Situation selection involves choosing or avoiding some activities, people or places as a function of their expected emotional impact. It is placed at the left-most point in Figure 1 because it affects the situation to which a person is exposed, and thus shapes the emotion trajectory at the earliest possible point. Adaptive situation selection involves knowing oneself and one's needs, forecasting the emotions that various situations are expected to produce (which is not as simple as it may sound; Wilson and Gilbert, 2005), and taking these emotions into account when selecting the situations to which one will be exposed (Loewenstein, 2007; Gross, 2008). Naturally, the costs and benefits in the short and long term must be carefully weighed before making the choice. Two situation selection strategies are confrontation and avoidance.

Confrontation involves choosing to face a situation in spite of the negative emotions it might potentially elicit. This strategy is particularly efficient if the situation is likely to bring long-term benefits. Speaking in public often induces negative emotions in the short term, but avoiding oral presentations in front of one's team might turn counterproductive for future promotions. Two meta-analyses indeed confirm that while confrontation often produces negative emotion in the short-term, it is an efficient strategy to maximize long-term happiness and mental health (Suls and Fletcher, 1985; Aldao et al., 2010).

Avoidance refers to the escape of the situation as a whole. If a situation is unlikely to bring future benefits and if there are no avoidance-related side effects (or if a situation has more detrimental than beneficial effects), then avoidance is often the best strategy. In the other cases, avoidance is likely to become dysfunctional. Research has shown that the chronic use of avoidance is associated with poor indicators of long-term well-being and health (see Suls and Fletcher, 1985; Penley et al., 2002; Aldao et al., 2010 for meta-analyses and reviews).

Although situation selection may be an efficient strategy, one cannot always avoid negative situations/emotions. The families of ER strategies presented next are useful in three main situations: (1) situations that are expected to induce negative emotions but that cannot be avoided due to positive long-term benefits, (2) unexpected situations that cause an unwanted emotion, and (3) situations that induce a conditioned negative emotion.

Situation modification, shown in Figure 1, encompasses the strategies aimed at modifying the situation so as to alter its emotional impact. Three strategies have received special attention: direct situation modification, support-seeking, and conflict resolution.

Direct situation modification (also called problem-focused coping in the stress tradition; Folkman and Lazarus, 1980) involves taking practical actions that impact directly on the situation (e.g., fixing the broken printer; rehearsing one's presentation). This strategy is usually associated with increased well-being and less psychological disorders (see Aldao et al., 2010 for a meta-analysis) as well as better health outcomes (see Penley et al., 2002 for a meta-analysis).

Help/support-seeking involves seeking others' assistance in modifying the situation (e.g., asking a classmate for some help in order to finish homework by the deadline or seeking the help of a counselor to deal with a difficult child). Although there are situations in which instrumental benefits come at a certain psychological cost (Nadler, 1991), the ability to seek—and obtain—help from others has long been judged adaptive by clinical and educational psychologists (Wills, 1987; Newman, 1994).

Conflict resolution involves taking steps to modify (defuse) a conflict situation (e.g., the conflict that my husband and I are having concerning whether to send our daughter to boarding school). While many different strategies can be used (Wall and Callister, 1995), they are not all efficient in reducing conflict (Deutsch et al., 2011). Moreover, while some techniques make it possible to achieve one's ends (e.g., I won: my daughter will stay at home), they can be unsuccessful at addressing the relationship dimensions of the conflict (my husband now agrees with my decision but remains bitter). The opposite could also be true (I may finally agree with my husband but be resentful). Only a few techniques engender satisfaction with both the decision and the relationship (Demoulin, 2014).

Although situation modification strategies make an early impact on the emotion generation process, it is not possible to modify every emotion-eliciting situation: one can neither prevent a sick colleague from coughing loudly every 5 min, nor easily get rid of a tyrannical boss. Other strategies must thus be considered.

Attentional deployment involves altering how we feel by selecting the information we attend to. It comes after situation modification in Figure 1. Existing literature has mainly focused on three forms of attentional deployment: distraction, rumination, and mindfulness.

Distraction involves a shift in attention, either away from the situation altogether or away from emotional aspects of the situation. It therefore includes the physical withdrawal (such as covering the eyes in front of a severely injured body) or the internal redirection of attention (such as focusing on the non-emotional aspects of the situation, or thinking about something else). Distraction has been found to decrease negative emotions (see Webb et al., 2012 for a meta-analysis), especially when associated with problem-focused coping (Shimazu and Schaufeli, 2007).

Rumination refers to a perseverative focus on thoughts and feelings associated with a negative emotion-eliciting event. It has been found to increase the duration and intensity of negative emotions (Morrow and Nolen-Hoeksema, 1990; Bushman, 2002) and to predict the onset, number and duration of depressive episodes over a 2.5 year follow-up of initially non-depressed individuals (Robinson and Alloy, 2003). Accordingly, this emotion regulation strategy is highly prevalent in clinical populations (see Aldao et al., 2010 for a meta-analysis).

