Dynamic self- and other-focused emotional intelligence: A theoretical framework and research agenda Journal of Research in Personality

This paper introduces a theoretical framework for research on the dynamics of self- and other-focused emotional intelligence (EI). The EI-framework focuses speciﬁcally on the interplay between different EI dimensions when individuals are processing their own emotions and the emotions of others. The frame- work captures different phases of processing self- and other-emotions. The ﬁrst phase consists of situational cues (e.g., an argument) that elicit interdependent emotions in the self and others. The next phase speciﬁes differential and interacting effects of EI dimensions when processing these emotions, and the third phase describes what proximal and distal consequences this processing may have. In addition, the framework includes candidate dispositional and contextual factors (e.g., emotion type, motivation) that may qualify the process. (cid:1) 2020 The Authors. Published by Elsevier


Introduction
Emotional experiences are an inseparable part of individuals' daily life. Emotions such as pride, anger, and shame influence experiences. They guide behavior, prioritize goals, and they communicate our mood states to others (Frijda, 1986). It is therefore not surprising that emotions influence both the self and the other when individuals are interacting. That is, emotions can be expressed towards others, they can elicit emotions in others, or they can be reactions to the emotions of others (Fischer & Van Kleef, 2010). During social interactions, individuals thus not only need to appraise and regulate their own emotions; they also need to keep track of the emotions of their interaction partner to facilitate the interaction and achieve what they want. Some individuals are better at this than others, and part of these individual differences is reflected in emotional intelligence (EI). EI can generally be described as the ability or knowledge to perceive, understand, and manage emotions (Mayer & Salovey, 1997;Petrides, 2011;Zeidner, Roberts, & Matthews, 2008). High-EI individuals tend to deal with emotions in such a way that their reactions are socially effective which may help them to reach their goals in various life domains. For example, the emotions that a sports coach expresses during a race may motivate and enable an athlete to reach new performance levels. The manifestation (i.e., enactment) of EI may even influence important life outcomes. To illustrate, EI is positively associated with a satisfying social life (Lopes et al., 2004;Schutte et al., 2001), health (Martins, Ramalho, & Morin, 2010), and job performance (Joseph & Newman, 2010;O'Boyle, Humphrey, Pollack, Hawver, & Story, 2011).
Although research has shown that EI is an important determinant of social behavior, virtually nothing is known about the more fundamental processes that are responsible for its effects (cf. Côté, 2014;Peña-Sarrionandia, Mikolajczak, & Gross, 2015). EI research typically focuses on individual differences and examines the construct at the global level using cross-sectional research designs or longitudinal designs with long time lags. Such approaches, however, discard the situational factors that may trigger the enactment of EI (Ybarra, Kross, & Sanchez-Burks, 2014), the interpersonal influence of emotions (Hareli & Rafaeli, 2008), and the temporal aspects of EI (Roe, 2008). Consequently, these traditional research designs cannot examine when EI is activated, how the emotions of one individual influence the emotional responses of another individual, and what the proximal and distal consequences are of dealing with one's own versus others' emotions.
In the present paper, we propose a theoretical framework for studying the dynamic processes involved in self-and otherhttps://doi.org/10.1016/j.jrp.2020.103958 0092-6566/Ó 2020 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/). focused EI. This framework can serve as a starting point for the empirical investigation of more detailed processes that may play a role in the enactment of EI. For example, our theoretical framework provides ideas on how the emergence of emotions in the self and others activates various EI facets or dimensions, how one's own and others' emotions are subsequently processed, and what consequences this processing has over time. The focal person in our framework is the self. Yet, we will also encourage EI researchers to include the expressed emotions of others as a significant element in their studies, as one's own and others' emotions have a mutual influence on each other in interpersonal situations and because people typically interact with others (Butler, 2011;Hareli & Rafaeli, 2008). Accordingly, the framework incorporates the following core components: (1) situational cues that elicit (interdependent) emotions in the self and the other, (2) differential and interacting effects of EI dimensions when processing these emotions, and (3) dispositional and contextual factors that may qualify this process. In addition, the theoretical framework allows looking at proximal consequences of processing emotions and more distal consequences of repeatedly processing emotions. We propose that integrating these understudied aspects into EI research may be crucial to better understand how it affects our daily lives.
The central time unit in the proposed theoretical framework is the emotional episode. An emotional episode starts when a situational cue (i.e., an event or emotional expression of another person) elicits emotions in the self and/or the other, and it ends when these emotions have been processed, and their experience has faded (Scherer, 2000). In other words, it is the entire period in which an individual emotionally deals with the situational cue. Emotional episodes usually consist of a mixture of different emotions that quickly follow one another, blend together, and are experienced by people as being a unity (Frijda, Mesquita, Sonnemans, & Van Goozen, 1991). One example of an emotional episode is the experience of the nervous, tense, and anxious feelings before presenting a public speech. These emotions may start to arise when a person wakes up on the day of the talk, and they usually elevate until the person is on stage. Feelings may peak until the moment the talk is over and the audience applauds. As of that moment, tension may drop abruptly, which marks the end of this emotional episode. Conceptually, emotional episodes can be distinguished using the following two criteria: (1) an emotional episode constitutes a continuous emotional impact elicited by one well-defined event, and (2) the feelings during the emotional episode are experienced by individuals as belonging together (Frijda et al., 1991). The end of an emotional episode is typically the moment in which an individual has successfully dealt with the situational cue, or when an individual decides to abandon efforts to emotionally deal with it (Frijda et al., 1991). Emotional episodes may vary in duration. Some episodes last a couple of seconds (e.g., annoyance because of the rainy weather), others may last hours (e.g., stress for an exam). This means, essentially, that emotional episodes can overlap, for example when a person is nervous for an interview (situational cue 1), and simultaneously is frustrated by the morning grumpiness of her daughter (situational cue 2).
One could say that emotional episodes capture a ''snapshot" of individuals' emotional life. Therefore, the current framework can be used to examine EI from a completely different perspective than the typical cross-sectional perspective that is dominating the EI literature; namely when individuals are actually using their EI. Such episodes are consistent with affective events theory (Weiss & Cropanzano, 1996) and the performance episodes model (Beal, Weiss, Barros, & MacDermid, 2005), as these theories use cues or regulatory goals to divide time in meaningful units. This episodic perspective is applicable to EI because it connects well with the lively nature of emotional experiences (Beal et al., 2005;Frijda, 1993;Weiss & Cropanzano, 1996). That is, emotions come and go, or they may evolve in different types of emotions. Furthermore, focusing on emotional episodes allows examining the manifestation of EI within time periods with a recognizable start and end, which may be helpful to reveal its underlying processes. This episodic perspective can also describe how the outcomes of one emotional episode may influence the next emotional episode (Barker, 1963;Weiss & Cropanzano, 1996), and eventually shape life outcomes.
The process-based approach in our theoretical framework also maps well onto conceptual and methodological developments in social science that enable studying psychological processes ''in vivo" (Kahneman, Krueger, Schkade, Schwarz, & Stone, 2004;Oerlemans & Bakker, 2013;Reis & Gable, 2000). Specifically, there is a shift in social science from traditional one-time survey methods to more dynamic process-based approaches that allow for a distinction in causes, mechanisms, and consequences. Such approaches have already become influential in the fields of stress (Lazarus & Folkman, 1984), emotion regulation (Gross, 1998(Gross, , 2015, and emotional labor (Grandey, 2000;Grandey & Melloy, 2017), and have resulted in inspiring and creative research agendas. Our proposed framework, like the aforementioned frameworks, also intends to clarify dynamic emotional functioning. It goes beyond them, however, in four important respects. First, the multilevel format of our framework allows examining individual differences in EI (which is characteristic of the EI literature) and within-person emotion processes (which is characteristic of the stress and emotion regulation literature; Peña-Sarrionandia et al., 2015), in tandem. This enables to examine what high-EI individuals do when they are confronted with emotions in daily life and in which phases of the emotion process they excel. As such, it may guide research that focuses on which specific steps in emotion processing distinguishes high-from low-EI individuals. In doing so, we connect two fundamental emotion research traditions that may mutually inform and complement each other. Second, our framework incorporates a broader scope of emotional abilities (i.e., emotion appraisal and emotion regulation) than earlier models. Hence, it is unique in offering direct and testable predictions of what happens when individuals appraise (negative) emotions but are not able to regulate them (or vice versa). Third, our framework explicitly focuses on two emotion processes: processing one's own emotions and processing others' emotions. As such, it offers a unique framework to examine how these processes may run parallel, interfere, or build upon each other. This addition may be a more realistic representation of the social nature that emotion processing has in daily life (Niven, 2016). A final value-added dimension of our framework is that it does not solely describe work-related emotion processes (Grandey, 2000;Grandey & Melloy, 2017) or personal emotion processes (Gross, 1998(Gross, , 2015, but may be applicable more generally to various life domains.

