Linking Cooperative Learning to Prosocial and Antisocial Behaviors in Adolescents: The Role of Affective Experiences

Abstract

Cooperative learning, a pedagogical approach emphasizing structured collaboration and mutual responsibility, has been associated with academic and social benefits, yet its influence on adolescents’ prosocial and antisocial behaviors remains underexplored. This study examined the relationships between cooperative learning, affective experiences, and social behavior in adolescent students. A total of 261 adolescent students (M = 13.06 years, SD = 0.86) from Greece participated in this cross-sectional study. Data were collected via questionnaires assessing perceptions of cooperative learning, enjoyment, vitality, and prosocial/antisocial behaviors toward teammates and opponents. Factor analyses confirmed the validity of the Greek versions of the instruments. Path analysis revealed that contextual elements of cooperative learning explained a significant proportion of variance in enjoyment and vitality, which in turn predicted prosocial behavior toward both teammates and opponents. Specifically, enjoyment and vitality were positively associated with prosocial orientations, whereas vitality negatively predicted antisocial behavior toward teammates. Unexpectedly, group reflection and heterogeneity emerged as negative predictors of enjoyment and vitality, while interaction showed marginal positive effects. Indirect effects indicated that cooperative learning dimensions influenced social behavior primarily through enjoyment and vitality. These findings suggest that positive affective states mediate the impact of cooperative learning on students’ social conduct in PE. Implications for practice highlight the importance of carefully designing cooperative structures to enhance enjoyment and vitality while fostering prosocial interactions during adolescence.

1. Introduction

An important mission of physical education in high school is to shape adolescents’ socio-emotional development [1]. In this respect, social behavior represents a crucial dimension of students’ experience in physical education. A persistent challenge for teachers involves managing antisocial behaviors such as aggression, cheating, and rule violations, which can disrupt the class climate and negatively affect students’ participation [2]. Physical education can support adolescents’ social behavior as it provides authentic contexts where values such as cooperation, respect, fairness, and inclusion are practiced and internalized [3]. Through structured physical education lessons, adolescents learn not only how to engage in physical activity but also how to make responsible choices, respect rules, and develop empathy toward peers. In this context, innovative, student-centered approaches to teaching and learning have gained increasing attention as they can transform both the philosophy and the practice of physical education. One such pedagogical approach is cooperative learning, which emphasizes collaboration, mutual responsibility, and the cultivation of social and moral competencies alongside cognitive and physical skills [4]. However, the potential of cooperative learning in physical education to shape adolescents’ social behavior has been underexplored.

1.1. Cooperative Learning in Physical Education

Over the past decades, cooperative learning has emerged as a promising framework for effective physical education lessons because it combines structured academic goals with opportunities for students to develop interpersonal skills and positive peer relationships [5]. Cooperative learning is defined as a structured form of group work in which students collaborate to achieve shared academic and social objectives [6]. Unlike unstructured group work, it is characterized by carefully planned activities that ensure interdependence and accountability among group members. Activities such as group projects, role-playing, simulations, and structured discussions exemplify this approach, as they encourage students to learn from one another and to build knowledge collectively [7].

Johnson and Johnson’s [8] framework has been especially influential in operationalizing cooperative learning. They identified five essential elements that determine its effectiveness: positive interdependence, individual accountability, promotive interaction, social skills, and group processing. Positive interdependence ensures that group members rely on each other for success, while individual accountability prevents free-riding and social loafing. Promotive interaction involves supportive peer exchanges, social skills are cultivated through practice, and group processing allows reflection on how effectively the group functions. Together, these elements help create structured environments where students learn academic content while also practicing communication, leadership, and conflict resolution skills [9]. Importantly, in physical education settings, cooperative learning has been shown to not only facilitate motor learning but also to encourage the development of broader social competencies [10,11]. In particular, Zach et al. [12] indicated that cooperative learning through promotive interaction, heterogeneous grouping, and structured reflection enhance students’ sense of social relatedness and perceived competence in PE. Engels and Freund [13], in addition, showed that through these mechanisms cooperative activity increases enjoyment. More specifically, Engels and Freund [13] suggested that cooperative learning’s structural features can propagate to sustained engagement, positive class climate, and learning.

1.2. Prosocial Behavior and Adolescents’ Experiences in Physical Education

Prosocial behaviors, such as helping an injured opponent, encouraging teammates, or showing respect for rules, contribute to a more positive and inclusive environment [14]. Prosocial behavior can be defined as voluntary, intentional actions aimed at benefiting others, whereas antisocial behavior is intentional conduct that harms or disadvantages others [15]. Bandura’s [16] social cognitive theory of moral thought and action emphasizes the role of self-regulatory processes in judging the acceptability of behavior. According to this framework, individuals evaluate their own actions against internalized moral standards, but these standards may be disengaged under certain social conditions, leading to antisocial behavior. The theory further highlights the importance of modeling, suggesting that adolescents are influenced by observing the behavior of peers, teachers, and even parents [17].

