Alongside an increase in the literature on the beneficial effects of EI, a body of research has developed on programs aimed at fostering these abilities, showing that EI can be effectively trained (Hodzic et al., 2018; Kotsou et al., 2019; Mattingly and Kraiger, 2019; Schutte et al., 2013). There are numerous examples of programs within the SEL framework that focus on promoting EI in school settings. The INTEMO program was developed in Spain (Ruiz-Aranda et al., 2013) and it consists of 12 sessions divided into four phases based on the ability model of EI (Mayer et al., 2016; Mayer and Salovey, 1997): perception and expression of emotions, facilitation of emotions, understanding of emotions, and regulation of emotions. The program has demonstrated effectiveness in improving adaptation, mental health, and social behaviors (Castillo et al., 2013b; Ruiz-Aranda et al., 2012a, 2012b). Another SEL evidence-based program is RULER (Recognizing, Understanding, Labeling, Expressing, and Regulating emotions; Nathanson et al., 2016), developed in the USA, it is grounded in the ability EI model (Mayer et al., 2016; Mayer and Salovey, 1997) and it aims at developing five main abilities which are included in the acronym: recognizing emotions, understanding emotions, labeling emotions, expressing emotions, and regulating emotions. Results showed that RULER had positive effect on student’s, adjustment and academic performance (Brackett et al., 2012; Nathanson et al., 2016; Rivers et al., 2013). teacher engagement, burnout, and teacher-student interaction (Castillo et al., 2013a). Recent evidence also highlights the importance of strengthening emotional skills during initial teacher education. There are examples of training programs based on Bar-On’s (2006) model in which interventions delivered to pre-service teachers have shown consistent improvements in EI (Gilar-Corbi et al., 2018; Özdemir Cihan and Dilekmen, 2024). These findings underscore not only the value of providing EI training to in-service teachers, but also the growing need to develop such competencies before teachers enter the school system.

In the Italian context, it was developed Meta-Emotions (D’Amico, 2018), a program grounded also on the ability EI model (Mayer and Salovey, 1997) and SEL framework, that aims at developing five main emotional abilities: recognition and non-verbal expression of emotions, emotional synesthesia, understanding emotions, emotional regulation, and use of emotions. While maintaining its main theoretical framework, the program is designed in two distinct versions based on its target audience and objectives: Meta-Emotions Test and Training, intended for psychoeducational and psychotherapy contexts, and Meta-Emotions at School, aimed at fostering EI and meta-emotional intelligence (MEI) in educational settings (D’Amico, 2018). While the Meta-Emotions at School program has primarily been applied to students and, more broadly, the school system, this study focuses on the application and evaluation of the Meta-Emotions Test and Training program exclusively for teachers. Meta-Emotions at School is conceived specifically for the school context and adopts a train-the-trainers approach in which teachers become facilitators of emotional literacy activities with their students. Unlike the Test and Training version, it does not typically include assessment and structured training phases; instead, teachers attend an 8-h experiential workshop and are then provided with online materials, activities, and guidelines to promote emotional awareness and create emotionally inclusive learning environments. The program emphasizes classroom-based activities, peer tutoring, and the integration of emotional competencies into everyday teaching practices, supporting teachers in becoming ambassadors of EI within their schools (D’Amico and Geraci, 2022a). Meta-Emotions Test and Training is an integrated program aimed primarily at developing and strengthening EI abilities and the awareness of those abilities. It usually includes an assessment phase of emotional abilities, followed by structured training sessions targeting the different abilities outlined in the Mayer and Salovey (1997) model, and concludes with a final focus-group session in which participants reflect on the process and evaluate the activities carried out. Additionally, a central component of the training program concerns the development of MEI abilities, defined as the awareness of one’s emotional abilities and measured using the IE-ACCME test (D’Amico, 2013). In particular, MEI is a multidimensional construct that is grounded in meta-cognition theory (Flavell, 1979) and EI theory (Mayer et al., 2016; Mayer and Salovey, 1997) and that takes into account beliefs about emotions, self-perceptions and self-evaluations about one’s own emotional functioning, and actual emotional abilities (D’Amico, 2013, 2018; D’Amico and Geraci, 2023). The awareness is critical because it may influence how individuals apply effectively their emotional abilities, as self-perceptions and actual abilities do not always correspond (Brackett et al., 2006; D’Amico and Geraci, 2023). Recent research on pre-adolescents (D’Amico and Geraci, 2021) has shown that a lack of awareness of one’s emotional abilities is associated with higher levels of peer rejection in the sociogram. Additionally, another study (D’Amico and Geraci, 2022b), consistent with research on gender differences in the field of EI, revealed also gender differences in MEI, with girls generally underestimating their emotional abilities and boys tending to overestimate them. Finally, findings from the first application of the training program demonstrated that, overall, MEI abilities can be improved as pupils who participated in the program exhibited a reduced tendency to overestimate their emotional abilities (D’Amico and Geraci, 2022a).

Given the crucial role of teachers’ EI in fostering students’ social and emotional growth, as well as managing the demands of the teaching profession, the need for targeted training programs for educators is increasingly evident. While numerous interventions within the SEL framework have focused on students’ EI and SEC, teachers have received comparatively less attention, despite their pivotal role in shaping classroom environments and student outcomes (Greenberg et al., 2016; Schonert-Reichl et al., 2017). To address this gap, the present pilot study investigates the effects of the Meta-Emotions program on enhancing EI and MEI in teachers. The program was adapted and tested for application in a teacher population. It was hypothesized that Meta-Emotions program would improve: 1) emotional abilities and 2) meta-emotional abilities. Specifically, for point 2, it was hypothesized that participants in the intervention group would increase their meta-emotional beliefs, meta-emotional knowledge, and self-evaluation by reducing the discrepancy scores between participants’ self-perceptions of abilities and actual abilities.

The study included a total of 46 teachers (44 females, 2 males) with age ranging from 32 to 63 years (M = 49.67, SD = 7.48). In terms of roles, 54.3% were subject teachers, while 45.7% were special education teachers. Regarding the educational level, most participants taught at the primary school level (47.8%), followed by 23.9% at the middle school level, 21.7% at the high school level, and 6.5% at the preschool level. Participants were divided into two groups: the intervention group (n = 24) and the control group (n = 22). The experimental group consisted entirely of female teachers, with an age range between 38 and 59 years (M = 49.08, SD = 5.24). In terms of teaching roles, 54.2% were special education teachers, and 45.8% were subject teachers. Most participants in this group taught at the primary school level (50%), followed by 25% at the high school level, 16.7% at the middle school level, and 8.3% at the preschool level. The control group included 20 females and 2 males teachers, with an age range between 32 and 63 years (M = 50.32, SD = 9.43). Regarding teaching roles, 63.6% of the control group were subject teachers, and 36.4% were special education teachers. In terms of school levels, 45.5% of the control group taught at the primary school level, 31.8% at the middle school level, 18.2% at the high school level, and 4.5% at the preschool level. All participants were teachers employed in public urban schools.

As above mentioned, the dedicated instrument for measuring MEI is the IE-ACCME test (D’Amico, 2013), a multi-method tool designed to assess emotional and meta-emotional abilities in adolescents. The test assumes the theoretical model of EI proposed by (Mayer et al., 2016; Mayer and Salovey, 1997) but utilizes various methodologies for capturing different aspects of emotional sphere. Specifically, the IE-ACCME test measures meta-emotional beliefs through the Meta-Emotional Belief Scale (CE), self-reported emotional abilities through the Emotional Self-Concept Scale (CME), emotional abilities through the Emotional Ability Test (AE), and self-evaluation of performance on the ability test through the Self-Rating of Performance Scale (AV). All the scales were developed by closely following the theoretical framework and structural organization of the Mayer and Salovey (1997) ability model of EI, and their validation resulted in the same latent EI dimensions. This allows to compare standardized scores obtained by adolescents in the four scales in order to obtain two additional meta-emotional scores: meta-emotional knowledge and meta-emotional self-evaluation (D’Amico and Geraci, 2023). The meta-emotional knowledge score (CMeta) represents the discrepancy between the standard scores obtained on the Emotional Self-Concept Scale (CME) and the Emotional Ability Test (AE). This score reflects the degree to which individuals’ emotional abilities correspond to their self-perceptions of these abilities in daily life (CMeta = CME – AE). The meta-emotional self-evaluation score (AVMeta) represents the discrepancy between the standard scores obtained on the Self-Rating of Performance Scale (AV) and the Emotional Ability Test (AE). It reflects the extent to which individuals’ emotional abilities align with their self-perception of performance in the ability test (AVMeta = AV – AE). Meta-emotional scores are expressed in absolute and relative values, providing different types of information. The absolute score indicates the overall discrepancy between self-perceptions and actual abilities, with higher scores reflecting greater discrepancies. Conversely, the relative scores reveal the direction of the discrepancy: negative values indicate a tendency to underestimate abilities (CME < AE; AV < AE), while positive values indicate a tendency to overestimate abilities (CME > AE; AV > AE).

Since the original IE-ACCME test was calibrated and standardized for adolescent samples and the items in AE test are not suitable for use with adults, we developed for the present study a new protocol by combining three scales drawn from the IE-ACCME test (CE, CME, and AV) with the Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT; Mayer et al., 2002), created for adult populations, and also validated in Italian population (Damico and Curci, 2010).

The CE, CME, and AV scales from the IE-ACCME test use wording and questions that may be also applied to adult population. However, as described above, we preliminarily ensured that they maintain the same structural organization of the Mayer and Salovey (1997) theoretical framework shared by the MSCEIT (Mayer et al., 2002). Therefore, the four scales used in our protocol for measuring MEI in adults are theoretically and structurally aligned with the original Mayer and Salovey (1997) theoretical framework.

In particular, the Meta-Emotional Beliefs Scale (CE; D’Amico, 2013) is a 16-item questionnaire that explores beliefs regarding the ability to perceive emotions from non-verbal and sensory stimuli, as well as the ability to understand, regulate, and use emotions to facilitate thought. Responses are rated on a five-point Likert scale ranging from 0 (not true) to 4 (definitely true). Higher scores indicate stronger meta-emotional beliefs. The adult adaptation of the scale, validated through exploratory and confirmatory factor analyses, confirmed the four-branch structural model (χ²₍₁₆₄₎ = 22.2, p = 0.075, CFI = 0.99, TLI = 0.99, SRMR = 0.02, RMSEA = 0.03, 90% CI [0.00, 0.04]). The normative sample consists of 963 adults (805 females, 158 males). The reliability coefficients (Cronbach’s α) for the scales at pre-test and post-test were as follows: EA = 0.69, 0.73; SA = 0.71, 0.70; Total = 0.70, 0.71.

The Emotional Self-Concept Scale (CME; D’Amico, 2013) is a 20-item self-report measure that assesses self-perceived abilities in recognizing, facilitating, understanding, and regulating emotions. Responses are rated on a five-point Likert scale ranging from 0 (not true) to 4 (definitely true). Higher scores reflect greater self-perceived emotional abilities. The CME scale evaluates individuals’ perceptions of their emotional capabilities in daily life. The adult adaptation confirmed the four-branch structure of the model (χ²(164) = 38.4, p < 0.001, CFI = 0.98, TLI = 0.97, SRMR = 0.02, RMSEA = 0.04, 90% CI [0.03, 0.06]). The normative adult sample included 963 participants (805 females, 158 males). The reliability coefficients (Cronbach’s α) for the CME scales at pre-test and post-test were: EA = 0.79, 0.67; SA = 0.68, 0.64; Total = 0.74, 0.79.

The Self-Rating of Performance (AV; D’Amico, 2013) is an 8-item scale that assesses individuals’ self-rated performance on the ability test. At the end of each task or section in the ability test, respondents are asked to evaluate their performance in solving the task by selecting a score on a six-point Likert scale ranging from 0 (not good at all) to 5 (very good). The reliability coefficients (Cronbach’s α) for the AV scales at pre-test and post-test were: EA = 0.88, 0.89; SA = 0.92, 0.86; Total = 0.92, 0.93.

The Mayer-Salovey-Caruso Emotional Intelligence Test (MSCEIT; Mayer et al., 2002) is a performance-based test comprising 141 items divided into eight tasks, which assess four key EI abilities: perceiving emotions, using emotions to facilitate thought, understanding emotions, and managing emotions. Scores are calculated using the consensus scoring method (Mayer et al., 2002), specifically using the general consensus score computed on the normative sample of 3,185 adults (1,718 females, 1,487 males). The reliability coefficients (Cronbach’s α) for the MSCEIT area and total scores at pre-test and post-test were: EA = 0.68, 0.70; SA = 0.69, 0.74; Total = 0.73, 0.69.

All CE, CME, AV, and MSCEIT scores are expressed as standardized scores with a mean of 100 and a standard deviation of 15. Since the study involved group comparisons, individual differences in ability EI were accounted for by weighting the differences between CME, AV, and AE scores based on the AE score. Accordingly, meta-emotional knowledge scores in absolute and relative values (CMeta_Abs and CMeta_Rel) were calculated as [(CME − AE)/AE]. Similarly, meta-emotional self-evaluation ability scores in absolute (AVMeta_Abs) and relative (AVMeta_Rel) values were computed as [(AV − AE)/AE].

The present study used a quasi-experimental pre-post design with one intervention group and one comparison group. Group assignment was based on voluntary participation in the training program. Teachers who enrolled in the intervention group were recruited online through an announcement shared on social media. To form the comparison group, teachers in the intervention group were asked to invite colleagues with similar professional characteristics. Overall, the activities took place from February to April 2024, and pre-test data were collected from both the experimental and control groups before the actual training began. Specifically, pre-test data were collected one week before the beginning of the first session of the training, whereas post-test data were collected from both groups one week after the end of the intervention. Both assessments were administered through the Giunti Testing online platform. The administration followed the structure of the IE-ACCME test, with the CE scale administered first, followed by the CME scale and then the ability test (i.e., MSCEIT). The AV scale was administered together with the MSCEIT, as participants were asked to self-rate their performance at the end of each of the eight MSCEIT tasks. The training sessions were conducted outside the school context and working hours and were held in the Department spaces at the University. All participants provided informed consent. The study was approved by the local ethics committee (Protocol No: 135883–28/22).

Prior to hypothesis testing, preliminary data screening was conducted to examine the distributions of the study variables using descriptive statistics and normality testing. These preliminary analyses indicated deviations from normality for several outcome variables. Based on these considerations and relatively small sample size, nonparametric statistical analyses were adopted. First, baseline differences between the intervention and comparison groups were examined using Mann–Whitney U tests to assess pre-test equivalence between intervention and comparison groups. Subsequently, intervention effects were evaluated by computing time change scores (Δ = post-test − pre-test) for each outcome variable in both groups. Differences in change scores between groups were then analyzed using Mann–Whitney U tests. Effect sizes were calculated using the rank-biserial correlation (r_rb), with values of 0.10, 0.30, and 0.50 interpreted as small, medium, and large effects, respectively, according to common guidelines like Cohen’s criteria for effect size (Cohen, 1992).

Despite the findings, this study should be regarded as an exploratory pilot investigation and, as such, it has several important limitations that preclude comprehensive conclusions. First, the small sample size and the pilot nature of the study limit the statistical power and the generalizability of the results. Given the exploratory nature of this pilot study, no priori power analysis was performed. However, the observed effect sizes provide valuable information for planning future studies with larger samples and sufficient statistical power. Larger, more diverse samples are needed to confirm these findings and explore whether they hold across different populations of teachers (e.g., preschool, primary school, and secondary school), educational settings, and cultural contexts. Additionally, another limitation concerns the sample distribution, as it consisted exclusively of female teachers. Although the teaching profession is predominantly composed of females both worldwide (UNESCO, 2024) and in Europe (Katsarova, 2020), individual differences—such as teaching experience, age, and gender—must be considered in future studies. This pronounced gender homogeneity represents a critical limit on the generalizability of the findings, as the present results cannot be directly extended to male teachers, given that emotional and meta-emotional processes may vary across genders as evidenced in previous studies (Cabello et al., 2016; D’Amico and Geraci, 2022b; Fernández-Berrocal et al., 2012). Moreover, individual and professional characteristics, such as years of teaching experience and prior training in social and emotional learning or EI, were not systematically collected in this pilot study. Future research should address these limitations by including more heterogeneous samples and by collecting detailed demographic, professional, and contextual information.

Second, another limitation might concern the methodology used for measuring MEI and calculate meta-emotional indices (CMeta and AVMeta) in adults, that combines two tests that have not been standardized in the same adult populations. The use of different normative sample might have contributed to variability in the derived meta-emotional indices and could have affected the results. In the future, a dedicated test for measuring MEI in adult would be beneficial.

Third, the study employed a relatively short intervention duration (21 h over seven weeks). While improvements were observed in meta-emotional dimensions, longer-term interventions may be necessary to produce significant changes in EI abilities, as evidenced by previous research (Hodzic et al., 2018; Kotsou et al., 2019; Wagstaff et al., 2013). Future studies should consider longitudinal designs to capture sustained changes over time and examine the retention of training effects. Moreover, no conclusions can be drawn regarding the long-term effectiveness of the intervention program due to the absence of follow-up assessments. Including follow-up measurements in future research would address this limitation.

Fourth, from a methodological perspective, participants were not randomly assigned to the intervention and comparison groups. The voluntary nature of the intervention may have introduced selection bias related to motivation. Although the two groups were largely comparable at baseline, some differences emerged on specific dimensions which should be considered when interpreting the findings. Thus, future research should prioritize randomized group allocation in order to reduce selection bias and strengthen causal inferences regarding the effectiveness of the training program (Kotsou et al., 2019). Additionally, including an active comparison group that receives an equivalent amount of training in terms of hours and engagement could help reduce potential motivation bias.

Finally, this study focused exclusively on direct measures of teachers’ emotional and meta-emotional abilities. Future studies should incorporate additional assessments, such as student or colleague evaluations of teachers’ skills. Furthermore, due to the pilot nature of this research, the focus was limited to EI and MEI dimensions. Future studies should also include other relevant outcomes, such as well-being and work-related variables (e.g., burnout, work engagement, and self-efficacy).

By addressing these limitations, future research can build on the findings of this pilot study, offering more robust evidence for the utility of targeted programs like Meta-Emotions in fostering emotional and meta-emotional abilities among educators. These insights could contribute to the development of more effective training initiatives aimed at promoting teachers’ well-being and improving educational outcomes for students.