Exploring emotional learning and its impact on student behavior, well-being, and resilience using structural equation modeling

Abstract

Students’ mental health in the context of emotional learning is essential to their academic and personal development. Emotional learning affects students’ emotional intelligence, social support, psychological capital, and educational environment. Emotions significantly impact learning, academic achievement, and overall well-being. Thus, students’ emotional needs must be met, and supportive learning environments must be created. This study examined the impact of emotional learning on student outcomes at the nexus of behavior, technological acceptance, mental well-being, cognitive engagement, and psychological resilience. This research was conducted using a convenience sampling technique across 10 major cities in nine provinces of China. A total of 5,313 students, comprising 2,633 males and 2,680 females, participated, and the data were analyzed using SmartPLS 3.2.9 to assess the relationships between key constructs. Out of the 11 direct correlations, 10 were confirmed with statistical significance (H1: t > 26.769, p < 0.000; H2: t > 25.226, p < 0.000; H3: t > 15.656, p < 0.000; H5: t > 11.334, p < 0.000; H6: t > 231.784 p < 0.000; H7: t > 34.375, p < 0.000; H8: t > 17.719 p < 0.000; H9: t > 19.060, p < 0.000; H10: t > 9.235, p < 0.000; H11: t > 10.307 p < 0.000), while the correlation for the 4th hypothesis was not statistically significant (H4: t > 0.248, p < 0.804). Students’ cognitive engagement is multifaceted and influenced by their prior knowledge, cognitive load, perceived value of the learning system, and instructional practices. Establishing effective learning environments that support students’ academic success and cognitive development requires understanding and fostering cognitive engagement.

Introduction

Educators and researchers have gradually recognized the importance of emotional learning in recent years for its role in student development and academic success. By examining the impact of emotional learning on various student outcomes, this study aims to help close the gap and provide policymakers and educators with actionable insights. Research shows that emotional learning affects learning interest, academic emotions, and technological adoption. A previous study highlighted the link between emotional learning, technology acceptability, and student results. It suggested that emotional learning affects students’ technological attitudes, learning pleasure, and engagement¹. Emotional learning is crucial for cognitive engagement and academic achievement, as research has shown that emotional intelligence and cognitive engagement significantly impact student learning outcomes². While emotional learning and cognitive engagement have been shown to affect academic performance³, noted the lack of research on the relationship between emotions and the adoption of education technology. It is implied that emotional learning influences students’ attitudes toward technology use and their enjoyment and engagement in learning. Since research shows that emotional intelligence and cognitive engagement enhance student learning outcomes, emotional learning is crucial⁴. Emotional learning and cognitive engagement significantly impact academic success; however, there is a notable lack of emotional education, as well as the adoption of technology^5,6. Personality and emotions may affect academic achievement. Learning methods modulate these factors and academic achievement⁷. Researchers have also investigated the impact of emotional regulation on learning engagement. For example⁸, found that psychological safety and self-efficacy influence emotion regulation and learner engagement. Positive emotions are essential for efficient learning, affecting approaches and performance⁹. Research also indicates that learning strategies impact students’ academic moods, online learning satisfaction, and the relationship between emotional experiences and learning processes¹⁰.

Studies have shown the importance of addressing students’ emotional needs throughout the learning process, as emotional well-being significantly impacts academic success and adjustment¹¹. Research indicates that favorable emotions motivate students to study, while negative emotions decrease their interest¹². Motivation, learning approaches, cognitive resources, learning autonomy, and academic success are linked to emotions affecting students’ learning¹³. This study also examined the complex impact of emotional learning on student behavior. Recent studies have shown that emotional learning can affect students’ academic behaviors, satisfaction with their learning, and achievements, while learning methods can affect students’ academic feelings and online contentment. It is therefore crucial to note that these emotional experiences can help shape the learning environment¹⁴. Recent studies have demonstrated that emotional intelligence can influence students’ academic performance, focusing on the relationship between emotional factors in the educational context¹⁵. Other research has shown that students’ emotions are crucial in driving their academic performance. For instance, positive learning emotions help students in university to concentrate on and grasp more from lectures, indicating that emotions can impact their studies¹⁶. Likewise, research emphasizes that students who acquire social and emotional competencies, such as managing emotions or collaborating with others, are more likely to succeed in school and remain engaged¹⁷. It has been discovered that when students become aware of the value of such skills, it enhances their learning, academic success, and even their behaviour in the classroom. Social skills and feelings are directly connected to student performance and behavior¹⁸.

Based on research, incorporating technology into classrooms enhances students’ enjoyment and engagement, making them more willing to utilize digital tools in subsequent activities. Another study proposed a model for successful technology integration in higher education¹⁹. Numerous studies have investigated the impact of students’ self-efficacy, motivation, and use of specific technologies on their academic performance. For instance, when students perceive technology as user-friendly and valuable, they are more inclined towards independent learning^19–21. It was also established that technology enhances academic success, fosters community ties, particularly in flexible learning contexts, and equips learners with the ability to handle unexpected situations, such as those encountered in distance learning^22,23. However, integrating technology in education is not always easy, with such issues as teacher training and access to resources among the challenges²⁴.

Students’ adoption of technology is complicated by their belief in their ability to use technology, their motivation, the extent to which the technology aligns with their tasks, and the unique effects of different technologies on their learning outcomes²⁵. Students’ mental health throughout emotional learning is essential to their academic and personal development. Work-integrated learning, public health disasters, online learning during the pandemic, and COVID-19 have highlighted student emotional and mental health issues^26–30. The research emphasizes that universities and educational institutions must prioritize student mental health, especially in crises³¹. During the COVID-19 pandemic, students prioritized their emotional well-being, focusing on positive emotions^32,33. Stressed educators’ and institutions’ vital involvement in student mental health wellness, and emphasized student mental health literacy and educators’ responsibility. The COVID-19 pandemic’s impact on middle schoolers’ emotional resilience and learning management showed their vulnerability to psychological distress when studying is difficult³⁴. Well-being extends beyond emotional health; thus, schools must promote it and adopt social-emotional learning frameworks³⁵. The link between student wellness and teaching and learning underscores the importance of implementing supportive practices and policies to enhance student mental health^36,37.

This study conceptualized student outcomes as a composite of behavior, technology acceptance, mental well-being, cognitive engagement, and psychological resilience. The study also supported a systematic approach to collecting student welfare and mental health data, emphasizing the need for evidence-based methodologies. These factors were selected based on their well-established roles in students’ academic success, personal development, and adaptability to different learning environments and systems³⁸. Moreover, students’ study behavior demonstrates their active participation and involvement in curricular and non-curricular activities, impacting their discipline, motivation, and overall engagement in the learning process. Psychological resilience also enables learners to recover from difficulties and failures, thus maintaining their trajectory towards long-term success. This research combined all these various elements into a holistic model aimed at explaining the contribution of emotions in shaping learning outcomes, aligning with accepted postulates in both the educational and psychological disciplines. Adopting technology in education is crucial, as it determines how well students are prepared and willing to utilize digital technologies for educational purposes, thereby affecting their potential to learn. Mental well-being is also essential because it keeps students emotionally balanced, enabling them to perform academically and form healthy relationships. Cognitive engagement involves attention and strategies learners can employ to understand and recall their acquired knowledge³⁹.

Literature review

The influence of emotional learning on students’ mental well-being

Emotional learning affects students’ emotional intelligence, social support, psychological capital, and educational environment. Research shows that emotional intelligence, social support, and psychological capital moderate the effects of emotional learning on mental health^40–42. Emotional intelligence enables individuals to manage their emotions and navigate social situations effectively. It also improves their academic performance, employment, and mental health^43,44. Due to the interrelationship of emotional learning, social support, and mental health, social support can mitigate the effects of childhood maltreatment on emotional intelligence and mental symptoms⁴⁵. The classroom environment and teaching methods also affect students’ mental health. Student well-being is improved by student-centred learning, intercultural understanding, and emotional intelligence⁴⁶. Optimistic psychology and mental health education enhance students’ mental well-being by fostering optimism and promoting emotional regulation—university students’ mental health benefits from emotional intelligence and social support⁴⁷. We, therefore, need to incorporate emotional learning into schools to enhance students’ mental health.

Students’ cognitive engagement

Learning requires cognitive engagement, involving intellectual effort and attention to understand complex concepts and apply them in real-world situations^48,49. Students participate in learning by trying to grasp the material, make connections, and use it⁵⁰. Cognitive engagement is linked to strategic learning and active self-regulation, which enables students to learn independently and solve problems⁵¹. It is also linked to students’ cognitive effort, desire to exceed minimal expectations and master advanced skills, and the belief that success in a sector would boost their self-esteem⁵². Several factors influence students’ cognitive involvement, and research indicates that the usefulness of the online learning system, instructor presence, and academic self-efficacy impact student engagement and satisfaction⁵³. Researchers have also examined how prior knowledge and cognitive load affect learner engagement. Students with previous knowledge and a lower cognitive load are more likely to seek instrumental support, resulting in improved learning engagement⁵⁴. In smart classrooms, students’ impressions of the learning environment, motivation, and self-efficacy foster deep cognitive engagement, encompassing cognitive, behavioral, and emotional factors^55,56.

Students’ behavior regarding cognitive engagement

Cognitive engagement refers to students’ intellectual effort and dedication to learning at school. The phrase refers to the close association between students’ opinions on the advantages of subject excellence; the research links cognitive and behavioural participation to student learning⁵¹. Student participation encompasses behavioral, emotional, and cognitive engagement. Cognitive involvement precedes behavioral engagement and affects learning; students’ cognitive participation may also impact teacher-student interaction^57,58. Research has also linked cognitive participation to academic success. Research shows that cognitive participation predicts academic performance⁵⁹. Academic involvement is related to cognitive reappraisal, indicating that academic engagement activities benefit students psychologically⁶⁰. Cognitive engagement mediates mental awareness and psychological and cognitive engagement among university students⁶¹. Teachers are crucial to cognitive engagement. Research shows that authentic-based multimedia learning enhances students’ cognitive and behavioural engagement, enabling them to master critical skills without distraction⁶². A reading nook project in elementary schools boosted students’ cognitive abilities and reading motivation⁶³.

The emotional learning influence on cognitive engagement

A study on junior middle school mathematics involvement identified cognitive engagement as a key component, encompassing behavioral, emotional, and cognitive inputs⁵¹. Numerous studies have indicated that emotional learning boosts cognitive engagement. Researchers have demonstrated that online course instructors significantly impact students’ behavioral, emotional, and cognitive engagement⁵³. The research has also highlighted the importance of perceived teacher support and instructional design on cognitive and emotional engagement and learner engagement⁶⁴. Additionally, cognitive engagement enhances the relationship between emotional intelligence and study habits, illustrating their interconnection⁶⁵. Additionally, cognitive engagement was found to promote social, behavioral, and emotional participation in mathematics learning during the COVID-19 pandemic. The tight association between cognitive and emotional involvement was highlighted⁶⁶. Moreover⁶⁷, found that diminished emotional involvement may lower students’ cognitive engagement in learning, highlighting the importance of emotional participation. Their research emphasized the link between emotional and cognitive development. Emotional intelligence integrates emotive and cognitive factors. It encompasses cognitive and emotional dimensions, promoting advanced development and personal and professional success⁶⁸.

Research gap

While growing evidence from research suggests that emotional learning yields benefits, we still lack a comprehensive understanding of its long-term impact on student outcomes, particularly in diverse educational contexts. Current research tends to focus on short-term behavior or academic gains, but there’s a lack of research into how emotional learning shapes students’ overall development. This investigation fills this gap by exploring the longer-lasting and widespread impacts of interventions that develop emotional learning across various aspects of student success.

Research objectives

The primary objective of this study is to evaluate the impact of emotional learning on students’ outcomes, including academic performance, emotional well-being, social behavior, and interpersonal skills. Specifically, the study aims to correlate emotional learning with students’ regulation of emotions, empathy, and their active role in the classroom. Moreover, the study also aims to examine whether differences exist in the impact of emotional learning programs on students across various educational environments, providing empirical data to inform the development of pedagogical interventions and policies.

Statement of the study

Emotional learning is a complex and vital process for students, as emotions significantly impact learning, academic achievement, and overall well-being. Therefore, meeting students’ emotional needs and creating supportive learning environments is essential. However, there is a gap in understanding how emotional learning influences technology adoption, highlighting the need for further research. The effects of emotional learning on students’ behavior, technological acceptance, mental health, cognitive engagement, and psychological resilience are complex and warrant further investigation. Emotional learning significantly impacts numerous student outcomes, underscoring the need for continued research to understand and leverage its effects in education. In light of the COVID-19 pandemic and the evolving landscape of technology-driven learning, understanding and actively addressing these issues are crucial to integrating technology into education. The conceptual framework, developed in reference to the literature, is illustrated in Fig. 1, and the proposed hypotheses are outlined below.

Fig. 1.

The conceptual model of the study variables.

Hypotheses

H1: Emotional learning positively influences mental well-being.

H2: Emotional learning positively influences students’ behavior.

H3: Emotional learning positively influences psychological resilience.

H4: Emotional learning positively influences technological acceptance.

H5: Emotional learning positively influences cognitive engagement.

H6: Students’ behavior during emotional learning has a positive impact on their mental well-being.

H7: Students’ behavior during emotional learning has a positive impact on their cognitive engagement.

H8: Technological acceptance has a positive association with mental health.

H9: Technological acceptance positively influences cognitive engagement.

H10: Mental well-being has a positive association with psychological resilience.

H11: Cognitive engagement has a positive association with psychological resilience.

Theoretical framework

The present research has its theoretical foundation in the interconnection of Self-Determination Theory (SDT), self-efficacy, and the Technology Acceptance Model (TAM). SDT clarifies how students’ learning and engagement behavior is shaped by intrinsic and extrinsic motivation, and how emotional learning is critical in developing self-motivation⁶⁹. Self-efficacy, as defined by⁷⁰, is one’s perception that one can accomplish a task. It also bridges affective learning and student motivation by influencing learners’ persistence, confidence, and flexibility. Furthermore, as discussed by⁷¹, the TAM examines students’ adoption and usage of technology in education. It perceives that their attitude towards embracing technology adoption hinges on the convenience and usefulness they consider it to have. Emotional learning influences what learners think regarding technology and, as a result, influences their engagement in online learning environments.

The present theoretical framework presents how students at a university incorporated technology-supported learning during the COVID-19 pandemic. Empirical evidence suggests that Task-Technology Fit (TTF) is a distinct factor in shaping students’ usage of e-learning, influencing their satisfaction and academic performance, thereby making it effective^72–74. The TAM may also consider emotional well-being as a crucial factor in understanding the emotional lives of students, given that emotions are becoming increasingly integrated as determinants of technology adoption⁷⁵. In this regard, the personality characteristics of narcissism and psychopathy were identified as determining factors for technology adoption, establishing the significance of emotional factors in technology adoption⁷⁶. Emotional factors also influence the adoption of technologies and help overcome learning issues, such as techno-stress^77,78. As explored in this research, emotional learning plays a significant role in an individual’s readiness to adopt new technologies. The Unified Theory of Acceptance and Use of Technology (UTAUT) also outlines how emotional factors affect technology adoption, particularly among older adults⁷⁹. Furthermore, learners’ value of e-learning influences their intention to adopt it, illustrating the intricate link between affective states and technology adoption⁸⁰. In summary, this model demonstrates how emotional learning influences motivation, self-efficacy, and the adoption of digital learning materials, ultimately impacting students’ academic performance and overall well-being.

Research question

The research question that this study aimed to address is as follows:

1. How does emotional learning affect student outcomes, particularly regarding their behavior, mental well-being, technology acceptance, psychological resilience, and cognitive engagement?

By answering this question, the primary objective of this study was to discuss the impact of emotional learning on students’ outcomes and to provide in-depth insights into the mechanisms by which emotional learning can influence students’ personal and academic development.

Research method

Study locale

This research was conducted in 10 major cities across nine provinces in China, with detailed information on the number of participants from each city provided in Figs. 2 and 3.

Fig. 2.

Study population.

Fig. 3.

Targeted cities.

Data collection

A quantitative approach was employed in this study, and a structured questionnaire was developed to collect data on emotional learning and its impact on students’ outcomes. This survey questionnaire was validated using the Index of Item-Objective Congruence (IOC), ensuring the reliability and clarity of each question⁸¹. The administration and design of our survey followed the CHERRIES criteria⁸². The study is approved by the ethical review board of Shenzhen Technology University, numbered (SZTU0236) dated October 20, 2024, and written informed consent was received from participants by ensuring that their responses will be used for research purposes only. To meet the research goals, the final version of the online survey comprised 38 items, covering the key constructs of behaviour, technology acceptance, mental well-being, cognitive engagement, and psychological resilience. A 5-point Likert scale was used, with 1 indicating strong agreement and 5 indicating strong disagreement.

We assessed the data quality during collection and ensured that participants’ privacy and data were protected. Initially, 350 responses were collected from participants as a pretest. After some modifications, the final version of the questionnaire was distributed for data collection.

Survey measures

School-based emotional learning impacts mental health. Student and teacher socio-emotional well-being are crucial because trauma and emotional distress can influence cognitive, social, and emotional development, academic performance, and classroom dynamics^27,29,35,83. School mental health relies on emotional intelligence and social support^84,85. Behavior is students’ active involvement in academic and non-academic activities, whereas emotional engagement is their agreement or rejection of instructors’ directives. Cognitive engagement is students’ endeavor to comprehend and apply complex ideas⁸⁶. Ecological, cognitive, motivational, and emotional factors may impact academic achievement^87,88. A study examining how systems engineering courses promote systems thinking found that training enhanced cognitive systems thinking, but not emotional involvement.

A study found that behavioral and emotional involvement significantly influenced online student satisfaction, highlighting the importance of emotional learning in adopting technology and student behavior²⁴. Furthermore⁵, stressed the role of emotions in students’ e-learning and technology adoption, while⁸⁹ evaluated how positive learning emotions affect college students’ academic performance, emphasizing the importance of emotional states in learning outcomes and educational quality. Highlighting the importance of emotional learning for mental health and cognitive engagement, research shows that positive emotions improve student and teacher learning. Through emotions and social relationships, learners combine cognitive and emotional components⁹⁰. Psychology and education emphasize emotional and cognitive development. The findings demonstrate how emotional learning impacts cognitive function and links emotional and cognitive processes. Cognitive emotion management affects learning and engagement⁹¹. showed that cognitive emotion management improves neuroplasticity, emotional intelligence, and behavior, while cognitive emotion management improves self-esteem, effectiveness, and originality by producing positive emotions and avoiding negative ones.

Procedure

The data were collected through an online survey from October 20, 2024, to February 20, 2025. The procedure is illustrated in Fig. 4. First, 350 participants completed our 5-point Likert scale questionnaire to ensure a reliable sample size. Pilot test results were used to refine the questions, providing the most relevant and accurate answers. To reach more participants beyond the initial contact point, we utilized WeChat to distribute the final questionnaire link in major cities in the selected provinces. Participants were asked to share the survey with other potential respondents. The questionnaire was accessible anonymously via a website https://www.wjx.cn/ on mobile and PC platforms. Each IP address, computer, or mobile phone can contribute only once to avoid duplication. Over 5,500 people received a link to the online survey questionnaire, research information, and instructions on completing and submitting it. After a rigorous quality evaluation, 5,313 valid responses were included in the analysis.

Fig. 4.

Research procedure.

Results

Study participants

The study recruited volunteers aged 23–42 from 10 major Chinese cities who held at least an undergraduate degree. Table 1 shows the demographic details of all survey participants. The sample included 2,633 males and 2,680 females. There were 927 respondents under 27 years old, 1,984 were between 28 and 32 years old, 1,207 were between 33 and 37, 998 were between 38 and 42, and 197 participants were above 42 years old. Regarding education, 886 respondents held doctorates, 1,492 held master’s degrees, 1,731 held bachelor’s degrees, and 991 held undergraduate degrees. The study comprised students in the humanities, psychology, sociology, education, and social sciences. As stated in Table 1, participants from different majors, including both male and female, were selected to ensure representative data.

Table 1.

Demographics of study participants (N = 5,313).

Demographics	Category	Overall (N = 5,313)
Age	23–27	927(17.44%)
	28–32	1,984(37.34%)
	33–37	1,207(22.71%)
	38–42	998(18.78%)
	Above 42 years	197(3.70%)
Gender	Male	2,633(49.56%)
	Female	2,680(50.44%)
Marital Status	Single	1,871(35.21%)
	Married	3,053(57.46%)
	Other	389(7.32%)
Education	Undergraduate	991(18.65%)
	Bachelor	1,731(32.58%)
	Master	1,492(28.08%)
	Doctorate	886(16.67%)
	Other	213(4.00%)

Data analysis and results

The data were analyzed using SmartPLS 3.2.9 without making any assumptions about the distribution. This study included multi-construct, structural route, and indicator variable models^92,93. The Partial Least Squares (PLS) statistical approach is a component of structural equation modeling. The goal is to anticipate outcomes and estimate the empirical relationship between variables to find credible explanations⁹⁴. The study presents the average scores, standard deviations, excess kurtosis, and skewness, as shown in Table 2.

Table 2.

Descriptive statistics.

Item No.	Mean	Standard Deviation	Excess Kurtosis	Skewness
EL1	2.853	1.4	−1.303	0.178
EL2	3.54	1.428	−0.69	−0.483
EL3	2.676	1.322	−1.086	0.321
EL4	3.658	1.212	−0.172	−0.81
EL5	2.953	1.333	−1.215	0.012
EL6	2.65	1.465	−1.476	0.111
SB1	2.604	1.352	−0.785	0.354
SB2	2.033	1.112	−0.902	0.701
SB3	2.035	1.056	−0.705	0.722
SB4	2.334	1.226	−1.337	0.303
SB5	1.908	1.086	−0.446	0.954
TA1	2.12	1.134	−0.759	0.595
TA2	2.643	1.397	−1.377	0.108
TA3	2.663	1.399	−1.378	0.089
TA4	2.616	1.356	−1.314	0.138
TA5	2.73	1.394	−1.391	0.019
TA6	2.732	1.369	−1.379	−0.023
MW1	3.607	0.762	0.895	−0.773
MW2	3.648	0.75	0.87	−0.662
MW3	3.599	0.782	0.63	−0.647
MW4	3.768	0.794	1.181	−0.787
MW5	3.619	0.847	0.533	−0.598
MW6	3.552	0.85	0.498	−0.619
CE1	3.678	0.816	0.901	−0.73
CE2	3.714	0.811	0.846	−0.712
CE3	3.736	0.796	0.887	−0.684
CE4	3.639	0.784	0.722	−0.651
CE5	3.327	1.085	−0.195	−0.671
CE6	3.41	1.078	0.049	−0.784
CE7	3.42	1.032	0.306	−0.91
PR1	3.782	0.759	1.028	−0.642
PR2	3.802	0.718	1.655	−0.992
PR3	3.667	0.698	1.381	−0.975
PR4	3.612	0.716	1.52	−1.024
PR5	3.757	0.707	2.184	−1.19
PR6	3.765	0.689	2.562	−1.235
PR7	3.642	0.738	1.351	−1.02
PR8	3.695	0.674	2.214	−1.125

Abbreviations: Emotional Learning (EL), Students’ Behavior (SB), Technological Acceptance (TA), Mental Well-being (MW), Cognitive Engagement (CE), Psychological Resilience (PR).

Model measurement

The scale’s validity and reliability for SEM analysis were evaluated by examining convergent validity, discriminant validity, and scale reliability. Table 3 presents the scale’s reliability, measured by Cronbach’s alpha (CA) and composite reliability (CR). These are above the benchmark of 0.70 recommended by⁹⁵. The range of the CA values was from 0.790 to 0.956, while the CR values ranged from 0.845 to 0.965. Convergent validity was checked by estimating factor loadings (FL) and the average variance extracted (AVE) for all scale items. The results indicate that all item loadings exceeded 0.5, while all constructs’ AVE exceeded 0.5. In addition, multicollinearity was assessed using the Variance Inflation Factor (VIF), with a minimum value of 0.500. All values confirmed the absence of multicollinearity. Figure 5 presents the results, whereas Table 3 presents evidence supporting convergent validity.

Table 3.

Factor loadings, reliability, and validity of constructs.

Constructs	Items	Loadings	VIF	Alpha	CR	AVE
Emotional Learning	EL-1	0.892	3.309	0.889	0.918	0.657
	EL-2	0.618	1.789
	EL-3	0.878	3.052
	EL-4	0.627	1.706
	EL-5	0.883	3.922
	EL-6	0.904	4.226
Students’ Behavior	SB-1	0.622	1.120	0.860	0.887	0.615
	SB-2	0.803	4.004
	SB-3	0.828	4.103
	SB-4	0.895	2.936
	SB-5	0.746	3.739
Technological Acceptance	TA-1	0,796	2.074	0.956	0.965	0.824
	TA-2	0.936	4.499
	TA-3	0.940	4.509
	TA-4	0.927	4.485
	TA-5	0.942	4.577
	TA-6	0.897	3.643
Mental Well-being	MW-1	0.720	1.858	0.849	0.888	0.570
	MW-2	0.753	2.027
	MW-3	0.735	1.976
	MW-4	0.786	2.005
	MW-5	0.778	2.159
	MW-6	0.756	1.982
Cognitive Engagement	CE-1	0.753	2.003	0.790	0.845	0.543
	CE-2	0.759	2.133
	CE-3	0.721	1.996
	CE-4	0.678	1.670
	CE-5	0.592	3.014
	CE-6	0.568	3.178
	CE-7	0.550	2.645
Psychological Resilience	PR-1	0.675	2.052	0.857	0.888	0.554
	PR-2	0.714	1.645
	PR-3	0.680	1.546
	PR-4	0.689	1.549
	PR-5	0.747	1.703
	PR-6	0.734	1.581
	PR-7	0.676	1.509
	PR-8	0.732	2.250

Fig. 5.

PLS-SEM.

Fornell lacker

The Fornell-Larcker criterion is used to determine the extent to which constructs differ. This criterion verification is the first step toward achieving discriminant validity. As^96,97 recommended, the square root of a construct’s average variance extracted (AVE) should be greater than its inter-correlation value. To differentiate from other constructs in the model, the items of a construct must exhibit greater variability. The square root of AVE for all constructs was greater than the respective inter-correlation values, demonstrating discriminant validity (See Table 4)⁹⁶.

Table 4.

Discriminant validity.

Constructs	Cognitive Engagement	Emotional Learning	Mental Well-being	Psychological Resilience	Students’ Behavior	Technological Acceptance
Cognitive Engagement	0.665
Emotional Learning	0.449	0.810
Mental Well-being	0.514	0.274	0.755
Psychological Resilience	0.625	0.303	0.676	0.706
Students’ Behavior	0.296	0.198	0.267	0.272	0.784
Technological Acceptance	0.33	0.873	0.152	0.182	0.17	0.908

To evaluate discriminant validity, we employed the Heterotrait-Monotrait Ratio (HTMT) approach, which assesses the degree to which each variable exhibits discriminant validity. The HTMT values fall below the minimal criterion of 0.90 (See Table 5).

Table 5.

HTMT.

Constructs	Cognitive Engagement	Emotional Learning	Mental Well-being	Psychological Resilience	Students’ Behavior	Technological Acceptance
Cognitive Engagement
Emotional Learning	0.556
Mental Well-being	0.602	0.312
Psychological Resilience	0.711	0.356	0.775
Students’ Behavior	0.371	0.234	0.265	0.263
Technological Acceptance	0.41	0.938	0.171	0.211	0.218

Model fit summary

Before the SEM estimation, the model’s efficiency was examined (See Table 6). The fitness of the model was evaluated using several key metrics, including the standardized root mean square residual (SRMR), geodesic distance (d_G), squared Euclidean distance (d_ULS), chi-square, and normed fit index (NFI). These measures collectively provide an in-depth examination of the internal path model, establishing a well-established basis for SEM⁹⁸. The structural model being researched in the current study was then tested using critical criteria, including the coefficient of determination for the endogenous variable, path coefficients, predictive validity, effect size, and multicollinearity⁹⁹. Table 6 presents the initial values along with explanations for each criterion.

Table 6.

Summary of the model fitness.

Fit Indices	Estimated Model
SRMR	0.08
d_ULS	2.775
d_G	0.639
Chi-Square	9451.932
NFI	0.702

Hypothesis testing

Table 7 presents the data relevant to our specific hypotheses. Of the 11 direct correlations, 10 were confirmed with statistical significance (H1: t > 26.769, p < 0.000; H2: t > 25.226, p < 0.000; H3: t > 15.656, p < 0.000; H5: t > 11.334, p < 0.000; H6: t > 231.784 p < 0.000; H7: t > 34.375, p < 0.000; H8: t > 17.719 p < 0.000; H9: t > 19.060, p < 0.000; H10: t > 9.235, p < 0.000; H11: t > 10.307 p < 0.000), while the correlation for the 4th hypothesis was not statistically significant (H4: t > 0.248, p < 0.804). All the current connections are included in Fig. 6; Table 7.

Table 7.

Standard Beta, t-statistics, and p-values.

No	Hypothesis Correlations	Sample Mean	Std. Deviation	t -Statistics	p -Values	Status
1	Cognitive engagement -> Psychological Resilience	0.376	0.014	26.769	0.000	Confirmed
2	Emotional learning -> Cognitive Engagement	0.629	0.025	25.226	0.000	Confirmed
3	Emotional learning -> Mental Well-being	0.545	0.035	15.656	0.000	Confirmed
4	Emotional Learning -> Psychological Resilience	0.003	0.01	0.248	0.804	Not Confirmed
5	Emotional learning -> Student Behavior	0.197	0.017	11.334	0.000	Confirmed
6	Emotional learning -> Technological Acceptance	0.873	0.004	231.784	0.000	Confirmed
7	Mental Well-being -> Psychological Resilience	0.482	0.014	34.375	0.000	Confirmed
8	Students’ behavior -> Cognitive Engagement	0.216	0.012	17.719	0.000	Confirmed
9	Students’ behavior -> Mental Well-being	0.221	0.012	19.06	0.000	Confirmed
10	Technological acceptance -> Cognitive Engagement	−0.255	0.028	9.235	0.000	Confirmed
11	Technological acceptance -> Mental Well-being	−0.361	0.035	10.307	0.000	Confirmed

Fig. 6.

Bootstrapping, t-statistics.

