Experiential and rational processing styles, emotional intelligence and wellbeing

Abstract

This research examined a model specifying precursors of emotional intelligence and subjective wellbeing. The model proposes that experiential and rational processing styles influence level of trait emotional intelligence, which in turn influences subjective wellbeing. One hundred and fifty-four adult Australian participants completed measures of preference for rational and experiential processing, trait emotional intelligence, and indicators of subjective wellbeing (affect and life satisfaction). Structural equation path analysis identified a model for predicting wellbeing. Preference for experiential and rational processing both significantly predicted higher levels of emotional intelligence, which in turn predicted greater subjective wellbeing. Emotional intelligence fully mediated the effect of rational processing on wellbeing, and partially mediated the effect of experiential processing.

Keywords:

• Dual processing
• emotional intelligence
• rational–experiential processing
• wellbeing

Emotional intelligence has consistently been associated with a variety of adaptive outcomes, as shown, for example, in meta-analyses by Schutte, Malouff, Thorsteinsson, Bhullar, and Rooke (2007) and Van Rooy and Viswesvaran (2004). Emotional intelligence can be conceptualised as characteristic adaptive emotional functioning that stems from a mix of cognitive processes and emotional processes (Salovey & Grewal, 2005). Trait emotional intelligence consists of typical emotion-related self-perceptions and tendencies. Dual process models of cognition and learning may provide insight into functions underlying trait emotional intelligence and into how higher emotional intelligence may lead to adaptive outcomes.

Dual processes

Dual process models attempt to explain the basic mechanisms through which humans react to information and benefit from experience. Most dual process models distinguish between two systems, one consisting of associative, experiential processes and one consisting of analytic, explicit processes (Schroyens, Schaeken, & Handley, 2003; Stanovich & West, 2000). The somatic marker hypothesis (Damasio, 1994) emphasises the importance of emotion in associative, experiential processes. Damasio proposed that individuals' decisions are influenced by the emotional valence of outcomes associated with alternative decisions. Damasio described this emotion-based process as rapid, largely unconscious and not governed by rational cost–benefit analyses. In contrast to emotion-based processing, explicit, rule-based and conscious processing characterises the analytic system (Stanovich & West, 2000). Dual process models suggest that these two systems coexist in determining much human functioning.

The Epstein (1994) cognitive experiential self theory provides detail regarding how the systems may operate. This theory postulates an experiential system that is largely unconscious, effortless, and rapid and tied to intuition and affect, and a rational system that is intentional, effortful, logic based, and for the most part affect free. The rational system is under the conscious control of the individual and is assumed to operate in a manner similar to that proposed by traditional decision models. The experiential system, in contrast, operates automatically with little conscious input from the decision maker. This system results in an affective response that is assumed to guide individuals, often at a preconscious level, towards reinforcing choices and away from punishing ones.

Individual differences in experiential and rational processing

Cognitive experiential self theory suggests that individuals differ in their preferences for using each system. To assess these individual differences, Pacini and Epstein (1999) developed a measure, the Rational–Experiential Inventory (REI), which assesses preferences for experiential and rational processing. Research with this inventory indicated that some individuals exhibit a strong preference for one system over the other and that preferences for experiential and rational processing tend to be uncorrelated (Pacini & Epstein, 1999). Individuals scoring high on experiential processing are likely to be agreeable, extroverted, and emotionally expressive; individuals scoring high on rational processing are more likely to believe that they can exert control over the world and to be conscientious; and higher levels of both experiential processing and rational processing are associated with a more positive view of the self, better relationships, a more positive view of the world, and more perceived meaning in the world (Pacini & Epstein, 1999). These results suggest that both experiential and rational processing may serve to enhance wellbeing.

Further, the characteristic that has been described as emotional intelligence may build on an optimal level of rational and experiential processing.

Emotional intelligence

The study of emotional intelligence has focused on operationalising the nature of adaptive emotional functioning and on identifying the outcomes that are associated with adaptive emotional functioning. Salovey and Grewal (2005) described emotional intelligence as resulting from interactions between an individual's emotions and cognitions that lead to adaptive functioning. This conceptualisation provides a point of contact with the dual processing models, which focus on both emotional associations and rational cognitive processes. Models of emotional intelligence propose that adaptive emotional functioning is a characteristic or integrated set of characteristics that comprise an individual difference dimension that can be assessed on a continuum.