Both prosocial and antisocial behaviours are not simply traits residing within individuals but are strongly shaped by social and motivational contexts. For example, recent work by [18] showed that the ways behavior unfolds depends heavily on how motives are structured in the environment. Meanwhile, Pavin Ivanec et al. [19] emphasized that peer norms, teacher behaviors, and the perceived climate of the learning setting serve as crucial social cues. Students are more likely to engage in positive or negative social behavior in line with what they observe or believe is expected in their peer group and broader classroom environment. Together, these findings underscore that social behaviour (whether constructive or harmful) arises not only from individual disposition, but critically from the surrounding social-motivational “architecture”, suggesting that altering norms, teacher strategies, or classroom climate can shift behavioral outcomes in meaningful ways. Similarly, Gano-Overway [20], showed that perceiving one’s physical education environment as caring is associated with greater cognitive empathy and increased prosocial behavior, and also with lower levels of antisocial behavior (such as bullying). Fostering a caring climate in physical education may be a promising strategy for encouraging prosocial conduct and reducing negative behaviors among adolescents.

In this regard, affective constructs such as enjoyment and vitality are particularly important. Enjoyment refers to the positive emotional state of satisfaction, pleasure, and interest during an activity, reflecting intrinsic motivation [21]. Vitality, on the other hand, captures the subjective sense of energy, aliveness, and enthusiasm that sustains both cognitive and emotional engagement [22]. Empirical research shows that in collaborative physical education environments, adolescents report higher enjoyment and vitality, which in turn positively influences their behavior during the lesson by demonstrating greater effort, persistence, and long-term participation in physical activity [23,24]. A recent systematic review and meta-analysis revealed that game-based physical education interventions, that inherently involve cooperative learning, enhanced enjoyment among children and adolescents [25].

1.3. Cooperative Learning and Social Behavior

Cooperative learning has been consistently associated with prosocial behavior. Choi et al. [26] suggested that cooperative learning activities can serve as an effective approach to enhance adolescents’ willingness to collaborate, which in turn will lower the occurrence of intentional aggressive acts and promote more frequent prosocial behaviors. This was further corroborated by experimental studies indicating that cooperative learning experiences can lead to higher prosocial and lower antisocial behaviors in childhood and adolescence [11,27,28,29].

The effect of cooperative learning on social behavior has been largely due to its impact on cognitive and affect variables such as learning motivation, gratitude and emotional intelligence [19,30]. Importantly, enjoyment has also been identified as a potential mediator between cooperative learning and social behavior. For instance, Engels and Freund [13] found that cooperative games in physical education enhanced enjoyment, social relatedness, and perceived competence, with social relatedness partially mediating the relationship between cooperative activities and enjoyment. Dewaele and Li [31] similarly demonstrated that enjoyment mediated the effect of teacher enthusiasm on students’ social engagement, suggesting that positive emotions play a central role in shaping classroom interactions. In this line, Al-Yaaribi and Kavussanu [32] indicated that enjoyment was correlated with prosocial and antisocial behavior in sport settings. This evidence collectively indicates that enjoyment may serve as a mediator in the cooperative learning-social behavior relationship as it is enhanced from cooperative learning and is correlated with social behavior.

1.4. The Present Study

Building on this background, the present study aims to investigate how students’ perceptions of cooperative learning in physical education influence prosocial and antisocial behaviors, with particular attention to the mediating roles of enjoyment and vitality. By integrating theoretical perspectives from cooperative learning, social behavior, and motivational psychology, this cross-sectional study seeks to shed light on the processes that underpin positive peer interactions in school-based physical education.

Based on the existing theorizing and evidence [8,10,11,13,14] cooperative learning influences the social development of students. In addition, affective experiences, such as enjoyment, are influenced by cooperative learning and in turn can be associated with social behaviors. However, while enjoyment has been examined in this way, the mediating role of vitality in the relationship between cooperative learning and social behavior has received little empirical attention. Therefore, the existing literature has not adequately investigated how cooperative learning shapes social behaviors. Affective experiences seem to influence this relationship, but still there is only limited evidence. This gap highlights the need for research that investigates how affective experiences link cooperative learning with social outcomes. To address this gap, in the present study we examined the following hypotheses: (a) dimensions of cooperative learning will be positively associated with prosocial behavior and negatively with antisocial behavior, and (b) the effect of cooperative learning dimensions on prosocial and antisocial behavior will be mediated by affective experiences such as enjoyment and vitality.