Mindfulness (also called “mindful attention awareness” to distinguish it from the ritualized practice of mindfulness meditation) involves purposefully paying attention to the present moment in a non-judgmental way. It consists in observing what is happening moment by moment in one's internal (thoughts, motives, emotions, bodily sensations) and external world, without judging it. Numerous cross-sectional, experience-sampling, experimental and intervention studies show that mindful attention increases happiness (Killingsworth and Gilbert, 2010) and decreases negative affects such as stress, anxiety, or depression (Brown and Ryan, 2003; Hofmann et al., 2010). It is worth noting that the balance between internal attention (what I think/feel) and external attention (what I do) is essential. Simply focusing on one's emotions and sensations does not produce the same benefits and can even be counter-productive (see Webb et al., 2012 for a meta-analysis).

Cognitive change (shown fourth in line in Figure 1) refers to changing the way we think in order to change the way we feel. We can either change how we think about the situation itself or about our capacity to manage its demands. Like the other strategies, cognitive change can be automatic or effortful. As it is difficult to determine whether the conscious appraisal of a situation is the spontaneous initial appraisal of the situation or the product of an automatic reappraisal, we chose to include both appraisals and re-appraisals in this section. The four forms of (re)appraisals that have received the most attention are self-efficacy appraisals, challenge/threat appraisals, positive reappraisal, and acceptance.

Self-efficacy appraisal captures an individual's confidence that s/he is able to deal with the situation (Bandura, 1997). Higher levels of self-efficacy lead to both lower subjective stress and increased cellular immunity (e.g., Wiedenfield et al., 1990).

Challenge and threat appraisals (Lazarus and Folkman, 1984; Tomaka et al., 1997) refer to the gains and losses perceived in an adverse situation. Threat appraisal is thought to occur when an individual appraises a given situation as exceeding his or her resources and/or focuses on the potential/actual losses inherent to the situation (e.g., loss of love and security in the case of a divorce). By contrast, challenge appraisal occurs in situations appraised as taxing resources but in which the individual, while recognizing the potential or actual losses—which makes it different from unrealistic optimism—focuses on the potential or actual gains inherent to the situation (e.g., increase in autonomy and decrease in quarrels in the case of a divorce). Challenge appraisals lead to less subjective stress and less hypothalamic-pituitary-adrenal axis activation than threat appraisals (e.g., Tomaka et al., 1993; Gaab et al., 2005).

Positive reappraisal refers to reappraising a situation or one's response to it in a more positive way. It may consist for instance in looking for the silver lining in the situation, in putting things into perspective or in reinterpreting one's negative emotional response as normal given the circumstances. Studies have shown that reappraisal strategies generally lead to a decrease in negative emotion experience and expression (see Webb et al., 2012 for a meta-analysis). Findings regarding its efficiency to modify physiology are mixed (Webb et al., 2012). While some studies show that reappraisal decreases neuroendocrine and autonomic responses (e.g., Dandoy and Goldstein, 1990; Abelson et al., 2005; Jamieson et al., 2012; Ben-Naim et al., 2013), other studies show that reappraisal increases them (e.g., Denson et al., 2014. Alternative studies have found that reappraisal neither increases nor decreases these responses (e.g., Gross, 1998a). Future studies should uncover the moderators of its effect.

Acceptance refers to accepting the situation and/or one's incapacity to deal with it. It is especially useful in situations that cannot be easily modified or reappraised (e.g., abuse as a child; incurable disease). The acceptance of uncontrollable negative events and the emotions that they elicit has been found to be protective at both psychological (decreases negative emotions) and physical (provides immunity and decreases pain) levels (e.g., Burns et al., 2002; McCracken and Eccleston, 2003). This strategy is not frequently used by people suffering from psychological disorders (see Aldao et al., 2010 for a meta-analysis).

Response modulation is shown on the right side of Figure 1. As this placement indicates, it occurs late in the emotion-generative process, after response tendencies have developed. These strategies can target the experiential (i.e., when one shares one's emotion or drinks a glass of alcohol in order to decrease feelings of anxiety), physiological (i.e., when one smokes marijuana in order to decrease the heart rate, or acts aggressively to defuse physical tension), and/or behavioral (when one attempts to hides one's emotion from others) components of the emotional response. Among response modulation forms, emotion sharing, aggression, substance abuse, and expressive suppression have received the most attention.

Emotion sharing refers to expressing one's emotions in a socially shared language (Rimé, 2007). It typically consists in describing the emotional event that one has just experienced or witnessed. If I just saw a pedestrian hit by a car, I will probably call someone I know to tell him or her about it. People share emotion primarily because they expect it to foster emotional recovery (catharsis effect). Research has however shown that sharing per se does not foster emotional recovery. Nevertheless, sharing emotion is beneficial to mental health due to several indirect effects such as the construction or reinforcement of social bonds and the transference of affection and warmth (see Rimé, 2007 for a review).

Verbal/physical aggression is a strategy used to reduce the bodily tension that arises from an emotional situation (which may or may not be related to the person targeted). Although expressing one's emotions is generally beneficial for both mental and physical health (e.g., Taylor et al., 1997), a number of studies suggest that hostility—and especially its “expressive” dimension—leads to exaggerated cardiovascular reactivity in response to provocative stressors (Suls and Wan, 1993). It also increases the possibility of developing coronary-heart disease (see Miller et al., 1996 for a meta-analysis). Needless to say, this strategy is also very detrimental to social relationships.

