciting. These items have been widely used in past research (Berlyne & Peckham, 1966; Evans & Day, 1971; Silvia, 2005a, 2005c, 2006a). Results Data reduction A principal-axis factor analysis found that the measures of trait curiosity, openness, and perceptual curiosity loaded highly on a single factor. Factor scores for this factor were thus computed and used as a composite traitcuriosity score. This enables an analysis of the scales' shared variance. The items measuring interest and the items measuring appraised ability were averaged to form interest and ability scores. Higher values indicate higher ratings of curiosity, interest, and appraised ability. The path analyses were conducted with AMOS 5 (Arbuckle & Wothke, 1999; Byrne, 2001) using fullinformation maximum-likelihood estimation. Several variables had skew that was resistant to transformation. Bootstrapped estimates (resampling n=1000) were thus conducted for all parameters. The two analyses were essentially identical, so the bootstrapped estimates are not reported. Table 1 displays the descriptive statistics. TABLE 1. Descriptive statistics and correlations between trait curiosity, interest, and appraisals: