Developing Children’s Socio-Emotional Competencies Through Drama Pedagogy Training: An Experimental Study on Theory of Mind and Collaborative Behavior

Drama pedagogy training (DPT; Celume, Besançon, & Zenasni, 2019a) can be defined as any drama-based training or workshop, created through drama activities and techniques (such as pretend play, improvisation, role-play, etc.) that follows the characteristics of the liberal progressivism tendency from drama pedagogy (García-Huidobro, 1996). This approach focuses on the work of internal world of the participant through drama games instead of searching an academic or artistic achievement. Moreno’s (1943, 1947) socio-drama, Boal’s (1978, 1989) theatre of the oppressed and Slade’s (1967, 1998) drama education were some of the first approaches using representational games in the work with people focusing on competencies development instead of an artistic training, setting the beginnings of DPT as a methodology. Those ideas were also outlined by Karakelle (2009), Libman (2001) and Woodson (1999) who defended the idea of focusing on the process and experience of learning competencies without any intention of showing off. In current literature, this methodology turned into several different names, such as creative drama (Rosenberg, 1981), process drama, drama pedagogy (García-Huidobro, 2004), applied theatre (Holland, 2009) or even child drama, play making, child play, or educational drama. What they all have in common, is that their focus is non academic, nor artistic, even if some of these concepts describe drama practices used in education for educational objectives (Freeman, Sullivan, & Fulton, 2003). DPT could be also considered as a specific kind of training within drama based pedagogy (for a review see Lee, Patall, Cawthon, & Steingut, 2015), based on drama pedagogy theory (García-Huidobro; 2004) and which is focused only on participant’s socio-emotional and creative development, and thus, not created to enhance any academic field, or to contribute to any subject in school, neither to train participants in any artistic way. According to Celume et al. (2019a) DPT is a methodology that encompasses drama activities and techniques through the use of embodiment and narratives, under a particular psycho-pedagogical framework that works with the use of narrative and the construction of a dialogic space (Celume, Besançon, & Zenasni, 2019b). DPT can be considered as an active methodology, that works with the cognitive and affective dimensions of the participants emphasizing the importance of playfulness and positive emotions, promoting a safe and caring environment and a positive student-pedagogue relationship (Garaigordobil & Berrueco, 2011; García-Huidobro, 2004; Hammond, 2015; Rosen, 1974) that would develop social and emotional competencies and skills.

An important part of successful socio-emotional competencies programs are based on DPT elements such as role playing, improvisation, pretend play, storytelling and others. These conclusions can be consulted in recent meta-analyses (Durlak, Weissberg, Dymnicki, Taylor, & Schellinger, 2011; Lee et al., 2015) that showed strong relationships between the trainings they reviewed and the socio-emotional outcomes measured. Nevertheless, the meta-analysis conducted by Durlak et al. (2011), is identified as a socio-emotional learning programs review, and thus is not focused on drama pedagogy trainings, but on different socio-emotional school-based interventions, even though several of the successful interventions reviewed used drama pedagogy elements within their trainings. Durlak et al. (2011) showed that in general, the programs reviewed had positive effects on the socio-emotional competencies willing to develop, and that these interventions worked better on children than adolescents. They also observed that the use of a specific protocol (ref. SAFE protocol) moderated positive students’ outcomes, which corresponds with the description of the psycho-pedagogical framework of DPT (Celume et al., 2019a). In the same line, interventions that were identified as more interactive and that, for example, used several DPT elements, such as role playing, were the most successful among young students. Similarly, Garaigordobil et al. (1996), compared 6 experimental groups versus 2 control groups, in order to see the effect of a game-based intervention (that included drama-based games) on collaborative behavior. Results showed significant differences between the experimental and the control groups, t(34) = 5.65, p < .001, meaning that an intervention of this kind was able to develop competencies in the field of reciprocal help behaviors. Regarding the development of other socio-emotional competencies such as empathy or theory of mind (ToM) through drama-based pedagogies, Goldstein and Winner (2012) conducted two studies in children and adolescents looking for relationships between drama, empathy and ToM, and finding that these skills could be enhanced through drama. Results for the elementary school children sample showed a higher impact of acting training on empathy over controls, F(1, 64) = 4.26, p = .043, d = 0.53, but no significant scores were found for ToM, probably due to a possible issue related to the measurement tool chosen for assessing ToM. In the same line, Celume and Zenasni (in press) conducted an observational study where they observed and analyzed DPT sessions, outlining perspective taking as un underlying mechanism that permitted the development of theory of mind and creativity in elementary school children.