Substance use refers to exaggerated consumption of alcohol, drugs, or medicines in order to anesthetize thoughts, feelings, and/or physiological arousal. Although moderate alcohol consumption can have health benefits (for a review, see Baum-Baicker, 1985), the regular use (and thus abuse) of alcohol and drugs as a coping style is associated with poor outcomes in terms of mental and physical health (e.g., Single et al., 2000; Teesson et al., 2000).

Expressive suppression consists of inhibiting the behavioral expression of unwanted emotions (e.g., hiding one's anger). Research has found that suppression is highly prevalent in several psychological disorders (see Aldao et al., 2010 for a meta-analysis). While it does decrease the observable emotion, suppression rarely changes the negative emotion experience (although it may decrease positive emotion). Moreover, it actually increases sympathetic activation of the cardiovascular system (e.g., Gross and Levenson, 1993, 1997; Gross, 1998a; Harris, 2001; Demaree et al., 2006; see Webb et al., 2012 for a meta-analysis). This may explain why expressive suppression decreases well-being (Gross and John, 2003) and increases vulnerability to cardiovascular diseases (see Mauss and Gross, 2004 for review).

Finally, it is worth noting that although it has received little attention, physical exercise can also be used as an emotion regulation strategy.

The emotional intelligence tradition is outcome-oriented rather than process-oriented in the sense that it seeks to capture the outcome of emotion regulation. Individuals are said to display intelligent emotion regulation if they are able to use emotion regulation in a flexible manner and in a way that is consistent with their goals and thus adaptive (Mayer and Salovey, 1995; Bar-On et al., 2003; Mayer et al., 2008). Put differently, emotionally intelligent individuals are those who carefully review the context before deciding whether and how they should regulate their emotion. These individuals take into account and maximize intra-individual and inter-individual long-term survival and welfare. For instance, a boss who up-regulates his anger in order to have his subordinates work overtime will be said to be manipulative but not emotionally intelligent because his emotional behavior will most likely impair his social adaptation in the long-term and lead his subordinates to burn-out. Another feature of emotionally intelligent regulation is that it takes into account the emotional display rules of the culture in which one lives.

In accordance with this adaptation-oriented view, EI has been associated with indicators of superior adaptation in many domains of life. Examples of this are higher life satisfaction (e.g., Petrides et al., 2007b; Di Fabio and Saklofske, 2014b), better health, both objectively measured (Mikolajczak et al., in press) and subjectively reported (see Schutte et al., 2007; Martins et al., 2010 for a meta-analyses), increased social support (e.g., Mikolajczak et al., 2007a), better quality of social and marital relationships (e.g., Schutte et al., 2001a; Lopes et al., 2004; Petrides et al., 2006b; Malouff et al., 2014), and enhanced academic and work performance (see Van Rooy and Viswesvaran, 2004; O'Boyle et al., 2011 for a meta-analysis).

As we previously mentioned, selecting situations depending on the expected emotional impact requires being able to accurately forecast how one is likely to feel in different situations, and then use that information to confront situations that are likely to bring long-term benefits, while avoiding the others. Accordingly, we hypothesized that high EI people would make more accurate affective forecasts comparing to their low EI peers. We also hypothesized that high EI individuals would do their best to prevent negative situations to occur (which should result in less daily hassles) but that they not avoid negative/stressful situations if they were to bring long-term benefits. In the latter case, high EI people would on the contrary struggle to deal with the situation rather than giving up their goal.

Empirical evidence suggests that trait EI increases affective forecasting accuracy (Hoerger et al., 2012). It must nonetheless be noted that Dunn et al. (2007) found only a non-significant trend in a smaller sample. Schutte et al. (2009) have shown that people with high trait EI use that knowledge to select situations more effectively (i.e., spend time in situations that boost wanted emotions or prevent unwanted emotions, seek out situations that inspire positive emotions or avoid those that arouse negative emotions). This may explain why two studies (Ciarrochi et al., 2002; Day et al., 2005) found negative correlations between trait EI and the self-reported frequency of various daily hassles such as marital, professional, health, financial and relationship hassles. It is also congruent with the findings that high trait EI people take more time to relax and wind down (Tsaousis and Nikolaou, 2005).

It is noteworthy that high EI individuals do not simply avoid all situations that could cause them negative emotions. As shown in Table 2, they actually report using less avoidant coping strategies than their peers (Gerits et al., 2004; Petrides et al., 2007a,b; Mikolajczak et al., 2009a, Studies 1 and 2; Rogers et al., 2006; Velasco et al., 2006; Shah and Thingujam, 2008; Kim and Agrusa, 2011; although see Monaci et al., 2013 for insignificant results). This is probably because many situations that cause negative emotions in the short-term have long-term benefits (academic exams for instance). Experimental and field studies support the idea that high trait EI individuals tend to struggle (confront) rather than give up (escape/avoid) in the face of adversity if the confrontation is likely to bring about more substantial long-term benefits. For example, an experimental study by Schutte et al. (2001b) showed that participants who scored higher in trait emotional intelligence solved more problems after encountering a very difficult and frustrating set of problems. This held true even when initial performance was held constant. In the same vein, a field study by Petrides et al. (2006a) showed that trait EI was positively related to the length of musical training among music school students, suggesting that high trait EI individuals do not let themselves be discouraged by the obstacles that stand in their way. Consistent with these results, Bastian et al. (2005)⁷ found that EI was associated to “behavioral disengagement” as people with high EI made an effort to deal with the stressor, and did not give up the attempt to attain goals with which the stressor was interfering.