Emotional intelligence
The literature on EI can roughly be divided in two main approaches that differ in their conceptualization and measurement of EI (Siegling, Saklofske, & Petrides, 2015). The ability-approach is largely based on the Four-Branch Model of EI (Mayer & Salovey, 1997). In this model, EI is conceptualized as a set of interrelated emotional abilities organized in four branches. The branches consist of (1) the ability to perceive emotions, (2) the ability to use emotions to facilitate thinking, (3) the ability to understand emotions, and (4) the ability to manage emotions to reach (interpersonal) goals. Characteristic of this approach is its measurement with performance-based EI tests, which is comparable to the way cognitive abilities are measured. In contrast, the trait-approach conceptualizes EI as a set of emotion-related traits and uses selfreport instruments to measure EI. This approach is more similar to research conducted in the personality field. Next to these two main approaches there also is a more hybrid line of research in which EI is conceptualized as ability (i.e., based on the fourbranch model), but measured with self-report questionnaires (i.e., similar to the trait-approach; O'Boyle et al., 2011). Correlations between the performance-based and the self-reported measurement of EI are low to moderate (Brannick et al., 2009;Petrides, 2011), which either confirms their theoretical difference or reflects methodological variance.
Despite the unique strengths and weaknesses of each approach, a relevant limitation in both approaches is their strong reliance on ''traditional" methodologies to examine EI. That is, scholars typically use global EI scores that mask the unique role of selffocused EI (dealing with emotions of the self) versus otherfocused EI (dealing with the emotions of others). However, recent research suggests that self-focused EI is particularly relevant to remain healthy, whereas other-focused EI contributes particularly to social and performance outcomes (Brasseur, Grégoire, Bourdu, & Mikolajczak, 2013;Mikolajczak et al., 2015;Pekaar, Bakker, Van der Linden, & Born, 2018). It is therefore likely that the manifestation of EI involves a dual process. On the one hand, individuals appraise and regulate their own emotions; and on the other hand, they appraise and regulate the emotions of others. In line with this dual-process-notion, several modern EI instruments allow to make this distinction. Studies in which these instruments were used show that the observed correlations between self-and otherfocused EI typically range between 0.17 and 0.54 (average r = 0.34) across different samples (Brasseur et al., 2013;Pekaar et al., 2018), suggesting that self-and other-focused EI can also empirically be delineated. A further limitation of using global EI scores is that they do not clearly reveal the different steps involved in processing an emotion. Yet, in real life, an emotion needs to be appraised first before it can be regulated (Joseph & Newman, 2010). In fact, research has shown that the final step in emotion processing, emotion regulation, is crucial to influence outcomes such as job performance (Joseph & Newman, 2010). Hence, global EI scores may not be optimal indicators to predict whether and how the process of dealing with emotions actually unfolds.
In order to capture this stepwise and dual process, the framework that we introduce in the current paper proposes to use four distinct EI dimensions, namely self-focused emotion appraisal and emotion regulation, and other-focused emotion appraisal and emotion regulation. These four EI dimensions emphasize the aforementioned distinction between self-and other-focused EI that we find relevant as this distinction may reflect different psychological processes related to health or social effectiveness (Brasseur et al., 2013;Pekaar, Bakker, Born, & Van der Linden, 2019). The four included EI dimensions also differ in their level of complexity of emotion processing. Specifically, emotion appraisal is considered the most basic level of emotion processing, whereas emotion regulation is the most complex level (Joseph & Newman, 2010;Mayer & Salovey, 1997). Hence, a distinction in these four EI dimensions may be useful to design studies on the effectiveness of individuals' response to self-and other-emotions in daily situations. Given that our framework aims to examine typical performance in EI (i.e., how people typically deal with emotions during emotional episodes in their daily life), it may be more feasible to use self-report EI instruments to capture the phenomenon and process. Although, different scholars clearly seem to have different ideas about the relative strengths and weaknesses of self-report versus performancebased EI measures, in the literature it often is assumed that selfreport EI measures maybe more in accordance with typical EI performance. As such, self-report measures may conceptually but also methodologically fit well with our model. The methodological aspect refers to the point that self-report measures may interfere less with an ongoing situation than performance-based tests.
The following sections introduce our framework in which selfand other-emotions, different EI dimensions, and dispositional and contextual factors interact to allow for a more real-life examination of EI than the contemporary EI literature has offered. An important asset of the current framework is the placement of emotion processing at the episodic level. This episodic perspective allows to explain the actual manifestation of EI as a process that is not only confined to those who excel in emotional skills or knowledge (high levels of EI), but as a process that people in general go through when they are confronted with emotions. In other words, our framework may also inspire research into the ways low-EI individuals process emotions in their daily life.