Nevertheless, a review on pretend play and its effects on child development (Lillard et al., 2013), found that correlations between pretend play and social skills, and pretend play and ToM were inconsistent, although most of the studies reviewed showed “important failures to replicate” (p. 16), which could be an explanation to the lack of positive results. In a similar way, the meta-analysis of Lee et al. (2015) studied the effect of drama based pedagogies (DBP) on children from preschool to adolescence, finding that DBP had a significant impact on educational achievement outcomes but that the impact on social skills was not significant, concluding that the belief that DBP might promote prosocial behaviors was unfounded.

On the contrary, it was supported that DBP had a significant positive effect on other children’s social skills, like attitudes toward achievement and the so-called 21st century skills (which includes collaborative behavior). Moreover, several drama-based trainings were considered to be out of the scope and thus weren’t examined in the meta-analysis, we missed, for example, some well-known drama-based methodologies, such as the Speech Bubbles project or the Drama for Learning and Creativity (D4LC), which we believe could have influenced some of the negative results regarding DBP and its impact on social competencies. In the same line, the meta-analysis of Lee et al. (2015) explains that they had to exclude several studies from their analysis due to a lack of reporting information, demanding that researchers must better report their methods, as it is fundamental for replicability. They found that most of the studies analyzed presented important bias as they were mostly quasi-experimental studies that rarely presented matching control groups, being difficult to assure that DBP was or was not the cause of social outcomes enhancement, concluding that there is a need of better reporting the trainings and strategies used in order to be able to have consistent material to analyze and thus have consistent results.

To our knowledge, and based on these meta-analyses, except for the recent work from Goldstein and Lerner (2018) and a study on fantasy play (e.g. Thibodeau, Gilpin, Brown, & Meyer, 2016), studies that have been done in order to examine the impact of DPT methods and elements on socio-emotional competencies relies on studies a) without an active control group well matching the experimental one, b) where children volunteer to participate in the experimental group and so there’s no randomised assignment, and c) where the researchers are involved in the training. In this line, several interventions and trainings are excluded from scientific analyses and thus, it remains difficult to provide consistent evidence of the impact of DPT for the development of socio-emotional competencies. The study of Goldstein and Lerner (2018) conducted on 97 pre-school children, aimed to provide evidence on this and thus controlled the variables shown before, seeking the impact of a pretend play group (PPG) on ToM, altruism, distress towards others and helping behavior, emotion matching and social behavior. The results showed that children in PPG group had lower levels of neutral social behaviors and higher levels of positive social behaviors than block play or reading groups. They also concluded that as children had lower levels of emotional control than the average, the pretend play training did not help with ToM and compassion, but it did help them with emotional regulation and social behaviors.

The importance of developing socio-emotional competencies through drama can be found in the origins of the emotional intelligence concept and in the fundamental role that socio-emotional competencies play in nowadays’ education. Today, there are still several approaches when defining socio-emotional competencies, although they are all coming from the concept of emotional intelligence (EI; Salovey & Mayer, 1990; Mayer & Salovey, 1997). According to Mikolajczak, Quoidbach, Kotsou, and Nelis (2014), the definition of EI still presents some issues within the scientific community, as there is still no consensus among the researchers to identify what means to be emotionally intelligent. Consequently, they proposed to gather these capacities under the concept of emotional competencies instead of talking about an intelligence (emotional intelligence). In this line, they defined emotional competencies as “the practical capacity of identifying, understanding, expressing, regulating and using personal and other’s emotions”ⁱ (Mikolajczak et al., 2014, p. 3) according to a particular context.