The studies that investigated ability EI in relation to situation selection strategies show that people with high ability EI make more accurate affective forecasts (Dunn et al., 2007; Hoerger et al., 2012), use less avoidant coping strategies (MacCann et al., 2011) and strive more to attain their goals (Bastian et al., 2005)⁸.

We hypothesized that high EI individuals would take steps to modify a disadvantageous situation. We also expect them to make use of both their social skills and their aptitude to express emotions to prompt situation modification when this requires the intervention of a third party. Therefore, we predicted that high EI individuals would use constructive conflict resolutions rather than avoidant ones. Finally, we hypothesized a positive relationship between EI and restraint (i.e., waiting for the appropriate moment to act, avoiding to act prematurely).

Empirical evidence indicates that when confronted with a negative situation, high trait EI individuals are more likely to modify the situation (Schutte et al., 2009) and take action to change things than their low trait EI counterparts (Salovey et al., 2002 (Study 3); Rahim and Minors, 2003; Gerits et al., 2004; Bastian et al., 2005⁹; Goldenberg et al., 2006; Rogers et al., 2006; Velasco et al., 2006; Petrides et al., 2007a,b; Saklofske et al., 2007, 2012; Almran and Punamaki, 2008; Kluemper, 2008; Mikolajczak et al., 2008, 2009a; Shah and Thingujam, 2008; Austin et al., 2010; Noorbakhsh et al., 2010; Kim and Agrusa, 2011; Moradi et al., 2011; Zomer, 2012; Monaci et al., 2013; Tsarenko and Strizhakova, 2013; although see Montes-Berges and Augusto, 2007 for null results). The relationship between trait EI and the restraint coping strategy does not reach significance though. Namely, people with high trait EI are not significantly more able than people with low trait EI to await the appropriate opportunity before taking action (that is, holding one-self back and not acting prematurely) (Bastian et al., 2005)¹⁰.

Direct problem solving is not the only strategy used by high trait EI individuals when it comes to modifying a situation. As expected, these individuals also make use of indirect modification strategies. First, high trait EI individuals report being more willing to seek help from friends, family, and health professionals in case of problems (Ciarrochi and Deane, 2001; Gerits et al., 2004; Bastian et al., 2005¹¹; Goldenberg et al., 2006; Velasco et al., 2006; Zomer, 2012; Monaci et al., 2013; although see Shah and Thingujam, 2008, for insignificant results). Note that obtaining adequate support may also be easier for them because EI has been associated with increased perceived quantity and quality of social support (e.g., Austin et al., 2005; Mikolajczak et al., 2007a).

Second, high EI people are, by definition, more able and inclined to freely express both positive and negative emotions which, according to Petrides and Furnham's model (2003), translates into higher assertiveness. Thus, instead of fuming at the cigarette smoke of their neighbor at table, high EI individuals would typically ask to have the cigarette put out. However, studies are needed to further test this claim. Third, EI seems to be related to more constructive conflict resolution strategies, although the pattern of results is not entirely clear (Jordan and Troth, 2002, 2004; Salami, 2010b).

Regarding ability EI, the results obtained are consistent with those that use trait measures. Higher ability EI is associated with greater use of problem-focused coping (Goldenberg et al., 2006; MacCann et al., 2011; but see Bastian et al., 2005¹² for non-significant results), although there is no significant relationship between an individual's ability to restrain him or herself (wait for the appropriate moment to act and avoid acting prematurely) and EI (Bastian et al., 2005)¹³. Ability EI also relates to more social support seeking, although effect sizes are nearly null (see Bastian et al., 2005¹⁴; Goldenberg et al., 2006).

Given that high EI people are characterized by greater positive trait affectivity, we hypothesized that they would pay greater attention to positive stimuli/events and ruminate less about negative events. It has indeed been shown that rumination on sad or angry events increases the duration and intensity of negative emotions (Morrow and Nolen-Hoeksema, 1990; Bushman, 2002). Unlike rumination, it is not clear how EI should be related to distraction.

Empirical evidence indicates that high trait EI people rely heavily on attentional deployment techniques to regulate their emotions. First, people with higher trait EI report more mindful attention awareness and pay greater, non-judgmental attention to the present moment (Brown and Ryan, 2003; Baer et al., 2004 (Study 4); Kokinda, 2010; Schutte and Malouff, 2011; Charoensukmongkol, 2014; Wang and Kong, 2014). Consistent with this finding, they report ruminating less about negative/stressful events than their lower EI counterparts (Salovey et al., 2002 (Study 3); Petrides et al., 2007a,b; Ramos et al., 2007; Salguero et al., 2013, although see Mikolajczak et al., 2008 for non-significant results). Schutte et al. (2009) show that higher trait EI people report paying greater attention to things that either help arouse the emotions they desire or prevent them from experiencing emotions they seek to avoid. Similarly, Totterdell and Holman (2003) have shown that high trait EI individuals focus more on positive things than people with low trait EI. These individuals also report that they make more use of distraction to regulate their emotions than their peers (Salovey et al., 2002 (Study 3); Bastian et al., 2005¹⁵; Mikolajczak et al., 2008; Saklofske et al., 2012; although see Austin et al., 2010 for non-significant results).

Results concerning ability EI seem to be less consistent. While Lanciano et al. (2012) found that high ability EI individuals ruminate less than their equivalents, Bastian et al. (2005)¹⁶ showed no significant relationship between ability EI and mental disengagement. Future studies need to examine further the relationship between ability EI and attentional deployment strategies.

We predicted a positive relationship with challenge appraisal because high EI individuals' positive dispositions should lead them to consider not only the potential losses but also potential gains inherent to a situation. Likewise, we predicted a positive link between EI and self-efficacy because high EI individuals' previous successful attempts to regulate their emotions should enhance their confidence to deal with negative events. We also predicted a positive relationship with strategies such as reappraisal and adaptive humor which are useful when the situation cannot be prevented, directly modified, or cognitively avoided (e.g., an academic exam). Finally, we predicted a positive relationship between trait EI and acceptance, but only for problems that could not be modified or reappraised (e.g., being diagnosed with terminal cancer).

Empirical studies show that trait EI is related to how one views a given situation. When asked in January (exam period) how threatening/challenging they appraised the exam session, high EI freshmen judged it as much less threatening than their low EI peers (Mikolajczak et al., 2006). These findings were replicated twice in a laboratory setting. Mikolajczak and Luminet (2008) found that high EI individuals tended to appraise an upcoming arithmetic task as a challenge whereas low trait EI individuals tended to appraise it as a threat.

In addition to influencing how people perceive a given situation, EI also affects how one views one's ability to manage the demands the situation poses (naturally, the two are related). In the above-mentioned study (Mikolajczak et al., 2006), when asked at the beginning of the academic year about their self-efficacy to manage the January exam session, high EI freshmen reported higher self-efficacy than low EI individuals. This was all the more interesting as there was no correlation between trait EI and general cognitive ability (IQ; e.g., Mikolajczak et al., 2007a). In order to rule out potentially confounding variables such as prior knowledge of the subjects or social support, the study was replicated within a laboratory setting (two studies). High trait EI people reported greater self-efficacy to deal with the tasks (analysis of the psychological profile of a movie character in Study 1; an arithmetic task in Study 2) than their low trait EI peers (Mikolajczak and Luminet, 2008). These results are consistent with many others that reveal a relationship between trait EI and self-efficacy (see Adeyemo, 2007 for self-efficacy to pass university exams; Brown et al., 2003 for career decision-making self-efficacy; Animasahun, 2008 for generalized self-efficacy; Chan, 2004 for general self-efficacy; Charoensukmongkol, 2014 for general self-efficacy; Di Fabio and Palazzeschi, 2008 for self-efficacy of teachers to deal with their classroom; Di Fabio and Saklofske, 2014a,b for core self-evaluation and career decision self-efficacy; Durán et al. (2006) for general self-efficacy and self-efficacy to succeed at university; Kaur et al., 2006 for self-efficacy to restrain from gambling; Kirk et al., 2008 for emotional self-efficacy; Kluemper, 2008 for core self-evaluation; Martin et al., 2004 for self-efficacy in counseling; Mouton et al., 2013 for self-efficacy among physical education teachers; Tsarenko and Strizhakova, 2013 for consumer self-efficacy; and Villanueva and Sanchez, 2007 for self-efficacy to perform a laboratory task and to coach and lead followers to perform a laboratory task; although see Salami, 2010a for a non-significant effect of trait EI on general self-efficacy).

When the initial¹⁷ appraisal of a situation and of one's ability to manage it is not sufficient to achieve the desired emotional state, individuals with higher trait EI report being more prone to change the way they think in order to feel or prevent particular emotions (Schutte et al., 2009). In line with this, high trait EI people report greater use of reappraisal strategies than low EI people (Totterdell and Holman, 2003; Bastian et al., 2005¹⁸; Velasco et al., 2006; Mikolajczak et al., 2008; Shah and Thingujam, 2008; Schutte et al., 2009; Moradi et al., 2011; Cabello et al., 2013 although see Kafetsios and Loumakou, 2007, for unsignificant correlations). It must be noted that these results were not replicated by Coumans (2005) who asked participants about the utilization of several coping strategies after a forced failure in a cognitive task. She found that high EI individuals were not more likely than their low EI peers to report using positive reappraisal strategies such as self-esteem rebuilding, putting things into perspective, trying to accept the situation as a part of life, or looking for the silver lining. Future studies are clearly needed to delineate the moderators and boundary conditions of this effect. It is worth noting that the tendency of high EI people to change the way they think in order to modify their feelings does not go as far as denying the problem. Indeed, people with high trait EI report using less denial strategies (Bastian et al., 2005¹⁹; Zomer, 2012; Tsarenko and Strizhakova, 2013). Besides positive reappraisal, it seems that higher trait EI is associated with greater use of humor (Bastian et al., 2005²⁰; Greven et al., 2008; although see Zomer, 2012 for non-significant results) and less maladaptive humor (Greven et al., 2008). Finally, it must be noted that to date, how trait EI relates to acceptance remains unclear. Although Bastian et al. (2005)²¹ found that high EI individuals reported accepting life events more easily, two other studies found no significant correlation between high EI and acceptance (Mikolajczak et al., 2008; Zomer, 2012).

