Methodological Strategies to Enhance Motivation and Academic Performance in Natural Sciences Didactics: A Systematic and Meta-Analytic Review

Abstract

Learning Natural Sciences represents a key opportunity to spark scientific interest and foster fundamental skills across different educational stages. This study aimed to analyze the influence of motivation on academic performance in the learning of Natural Sciences at various educational levels. To this end, a systematic review method was employed following PRISMA guidelines, consulting the Web of Science and Scopus databases, identifying four relevant studies. The results showed that high levels of motivation were associated with a more positive classroom attitude and better conceptual understanding, which enhanced academic performance. The use of innovative methodological strategies, such as implementing immersive virtual reality in the classroom, PhET simulations (Physics Educational Technology), and the use of hypertext, significantly increased both student motivation and academic performance. The meta-analysis revealed a favorable effect in experimental groups, showing moderate heterogeneity (I² = 49) and significance of p = 0.0001. The concurrence analysis reported that current pedagogical practices should focus on strengthening student autonomy and active engagement, integrating critical reflection, the use of innovative methodological strategies, and technological resources that enhance meaningful learning in scientific literacy. Among the instruments used to measure motivation, the Motivation to Learn Science Questionnaire was identified, and for academic performance, the Motivated Strategies for Learning Questionnaire. In conclusion, the importance of implementing the identified methodological strategies across different educational stages is emphasized, in order to promote competency-based learning through meaningful and innovative acquisition of content in Natural Sciences.

1. Introduction

In recent years, advances in neuroscience have transformed our understanding of the human brain, revealing the processes that drive learning, memory, and adaptation to dynamic situations (Duncan, 2025). Within this context, neuroeducation has emerged as a novel field that combines findings from cognitive science with educational practice, offering strategies to improve the quality and equity of education in the twenty-first century (Sánchez et al., 2025). This perspective aligns with Sustainable Development Goal (SDG) 4 of the United Nations 2030 Agenda, which promotes inclusive and high-quality education focused on the development of crucial skills such as critical thinking, problem-solving, and global awareness (United Nations, 2023). Accordingly, the Didactics of Natural Sciences, from preschool to initial university teacher training, requires a gradual approach that stimulates curiosity, encourages environmental observation, and fosters the development of scientific thinking and understanding of natural phenomena (Cañal, 2018; Ospankulova et al., 2024). Preparing education professionals to cultivate students’ interest in science not only strengthens scientific literacy but also contributes to creating a reflective citizenship committed to sustainable development and environmental protection, in line with SDG 13 (Climate Action) and the foundations of Education for Sustainable Development (United Nations, 2023).

1.1. Motivation, Self-Efficacy, and Active Learning in the Science Classroom

Motivation emerges as a central element in the teaching-learning process, as it directly affects academic performance and the quality of knowledge constructed by students (Mendoza et al., 2025). Studies such as those conducted by Ábalos-Aguilera et al. (2024) and Stieha et al. (2024) show that students with high levels of motivation not only achieve better academic results but also develop a more autonomous, persistent, and positive attitude toward studying. Learning Natural Sciences requires active student engagement, involving processes of exploration and observation (Chengere et al., 2025), scientific reasoning (Abate et al., 2024), and a meaningful connection with the natural environment (Sultana & Hawken, 2023). This engagement is deeply influenced by emotional well-being and the student’s level of involvement in their learning process, factors closely linked to motivation (Pich, 2020).

Self-Determination Theory, proposed by Deci and Ryan (1985), asserts that sustainable motivation, capable of fostering ecological responsibility and social commitment, depends on the satisfaction of three basic psychological needs: autonomy, competence, and relatedness. This theory distinguishes between intrinsic motivation (driven by interest or enjoyment in the activity) and extrinsic motivation (derived from external rewards or social pressure). This presents the challenge of translating the theory into concrete pedagogical strategies that promote motivation in educational contexts. In education, this distinction is key to understanding the degree of student engagement with academic challenges, which, according to Wang et al. (2023), is related to cognitive flexibility and academic achievement. Intrinsic motivation is associated with sustained improvement in academic performance, the development of self-efficacy, the construction of student identity, and continuous effort in learning (Y. Liu et al., 2019). Authors such as Bandura (1997) indicate that perceived self-efficacy is a significant predictor of academic performance, as students with greater confidence in their abilities tend to exert more effort, persist in the face of difficulties, and achieve better academic results.

Learning Natural Sciences requires flexibility, creativity, and responsibility to adapt the educational environment to individual student needs (Triyanto & Handayani, 2018). Pujol (2010) emphasizes the importance of transcending mere content transmission, promoting comprehensive learning that combines thought, action, and communication. She proposes that students view science as a tool to interpret reality, developing scientific competencies, critical thinking, and social awareness. Moreover, she advocates fostering responsible and engaged citizenship, overcoming stereotypes, and promoting values such as autonomy, cooperation, and active learning in collaborative classroom environments where students interact and exchange ideas (Gerard et al., 2019).