Theoretical framework for research on dynamic self-and other-focused emotional intelligence
Below, the full model of our proposed framework is presented (see Fig. 1). The first phase of the framework consists of a situational cue eliciting emotions in the self and/or in others. The nature of this situational cue may differ widely and can be adapted to a specific study domain. Examples may range from an angry outburst of a friend eliciting feelings of stress or fear, to the smell of a specific perfume that makes one nostalgic and yields slight feelings of sadness. These emotions, in turn, activate EI-related processes in order to deal with the experienced emotions (cf. Grandey & Melloy, 2017;Gross, 1998). Importantly, this emotion processing phase involves a dynamic interplay between different EI dimensions. For example, the effectiveness of regulating one's own emotions may affect the effectiveness of appraising and regulating others' emotions. The processing of emotion is further qualified by dispositional and contextual factors such as motivation, emotion type, and the relationship between the self and the other. Furthermore, the framework provides ideas on how the process may spiral over time and how it may influence proximal (i.e., episodic performance) and distal outcomes such as health and overall job performance. In the following sections, we will elaborate on the main components as depicted in our framework (see Fig. 1).

Situational cues
The emotional episode in this framework starts with a situational cue that elicits emotions. This situational cue can either be an external or internal event eliciting emotions (labelled ''Event" in Fig. 1), or it can be the emotional expression of another person. In both scenarios, the key is that there needs to be input that generates emotions in the self and/or the other to activate an individuals' emotion processing. For example, smelling a specific perfume may produce feelings of nostalgia that need to be regulated to not interfere with (work) tasks, or, witnessing a friend's grief may produce feelings of pity in the self (Hareli & Rafaeli, 2008). The prominent place of the situational cues in the framework emphasizes that the enactment of EI depends on them. For example, it may be interesting to manipulate such emotional cues in a lab experiment, or, alternatively, to focus on specific emotional demands that individuals face in a field setting.
Emphasizing the role of context is in itself not new in the EI literature (Côté, 2014;Jordan, Dasborough, Daus, & Ashkanasy, 2010;Ybarra et al., 2014). Previous research already suggested that the context might determine what consequences EI has (Joseph & Newman, 2010). However, approaching situational cues as the initiator of EI responding is new, and may thus inspire innovative creative research designs. In doing so, we draw from Trait Activation Theory (Tett & Guterman, 2000), which states that (personality) traits are only enacted in situations that provide the opportunity for trait expression. According to this theory, which was developed in the context of work, there are five situational features that potentially activate or deactivate a trait (Tett & Burnett, 2003). These features are job demands (prescribed behaviors that may go along with the expression of specific traits), distractors (cues that evoke spontaneous trait expressions that interfere with tasks or duties), constraints (situations that do not allow for the expression of a trait), releasers (chances to express a trait in constraining situations), and facilitators (emphasis on trait-relevant information that makes the expression of a trait more likely). Following this theory, we propose that such situational features may also boost or inhibit the enactment of EI (i.e., also in non-work-related situations). Indeed, both the emotion regulation literature (Gross, 1998) and the emotional labor literature (Grandey, 2000;Grandey & Melloy, 2017) have proposed that regulatory efforts start with a relevant emotional event. Consequently, the situational features that characterize this emotional event, such as social norms (i.e., equivalent to job demands), facilitators, and constraints, may determine whether and how people process their own and others' emotions. Hence, we argue that the trait activation principle of situational cues may well apply to the manifestation of EI as it could explain why EI matters in some, but not all, situations (Ybarra et al., 2014).
The framework distinguishes two scenarios. In the first scenario, there is only the individual, not interacting with another individual. This scenario is depicted in the darker grey pathway in Fig. 1. In this scenario, the situational cue may consist of any stim-ulus that is (potentially) important for this individual such as a thought, a bodily state, or a thunderstorm. It is relevant to note that it is usually not the stimulus in itself that evokes emotions, but rather the meaning that the individual assigns to it (Frijda, 1988). When the individual is alone, the emotion processing solely focuses on emotions of the self. Most contemporary emotion theories, including EI, have taken such an intrapersonal perspective, which has yielded valuable insights in how individuals emotionally respond to a stimulus. However, this perspective may not be applicable to individuals in interpersonal situations because such situations comprise emotions of the self and others. Remarkably, these interpersonal situations tend to be the type of situations in which EI is most relevant (Joseph & Newman, 2010;Lopes et al., 2004).
Therefore, the second, and more intriguing, scenario of the framework is a situation in which an individual is interacting with another individual. In this social interaction, the emotions expressed by the other individual may function as a situational cue that elicits emotions in the self. Hence, there are two emotions; emotions of the self and emotions of the other (see the light and darker grey pathways in Fig. 1). This scenario is put central in our framework because EI involves the knowledge or ability to effectively deal with one's own and others' emotions (Mayer & Salovey, 1997;Petrides, 2011;Zeidner et al., 2008). That is, the inclusion of self-and other-emotions may actually be a vital element in research on the manifestation of EI as both sources of emotions may activate different EI dimensions. Furthermore, as humans are social by nature, emotions most frequently originate from interpersonal contexts (Butler, 2011;Hareli & Rafaeli, Fig. 1. Theoretical framework for research on dynamic self-and other-focused emotional intelligence. The proposed framework includes different phases that may be relevant when studying EI processing; the situational cue phase, the emotion processing phase, and proximal consequences that may develop into distal consequences. The light grey pathway symbolizes the process of dealing with others' emotions and the darker grey pathway symbolizes the process of dealing with one's own emotions. During the emotion processing phase, specific EI dimensions interact which is depicted with the dashed arrows. The contextual and individual differences that are displayed above the episodic process incorporate candidate moderating factors that may influence the process. 2008). That is, individuals adapt or respond emotionally to the emotions of others through processes as emotion contagion (Hatfield, Caciappo, & Rapson, 1994) or emotional interpretation (Hareli & Rafaeli, 2008). There is an emerging research field devoted to these social effects of emotions that may, using our theoretical framework, be elegantly combined with the literature on EI (see for example Barsade, 2002, Parkinson & Manstead, 2015Van Kleef, 2016). For instance, it could be interesting to study whether interacting with individuals who are experiencing negative emotions affects one's own emotion regulation capability.
To recapitulate, situational cues such as internal or external events that are salient for the self or that relate to expressed emotions by others seem to be conditional factors that initiate emotion processing. Our framework proposes to include situational cues in EI research because without such cues, EI may not be enacted in the first place. This notion may inspire scholars to formulate innovative hypotheses, such as the prediction that individual differences in (enacted) EI may become more salient during emotionally demanding episodes, because such episodes provoke the use of EI. Alternatively, it may be interesting to test whether during social interaction, the processing of others' emotions may occur at the expense of the processing of one's own emotions, because the presence of others may take attentional resources from own emotional experiences.