Two basic approaches to defining and assessing emotional intelligence are the trait and the ability approach. The trait approach (e.g., Neubauer & Freudenthaler, 2005; Petrides & Furnham, 2003; Petrides, Pita, & Kokkinaki, 2007) holds that emotional intelligence is a stable aspect of personality that manifests itself through typical emotional functioning. Petrides et al. (2007) found that trait emotional intelligence consists of a mix of characteristics identified in established personality taxonomies, such as the Big Five and Giant Three as well as qualities that fall outside of these taxonomies and that trait emotional intelligence has incremental validity over other major established personality characteristics.

The ability conceptualisation (e.g., Mayer, Roberts, & Barsade, 2008) holds that emotional intelligence consists of latent functions that may or may not manifest themselves in daily life. The Salovey and Grewal (2005) characterisation of emotional intelligence as resulting from a confluence of emotional and cognitive processes and the basic functions they described as comprising emotional intelligence may apply to both the trait conceptualisation of emotional intelligence and the ability conceptualisation. When the construct of interest is everyday or typical adaptive emotional functioning, an individual difference approach focusing on emotional intelligence trait functioning may be most appropriate.

Higher levels of emotional intelligence, both measured as a trait and as an ability, have been found to be associated with various positive outcomes, and most especially with indices of wellbeing (Austin, Saklofske, & Egan, 2005; Brackett & Mayer, 2003; Brackett, Mayer, & Warner, 2004; Extremera & Fernádez-Berrocal, 2005; Schutte, Malouff, Simunek, McKenley, & Hollander, 2002; Schutte et al., 2007; Van Rooy, & Viswesvaran, 2004; Wing, Schutte, & Byrne, 2006). There are alternative approaches to operationalising wellbeing (e.g., Csikszentmihalyi & Csikszentmihalyi, 2006; Ryan & Deci, 2001); one important approach emphasises subjective wellbeing. Numerous studies have used life satisfaction, higher levels of positive affect, and lower levels of negative affect as indicators of subjective wellbeing (Lyubomirsky, King, & Diener, 2005). Higher trait emotional intelligence has been found to be associated both with greater life satisfaction (Extremera & Fernádez-Berrocal, 2005; Gallagher & Vella-Brodrick, 2008; Wing et al., 2006; Schutte, Manes, & Malouff, 2009) and with more positive affect (Gallagher & Vella-Brodrick, 2008; Kirk, Schutte, & Hine, 2008; Schutte et al., 2002; Schutte, Manes, et al., 2009).

It should be noted that even though the bulk of research on emotional intelligence has found emotional intelligence to be associated with positive characteristics, in a few studies emotional intelligence has also been found to be associated with potentially maladaptive characteristics, such as lowering of mood after remembering a poor decision (Sevdalis, Petrides, & Harvey, 2007).

Aim

Examining subjective wellbeing in the context of rational–experiential processing and emotional intelligence may provide initial support for a model of the interaction between rational–experiential processing and emotional intelligence, and the role of these as subjective wellbeing precursors. This model proposes that experiential and rational processing styles are a basis for level of emotional intelligence, which in turn influences wellbeing. Based on the premise that emotional intelligence consists of an integration of emotional and cognitive processes, we expect that both more experiential and more rational processing may contribute to higher emotional intelligence. Higher emotional intelligence in turn may promote positive affect and life satisfaction.

Method

Participants

One hundred and fifty-four individuals from various regions of the States of New South Wales and Victoria in Australia participated in the study. Participants were recruited from community populations, including nursing staff, office workers and real estate agents, and from university student populations, including traditional-aged students and mature-aged students. Participants' average age was 40.42 years (SD = 13.90) and 93 of the participants were women.

Materials and procedure

Participants anonymously completed the REI, the Assessing Emotions Scale, the Satisfaction with Life Scale, and the Positive and Negative Affect Scales (PANAS). They either returned the measures in an unmarked envelope to an experimenter in person or mailed back the measures in a pre-paid envelope.