Thus, regarding children’s development and its implication in educational settings, emotional competencies would be the skills that children should have in order to identify, understand, express, regulate and use their own emotions (intrapersonal level) and those of their classmates, teachers, friends and people who they interact with (interpersonal level), within the school setting. According to Mikolajczak et al. (2014), the interpersonal variant of the “understanding” level of emotional competencies, can be described as the practical capacity of understanding the causes and consequences of others’ perceived emotions. In this line, this capacity would be directly related to the ability of understanding and reacting, in a cognitive, social or affective level, the emotion expressed by another, which is known in academic and popular literature as empathy. According to Davis (1980, 1983) empathy is a complex construct that may be divided into (a) emotional empathy, and (b) cognitive empathy. The emotional branch or side, is related to the emotional process and response of an individual facing another’s emotional reaction (e.g. Eisenberg & Miller, 1987), while the cognitive branch answers to the ability of understanding others’ mental states (e.g. Baron-Cohen & Wheelwright, 2004). The former, is considered by some authors as emotional empathy or just empathy (de Vignemont & Singer, 2006), while the latter, considered as a social cognitive ability to adscribe mental states, such as thoughts, beliefs and emotions to another person is known in literature as ToM (Premack & Woodruff, 1978).

In this line, the development of socio-cognitive skills, such as ToM, and socio-emotional competencies, is intrinsically related to social functioning behaviors and competencies, such as teamwork/collaboration. The latter is considered to be part of the main 21st century skills to develop in education (e.g. Trilling & Fadel, 2009) altogether with creativity and other socio-emotional competencies. In fact, already in 1996, Garaigordobil, Maganto, and Etxeberría (1996) presented evidence on how the experiences lived by children at school had an important impact on children’s prosocial behavior. Social competencies, are detrimental for children development and school is an important modulator of them. According to Burrus and Brenneman (2016), most teachers agree that teamwork/collaboration is a fundamental skill to be learned in the classroom; nevertheless, they are not always sure how to define it, and thus, it gets difficult to find useful tools in order to develop it.

In summary, DPT in schools is an active pedagogy focused on the social and emotional world of children. Studies have presented evidence of the efficacy of DPT related tools and elements in the development of socio-emotional competencies such as ToM and collaborative behavior in elementary school children, although the presented literature regarding the different studies analyzed, concerning drama-based trainings and their impact on socio-cognitive and socio-emotional competencies, do not always present complete reporting, didn’t provide a substantial randomised control, and the only one that did, was focused on preschool children (Goldstein & Lerner, 2018). Teachers agree with the fact of the importance to develop socio-emotional competencies and a collaborative behavior in children, although they are not wide aware of the kind of training that could help them reach this objective. Thus, the main objective of the current study is to conduct a randomised pre-post test study with matching control group in order to present evidence of the efficacy of DPT in promoting socio-emotional competencies, such as ToM skills and a collaborative behavior in elementary school children.

Method

Results

Discussion and Conclusion

Studies have shown that socio-emotional learning competencies can be taught and learnt at school (Durlak et al., 2011; Ebert, Hoffmann, Ivcevic, Phan, & Brackett, 2015) and at the same time, some researchers insist on the importance to train children as well as teachers on socio-emotional learning competencies, so the benefits can be positive in a personal way but also in a collective way (Fernández-Berrocal, Cabello, & Gutiérrez-Cobo, 2017). More specifically, some authors sustain that ToM and prosocial behavior are two main domains in the field of socio-emotional learning that are important to be developed among children and youngsters (Beauchamp & Anderson, 2010; Rose-Krasnor, 1997). As a recent meta-analysis showed, dramatic play seems to have a positive impact on both of these competencies, even though no rigorous studies have been conducted yet in order to corroborate these effects (Lillard et al., 2013). In the present study, we carried out a randomised protocol in which specific measures for both ToM and collaborative behavior were applied.

ToM can be defined as the ability to read and understand another's mental states through body, face and vocal expression as well as previous knowledge of the other (Goldstein, 2009). In our study, this ability was measured through the RMET-G in which children had to infer mental states through the observation of eyes zones pictures.