Emotion processing
The next phase of the theoretical framework describes the emotion processing, which is important to adapt to situational demands and has been associated with higher well-being and social effectiveness (Gross & John, 2003;Van der Linden et al., 2017). In fact, research showed that people who do not effectively process and flexibly align their emotions with their goals and the context tend to suffer from psychological maladjustment (Kuppens, Allen, & Sheeber, 2010). Moreover, a lack of emotion processing in interpersonal situations creates a risk to become trapped in maladaptive emotional patterns that may escalate and ruin the quality of social relationships (Snyder, Stoolmiller, Wilson, & Yamamoto, 2003). Hence, self-and other-focused emotion processing is vital to meet situational demands and to smoothen social interactions. During social interactions, various specific EI dimensions may be activated. The emotion processing phase of our framework describes how effective these EI dimensions are enacted. The more accurately an emotion is recognized and interpreted, the more effectively self-or other-focused emotion appraisal will be (Bänziger, Grandjean, & Scherer, 2009). In contrast, when an emotion is inaccurately recognized, falsely labeled, or misunderstood, (self-or other-focused) emotion appraisal is assumed to be ineffective, possibly leading to maladaptive responses. The effectiveness of (self-or other-focused) emotion regulation depends on the success of the regulatory actions to change the intensity of an emotion in the desired direction (Pekaar et al., 2018;Sheppes & Gross, 2012). For example, selffocused emotion regulation is effective when an individual who feels stressed, starts to feel calmer after an intended walk outside.
One contribution of the current framework is its emphasis on interactive effects that may occur when various specific EI dimensions are used. To illustrate, consider the situation in which a cranky manager frustrates an employee. In this case, a first step for the employee is to appraise his own frustration, but in order to quickly reduce these negative emotions, subsequent regulatory actions are required such as counting to ten or reappraising the situation. Moreover, in order to effectively communicate with the manager, it may be helpful to down regulate the emotions of the manager first. This example shows how the appraisal and regulation of self-and other-emotions can be simultaneously or sequen-tially activated, and can influence each other. Accordingly, we propose that, during an emotional episode, various combinations of EI dimensions may evolve, which, in turn influences the effectiveness of one's response. Those combinations may occur at two levels: (1) The intrapersonal level, in which engaging in emotion appraisal and emotion regulation affect each other, and (2) the interpersonal level, where the processing of one's own emotions interacts with the processing of others' emotions. In the following sections, we first focus on the intrapersonal interplay between EI dimensions before elaborating on the interpersonal interplay between EI dimensions.

The intrapersonal interplay between EI dimensions
The interplay between EI dimensions at the intrapersonal level is based on the Modal Model of Emotion (Gross & Thompson, 2007) positing that processing emotions is a stepwise process in which emotion regulation is the ultimate step through which emotions are modified. Consequently, it is mainly this regulation that may influence outcomes (George, 1991). Previous meta-analytic studies confirmed that the effect of emotion perception on job performance is mediated by emotion regulation (Joseph & Newman, 2010). However, this stepwise process also implies that when the final step is missing, for instance when emotions are only appraised but not regulated, the subsequent response may be less adaptive. A typical example would be a person who is very able to recognize emotions, yet does not have the ability to adequately regulate these emotions. Hence, we propose that emotion processing is more effective when emotion appraisal is followed by emotion regulation.

The interpersonal interplay between EI dimensions
Another aspect of the framework that goes beyond previous research is that it emphasizes conflicting or facilitating effects of processing one's own and others' emotions. One facilitating effect is that processing the emotions of others may be more effective when one's own emotions have already been regulated. In such a situation, having finished the step of regulating one's own emotions is assumed to free energetic and processing resources that can be used to focus on the emotions of the other. This proposition is consistent with lay person's beliefs that ''You first need to help yourself before you can help the other". Accordingly, previous research showed that employees who had effectively processed their own emotions were better able to process the emotions of their customers and sold more products (Pekaar, Van der Linden, Bakker, & Born, 2017).
On the other hand, the interplay between emotion processing of the self and others can also be conflicting: When the energetic and processing resources that are invested in one type of emotion (e.g., regulating others' emotions) occur at the expense of resources that would be required for other type of emotion (e.g., appraising one's own emotions). This has been illustrated in a diary study in which employees' emotional exhaustion that resulted from witnessing unpleasant interactions between coworkers was affected by the extent to which they controlled their own emotions (Totterdell, Hershcovis, Niven, Reich, & Stride, 2012). The unique combination of appraising others' emotions and controlling their own emotions made witnesses more emotionally exhausted. Another study found that lawyers who tended to manage the emotions of both themselves and their clients performed worse than lawyers who tended to manage the emotions of either themselves or their clients . Following this, we propose that during emotional episodes in which individuals have limited energetic and processing resources, the handling of others' emotions may diminish their effectiveness in dealing with their own emotions and vice versa.
In short, emotion processing appears to be a stepwise process that starts with the appraisal of an emotion and ultimately ends with its regulation. Emotion processing may simultaneously occur at the intrapersonal and the interpersonal level. This processing may be facilitating or conflicting -depending on individuals' allocation of their energetic and processing resources. Drawing from this conceptualization, there are several interesting hypotheses to test. For example, individuals who fail to appraise negative emotions may be better off than individuals who appraise these emotions, but are not able to regulate them. Or, regulating one's own emotions first (versus later), leads to more effective otherfocused emotion regulation. Another interesting hypothesis could be that the appraisal of others' emotions will be more accurate when individuals are experiencing positive emotions (i.e., requiring less regulation efforts) than when they are experiencing negative emotions (i.e., requiring more regulation efforts).