The REI (Pacini & Epstein, 1999) assesses rational processing and experiential processing as two separate dimensions. The REI consists of two 20-item subscales, one assessing ability and engagement for rational cognition (e.g., “I have a logical mind” and “I enjoy intellectual challenges”) and the other ability and preference for experiential cognition (e.g., “I believe in trusting my hunches”, and “I like to rely on my intuitive impressions”). Each response was assessed on a 5-point scale, with end-points of 1 (definitely not true of myself) to 5 (definitely true of myself). Both subscales exhibited a high level of internal consistency for the current sample. In the present study the internal consistency of the Rational Processing Scale, as assessed by Cronbach's alpha, was .89 and the internal consistency of the Experiential Processing Scale was .91. These values are very similar to those reported by Pacini and Epstein (1999) in their development sample.

The Assessing Emotions Scale (Schutte et al., 1998, Schutte, Malouff, & Bhullar, 2009) was used to assess characteristic emotional intelligence. The scale is based conceptually on the Salovey and Mayer (1990) original emotional intelligence model. The 33-item scale assesses how well respondents typically identify, understand, regulate, and harness emotions both in themselves and others. Sample items include “By looking at their facial expressions, I recognise the emotions people are experiencing” and “I know why my emotions change”. Items are rated on a 5-point scale from 1 (strongly disagree) to 5 (strongly agree). In previous research the scale had internal consistency of between .87 and .90, 2-week test–retest reliability of .78, and evidence of construct validity (Schutte et al., 1998, Schutte, Malouff, et al., 2009). In the present study the internal consistency of the scale, as assessed on Cronbach's alpha, was .90.

The PANAS (Watson, Clark, & Tellegen, 1988) measure positive and negative affect. The Negative and Positive Affect Scales consist of 10 emotion descriptors each (such as “interested” and “enthusiastic” for the Positive Affect Scale and “guilty” and “scared” for the Negative Affect Scale). Respondents rate themselves on each descriptor from 1 (very slightly or not at all) to 5 (extremely). The PANAS has different instructions for different time periods, ranging from how the respondent feels at the moment (state affect) to how the respondent generally feels (characteristic or trait affect). The present study used the general (trait) instructions. In previous research internal consistency for the scales using the trait instructions ranged from .85 to .88; 8-week test–retest reliability ranged from .68 to .71, and the scales showed evidence of validity (Watson et al., 1988). In the present research internal consistency, as assessed by Cronbach's alpha, was .83 for the Positive Affect Scale and .88 for the Negative Affect Scale.

The Satisfaction with Life Scale (Diener, Emmons, Larsen, & Griffin, 1985) is a five-item measure of global life satisfaction. Sample items include “In most ways my life is close to my ideal” and “The conditions of my life are excellent”. In previous research the internal consistency of the measure ranged from .82 to .87 and the scale showed evidence of construct validity (Pavot & Diener, 1993). In the present study the scale had an internal consistency of .86.

Results

Out of the original 154 participants four multivariate outliers were identified and deleted from further analyses. Table I lists correlations between measures, means and standard deviations. Higher scores for both experiential and rational processing were associated with higher levels of emotional intelligence. Higher levels of emotional intelligence were associated with greater life satisfaction, more positive affect and less negative affect.

Table I. Means, standard deviations, and correlations (N = 150)

Measures	1	2	3	4	5	6
1. Rational Processing	–
2. Experiential Processing	.02	–
3. Emotional Intelligence	.38*	.50*	–
4. Positive Affect	.28*	.37*	.53*	–
5. Negative Affect	−.25*	−.07	−.29*	−.07	–
6. Life Satisfaction	.08	.31*	.46*	.39*	−.30*	–
M	101.19	95.36	126.07	34.89	17.68	24.84
SD	18.52	18.41	15.16	5.64	6.13	6.53
Note: *p < .001.

Path analysis, using AMOS (SPSS Inc., Chicago, IL, USA), tested the theoretical model positing that both experiential and rationale processing would predict level of emotional intelligence, and that emotional intelligence in turn would predict wellbeing, with emotional intelligence mediating between processing styles and wellbeing. Wellbeing was conceptualised as a latent construct consisting of high life satisfaction, high positive affect and low negative affect. Figure 1 shows the path coefficients.

Figure 1. (a) Original and (b) best fit models of processing style, emotional intelligence and wellbeing.

A structural equation modelling (SEM) model exhibits good fit if the chi-square test is non-significant, the fit indices (comparative fit index [CFI] and Tucker-Lewis coefficient [TLI]) are >.90, and the root mean square error of approximation (RMSEA) does not exceed .05. The model testing showed that for the original model chi-square equalled 25.49 (df = 9), p = .002, CFI = .91, TLI = 85, and RMSEA = .11. Thus, the original model did not reach good fit.