2. Materials and Methods

2.1. Participants

A total of 261 adolescent students (105 males, 40.2%; 156 females, 59.8%) from the regional unit of Serres, Greece, participated in this cross-sectional study. Participants were recruited from the first (Grade 7, n = 90, 34.5%), second (Grade 8, n = 63, 24.1%), and third (Grade 9, n = 108, 41.4%) grades of public junior high school. A random sampling approach was employed with school being the unit of selection. A list of schools of the regional unit of Serres was created and two schools were randomly selected. Both schools were located within the city of Serres and the students attending these schools shared similar demographic profiles, characterized by a medium socioeconomic status. Selected schools that provided approval from their principals and teaching boards took part in the study. The two schools provided consent and participated in the study.

2.2. Measures

Data were collected using a multi-section online questionnaire administered in Greek. All scales demonstrated good internal consistency in the present sample (See

Table 1. Descriptive statistics and analysis of correlation.

	1	2	3	4	5	6	7	8	9	10	11	12	13
1. Enjoyment	1	0.82 **	0.58 **	0.52 **	−0.16 **	−0.19 **	−0.31 **	−0.43 **	−0.43 **	−0.36 **	−0.44 **	−0.41 **	−0.49 **
2. Vitality		1	0.50 **	0.46 **	−0.16 **	−0.15 *	−0.24 **	−0.35 **	−0.30 **	−0.28 **	−0.32 **	−0.33 **	−0.39 **
3. Pro-teammate			1	0.60 **	−0.16 **	−0.14 *	−0.28 **	−0.38 **	−0.30 **	−0.30 **	−0.31 **	−0.33 **	−0.36 **
4. Pro-opponent				1	0.02	0.10	−0.23 **	−0.35 **	−0.24 **	−0.21 **	−0.25 **	−0.25 **	−0.34 **
5. Anti-teammate					1	0.804 **	0.19 **	0.25 **	0.24 **	0.30 **	0.32 **	0.28 **	0.17 **
6. Anti-opponent						1	0.26 **	0.27 **	0.29 **	0.31 **	0.32 **	0.35 **	0.20 **
7. Interaction							1	0.74 **	0.74 **	0.59 **	0.73 **	0.74 **	0.69 **
8. Heterogeneity								1	0.76 **	0.66 **	0.76 **	0.77 **	0.77 **
9. Social skills									1	0.72 **	0.87 **	0.80 **	0.81 **
10. Assessment										1	0.70 **	0.76 **	0.73 **
11. Tutoring											1	0.77 **	0.79 **
12. Interdependence												1	0.80 **
13. Group reflection													1
Mean	5.69	5.30	4.02	3.26	1.66	1.56	1.63	1.79	1.73	1.84	1.76	1.81	1.97
SD	1.41	1.22	0.95	0.96	0.81	0.74	0.71	0.73	0.68	0.70	0.72	0.65	0.75
Cronbach’s alpha	0.93	0.92	0.72	0.88	0.86	0.85	0.82	0.77	0.85	0.79	0.89	0.84	0.86

* p < 0.05, ** p < 0.01.

for Cronbach’s α values).

2.2.1. Perceptions of Cooperative Learning

Students’ perceptions of cooperative learning in their physical education (PE) classes were assessed using the Cooperative Learning Questionnaire (CLQ) [33]. CLQ is a 44-item self-report instrument developed to assess the extent to which cooperative learning principles are perceived in educational and sport contexts. All items are rated on a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree), with higher scores indicating stronger endorsement of cooperative learning practices. The questionnaire measures seven factors, corresponding to key components of cooperative learning: positive interdependence assessing the perception that group members’ outcomes are linked (e.g., “When we work in groups, we need the ideas of all of us in order to achieve success.”), interaction referring to the degree of mutual support, encouragement, and face-to-face interaction (e.g., “This subject allows me to interact with my fellow group members.”), tutoring measuring the support provided by the teacher in order to foster the achievement of the group’s activities (e.g., “The teacher provides us with guidelines to solve conflicts that might arise in the group.”), social skills measuring the extent to which students use communication, leadership, trust-building, and conflict-resolution skills (e.g., “In this subject, respect in group relationships is encouraged.”), group reflection referring to the group’s ability to reflect on its functioning and plan improvements (e.g., “During the lesson we have time to reflect on our ways of working in the group and how to improve.”), heterogeneity assessing the recognition of diversity within the group as a resource for learning (e.g., “The members of my group have skills and abilities that complement each other.”), assessment measuring the extent to which evaluations consider both individual and collective contributions (e.g., “In this subject the students receive the grades they deserve, no more, no less.”). Psychometric validation by Fernández-Río et al. [33] supported the seven-factor structure with satisfactory reliability (α ranging from 0.73 to 0.84) and evidence of construct validity. CLQ was translated into Greek following established back-translation guidelines [34,35]. First, two independent bilingual translators, whose native language was Greek and who were familiar with the terminology of education and sport sciences, translated the original English items into Greek. A synthesis of the two translations was then created through discussion between the translators and the research team, ensuring semantic, idiomatic, and conceptual equivalence. Next, the synthesized Greek version was independently back-translated into English by two additional bilingual translators, blinded to the original version. The back-translations were compared with the original English items to identify discrepancies. Any inconsistencies were resolved through consensus among the translators and the research team. Finally, the pre-final Greek version was piloted with a small sample of students (n ≈ 7) to assess clarity, readability, and cultural appropriateness. Feedback from the pilot testing was used to make minor adjustments, resulting in the final Greek version of the CLQ used in the present study.