The bootstrapping results (see Fig. 6; Table 7) are further clarified by rewriting the significant paths as structural equations. These equations specify the tested relationships, with standardized β coefficients, and directly indicate their confirmation status.

SEM Equations with Standardized Coefficients.

SB = 0.197 EL + ε1SB = 0.197EL + ε1 (Confirmed)

TA = 0.873 EL + ε2TA = 0.873EL + ε2 (Confirmed)

MW = 0.545 EL + 0.221 SB − 0.361 TA + ε3MW = 0.545EL + 0.221SB − 0.361TA + ε3 (All confirmed)

CE = 0.629 EL + 0.216 SB − 0.255 TA + ε4CE = 0.629EL + 0.216SB − 0.255TA + ε4 (All confirmed)

PR = 0.003 EL + 0.482 MW + 0.376 CE + ε5PR = 0.003EL + 0.482 MW + 0.376CE + ε5 (EL→PR not confirmed; MW and CE confirmed )

This presentation highlights that of the 11 hypothesized paths, 10 were confirmed, with the exception being the non-significant direct link between Emotional Learning and Psychological Resilience. Indirect effects through Mental Well-being and Cognitive Engagement, however, remain strong predictors of resilience.

Discussion

The results of this study provide significant insights into the relationships between emotional learning and its related aspects, such as students’ psychological resilience, cognitive engagement, mental health, acceptance of technology, and behaviour. The results offer theoretical and practical implications for understanding how these constructs interact within an educational context. Psychological resilience enables individuals to adapt and thrive in challenging situations. Emotional learning, emotional intelligence, and control have been demonstrated to boost psychological resilience. These findings indicate that psychological resilience mitigates negative feelings and enhances creativity. According to numerous studies, strong emotional resilience is associated with excellent learning management skills¹⁰⁰. Individuals with high emotional intelligence tend to have stronger psychological resilience. Research indicates that psychological resilience has a direct impact on emotional stability and an indirect effect through pleasant and negative mood states^101–103. Researchers have also examined psychological resilience and unpleasant emotions. Psychological resilience relates to college students’ self-concept and negative emotions¹⁰⁴.

Furthermore, psychological resilience partially mediates the link between negative emotions and creativity. Post-traumatic development affects this connection¹⁰⁵. Psychological resilience and emotion management are also crucial in understanding the relationship between psychological resilience and examination anxiety, especially for medical students¹⁰⁶. Emotion regulation mediates the effect of emotion management on the psychological capital of university students¹⁰⁷. The impact of personal attributes and learner emotions on resilience has also been recognized, with emotional intelligence being found to increase the resilience of second language learners¹⁰⁸. The current study’s findings have also revealed that emotional intelligence directly impacts job success and employee resilience, highlighting its relevance to professional resilience. The COVID-19 pandemic showed the relevance of psychological resilience and emotional self-efficacy to regulating the link between college students’ stress and anxiety¹⁰⁹. Research has examined how emotional intelligence affects resilience in various contexts, including the parenting of children with special needs. Research shows a substantial positive link between emotional intelligence and resilience¹¹⁰. Research also shows that emotional intelligence enhances resilience, particularly regarding work engagement during the pandemic¹¹¹.

Furthermore, emotional intelligence and resilient character traits significantly impact school teachers’ psychological well-being, underscoring the role of emotional intelligence in educational resilience¹¹². Additionally, studies have examined how emotional intelligence affects nurses’ happiness. Self-efficacy and resilience mediate this connection¹¹³. Emotionally intelligent people may also exhibit psychological resilience¹¹⁴. University students’ psychological resilience is linked to emotional intelligence sub-dimensions¹¹⁵. Emotional regulation and adaptive-feedback emotional computing technologies improve learning effectiveness and reduce learning anxiety, demonstrating the impact of emotional learning on educational outcomes^116,117. Stressing the necessity of emotional regulation in transforming emotionally intense experiences into optimal learning opportunities underscores the importance of emotional learning in education¹¹⁸. Psychological resilience is linked to well-being and emotional expressiveness, highlighting its importance in emotional learning¹¹⁹. Emotional intelligence and resilience are also crucial for self-perception, self-control, self-motivation, modeling, and social practices, particularly among athletes^120,121.

Numerous studies have examined the intermediate factors that link students’ cognitive engagement with psychological resilience. Positive coping techniques boosted middle schoolers’ cognitive reappraisal abilities and psychological resilience¹²². Cognitive reappraisal mediates the relationship between coping mechanisms and psychological resilience. Additionally, higher post-traumatic development was connected to increased use of constructive coping strategies, which promoted cognitive reappraisal and psychological resilience¹²³. Cognitive reappraisal was also explored as a mediator between thinking style and resilience. Prior research found a positive association between cognitive evaluation and resilience¹²⁴. The study revealed that cognitive systems and thinking styles may impact student resilience. Student participation was found to be negatively correlated with mental health symptoms and positively correlated with resilience and other positive psychological traits¹²⁵. These results suggest that active engagement may help students build psychological resilience.