Consequences of emotion processing
Our framework for research on dynamic self-and other-focused EI does not only emphasize how emotions of the self and others may be processed but also describes the potential outcomes of emotion processing. That is, the extent to which one succeeds or fails in appraising and regulating one's own and other's emotions is assumed to have proximal (episodic) consequences, and more distal consequences (see Fig. 1). In the following sections we start with addressing the (potential) proximal consequences from three different angles. First, we will explain how the quality of the emotion processing steps may contribute to direct episodic performance. Next, we will elaborate on possible affective spillovers to next emotional episodes. And third, we will describe how proximal consequences may, over time, develop into distal consequences. The subsequent sections will address the distal consequences in more detail.
3.3.1. Proximal consequences 3.3.1.1. Episodic performance. The processing of emotions may have immediate consequences for the self and others involved. It is assumed that after the emotion processing phase, the self and the other would often experience different emotions than at the beginning of this phase (i.e., less or more intense or even a different type of emotion). These modified emotions are labelled ''Secondary emotions" in our framework. The idea is that the secondary emotions help to achieve one's short-term goals Mayer and Salovey, 1997), for example selling a product or having a nice conversation with someone. The achievement of these shortterm goals is therefore named ''Episodic performance" in the framework. The specific content of this episodic performance depends on individual preferences and desires and is usually embedded in the (social) context. We therefore encourage researchers to operationalize this episodic performance in relation to the specific setting of their study. Examples are being polite to a customer, manipulating others, or being able to concentrate on a difficult task. These examples imply that episodic performance does not necessarily have to be pro-social, but can also include pro-self or so-called ''darker" elements (Côté, DeCelles, McCarthy, Van Kleef, & Hideg, 2011).
Previous research has suggested that processing otheremotions is particularly relevant for interpersonal episodic performance, for example when one wants to change the behaviors of others. To illustrate, a negotiation study showed that high-EI individuals induced higher negotiation satisfaction in their negotiation partners by inducing a positive mood in their partners (Mueller & Curhan, 2006). Furthermore, salespersons who appraised the emotions of their customers sold more products ). In contrast, processing self-emotions is considered particularly rele-vant for intrapersonal episodic performance because it directly changes one's own mood state which could contribute to remaining happy and healthy. For example, self-focused emotion processing may help to decrease (work-related) stress (Jordan, Ashkanasy, & Hartel, 2002). Research showed that academic teachers who were good regulators of their own emotions experienced less stress during their work than teachers who were less effective regulators (Pekaar et al., 2018).
3.3.1.2. Spill-over effects. One asset of the current framework is the incorporation of spill-over effects, which means that the emotions one experiences at the end of an emotional episode may influence following emotional episodes. Consequently, they may influence the subsequent interaction with the person involved (Hareli & Rafaeli, 2008), or even a next interaction with another person (Bakker & Demerouti, 2013). The incorporation of spill-over effects in our framework draws attention to a parameter that has not received much attention in the EI literature yet, namely the role of time. This parameter seems of particular importance for the episodic manifestation of EI as it responds to the dynamic experience of emotion (Mesquita & Boiger, 2014). Recently, this lively nature of emotions has gained a more prominent place in emotion research as scientists have started to acknowledge that time constitutes a vital aspect of emotional phenomena (Kuppens & Verduyn, 2017). Hence, feedback loops and spirals over time now are a relevant element of the emotion regulation (Gross, 2015) and the emotional labor literature (Grandey & Melloy, 2017). The spill-over effects in our framework follow this trend as they allow describing how one's emotional response to specific situational cues may have a significant impact on one's emotional functioning over time.
One of the ways the proposed spill-over effects work is via episodic performance. Episodic performance may elicit positive emotions that can transfer to a subsequent emotional episode. For example, a diary study showed that police officers who regulated their emotional display during their shift, reached more work goals, which, in turn, enhanced their work engagement at the end of the shift (Van Gelderen, Bakker, Konijn, & Binnewies, 2014). A related study found that employees who regulated their emotional display during social interactions were more successful in achieving their goals in these interactions, which predicted their well-being after these interactions (Wong, Tschan, & Semmer, 2017). Moreover, this latter study showed that the physical and energetic costs that were associated with the regulatory efforts were replenished when individuals achieved their goals (Wong et al., 2017). Hence, the proximal outcomes of emotion processing may reach beyond one emotional episode and can extend into the next one. Understanding and acknowledging these phenomena in psychological research may help to describe real-life emotional experiences better.

Distal consequences
The final phase of the framework includes the distal consequences of emotion processing. The idea is that the increased probability of achieving short-term (episodic or daily) goals will accumulate in enhanced performance in several major life domains that have been linked to EI in previous research, namely health and well-being (Martins et al., 2010), social relationships (Lopes et al., 2004;Schutte et al., 2001), and job performance (Joseph & Newman, 2010, O'Boyle et al., 2011. In the sections below we will elaborate on how effectively going through emotional episodes may translate into long-term positive outcomes in these areas. In Fig. 1, this influence is depicted with the upwards arrow between the ''Proximal consequences" on the episodic level to the ''Distal consequences" on the person level.

Health and well-being.
We propose that there may be at least three potential pathways by which episodic emotion processing influences health and well-being (Matthews, Zeidner, & Roberts, 2017). The first runs through the ability to regulate one's own emotions in an adaptive and positive way with direct health benefits (DeSteno, Gross, & Kunzansky, 2013;Friedman & Kern, 2014). The second potential pathway is interpersonal in nature and constitutes social support from others (Matthews et al., 2017). Individuals who process the emotions of others in a socially effective way are more likely to receive positive emotional feedback from their interaction partners, which may contribute to a feeling of being understood and supported (Zeidner, Matthews, & Shemesh, 2016). A third potential pathway that may be worthwhile to examine is that the repeated successful processing of emotions during emotional episodes may lead to a more frequent achievement of one's smaller and larger life goals (Wong et al., 2017). This may enhance overall feelings of efficacy and confidence and a general sense of accomplishment.