2.2.2. Enjoyment

Enjoyment in PE was measured using a 5-item subscale adapted from the Intrinsic Motivation Inventory (IMI) [36,37]. Participants responded on a 7-point Likert scale (1 = Not at all to 7 = Very much) to items such as “The lesson is enjoyable” and “The lesson is an interesting experience”. The scale has been translated into Greek and demonstrated acceptable psychometric properties [38].

2.2.3. Vitality

Subjective vitality, representing a feeling of energy and aliveness, was measured using the Subjective Vitality Scale [39]. Students indicated their responses on a 7-point Likert scale (1 = Not at all to 7 = Very much) to items including “I feel I have a lot of energy” and “I feel so energetic and alive that I could do anything”. The scale has been used with Greek students in the past and is considered psychometrically sound [40].

2.2.4. Prosocial and Antisocial Behavior

Prosocial and antisocial behaviors in PE were assessed using the adapted Greek version of the Social and Antisocial Behavior in Sport Scale developed by Kavussanu and Boardley [41]. The instrument consists of 20 items grouped into four subscales, which distinguish between behaviors directed toward teammates and opponents. More specifically, the scale measures prosocial behavior toward teammates (5 items; e.g., “I encouraged a teammate”), prosocial behavior toward opponents (5 items; e.g., “I helped an injured opponent”), antisocial behavior toward teammates (5 items; e.g., “I verbally abused a teammate”), and antisocial behavior toward opponents (5 items; e.g., “I intentionally tried to injure an opponent”). Students indicated how often they had engaged in each behavior on a 5-point Likert scale ranging from 1 (Never) to 5 (Very often). Higher scores on the prosocial subscales reflect more frequent engagement in helping, supporting, and cooperative actions, while higher scores on the antisocial subscales indicate more frequent engagement in aggressive, harmful, or unsportspersonlike behaviors. The scale has been used with Greek students in the past and is considered psychometrically sound [42].

2.3. Procedure

The study procedures were approved by the institutional review board of the Department of Physical Education and Sport Science of the Aristotle University of Thessaloniki and were conducted in accordance with the ethical standards laid out in the Declaration of Helsinki. Prior to data collection, written informed consent was obtained from the parents or guardians of all participating students, and assent was obtained from the students themselves. The entire data collection procedure was conducted online. Researchers first obtained official permission from the school principals and the teachers’ board of each participating school. Subsequently, physical education teachers and school administrators were briefed on the study’s objectives and procedure. They then facilitated the process by informing their students about the study and distributing the digital consent forms to parents via the schools’ official online communication platforms. The online questionnaire, created using Google Forms, was distributed to students via their teachers. Participants completed the questionnaire anonymously during a scheduled school hour under the supervision of their PE teacher, who ensured a standardized administration procedure. The teachers were instructed to clarify that participation was voluntary and that answers were confidential. The questionnaire took approximately 15–20 min to complete.