As regards ability EI, these studies are consistent with the results obtained using trait measures in various ways. First, there is a positive relationship between high ability EI and self-efficacy (Kirk et al., 2008; Di Fabio and Saklofske, 2014a,b). Second, people with a high ability EI score report using less denial compared to their peers (Bastian et al., 2005)²². However, the relationship between ability EI and positive interpretation, acceptance and humor is not significant (Bastian et al., 2005)²³.

Because high EI people are expected to have achieved their regulatory goal (namely, the desired emotional state) through the use of the four families of strategies reviewed above, we predicted a negative relationship between trait EI and response-modulation strategies.

As predicted, there is a negative relationship between EI and most response modulation strategies, or at least those strategies whose relationship with EI has been investigated. The only exception concerns “exercise as a mood-regulation strategy,” which shows a positive correlation with trait EI (Solanki and Lane, 2010). To the best of our knowledge, no study has yet examined the relationship between EI and the social sharing of emotions. We would nevertheless expect a negative relationship between EI and the number of times a given emotional episode is shared. Indeed, the more intense the emotion, the greater the extent of sharing (Rimé, 2009). As high scoring EI people have more (and more functional) strategies at their disposal to regulate their emotions, we would expect that they would have less need for emotion sharing. Moreover, we would expect them to better choose the recipient and moment for sharing; they would thus feel better listened to, thereby reducing the need for further sharing.

While no study on trait EI and social sharing exists, two studies have examined the relationship between trait EI and venting (i.e., expressing negative feelings, usually one's anger). Consistent with our expectations, one study found that higher trait EI participants were less likely to vent (Zomer, 2012). However, the other study found no relationship between trait EI and venting (Bastian et al., 2005)²⁴.

A number of studies have analyzed EI and other response-modulation strategies such as expressive suppression, aggression, and substance abuse. As far as expressive suppression is concerned, high trait EI people report suppressing their emotions less (Totterdell and Holman, 2003; Velasco et al., 2006; Johnson and Spector, 2007; Mikolajczak et al., 2007b; Austin et al., 2008; Schutte et al., 2009; Lee and Ok, 2012; Cabello et al., 2013; although see Kafetsios and Loumakou, 2007 for null findings), probably because they have already achieved their desired emotional state and/or because they value genuine emotion expression. Two pieces of evidence support the hypothesis that high trait EI individuals' lesser use of suppression is attributable to a lesser need rather than a lesser capacity. First, although high trait EI people report making less use of suppression, they do not report feeling less capable than their low EI peers to use it when required (Mikolajczak et al., unpublished data). Second, in a study of service workers' emotional labor²⁵, high trait EI employees reported experiencing less emotional dissonance (i.e., dissonance between their inner feelings and the feelings required by organizational display rules) than low EI people. High EI individuals thus needed to perform reappraisal and suppression less frequently than low EI individuals (Mikolajczak et al., 2007b).

As expected, trait EI was also negatively linked to both verbal and physical aggression (Brackett and Mayer, 2003; Shahzad et al., 2013), suggesting that high EI people can defuse their anger and frustration through other means. We must draw attention to the fact that people with high trait EI do not seem to turn their aggressiveness against themselves. There is indeed a negative correlation between trait EI and self-harm (Mikolajczak et al., 2009a), including suicidal attempts (Aradilla-Herrero et al., 2014). High EI people do not therefore need to resort to self-harm to reduce unwanted emotions.

Finally, trait EI is negatively associated with the consumption of alcohol (Brackett and Mayer, 2003; Austin et al., 2005; Schutte et al., 2011; although see Tsaousis and Nikolaou, 2005; Saklofske et al., 2007; Ghee and Johnson, 2008; Monaci et al., 2013 for null results), suggesting that high trait EI people do not need alcohol to anesthetize their feelings. Moreover, they do not seem to (ab)use food to regulate their emotions as there is a negative relationship between trait EI and both Body-Mass Index (Swami et al., 2010), binge eating (Gardner et al., 2014), and bulimia (Markey and Vander Wal, 2007; Pettit et al., 2010; Gardner et al., 2014). It must be noted that the findings on the relationship between trait EI and the use of cannabis and other drugs are inconsistent. While some studies have found the expected negative relationship (Brackett and Mayer, 2003; Riley and Schutte, 2003; Tsaousis and Nikolaou, 2005; Limonero et al., 2006; Rivers et al., 2013), other studies have found non-significant correlations (Bastian et al., 2005²⁶; Limonero et al., 2006; Saklofske et al., 2007; Zomer, 2012).