Relationships.
Being generally successful in dealing with emotions, on average, tends to favor the quality of social contact (Lopes et al., 2004;Schutte et al., 2001). Proposed mechanisms are the capacity of high-EI individuals to enhance positive emotions in the self and in others when interacting, and their capacity to process self-and other-emotions in ways that limit tension and conflict (Lopes et al., 2004). If this is done repeatedly, it may contribute to long-term relationship quality.

Job performance.
A similar line of reasoning goes for EI's positive association with job performance. As this association appears to be predominantly apparent in jobs in which employees frequently interact with others (Joseph & Newman, 2010), it is likely that it stems from the adaptive and positive ways in which high-EI employees deal with emotions during such interactions (Joseph & Newman, 2010;Tsai, Chen, & Liu, 2007) leading to episodic performance outcomes. This point is particularly salient in the link between EI and leadership performance. EI is related to leadership performance via a transformational leadership style which is characterized by inspiring, motivational, and emotional behaviors of the leader towards his/her followers (Hur, Van den Berg, & Wilderom, 2011). Furthermore, a repetitive achievement of one's (work) goals (i.e., accumulating episodic performance) during emotional episodes may logically also contribute to long-term job performance outcomes.
To sum up, emotion processing may have proximal and distal consequences. The proximal consequences of emotion processing can be operationalized as episodic performance, which is the achievement of (interpersonal) goals. As daily life consists of an ongoing stream of emotional episodes, emotion processing could be seen as a continuous process in which the immediate consequences of one emotional episode spill over to the next emotional episode. Over time, the repetitive achievement of proximal (interpersonal) goals by adequately dealing with emotions may accumulate to affect more distal outcomes such as health and well-being, social relationships, and job performance. The hypotheses that one could test based on this proposed process are intriguing. A hypothesis, focusing on doctors for example, is that doctors need to invest more effort to be empathetic to a patient, when their previous patient was not satisfied (i.e., low episodic performance) than when the previous patient was satisfied (i.e., high episodic performance). Another hypothesis, related to salespersons for example, could be that a talented salesperson will become more selfconfident over the years when she can use this strength in her job (i.e., a frequent high episodic performance), in comparison to a job in which she cannot use her talents.

Process moderators
Thus far, we have focused on the linear process between situational cues, emotion processing, and its proximal and distal consequences, yet it may be clear that this is a simplified process. As stated by Côté (2014), dispositional and contextual factors need to be considered to fully understand the workings of EI. Hence, for a more comprehensive and accurate representation of the real-life manifestation of EI, candidate individual and contextual differences are included in the framework as moderators of the process (see Fig. 1).

Individual differences
The existing EI literature suggests a range of individual differences that may influence the process between emotional events and their outcomes. Those individual differences include cognitive intelligence , personality (Van der Linden et al., 2017), gender (Fischer, Kret, & Broekens, 2018), and age (Doerwald, Scheibe, Zacher, & Van Yperen, 2016). In the following, we will list a few related concepts that received less research attention and that may be especially important for the episodic manifestation of EI.
3.4.1.1. Individual differences in emotional intelligence. One asset of the current framework is that it specifies the steps involved in the episodic enactment of EI. An implicit assumption is that these steps describe how people generally deal with self-and otheremotions during emotional episodes. However, the literature on individual differences in EI suggests that not every individual tends to be equally successful in dealing with emotional episodes in daily life. Specifically, compared to low-EI individuals, high-EI individuals are more likely to adequately process the emotions that emerge in a particular situation (e.g., more accurate emotion appraisals, more effective emotion regulation efforts). In our framework, the moderating role of individual differences in EI in dealing with emotions on the episodic level creates new research opportunities. For example, are high-EI individuals more successful in all phases of the process? Do they excel in other-focused emotion processing, self-focused emotion processing, or both? One would expect that individuals with a high general level of EI are, on average, relatively successful in all these processes. Nevertheless, within the individual difference domain of general EI, people may still display differential patterns of specific EI dimensions. That is, two persons may approximately have the same general level of EI (whether high or low) but may differ in their relative strength of dealing with their own versus other's emotions. Such a phenomenon can be compared to the domain of cognitive ability. In that domain, individuals differ on general intelligence, yet it is also clear that within the domain of general intelligence, differential patterns exist. For example, some smart people have better verbal abilities than mathematical abilities, while others show a reversed pattern. Similarly, in the domain of EI, people may have unique patterns of selfversus other focused emotional skills. In order to examine the expression of these unique patterns in real situations, it would indeed be necessary to precisely analyze the different aspects of (enacted) EI.
There are some initial studies that already shine some light on this issue. A meta-analysis examining own emotion processing showed that high-EI individuals tend to adapt their emotional response to a situation as soon as they can. When confronted with a negative emotional cue, they use all their resources to modify the situation. Besides, they tend to seek social support, and manage the way they perceive a situation in order to change their feelings about it. As such, high-EI individuals do not have to use maladaptive response modulation strategies like aggression or expressive suppression, because they already modified their emotional expe-rience early on in the emotional episode (Peña-Sarrionandia et al., 2015). This conclusion suggests that individual differences in EI play a significant role, particularly in the first phases of the current framework (see Fig. 1) when processing one's own emotions. There is also previous research examining how high-EI individuals process the emotions of others. For example, high-EI persons more often use mood-enhancing techniques (e.g., by offering help or reassurance), or mood-diverting techniques (e.g., with humor; Austin & O'Donnell, 2013;Austin, Saklofske, Smith, & Tohver, 2014) to regulate the emotions of others. Moreover, when dealing with an ostracized person, high-EI individuals more likely try to improve this person's mood, even if he or she did not yet explicitly express sadness or disappointment: They can infer the person's emotional state already from situational and non-verbal cues (Nozaki, 2015). Although these insights are valuable and contribute to a better understanding of how high-EI people behave when they face emotions in real daily situations, they only show just one side of the medal; namely processing one's own emotions or processing others' emotions. It is our aim that the current model will inspire scholars to integrate these perspectives and to start examining how high-EI individuals process their own and others' emotions simultaneously.
3.4.1.2. Motivation. Another relevant individual difference factor is motivation. Motivation may be a key aspect in the dynamic EI process because it fluctuates from day to day and it determines if people engage in emotion processing and for what reasons they do so. Regarding the ''if-question", only individuals who are motivated to use their emotional skills and knowledge may profit from its favorable consequences (Ybarra et al., 2014). Rode et al. (2007) showed that EI only contributed to academic performance among students with high levels of conscientiousness and not among students with low levels of conscientiousness. This finding suggests that students' motivation for academic achievement (as measured by their trait conscientiousness) determines whether they utilize their EI to successfully complete their school assignments. One's motivation may also influence what sort of goal(s) individuals want to achieve and thus what purpose emotion processing serves. For example, a waiter may act friendly to receive a higher tip, whereas a friend may act friendly because he genuinely likes you. Côté et al. (2011) showed that emotion regulation knowledge can facilitate both prosocial and self-serving goals depending on participants' motivation. In light of the current framework, one could for example study whether motivation is needed throughout all the phases of the process. It could be that the appraisal of emotions following a situational cue is rather independent of one's motivational state, whereas the active regulation of emotions demands high motivation. Another interesting possibility is that one's motivation could also influence whose emotions an individual will process first (selfor other-emotions), which may ultimately affect the effectiveness of this individuals' response.