2.4. Data Analysis

Descriptive statistics and analysis of correlation was performed with SPSS version 29 (IBM Corp, Armonk NY, USA). The factorial validity of the CLQ was examined using Mplus 8.10 (Muthen & Muthen, Los Angeles, CA, USA). Model fit was evaluated with multiple indices: the Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI), with values above 0.90 indicating acceptable fit and values above 0.95 reflecting excellent fit; the Root Mean Square Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMR), with values below 0.08. These indices jointly provide support for the factorial validity of a scale [43,44]. A factor score path analysis was conducted in Mplus 8.10 to examine the relationships between contextual factors (heterogeneity, interaction, social skills, assessment, tutoring, interdependence, and group reflection), motivational experiences (enjoyment and vitality), and social behavior outcomes (prosocial to teammate, prosocial to opponent, antisocial to teammate, and antisocial to opponent orientations). The model was estimated using maximum likelihood (ML). Prior to testing the structural model, the latent constructs were derived using Exploratory Factor Analysis (EFA) with Varimax rotation in SPSS version 29. Separate EFAs were conducted for each theoretical construct to ensure the unidimensionality of each scale and to compute composite factor scores. The number of factors to extract for each scale was determined via parallel analysis and examination of the scree plot. Items with primary loadings below 0.40 were considered for removal; in none of the scales were there items below this threshold. Factor scores for each participant on each construct were saved using the Regression method in SPSS. These scores represent each individual’s estimated standing on the latent variable. Then, the hypothesized structural model was tested using path analysis in Mplus 8.10, where the factor scores derived from the EFAs were imported as observed variables. This two-step approach [45], deriving factor scores independently and then using them in a path model, was employed as a pragmatic alternative to full Structural Equation Modeling (SEM) due to the complex nature of the hypothesized model and the available sample size (N = 261). This approach significantly reduces the number of parameters to be estimated simultaneously, enhancing the stability of the estimates and statistical power, which is particularly advantageous with smaller samples. Large-sample Monte Carlo simulation studies suggest that sample sizes below N = 300 can lead to issues such as non-convergence, improper solutions, and low statistical power in complex SEMs [46,47]. This two-step approach effectively decouples the measurement model from the structural model [45]. While treating factor scores as observed variables does not account for measurement error within the path model itself, the use of the regression method for score calculation and the strong simple structure achieved in the EFAs help to ensure that these scores are robust proxies for their underlying constructs.

3. Results

3.1. Preliminary Analyses

Descriptive statistics, internal consistency coefficients and correlations among the variables are presented in Table 1. The confirmatory factor analysis (CFA) of CLQ indicated that the hypothesized 7-factor model demonstrated an acceptable fit to the data: χ²(df = 875) = 1258.85, p < 0.001; CFI = 0.962; TLI = 0.959; RMSEA = 0.041 (90% CI [0.036, 0.046]); SRMR = 0.055. All standardized factor loadings were statistically significant (p < 0.001) and ranged from 0.567 to 0.866, providing support for the construct validity of the scale.

3.2. Effect of Cooperative Learning on Social Behavior

The overall fit indices suggested that the model provided an acceptable representation of the data. Although the chi-square test of model fit was significant, χ²(27) = 64.28, p < 0.001, incremental fit indices indicated good fit: RMSEA = 0.073, 90% CI [0.050, 0.096], CFI = 0.961, TLI = 0.917, and SRMR = 0.076. Taken together, these values meet conventional thresholds for acceptable model fit, with RMSEA and SRMR falling within the moderate range and CFI exceeding 0.95. The model explained a moderate proportion of variance in motivational and outcome variables. Specifically, 26.6% of the variance in enjoyment and 23.3% in vitality were accounted for by contextual factors. Prosocial to teammate and prosocial to opponent orientations showed higher explained variance (35.6% and 37.8%, respectively), whereas antisocial to teammate (4.8%) and antisocial to opponent (6.9%) orientations were only modestly explained.

3.2.1. Direct Effects

Enjoyment and vitality were strongly predicted by several contextual factors (

Table 2. Direct and indirect effects.

Direct Effects			Indirect Effects
Predictor → Outcome	β	p	Indirect Path	β	p
Group → Enjoyment	−0.38	<0.001	Group → Enjoyment/Vitality → Pro-teammate	−0.21	0.002
Group → Vitality	−0.29	0.007	Group → Enjoyment/Vitality → Pro-opponent	−0.22	0.001
Heterogeneity → Vitality	−0.21	0.040	Heterogeneity → Enjoyment/Vitality → Pro-teammate	−0.11	0.056
Interaction → Enjoyment	0.17	0.060	Heterogeneity → Enjoyment/Vitality → Pro-opponent	−0.12	0.055
Enjoyment → Pro-teammate	0.32	<0.001	Interaction → Enjoyment/Vitality → Pro-teammate	0.09	0.115
Vitality → Pro-teammate	0.3	0.001	Interaction → Enjoyment/Vitality → Pro-opponent	0.09	0.114
Enjoyment → Pro-opponent	0.33	<0.001
Vitality → Pro-opponent	0.31	<0.001
Vitality → Anti-teammate	−0.28	0.011
Interaction → Anti-opponent	0.15	0.001

pro-teammate = prosocial to teammate; pro-opponent = prosocial to opponent; anti-teammate = antisocial to teammate; anti-opponent = antisocial to opponent.