Regarding ability EI, results are more or less consistent with studies that have used trait measures. There is no significant association between ability EI and venting emotions (Bastian et al., 2005)²⁷. Subjects with high EI ability display less aggressive behavior (Brackett et al., 2004; Rivers et al., 2013), less suicidal ideation, fewer suicidal attempts (Karim and Shah, 2014) and are less likely to use drugs (Trinidad and Johnson, 2002; Brackett and Mayer, 2003; Brackett et al., 2004; Rossen and Kranzler, 2009; Schutte et al., 2011; Rivers et al., 2013; although see Bastian et al., 2005²⁸ for null results). The correlation between ability EI and bulimic symptoms and binge eating is non-significant (Gardner et al., 2014).

The studies reviewed above have focused on the down-regulation of negative emotions, by far the most common target of emotion regulation efforts (Gross et al., 2006). However, although this is a reasonable starting point, much remains to be done. As emphasized elsewhere (Gross, 2014), emotion regulation can occur anywhere in the 2 × 2 matrix formed by crossing negative and positive emotions with up- and down-regulation. Consequently, people not only try to decrease negative emotions (Gross et al., 2006), they also try to increase positive ones (Quoidbach et al., 2010, 2015). On some occasions, they also strategically increase negative emotions (e.g., anger when collecting debts, Sutton, 1991; anger before engaging in a confrontational task, Tamir et al., 2008; worry and fear when anticipating a threatening task, Tamir et al., 2007), or decrease positive emotions (e.g., amusement during a serious meeting; Gruber et al., 2011).

We believe that research on individual differences in emotion regulation would greatly benefit from integrating research all four facets of the emotion regulation goals matrix. For instance, are high EI individuals are as efficient in upregulating negative emotions as in downregulating them? The up- and down-regulation of positive emotions must also be examined. This kind of research is especially warranted as it is not obvious, for instance, that EI will be conductive of efficient down-regulation of positive emotions. It may well be the case that high trait EI individuals experience difficulties in down-regulating positive emotions and up-regulating negative emotions. Studies that delve deeper into these questions are crucial as this issue has both theoretical and practical implications.

At the theoretical level, it is important to know whether an individual can perform at above average levels in each of the 2 × 2 ER matrix cells, or whether the price to pay for above average down-regulation of negative emotions/up-regulation of positive emotions is below average up-regulation of negative emotions/down-regulation of positive emotions. If the latter is true, EI definitions and theories should be amended accordingly. At the practical level, if measures of trait EI capture people who are skilled in the down-regulation of negative emotions/up-regulation of positive emotions but whose down-regulation of positive emotions/up-regulation of negative emotions is impaired, then those scoring high on EI measures would not be suitable candidates for positions such as bill collectors, funeral directors and so on. Research is also needed to understand how EI affects composite instances of emotion regulation. Indeed, individuals sometimes try to regulate multiple and conflicting emotions such as simultaneous feelings of pride about one's achievement and sadness/concern about the failure of a friend, or both joy and guilt when one indulges in some forbidden pleasure.

We have provided some evidence that high EI individuals are open to emotions and leave room for their emergence. Thus, they do not always regulate their emotions, but appear to know how to do so efficiently when necessary. Although findings reported so far seem to associate EI with the wise utilization of emotion regulation, the ultimate evidence in favor of the adaptive nature of EI would be to show that it correlates with superior indicators of adaptation, such as better mental and physical health.

A recent meta-analysis of the relationships between EI and health (105 effect sizes, 19,815 participants; by Martins et al., 2010) showed that EI was positively associated with mental health (r = 0.36) and self-reported physical health (r = 0.27). Mikolajczak et al. (in press) recently confirmed the positive association between trait EI and health using objective indicators of physical health available from participants' Mutual Benefit Society records (i.e., doctor consultations, hospitalization, drug consumption) (r = 0.11). This relationship between trait EI and physical health is not surprising as mounting evidence suggests that different emotion regulation strategies exert a distinct influence on physical health (Gross, 2013). For instance, researchers found that reappraisal was associated with lower levels of C-reactivity protein, whereas suppression was associated with higher levels of C-reactivity protein (Appleton et al., 2013). In another study, Kubzansky et al. (2011) found that self-regulation (understood as a psychological asset that enables individuals to manage feelings, thoughts, impulses, and behavior, with the capacity to regulate emotions serving as a central component) predicted a decreased subsequent risk of heart attacks and coronary heart disease in men, even when controlling for traditional coronary risk factors.