Contextual differences
The situational cues in the current framework emphasize the importance of an individual's context for emotion processing. Specifically, the context produces cues that elicit emotions in the self and/or in others that need to be managed. We did not yet elaborate on the nature of these situational cues; however, there may be substantial contextual differences that influence whether and how individuals process emotions. A first interesting contextual factor for the enactment of EI consists of environmental demands. Research already suggested EI to be related to emotional labor demands (Joseph & Newman, 2010), stressful situations (Dong, Seo, & Bortol, 2014), and cultural values (Côté, 2014). Other potential contextual differences that might impact the episodic process of EI may be the type of emotions people are exposed to and their relationship with the other person involved.
3.4.2.1. Emotion types. Emotions exist in various flavors that differ in valence and intensity. Each discrete emotion prioritizes specific goals and has its own specific action tendency (Frijda, 1986;Roseman, Wiest, & Swartz, 1994). For instance, fear is associated with the urge to run away from the fear-evoking situation and anger motivates to confront the source of anger (Zeelenberg, Nelissen, Breugelmans, & Pieters, 2008). Because different emotions activate different goals and actions, they may also be processed in different ways. Intense negative emotions such as fear and anger typically provide the greatest call for emotion regulation (Barret, Gross, Christensen, & Benvenuto, 2001), because they signal potential threats (Quigley & Feldman Barrett, 1999). In contrast, positive emotions such as pride, love, or gratitude may have less emotion regulation demands because they signal that the situation is going well. Surprisingly, emotion type has not received much attention in the EI field and emotions are typically approached in a generic way. However, considering emotion types may be quite meaningful because different emotions may lead to different regulatory behaviors. Indeed, a diary study among policemen showed that the positive relation between EI and effective coping strategies was dependent on emotion type. Guilt evoked the strongest urge for emotion-focused coping whereas pride did not evoke coping at all (Gooty, Gavin, Ashkanasy, & Thomas, 2014).

Relationship types.
One potentially relevant interpersonal contextual factor in the framework is one's relation to the other person. The nature, length, and intimacy of this relationship may influence whether and how an individual engages in otherfocused emotion processing (Niven, 2017). An example of such a relational characteristic is the power distance between interaction partners. Interestingly, however, research findings on the link between power distance and other-focused emotion processing are mixed. On the one hand, there is research showing that people who have more power than their interaction partner are less attuned to the other person's emotions because they are relatively independent of the other (Anderson, Keltner, & John, 2003). On the other hand, research showed that people with more power have elevated levels of interpersonal sensitivity, which includes identifying and decoding emotions in others (Schmid Mast, Jonas, & Hall, 2009). A possible explanation for this discrepancy in the literature is that this latter study showed that the positive relationship between power and interpersonal sensitivity only held for leaders with an empathic leadership style and not for leaders with an egoistic leadership style. In other words, it may not be people's level of power that determines how the emotions of others will be processed, but rather their concern for other people. In fact, being responsive to others' emotions communicates respect, appreciation, and a concern for the well-being of the other (Berscheid & Reis, 1998). Therefore, it may not be surprising that partners in intimate relationships tend to use highly adaptive emotion regulation strategies to manage the feelings of the other (Debrot, Schoebi, Perrez, & Horn, 2013). In more strategic social interactions, it is mainly the length of the relationship between individuals that influences the value that is placed on others' emotional experience (Curhan, Elfenbein, & Eisenkraft, 2010).
To summarize, our framework for research on dynamic self-and other-focused EI points to several (episodic) contextual and individual factors that deserve attention in the EI literature. Beyond typical individual differences such as cognitive intelligence and personality, we explained how motivational processes and individual differences in EI seem conceptually important at the episodic level as they may explain whether and how EI is used. One could for example hypothesize that team boosters (i.e., individuals who are motivated to build a good atmosphere in a team) will allocate more effort to other-focused emotion processing than to selffocused emotion processing. Another hypothesis could be that individuals with a high level of EI always start to process their own emotions first in interpersonal situations. Context-wise, our theoretical framework calls to consider proximal features of the situational cue in research on EI because that may help to better understand people's daily emotional functioning. To illustrate, one could test the hypothesis that intense negative emotions (i.e., anger, fear) evoke more other-focused emotion regulation efforts than more subtle negative emotions (i.e., shame, disappointment). Another potential hypothesis is that individuals are better able to suppress their frustration in the company of their manager (i.e., stakes are high) than in the company of a friend (i.e., stakes are lower).

Discussion
Previous research has often adopted a relatively static perspective on EI by solely focusing on individual differences and its consequences in different life domains (cf. Côté, 2014;Peň a-Sarrionandia et al., 2015). As the EI research field is maturing, scholars have become more interested in the mechanisms that underlie the manifestation of EI which has resulted in an emphasis on the role of context, motivation, and other potential moderators that influence the use of EI Jordan et al., 2010;Ybarra et al., 2014). However, an integrated theoretical framework that helps to unravel the EI process over time has, thus far, been lacking. We sought to develop such a theoretical framework that can be used as a starting point to address questions such as ''What kind of inter-and intrapersonal processes underlie the manifestation of EI?" ''How do the different EI dimensions interact and what unique consequences do they have?" and ''How does EI interact with other individual factors?" (cf. Matthews et al., 2017;Petrides et al., 2016). It is our aim that the current framework may function as a blueprint for further empirical investigation of these questions.