). Group reflection negatively predicted both enjoyment (β = −0.38, p < 0.001) and vitality (β = −0.29, p = 0.007), whereas heterogeneity also negatively predicted vitality (β = −0.21, p = 0.040). Interaction showed a marginally positive effect on enjoyment (β = 0.17, p = 0.060). Other contextual predictors (social skills, assessment, tutoring, and positive interdependence) were nonsignificant. Downstream effects revealed that enjoyment (β = 0.32, p < 0.001) and vitality (β = 0.30, p = 0.001) both positively predicted prosocial to teammate orientation. Similarly, enjoyment (β = 0.33, p < 0.001) and vitality (β = 0.31, p < 0.001) predicted prosocial to opponent orientation. For antisocial behavior, vitality negatively predicted antisocial to teammate orientation (β = −0.28, p = 0.011), while interaction directly and positively predicted antisocial to opponent orientation (β = 0.15, p = 0.001). Enjoyment did not significantly predict either antisocial to teammate or antisocial to opponent orientations.

3.2.2. Indirect Effects

Several indirect effects emerged through enjoyment and vitality (Table 2). Group reflection exerted significant negative indirect effects on both prosocial to teammate (β = −0.21, p = 0.002) and prosocial to opponent orientations (β = −0.22, p = 0.001), mediated by enjoyment and vitality. Similarly, heterogeneity showed a trend toward negative indirect effects on prosocial to teammate (β = −0.11, p = 0.056) and prosocial to opponent orientations (β = −0.12, p = 0.055). Interaction showed a trend toward positive indirect effects on prosocial to teammate (β = 0.09, p = 0.115) and prosocial to opponent orientations (β = 0.09, p = 0.114), although these did not reach statistical significance. No meaningful indirect effects were observed for social skills, assessment, tutoring, or positive interdependence.

4. Discussion

The purpose of the present study was to examine the direct and indirect effect of cooperative learning dimensions in motivation and social behavior of junior high school students in PE classes. Overall, the model showed that the cooperative context has meaningful links with both student enjoyment/vitality and their prosocial or antisocial orientations. More specifically, group reflection (i.e., the group’s ability to reflect on its functioning and plan improvements) emerged as a significant negative predictor of students’ enjoyment and vitality in PE. In other words, students who perceived more frequent or intense group reflection reported lower enjoyment of the class and reduced feelings of vitality. This finding is counterintuitive given that group processing is traditionally considered an essential component of effective cooperative learning [48]. According to Johnson and Johnson’s model, structured group processing or reflection is meant to improve group functioning and consolidate learning. However, there might be possible issues in how group reflection is perceived or implemented in these PE classes. Maybe reflection sessions were not conducted in an engaging or supportive manner, leading students to view them as tedious or critical. For instance, if group reflection devolved into pointing out what went wrong or was simply experienced as a pause in the action, students might have felt bored or even defensive, thereby dampening their enjoyment of the lesson. Likewise, having to reflect on group dynamics or personal contributions could potentially make some students self-conscious or highlight conflicts, which might reduce their sense of vitality. Thus, instead of energizing students, poorly facilitated reflection could momentarily sap their energy and positivity. Another possible explanation might be that reflective activities can reduce students’ perceived autonomy by increasing pressure, public evaluation, or discomfort. As Zach and Yanovich [49] indicated, a key question is what type of choices should be provided to students; if wrong choices are provided then it is possible for enjoyment and vitality to deteriorate due to discomfort students may experience. In this respect, a thoughtful pedagogical structure (e.g., ensuring competence support, meaningful options, guidance) can be used to foster enjoyment and vitality. The current findings suggest that when group reflection resembles a high-stakes evaluative process, it operates in a manner akin to competitive or performance-oriented strategies that may detract from enjoyment [50]. and indicate that even widely recommended cooperative learning strategies should be applied with care to yield positive emotional outcomes in PE settings.

