Some studies have shown that AEI predicts job performance (O’Boyle et al., 2012), the quality of relationships (Lopes et al., 2004), and overall wellbeing (Di Fabio and Kenny, 2016; Sánchez-Álvarez et al., 2016). However, these associations tend to be relatively weak and inconsistent compared to the predictive power of TEI (Robinson, 2024; Zeidner and Olnick-Shemesh, 2010).

Recently, Robinson (2024) called on researchers to provide fresh insights to know a lot more about whether and how AEI manifests itself in daily life. According to him, AEI should significantly impact decision-making, social interactions, and overall wellbeing, but further research is crucial to “forge” these links. We follow this direction by conducting a latent profile analysis of AEI to identify beneficial and risk profiles in wellbeing, mental and physical health, and decision-making.

Nearly three decades after the publication of their foundational article on AEI (Mayer and Salovey, 1997), Mayer et al. (2016) began to enhance the theoretical framework surrounding the AEI model and its assessment. Following this approach, Haag et al. (2023) introduced the QEPro model, which extends Mayer and Salovey (1997) AEI framework through theory-based item development and scoring method, integrating it within the situational judgment test (SJT) framework. While the QEPro model was specifically developed for managers, the authors later introduced the QEg model, which serves as its generalized version intended for the broader population (Haag et al., 2024a). Both (QEPro and QEg) have the same factorial structure and are composed of the same dimensions. Here, we refer to the QEg model as our target population is a general population of adults.

According to the QEg model, AEI can be defined as the ability to identify (IE), understand (UE), and regulate (SME) emotions (Haag et al., 2024a):

Identification of Emotions (IE) represent the meta-competency to accurately identify emotions in self and others (Haag et al., 2024a). It consists of three abilities:

Although previous studies have highlighted the key role of emotional intelligence in everyday life, most of them used a variable-centered approach to investigate how individuals differed with regard to the mean of the sample but not how the specific dimensions of emotional intelligence interacted together (Pirsoul et al., 2022).

Moreover, most of these studies used a TEI (rather than AEI) approach (e.g., Martínez-Monteagudo et al., 2021; Pirsoul et al., 2022; Toyama and Mauno, 2016) that considers EI as a “trait” among other personality traits measured through self-report questionnaires (Haag et al., 2023; Matthews et al., 2004; Schlegel and Mortillaro, 2019).

As TEI and everyday life outcomes are commonly measured using self-report questionnaires, these concepts tend to show stronger correlations with each other due to their subjective nature and the consistency in how individuals evaluate their own traits or behaviors (Podsakoff et al., 2003).

In contrast, performance tests, which measure AEI more objectively, often show weaker correlations with self-report scales. This can be attributed to self-assessment biases, where individuals may overestimate or underestimate their actual abilities, leading to discrepancies between self-reported results and objectively measured performance (Kruger and Dunning, 1999; Paulhus and Reid, 1991). Adopting a different perspective to examine the relationship between AEI and life outcomes is thus required.

In this article, we explore AEI following a person-centered approach to identify—through a latent profile analysis (LPA)—AEI profiles regarding QEg’s seven dimensions and an Anova to observe their relationship to everyday life outcomes.

Recently, Haag et al. (2024b) identified two latent profiles of AEI—using the QEPro/QEg model—within a diverse population of managers. LPA, which is specifically structured to account for the presence of subpopulations characterized by different parameters (Meyer and Morin, 2016; Morin, 2016), offers a unique framework for exploring the quantity and nature of emerging AEI profiles.

Their analysis yielded two distinct profiles: The full emotional processing (FEP) and the minimal emotional processing (MEP). The FEP profile significantly outperformed the MEP profile on the following five AEI dimensions: Identifying Emotional Triggers, Understanding Emotional Timelines, Anticipating Emotional Outcomes, Selecting the Target Emotional State, and Emotion Regulation. The authors concluded that the FEP displays “the most complete processing of emotion” (p4).

Our approach, in line with Haag et al. (2024b), uses LPA to identify emerging AEI profiles, based on the QEg model and its questionnaire, within a general population, and subsequently examines their associations with real-life outcomes.

All participants (N = 2,877; Table 1), French native, volunteered in the study without financial incentives and being aware of the confidentiality of their answers. All participants were recruited with the support of a French popular science magazine and a French television program devoted to science which have published on their respective websites a call to participate in our study. They completed QEg and other questionnaires online (via the Qualtrics software package). Test administration was extended over 4 weeks.

QEg was administered online to all participants. All other measures were French adaptations of self-report questionnaires and were also online administered. We confirmed the psychometric qualities of each of these scales through confirmatory factorial analyses (CFAs) and verification of internal consistency using Cronbach’s alpha. All the CFA results showed a good fit with the presupposed models, and the alpha values were in line with those expected (see Table 2; Nunnally and Bernstein, 1994):

LPA is designed to determine sub-groups in an extent sample, called a profile. LPA was run with the open-source software Mclust package (Scrucca et al., 2016) to investigate one to five profiles. The optimal number of profiles has been selected in regard to the statistics criterion (see Table 3), substantive meaningfulness, and theoretical conformity (Gillet et al., 2019).

The three-profile to the five-profile solutions were examined more closely. This examination demonstrated that the three-profile solution was the most proper solution regarding its statistical and theoretical conformity. The three-profile solution was retained for further analyses (BIC = 294; Figure 1).