Contributions and limitations
Our framework may advance the EI literature in several ways. First, it can provide a more complete view of what actually happens when individuals are confronted with emotions. By encouraging researchers to incorporate the trigger, the response, and the consequences of emotion processing, we emphasize that the manifestation of EI may function as part of an emotional process that can be activated, intervened, or be built upon. This process-based approach does justice to the lively nature of emotions and to the complex ways in which people tend to respond to them. Moreover, it may provide researchers and practitioners opportunities to intervene in the different phases of using EI.
As a second contribution, we proposed that the context (i.e., situational cues) initiates the enactment of EI. Hence, the context may activate different EI dimensions and determine whether they will be effective, which may explain at least part of the mixed findings in the field (Joseph & Newman, 2010). This perspective responds to recent ''calls-to-context" in the EI literature (Côté, 2014;Jordan et al., 2010;Ybarra et al., 2014) and may inspire researchers to integrate the stimuli that activate emotion processing into their research models. Another novel aspect of the current framework is that it suggests incorporating individuals' motivation and individual differences in EI in the process-based examination of EI. Previous studies have already shown that the usage of EI is no isolated process, but it interacts with other abilities and beliefs Côté et al., 2011;Rode et al., 2007). The inclusion of these individual differences may better connect the EI literature to the actual ways different people deal with emotions in their day-to-day routines.
Third, our theoretical framework is unique in the sense that it explicitly emphasizes whether individuals are dealing with their own emotions or with the emotions of others, and that the accompanying consequences are also distinguished in intra-and interpersonal terms. Using self-and other-focused EI dimensions may be a conceptually different process and predictive of different kinds of criteria (Pekaar et al., 2019;Salovey & Mayer, 1990). Hence, a distinction in self-and other-focused processing seems relevant to better understand the underlying mechanisms of EI. Relatedly, our framework also includes combinations of EI dimensions. Research has suggested that not all individuals use all their EI dimensions to the same extent, but that a unique mixture of EI dimensions better resembles reality (Elfenbein, 2016). This specific interplay, in turn, further improves predictions of emotional responding (Papousek, Freudenthaler, & Schulter, 2008;Pekaar et al., 2017). The current framework incorporates these dynamics to encourage researchers to move beyond the global EI level (cf. Petrides et al., 2016).
Finally, our framework includes time, which is an oftenoverlooked factor in psychological research (Roe, 2008). Incorporating time allows to distinguish proximal consequences of emotional responding from distal consequences. For example, a manager who displays anger to get his followers to work harder may succeed in the short run because the followers would not dare to disobey. However, this practise may at the same time demotivate and alienate the followers from their work causing them to deliver lower quality work and/or to start looking for another job. In other words, using emotions to achieve short-term goals may not necessarily guarantee long-term goal achievement. Hence, it is important to consider time in research on EI as processing emotions is no isolated act but is often part of one's longer-term goals (Grandey & Melloy, 2017;Wong et al., 2017). In fact, research has suggested that high-EI responding can be characterized by adaptive actions that also take one's own and others' well-being and long-term consequences into account (Peň a- Sarrionandia et al., 2015).
Regarding the limitations, the focus on the four specific EI dimensions in the current framework may represent an incomplete capture of the EI construct (Brasseur et al., 2013;Mayer & Salovey, 1997;Wong & Law, 2002). This simplification, however, does not aim to discard any other EI sub-dimension that might play a role in between the initial appraisal of an emotion and its regulation. We acknowledge that, on an even more detailed time scale, inbetween appraisal and regulation several other micro processes of emotion regulation may be involved (e.g., emotion facilitation, emotion understanding). However, we believe that elaborating on those in the framework would lack parsimony and would not further clarify the point we aim to make, namely that the interplay between (self-and other-focused) EI dimensions may be a fruitful perspective to examine the effectiveness of EI responding. Nevertheless, we encourage broader inquiries of the interplay between EI dimensions than the ones we explicitly capture in our framework because it may represent the EI process in even more detail.

Implications for research
Our proposed framework has important implications for the research designs and methods that scholars may want to use to test aspects of the theory. To facilitate this research, validated measures of enacted EI dimensions and episodic performance must first be developed. Furthermore, we encourage researchers to use multilevel designs that enable to capture individual differences in EI as well as fluctuations in emotions and behaviors across emo-tional episodes. Useful methodologies are the experience sampling method in which individuals report on their experiences in the moment (e.g., using a smartphone), or the day reconstruction method were individuals report on their experience after chronologically reconstructing various preceding (emotional) episodes (Kahneman et al., 2004;Oerlemans & Bakker, 2013;Reis & Gable, 2000). Such methods allow modeling the focus areas of our framework, namely fluctuations in the effectiveness of emotion processing, spill-over effects of emotional experiences, and relations between emotion processes and outcomes over time. To illustrate, a five-day experience sampling study could be conducted in which dentists are asked to report on their emotion processing and realtime energy level before and after each patient. In addition, patients could be asked to rate their satisfaction with the treatment they received. Using this methodology, one could test the hypothesis that emotional episodes in which dentists invest more other-focused emotion regulation efforts are associated with more positive patient satisfaction ratings (i.e., higher episodic performance). Furthermore, one could model whether treating a satisfied patient may, in turn, conserve dentists' level of energy at the start of the next treatment. In addition, one could assess dentists' general level of EI before the experience sampling period and include general EI as a moderator. This enables testing the hypothesis that high-EI dentists (versus low-EI dentists) invest more in the emotions of their patients and therefore profit more from the beneficial process it may evoke. Hence, such a study offers innovative perspectives on the dynamics of daily emotional life that our framework describes.

Concluding remarks
The framework for research on dynamic self-and other-focused EI introduced in this paper provides a new theoretical perspective that describes which steps are required for the manifestation of EI and what the consequences of such behavior are. The framework is not meant to be tested in one comprehensive study but may rather function as a challenging agenda to move EI research forward. We believe that incorporating the role of time, context, and individual and contextual moderators that influence the enactment of EI is a promising avenue that may capture the lively nature of emotional processes better than classic and static EI models.