Several other factors may explain this relationship between EI and physical health. First, if high EI individuals regulate their emotions at the earliest possible point, they should experience less prolonged arousal in response to a negative situation/stimuli and thus be protected vis-à-vis the deleterious consequences of chronic arousal on physical health (e.g., coronary heart disease, gastro-intestinal disorders, asthma, psoriasis, migraine etc; Chrousos, 2000; Thurin and Baumann, 2003). Second, if high EI people do not need alcohol and drugs to anesthetize unwanted feelings, they should be consequently less at risk for substance-use related health problems such as cirrhosis of the liver, pancreatitis, and polyneuropathy. Third, their increased emotion regulation efficiency should have a positive effect on their sleep, which is known to be crucial to health (Belloc and Breslow, 1972; Pilcher and Ott, 1998; Gottlieb et al., 2006). Brown and Schutte (2006) indeed showed that higher trait EI is associated with better quality and more refreshing sleep.

Authors in the ER tradition have repeatedly warned against classifying strategies as irremediably (i.e., always) good or bad (Barrett and Gross, 2001; Gross and Thompson, 2007). They argue that the efficiency of a strategy depends on the context and the angle from which we see things (Aldao et al., 2015).

To illustrate, we can take the example of a man who prides himself on being strong and independent, and thus suppresses his anger and sadness when his girlfriend confesses that she cheated on him. The success or value of this strategy depends on the criteria adopted. In this case, his reaction could be deemed successful with respect to the regulator's goals (the strategy used allows him to meet his goal: appearing strong and independent in the eyes of his girlfriend), with respect to some social norms (i.e., his reaction is aligned with social norms concerning avoidance of unmanly reactions such as crying and violent reactions such as hitting), and with respect to the relationship from a short-term perspective (the strategy used makes it possible to avoid a fight). This same strategy could be deemed unsuccessful with respect to the relationship from a long-term perspective (his girlfriend does not get a chance to know how hurt he really is. She might therefore cheat on him again, leading eventually to a breakup), and possibly with respect to the man's long-term adaptation (by hiding his emotions, he prevents people from sensing his needs and exposes himself to further negative emotions, thereby endangering his long-term well-being and health).

Whereas we believe that it is crucial to understand that a given emotion regulation strategy cannot be considered as adaptive or maladaptive per se—namely, irrespective of the context, the temporal perspective, and the individual's goals—this review nonetheless supports the idea that different emotion regulation styles (i.e., the repeated use of a given emotion regulation pattern) carry different consequences for longer-term adaptation. Inasmuch as EI predicts successful adaptation in a number of domains (e.g., work performance, marital/social relationships), the fact that it is positively associated with some strategies (e.g., problem-focused coping, reappraisal) and negatively with others (e.g., rumination, substance use) suggests that some regulation styles are more adaptive than others. This view is consistent with preliminary empirical evidence obtained in the emotion regulation tradition (e.g., Gross and John, 2003; John and Gross, 2007) and the accumulated evidence in coping research (e.g., Zeidner and Endler, 1996) and in other specific domains (e.g., Nolen-Hoeksema et al., 1993 regarding rumination).

The foregoing theoretical consideration suggests that research in the ER tradition would benefit from delving more deeply into the investigation of the consequences of different emotion regulation styles for adaptation. For instance, what are the relative costs and benefits of each regulation family? Aldao et al.'s 2010) meta-analysis showed that some ER strategies are most represented in psychopathological groups, thereby suggesting that these may be less effective. The meta-analysis carried out by Webb et al. (2012) nicely complemented these findings by comparing the relative efficiency of the various ER strategies regarding experimentally-induced emotions. A meta-analysis that could compare the relative short- and long-term efficiency of ER strategies regarding real-life events is needed to complete the picture, but the current paper provides a preliminary idea of what the findings may look like (at least in the long-term perspective).

We further believe that the real-world contexts explored by EI researchers constitute a crucial testing-ground for the predictions made in the ER tradition. Gross (2002) and Gross and John (2003) have investigated the social consequences of the use of reappraisal and suppression but this work should be extended to other regulation strategies and other domains of life. One important issue to be addressed concerns the outcomes of different emotion regulation styles in family, friendship, educational, and work settings. For instance, do people who chronically use reappraisal have better social and marital relationships? Are they more efficient at work than those who use distraction or emotion expression? Another issue that needs to be addressed concerns the nature of the benefits/costs of each regulation style. Indeed, as we have emphasized above, strategies can be differentially successful depending on the criteria adopted (short term vs. long term, self vs. other vs. the relationship, emotion regulation vs. task efficiency). Accordingly, it is conceivable that reappraisal and distraction are equally efficient in decreasing the negative affect induced by a given task, but that the former is more efficient than the latter regarding one's performance of the task. Studies such as these are needed to determine whether some strategies are more successful than others in maximizing adaptation in a large number of domains.

Finally, it is possible that the key to successful adaptation lies not only in the use of effective strategies but also in the breadth of one's ER repertoire and in the flexibility with which one uses these regulation strategies. Studies are urgently needed to address strategic flexibility, namely the ability to modify one's regulation style when it is not appropriate for the situation at hand (Bonanno et al., 2004; Aldao et al., 2015). Strategic flexibility seems indeed essential for one's adaptation to a complex and changing environment such as ours, in which a usually functional regulation style can at times turn out to be dysfunctional (e.g., the rigid utilization of a usually functional regulation style such as problem-focused coping can have serious costs in uncontrollable situations such as an incurable disease).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.