Researchers have also investigated psychological resilience, undergraduate student attributes, and coping mechanisms, stressing the need to understand how personality and coping affect resilience¹²⁶. These data suggested that environmental factors and support networks boost student resilience. The study also recommended cognitive behavioral therapy to improve students’ mental health, particularly their resilience. It suggested that specific strategies could help students manage academic and personal issues. As mentioned earlier, the analysis of the results highlights the interconnectedness of students’ behaviour, cognitive engagement, and mental well-being, with positive student behaviour contributing to academic engagement and psychological health. These findings align with the existing literature emphasizing the importance of emotional intelligence and self-regulation in educational settings. A study highlighted the complexity of resilience and underlined the importance of considering multiple factors in assessing it. Moreover, standardized psychological therapy and university mental health education have enhanced college students’ psychological resilience, indicating that institutional support may improve student resilience^127,128.

Conclusion

Educational experiences are complicated by emotional learning’s effects on students’ behavior, and significantly contribute to shaping students’ academic and psychological outcomes. Many emotional, psychological, and environmental factors influence people’s attitudes and behaviors regarding technology, making emotional learning and technological adoption a complicated and ever-changing research topic. Emotional intelligence, academic emotions, and social-emotional competence affect students’ behavior, academic performance, and learning satisfaction. The findings suggest that enhanced emotional learning and cognitive engagement interventions can positively impact students’ mental well-being, resilience, and technology acceptance. The findings comprehensively understand the complex relationship between emotional learning and technical acceptance, emphasizing the need for more research. Students’ cognitive engagement is multifaceted and is influenced by their past knowledge, cognitive load, perceived learning system value, and instructional practices. Establishing effective learning environments that support students’ academic success and cognitive development requires understanding and fostering cognitive engagement.

Study limitations

While this study provides valuable insights and findings related to a crucial topic, it has limitations. First, this study relied on self-reported data generated through an online survey, which may be subject to biases such as social desirability, regional choice, or recall bias. Future research could benefit from incorporating objective measures of the constructs studied and could include other variables accordingly. Second, it was a cross-sectional study, limiting the ability to draw causal inferences. Longitudinal studies would better understand the causal relationships between these constructs. Finally, the study was conducted in a specific educational context based in nine regions of China, so the findings may not be generalizable to other settings.

Research implications

This study adds strong empirical evidence to understand more deeply the complex nature of emotional learning and school achievement, including students in higher education in China. The findings emerging from a multi-group SEM approach (applied to 5313 students’ responses) inverting the positive and significant influences of emotional learning. These results have theoretical and practical implications for pre-service teachers, policymakers, researchers of educational psychology, and technology use.

Theoretical implications

This study aims to integrate key theoretical constructs such as Self-Determination Theory (SDT), self-efficacy theories, and the Technology Acceptance Model (TAM), introducing emotional learning as a central driver for student outcomes. The confirmation of 10 of 11 hypotheses and the high number of significant associations illustrate, alternatively, that emotional aspects characterize cognitive, behavioral, as well as psychological functions (e.g., Pfaff & Adolf, 2012), which continue in their correlations (cf. meta-analyses on EI, SEC, or resilience). It is interesting that the non-significant effect among emotional learning and technology acceptance (H4) contrasts with TAM extensions models concepts that assumed emotional factors directly cause the use or continue use of a technology 7 which must be deeply contrasted, as seemingly there may also mediation roles by other constructs like perceived usefulness or barrier situations as access during emergency crisis to perform a school subjects as happens in COVID-19. The current discovery may further suggest that theoretical models should be revised to include moderators such as cultural context factors, because social support and emotional regulation in collectivist China may be more concerned about interpersonal than technological outcomes. The work presented here contributes to ecology’s student development model by identifying an emotional learning process between states of arousal and adaptive goals. It provides a foundation for continued interdisciplinary syntheses of psychology and education.

Future scope of the research

While this paper has provided a thorough SEM analysis, limitations associated with subjective measurement, cross-sectional data, and the China-based sample offer multiple promising research directions.

Longitudinal and causal assumptions

No assumptions were made about causality as this study was cross-sectional. Longitudinal studies may also be employed to track the time-priority and dynamic reciprocity regarding changes in emotional learning and student outcomes (e.g., before and after an intervention across academic semesters, or during crises).

Sample

The participants were urban Chinese students in the range of 23–42 years of age with a high level of education, which might limit generalization. Taking into consideration that the present study was situated in a collectivistic eastern culture, applying emotional learning to other populations (e.g., rural areas or younger populations like adolescents) might help to discover some cultural differences in how emotional learning affects both TA and resilience, investigating the generalizability of our model across different populations with various age ranges or cultures.

Author contributions

MY and DD written the original draft, YJ collected data and analyzed it and participated in discussion section. All authors participated equally in this manuscript.

Funding

The authors thank Prince Sultan University for funding this research project under the Language and Communication Research Lab grant [RL-CH-2019/9/1].

Data availability

Data are not publicly available due to ethics committee limitations; however, they can be provided to the corresponding author upon reasonable request.

Declarations

Competing interests

The authors declare no competing interests.

Ethics approval

The ethical guidelines specified in the World Medical Association’s Helsinki Declaration were followed throughout the research conducted for this study. The research’s goals were explained to the participants, examined, and authorized by the ethical committee review board of the School of Foreign Languages, Shenzhen Technology University, P.R. China. A signed informed consent form was received from the participants who voluntarily consented to this research and understood that personal information would be kept secret.

Informed consent

Before collecting the data, permission was received from participants to use the data for research purposes.