Heterogeneity of the group also showed a small negative association with students’ vitality, suggesting that working in more diverse teams was related to feeling less energetic. Cooperative learning often values heterogeneous grouping as a way to enrich learning (e.g., peers can learn from each other’s diverse strengths). Yet, in our sample, students who perceived their groups as more heterogeneous were more likely to feel a bit less energetic and alive during PE. One plausible explanation is that working with diverse teammates can introduce challenges that, if unaddressed, leave some students less engaged. In addition, if students feel socially exposed, or group demands exceed their comfort level, they may feel uncomfortable participating in cooperative learning activities. This aligns with Schulze and von Huth [51]’s findings suggesting that successful use of cooperative learning in PE requires adequate teacher training, careful planning, and ability to adapt to student heterogeneity. If teachers lack time, experience, or resources to implement cooperative learning properly, cooperative learning may degrade into unstructured or chaotic group activity, reducing its benefits or causing frustration. Another plausible explanation might be that friendships and familiarity play a role; early adolescents often prefer interacting with close friends or those with shared interests. Also, in mixed-ability teams, students with lower levels of physical skills may feel demotivated when compared with their higher-ability peers. This imbalance can lead to internalized feelings of inadequacy, which directly impact a student’s emotional energy and overall vitality [52,53]. Research has shown that when students perceive their abilities to be inferior to those of their peers, the resulting negative self-assessment can diminish not only their confidence but also their willingness to actively engage in collaborative tasks [53]. That means that if the value of diversity in teamwork was not explicitly cultivated, students could perceive heterogenous groups as a hurdle rather than a resource.

In addition, and jointly with the findings on group reflection, our findings highlight that cooperative-learning outcomes in physical education depend on the interplay between structural design, group processes, and teacher facilitation. As Zach et al. [12] highlight cooperative learning is not a uniform method, its success depends on how well the chosen structure creates positive interdependence, equitable roles, and meaningful peer interaction. Effective cooperative learning in PE requires teachers to manage group formation, scaffold interaction, ensure accountability, and support socially inclusive participation. Inconsistent or under-resourced facilitation may explain why cooperative learning sometimes produces weak or mixed results.

In contrast, interaction among students showed a positive trend to impact enjoyment, although was marginally not significant, aligning with the idea that opportunities for face-to-face encouragement and teamwork tend to make class more fun. The fact that this association was only marginally insignificant, might indicate that interaction alone, without the other four “core” elements of cooperative learning, has a mild effect. Maximum benefits might occur when interaction is coupled with positive interdependence, group goals, etc. Nonetheless, the positive direction of this association reinforces the idea that encouraging students to actively engage with one another is key to a motivating PE environment. This aligns with past research that emphasizes the fun and engagement generated by peer interaction in physical activity settings [13].

Other cooperative learning elements, such as perceived social skills use, positive interdependence, tutoring, and formative assessment, did not individually predict motivation in this model. This does not necessarily mean those elements are unimportant, but that they may be baseline conditions present in most classes, or their influence might manifest indirectly. Another possibility is that these elements are more directly tied to cognitive or skill outcomes rather than immediate motivational states. These affective outcomes, in turn, were strongly related to students’ social behaviors. Higher enjoyment of PE was associated with more frequent prosocial orientations toward both teammates and opponents. This finding aligns with the research of Al-Yaaribi and his colleagues [15] that demonstrated that prosocial behaviors, such as encouraging, helping, and giving positive feedback, directly enhanced athletes’ enjoyment of the game. This heightened enjoyment may not only reinforce positive effects but may also mediate the relationships between prosocial actions, increased effort, and improved performance. In PE settings, when the environment is structured to promote feelings of enjoyment through cooperative gameplay and supportive teacher behaviors, students are more likely to engage in interactions that exemplify prosocial orientations. Such interactions manifest both among teammates and in behavior toward opponents, thereby contributing to a culture of respect and fairness on the field [15].

Similarly, higher vitality (energy and aliveness) predicted greater prosocial intentions toward teammates and opponents. When students experience vitality in PE, they are more likely to be more capable of investing sustained effort and engaging in cooperative tasks. For example, students who feel energetically engaged during physical activities have been observed to participate more actively in group exercises and to assist their peers during challenging tasks, thus fostering an overall prosocial climate [13]. The presence of vitality, therefore, may not only augment individual performance but also may bolster collective effort by enhancing interpersonal interactions and promoting a shared sense of commitment to team goals [20].

Thus, when students felt happy and energized in class, they were more inclined to encourage, help, and support others, even those on competing teams. This connection between positive personal experience and positive social behavior is consistent with theoretical expectations. For example, the theory of positive emotions [54] suggests that feeling joy or interest broadens one’s mindset and fosters altruism and helping behavior. This also aligns with self-determination theory (SDT) propositions, where an autonomy-supportive environment should enhance well-being and intrinsic motivation, leading students to willingly assist and include others as a natural extension of their positive state [55].

In terms of antisocial behaviors, vitality was linked to lower antisocial orientation toward one’s teammates, implying that students who felt more alive were less likely to insult or act aggressively toward their peers in the same group. In SDT framework, vitality is often seen as an index of wellness resulting from need satisfaction (feeling effective, connected, and autonomous). Thus, the student might be more empathetic or have more emotional resources to handle conflict, thereby less inclined to exhibit antisocial behaviors. That aligns with previous findings where engagement in physical activity was found to enhance emotional well-being, and self-esteem, which in turn correlated with reduced aggressive behavior and bullying among peers [56].