Results before the intervention, showed significant differences among schools for ToM for the general sample, implying that depending on the school the children come from, their competencies in this ability were already different regardless the experimental group they belonged to (DPT or CSG). When comparing means, the greater differences were found between the public school and the priority school, with higher scores for the public school. This results could be in part explained by the socio-cultural background of children. In this line, the public school, located in the center of Paris, receives mostly French children with parents that were born in France, and from middle and upper-middle class families, while the priority school, located outside downtown, receives children from different nationalities and/or with parents that were born outside France, some of them from a lower socioeconomic level. Consequently, even if some specific vocabulary is taught at school, an important number of language is learnt at home through parent-child interaction (Hart & Risley, 1992), thus, foreign-parent(s) children would have more difficulties in “foreign” vocabulary acquisition (Mingat, 1991). A more homogeneous French socio-cultural background was present in the public school, with a more foreign socio-cultural background one in the priority school, thus, suggesting that public school children have a more direct access to French vocabulary because they speak the same language at home in contrast to priority school children whose parents (mother or/and father) do not speak French as a first language. We also support this idea based on the fact that private school, where children’s socio-cultural background was very heterogeneous, did not exhibit significant differences with neither of the two other schools.

It was also found a significant difference between both grades’ scores being higher for 5th grade than 4th grade. Nevertheless, this was kind of expected, and can be simply explained by the fact that as children grow up, they acquire more language skills, thus having 5th grade children a larger vocabulary than 4th graders.

Post-test analyses presented encouraging results regarding our hypotheses.

First, correlation analyses showed that there was a significant but weak correlation between ToM and collaborative behavior. This results suggest that it seems to be an important relationship between ToM and collaborative behavior competencies in children that occurs inside a DPT setting. Nevertheless, this relationship is not strong enough to ensure a clear correspondence of ToM with collaborative behavior, thus suggesting more investigation and precisions in this area that should be considered for further research addressing the links between the social competencies that might be enhanced in DPT.

Further on, when comparing the type of training, in order to see the impact of DPT on ToM, both ANOVA analyses and t-tests showed effects and significant differences of the type of training on ToM. These significant differences, were statistically higher for DPT group, suggesting that DPT enhanced ToM in both 4th and 5th grade children from all three schools. These results are consistent with previous studies that established a relationship between drama-related games and the development of ToM skills (e.g. Goldstein & Winner, 2012) even though, their results were achieved on a high-school sample. Type of training also showed an effect between pre-test and post-test, confirming post-test analyses. DPT group showed higher differences in scores than CSG, the latter presented a difference of pre-test and post-test scores that was almost invariable. These results are inside the scope of what we predicted, as children who participated in the DPT were trained in emotion and mental states identification through the different DPT methods and activities played. Sessions 1, 3, 4 and 5 were dedicated to emotional identification, expression and communication through play and mini-pretend-plays in where children had to choose an emotion or emotional state and to represent it in an invented collective scene. In this way, children were “obliged” to identify the different characteristics of emotions and the mental states they were to represent, and so learning was acquired through playing, using their minds and bodies to recreate, permitting embodiment, which helped with the understanding of the emotions and mental states. This idea, which defends the theory that embodiment is a direct contributor of the development of the cognitive process of empathy, in which ToM is a fundamental part, is supported by the work of several authors (e.g. Gieser, 2008; Sofia, 2016). Moreover, for some authors, mind cannot be separated from body, as the body is evidence of the mind (Bateson, 2000; Butterworth, 1994). In this means, body and mind cannot be divided because the body is the support of mind expressions.

Regarding the class level, the highest difference in ToM post-test is presented in 5th grade. While DPT group in 4th grade is slightly higher than CSG in 4th grade, scores get clearly higher for DPT in 5th grade. These results suggest that DPT would work better for 5th graders. In both cases, these results only reflect our specific sample and results should be taken carefully as they do not represent the whole French population. Besides, the fact that DPT worked better in 5th grade for developing ToM can be only due to the fact that understanding mental functioning and ascribing mental states to others is in part, a cognitive ability, thus, it should naturally evolve along with children’s cognitive development. In this means, 10-year-old children, that were already turning 11 by the time of the post-tests, have these cognitive abilities more developed. In fact, the theory of stages (Piaget, 1929) describes an evolutionary process of the intellectual capacities which establishes a transformation from concrete operational stage to formal operational stage at the age of 11. Moreover, regarding ToM, Perner and Wimmer (1985), confirm a natural evolution in the attribution of second-order beliefs in their work with children aged 5 to 10-year-olds.