We named the three profiles full emotional processing (FEP), minimal emotional processing (MEP), and partial emotional processing (PEP). The FEP outperforms the two other profiles on all EI abilities measured by QEg (the subscales appear in Figure 1). These results suggest that the FEP profile captures and processes emotional information in a broader and deeper way than the two other profiles (Scanning Physiological Manifestations, Interpreting Emotional Cues, Identifying Emotional Triggers, Understanding Emotional Timelines, Anticipating Emotional Outcomes, Selecting the Target Emotional State, and Emotion Regulation).

We also ran ANOVA in order to understand more deeply the benefits of full emotional processing on real-world outcomes such as perceived stress, home–work interaction, mental health, wellbeing, and decision-making.

We test differences between the three identified profiles (FEP, PEP, and MEP) on several variables through ANOVA. We then compare profiles means two by two with pairwise t-test. Our results show that FEP significantly outperforms the two other profiles (MEP performs the worst each time) on all key life variables measured (see Table 2).

FEP is associated with the lowest levels of perceived stress which is consistent with the literature (Shahin, 2020). High perceived stress has a negative impact on mental and physical health (Lu et al., 2019).

FEP is characterized by the lowest score on the home–work negative interaction. Previous research pointed out associations between negative HWI and fatigue (Geurts et al., 2003) and decreased psychological health (Van Hooff et al., 2005).

FEP exhibits and experiences greater gratitude, an emotion that helps individuals to deal with adversity, anxiety, and depression, and to build strong interpersonal relationships (Cunha et al., 2019; Diniz et al., 2023).

FEP has the highest scores on the satisfaction with life scale. Satisfaction with life is associated with better health, higher self-esteem, stronger resilience, lower depression, and anxiety (Diener et al., 2018; Cerezo et al., 2022).

FEP has the lowest score on the Emotional Exhaustion subscale which is considered the core meaning of Burnout (Lee and Ashforth, 1996). Thus, FEP tends to be more protective against depressive symptoms (Gerber et al., 2018).

Finally, our results indicate that FEP has a great consideration of future consequences (vs. immediate gratification) and a tendency to make decisions that are more future-oriented. Not considering the future consequences of our actions is highly risky for our health (Murphy and Dockray, 2018).

The different results suggest that FEP tends to be associated with better mental and physical health, wellbeing, satisfaction with life as well as decision-making.

While a widely read science magazine and a health-focused television program facilitated the recruitment of a substantial participant pool, this approach may have inadvertently resulted in a sampling bias. Our sampling could bias the generalizability of the findings by disproportionately including individuals who possess a pre-existing interest in science or health-related topics (Franzke et al., 2020). This demographic may limit the generalizability of the study’s conclusions. Future research should consider using more varied recruitment strategies, including outreach to less engaged audiences, to ensure a diverse participant representation that captures a broader spectrum of perspectives on the subject matter (Merriam and Tisdell, 2015).

Not controlling for pre-existing psychiatric and neurological disorders in this study could potentially impact the generalizability and interpretation of the findings. Individuals with such disorders may exhibit unique patterns of emotional processing and regulation that differ substantially from the general population (Aslan et al., 2024; Lezak et al., 2012; Pepping et al., 2024). Failing to account for these differences introduces the risk of confounding variables, obscuring the true relationships between AEI profiles and the measured outcomes. For example, individuals with anxiety disorders may exhibit heightened sensitivity to emotional cues and experience greater emotional reactivity, potentially leading to higher scores on certain QEg dimensions, even in the absence of a “Full Emotional Processing” profile (Barlow, 2002). Conversely, individuals with depressive disorders may show blunted emotional responses, resulting in lower scores regardless of their underlying emotional intelligence competencies (American Psychiatric Association, DSM-5 Task Force, 2013).

This limitation reveals the need for future research incorporating robust screening measures for psychiatric and neurological conditions and potentially implementing stratified analyses to account for their influence on AEI profiles and their associations with real-life outcomes.

Self-report measures as the ones used in this article to evaluate real-life outcomes, while valuable for assessing subjective experiences, are susceptible to biases such as social desirability response bias and acquiescence bias (Paulhus, 1991; Podsakoff et al., 2003). Future research could incorporate methods to mitigate these biases, such as using multiple informants or using techniques such as the Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991) to assess response styles. Further, exploring the potential for common method variance through techniques such as Harman’s single-factor test would enhance the robustness of the findings.

The Need for diverse sampling and methodological approaches. While LPA provides significant insights into patterns and relationships within data, future research will aim to integrate qualitative data and case studies to enhance the ecological validity of our findings. Incorporating qualitative methods can offer a more profound and nuanced understanding of the contexts and experiences underlying the patterns identified through LPA (Creswell and Poth, 2018). This multi-method approach will allow us to triangulate data sources, thus enriching our interpretations and providing a more comprehensive overview of the phenomena under investigation (Flick, 2018). By leveraging both quantitative and qualitative techniques, we aspire to refine our findings and ensure that they are not only statistically robust but also grounded in real-world experiences and perspectives (Yin, 2018).

As LPA shows, emotion regulation considered as meta-competence (Peña-Sarrionandia et al., 2015) is high when the scores on the six dimensions preceding “emotion regulation” are high too. This first insight encourages us in the future to examine if FEP is a “cascading” profile (Joseph and Newman, 2010). Future research could examine the specific cognitive and emotional regulation strategies used by individuals with the FEP profile and explore if a cascading model emerges. This could involve measures of cognitive appraisal, coping mechanisms, and different types of emotion regulation strategies (Gross and Ford, 2024).

Exploring contextual factors. Future research could explore the factors contributing to the emergence of the PEP profile, potentially focusing on contextual factors such as work environment, age, or personality traits. Such a comparative analysis underscores the utility of person-centered approaches for uncovering a more comprehensive understanding of the multifaceted nature of AEI and its impact on various life domains.