Examination of the modification indices showed that model fit could be improved by (a) removing negative affect as an indicator of wellbeing, and (b) adding a direct path from experiential processing to wellbeing. The modified model is shown in Figure 1. This model indicates that the relationship between rational processing and wellbeing is fully mediated by emotional intelligence, whereas the relationship between experiential processing and wellbeing is only partially mediated by emotional intelligence. For this optimal fit model chi-square equalled 4.89 (df = 4), p = .30, CFI = .99, TLI = .99; and RMSEA = .04. The Sobel test showed that there was significant mediation by emotional intelligence of the experiential process to wellbeing relationship (z = 4.90, p < .001). There remained a significant direct association between experiential processing and wellbeing. There was full mediation of the relationship between rational processing and wellbeing by characteristic emotional intelligence (z = 4.27, p < .001).

Discussion

The results provided information regarding rational–experiential processing style and emotional intelligence as possible precursors of subjective wellbeing. Life satisfaction, high positive affect and low negative affect are commonly used indices of subjective wellbeing (Lyubomirsky et al., 2005). In the present study, higher emotional intelligence was associated with greater life satisfaction, higher characteristic positive affect, and lower characteristic negative affect. A higher level of rational processing was associated with higher characteristic positive affect and lower characteristic negative affect, but was not associated with life satisfaction. A higher level of experiential processing was associated with higher characteristic positive affect and greater life satisfaction, but was not associated with characteristic negative affect. Higher levels of both experiential and rational processing were associated with greater emotional intelligence.

The finding that emotional intelligence was related to indices of subjective wellbeing is congruent with past research (e.g., Schutte et al., 2002; Wing et al., 2006) and with the notion that emotional intelligence, perhaps through facilitating affect management and establishing good relationships, may lead to greater wellbeing. The finding that both rational processing and experiential processing were each associated with two indices of subjective wellbeing identifies possible precursors of wellbeing not previously studied. The finding that both rational processing, which has an analytical cognitive basis, and experiential processing, which is largely unconscious and emotion based, were associated with emotional intelligence lends support to the Salovey and Grewal (2005) description of emotional intelligence as lying at the juncture of cognition and emotion. By identifying processing style as a potential explanatory variable, these findings provide a different perspective on those reported elsewhere in this special issue showing positive associations of emotional intelligence with adaptive coping (Austin, Saklofske, & Mastoras, 2010; Downey, Johnston, Hansen, Birney, & Stough, 2010) and academic success (Hogan et al., 2010).

The model testing of the present study provided some insights into the processes and individual differences that may be precursors of subjective wellbeing. There were significant paths from both experiential and rationale processing to level of emotional intelligence. Emotional intelligence in turn was associated with a latent wellbeing factor, which consisted of characteristic positive affect and life satisfaction. Emotional intelligence mediated between the processing styles and wellbeing, with experiential processing also having a direct effect on wellbeing.

It should be noted that when characteristic negative affect was included in the latent subjective wellbeing factor, the model fit was not optimal. This may be because even though rational processing and emotional intelligence were significantly associated with negative affect, the variance accounted for was relatively low, and experiential processing was not associated with negative affect. Furthermore, there was only a small association between negative and positive affect, suggesting two distinct concepts that cannot both contribute to a higher order factor such as wellbeing. Thus, the best-fit model identified in the present research does not include one of the outcomes often used as an indicator of subjective wellbeing.

The present results should be viewed as preliminary and to be confirmed through replication in other populations and with other measures. Studies of other populations could explore issues of emotional intelligence and processing style profile differences in different groups, as done for emotional intelligence profiles by Sánchez-Ruiz, Pérez-González, and Petrides (2010) in the current special issue. Examining other emotional intelligence measures would contribute to the suggested program of validity comparisons for different emotional intelligence instruments suggested by Gardner and Qualter (2010). Additionally, the number of participants was relatively low for structural equation modelling but power should not have been adversely affected given large effect sizes.

The present study focused on subjective wellbeing. Future research might explore the paths between processing styles, emotional intelligence and psychological wellbeing. Psychological wellbeing (Ryff & Keyes, 1995) includes characteristics such as autonomy and purpose in life. Through experimental methodologies future research might attempt to confirm the causal role of processing styles in the development of emotional intelligence and increases in wellbeing.