Regarding the difference between pre-test and post-test repeated measures ANOVA found that DPT on ToM task had differences considering the type of school. As shown in results, PU school showed almost no increase between pre and post ToM tasks in comparison with REP and PR schools which post scores were between 2 and 3 points of difference.

On the other hand, collaborative behavior may benefit children’s social relationships (Schellenberg, Corrigall, Dys, & Malti, 2015) as is encompassed in prosocial behavior which appears to be one of the key elements that benefits social interaction (Eisenberg, Spinrad, & Knafo-Noam, 2015). Moreover, in 2006, Barry & Wentzel (2006) established that collaborative behaviors develop a sense of security among friends and peers. In our study, we analysed collaborative behavior from two perspectives; an objective one through task performance and a more subjective one, through the observation of children’s behaviors.

The state of the art showed that there were already differences among children from different schools, being always PR school children who had the more higher scores in collaborative behavior. After questioning teachers and directors from the three schools involved, there’s still no clear answer that could explain this higher collaborative behavior in comparison to the other schools. No specific program was held before our intervention, and all three schools offered the same kind of extracurricular activities.

Girls showed to be more collaborative than boys. Observations in both pre-test and post-test showed that they tended to be more encouraging than boys in their groups, but also dynamics observed among groups constituted mostly of girls presented less anti-collaborative behaviors and showed a more federated way of work. This is consistent with Garaigordobil (2009) findings who established that girls are more collaborative than boys.

Even though group randomisation was made completely blinded and through an online free software, t-test results showed a significant difference between intervention groups. This means that there were already differences in collaborative behavior scores between the experimental (DPT) and the active control group (CSG). CSG started the interventions presenting an advantage over the DPT group. When teachers were asked about differences they could perceive between both groups of their class after they were constituted, some of them agreed that DPT group was composed by “more difficult students than the other group” and that they felt the difference of working in the class with the other half (CSG) while the DPT session was being carried out. They reported they were able to “work better without that part of the class”. Nevertheless, this wouldn’t be sufficient to explain these differences, so we rely on the fact that the evaluation of collaborative behavior was carried on one week after the groups were constituted. With this timing in mind, we might infer that after a week of intervention, children already had created some particular relationships, group identification or bonding links with their classmates in their corresponding groups (DPT and CSG), which would had interfered on the second week, when the evaluation of collaborative behavior was carried on.

Finally, when testing our hypothesis, for collaborative behavior we found significant differences between the two groups, being higher for DPT. Both t-test analysis and ANOVA confirmed the hypothesis. This means that children in the DPT group increased their collaborative behavior during the intervention showing important differences when compared to their CSG classmates. PD task measure how the children respond and solve a situation in which they have to decide whether to collaborate to their partner or no. Children are offered the possibility to escape from prison, even though this means that their partner will stay in prison for some years. At the beginning, DPT group presented less collaboration, showing answers that in great majority chose not to collaborate and so escape from prison. After DPT was conducted, children appeared to develop more consciousness on the need to collaborate with others, and they decided to collaborate in the PD task, staying a few years in prison in order to not let their partner alone in prison for even more years. This behavioral change confirms that a DPT enhances collaborative behavior, and that a training based on expressive art such as drama can facilitate some socio-emotional learning competencies. Regarding this, the International School for Interdisciplinary Studies, in Canada establishes that expressive arts might improve social learning by giving children the opportunity to work in group, collaborate, and create a social support network (ISIS, 2004). Our results are consistent with several studies that confirm the positive effect of drama and drama based trainings over collaborative behavior and social skills development (e.g. Garaigordobil, 2003; Joronen, Konu, Rankin, & Astedt-Kurki, 2012; Wright, John, Ellenbogen, Offord, Duku, & Rowe, 2006). Nevertheless, they are not consistent with Lee et al. (2015) who explained that even though drama-based pedagogies have been supported as being an effective way to foster prosocial behaviors studies analysed in their meta-analyses were not consistent.