Enjoyment showed no significant direct relation to antisocial behavior, but the interaction factor had a small positive direct association on antisocial orientation toward opponents. In other words, greater within-group interaction slightly correlated with more antisocial behavior directed at rival teams. This pattern suggests that simply having fun in PE may not be enough to prevent all forms of negative behavior. A student might thoroughly enjoy a competitive game yet still behave aggressively or selfishly in the heat of competition. Enjoyment can increase the likelihood of positive acts (as we saw with prosocial behavior), but it does not necessarily inhibit negative acts once other pressures (like a desire to win) come into play. By contrast, vitality represents a more enduring sense of vigor and “good energy” that could be linked with overall mood and self-regulation capacity. A student who feels high vitality likely also feels psychologically fulfilled and less stressed, which might make them less prone to frustration or hostility [56]. Finally, the analysis of indirect paths indicated that the cooperative context was associated with social behaviors largely through enjoyment and vitality. More specifically, group reflection had significant negative indirect relation on prosocial orientations toward both teammates and opponents by undermining students’ enjoyment/vitality. Heterogeneity showed a similar trend of indirectly reducing prosocial tendencies. Research by Nicolosi and Ancona [57] has pointed out that while cooperative learning generally enhances social skills, substantial heterogeneity in group composition can sometimes introduce friction or reduce collective efficacy. Such findings suggest that the negative indirect effects of heterogeneity observed in the outcome model are not anomalous but rather reflect real challenges in managing group dynamics within cooperative settings [58]. Furthermore, contextual or dispositional variables including peer norms, school-level climate, trait aggressiveness may have influenced the effect of affective experiences on antisocial behaviors. As Zachs et al. [59] pointed out, these factors may reflect overlooked complexities in applying cooperative learning methods in real classrooms, potentially diminishing their positive effects.

Meanwhile, the interaction dimension tended to boost prosocial orientations via increased enjoyment. These mediating pathways underscore that the way cooperative learning features are associated with behavior is largely through their impact on students’ motivational states. The mediating role of enjoyment in shaping prosocial behaviors is well supported by numerous studies [13,33]. This integrative perspective is consistent with contemporary theories, such as self-determination theory, which posit that fulfillment of basic psychological needs leads to enhanced well-being and social functioning.

The overall model fit was acceptable, and it explained about 36–38% of the variance in prosocial orientation toward teammates and opponents but much less in antisocial behavior. This discrepancy suggests that while a supportive cooperative context is strongly associated with prosocial behavior, it has only modest relation to antisocial behavior, pointing to other factors. This finding is supported by a range of studies which find that cooperative learning interventions yield stronger effects on enhancing positive behaviors than on directly mitigating negative ones [13,33].

5. Conclusions

This study is not free of limitations. It is acknowledged that the factor score path analysis treats the factor scores as error-free observed variables, which can lead to an underestimation of the path coefficients (attenuation bias). However, given the sample size constraints and model complexity, this method provides a conservative and statistically stable test of the proposed relationships. Secondly, self-reports were used, and social desirability might have influenced students’ responses. In addition, the study employed a cross-sectional design, and causal inferences cannot be derived. In addition, the study did not take into account the cooperative learning approaches used by the teachers. As Zach et al. [59] have pointed out, teachers may not always correctly implement cooperative learning strategies and this may influence students’ perceptions. Nevertheless, the study provided valuable information in explaining why students choose to behave prosocially, and that antisocial behaviors may arise from competitive group dynamics or individual predispositions outside the cooperative framework. This disparity underscores the importance of addressing antisocial behavior through additional avenues, possibly including direct teaching of ethics, anger management, or by structuring competition in ways that minimize intergroup animosity. The study also reaffirms that creating a positive social environment has a more straightforward impact on promoting good behavior than on suppressing bad behavior. Importantly, as Zach et al. [59] pointed out, teachers should intentionally scaffold cooperative learning environments by structuring tasks that offer meaningful choices and shared decision-making, thereby supporting students’ autonomy without overwhelming them. To reduce anxiety during individual and group reflection, teachers can provide guided reflection prompts, model constructive feedback, and normalize mistakes as part of collaborative learning. Managing heterogeneous groups can also be facilitated by establishing explicit role assignments, rotating responsibilities, and incorporating brief skill-building activities that prepare students to participate effectively in mixed-ability teams. These practical strategies translate cooperative learning principles into concrete pedagogical actions that address the implementation barriers documented in PE settings.