Student motivation is influenced by various factors, including the emotional climate of the classroom (Ma & Wei, 2022), teaching style (Abid et al., 2025), and the type of proposed activities, for example, through the integration of Information and Communication Technologies (ICT), which promote inclusion and personalized learning (Soriano-Sánchez, 2025). Additionally, perceived self-efficacy (Kans & Claesson, 2022) and the connection of functional content—that is, content linked to everyday life—significantly influence motivation and academic leadership (Zupanec et al., 2018).

1.2. Developmental Progression of Learning in Natural Sciences

The learning of Natural Sciences develops progressively throughout the educational system, adapting to the evolutionary and cognitive characteristics of students at each stage (Harlen, 2010, p. 12). In the Early Childhood Education stage, in the case of the curricular framework of Andalusia (Spain), this knowledge is introduced through the area of Discovery and Exploration of the Environment, where curiosity, observation, and contact with the nearby natural world are encouraged through learning situations (LdL) that involve activities, exercises, and tasks (Order of May 30, 2023, 2023a, establishing the curriculum for Early Childhood Education in the Autonomous Community of Andalusia, among other aspects). This stage covers ages 0–6 and encompasses the “Sensorimotor” and “Preoperational” developmental stages, making it necessary to promote symbolic play to interpret reality (Piaget, 1970).

In Primary Education, this area is linked to the area of Knowledge of the Natural, Social, and Cultural Environment, which covers ages 6–12, expanding students’ understanding of living beings, physical phenomena, and natural resources, thereby promoting an integrated view of the environment (Order of May 30, 2023, 2023b, establishing the curriculum for Primary Education in the Autonomous Community of Andalusia, among other aspects). During Primary Education, students reach the “Concrete Operations” stage, enabling them to comprehend reality (Piaget, 1970).

In the following stage, Compulsory Secondary Education (ESO), covering ages 12–16, the content is organized into specific subjects such as Biology and Geology, as well as Physics and Chemistry, allowing for a deeper understanding of scientific concepts and the development of critical thinking (Pérez et al., 2020), as established in the corresponding curricular framework (Order of May 30, 2023, 2023c, establishing the curriculum for Compulsory Secondary Education in the Autonomous Community of Andalusia, among other aspects). At this stage, students generally reach the fourth and final developmental stage proposed by Piaget (1970), “Formal Operations,” which enables cognitive development that allows the abstraction of concepts. Subsequently, in Upper Secondary Education (Bachillerato, ages 16–18), the Science and Technology track includes subjects such as Biology, Geology, Physics, and Chemistry, enabling students to develop advanced scientific competencies that allow them to interpret natural phenomena from a rigorous perspective (Order of May 30, 2023, 2023d, establishing the curriculum for Bachillerato in the Autonomous Community of Andalusia, among other aspects).

At the university level, initial teacher education covers degrees in Early Childhood Education and Primary Education. In this context, Natural Sciences content is integrated into the area of Didactics of Experimental Sciences, which aims to convey scientific knowledge and develop pedagogical competencies for the future teaching of Natural Sciences, through the acquisition of competencies in Biology, Geology, Physics, and Chemistry. This training is crucial for empowering future teachers with tools to design meaningful, rigorous, and contextually relevant learning experiences in scientific literacy (Acevedo, 2010; Pérez-Rodríguez & Baquero-Mendieta, 2025). In this way, students are encouraged to reflect on the epistemological and pedagogical relationships that underpin the teaching of Natural Sciences (Quijano, 2016), fostering positive attitudes toward education in sustainability (Rico et al., 2025).

In summary, the curricular framework of each educational stage in the learning of Natural Science content promotes the development of competency-based learning in students by addressing foundational knowledge composed of content, skills, and abilities that guide the achievement of the objectives specific to each educational level. To achieve this, it is necessary to implement active and collaborative methodological strategies that facilitate the acquisition of competencies.

1.3. Methodological Strategies and Motivation in Natural Science Teaching

The application of strategies is not limited to organizational and curricular aspects but also includes methodological approaches to promote quality education (Bolívar, 2019). Methodological strategies constitute the set of procedures, techniques, and teaching resources that the teacher selects and organizes to facilitate the teaching-learning process, and they must be adapted to the characteristics of the content, context, and students (Medina & Salvador, 2009). These strategies involve an organized set of operations and decisions carried out by the teacher to foster intentional learning among students (Díaz & Hernández, 2010, p. 103). Such strategies should be flexible and contextualized, encouraging student participation as well as promoting autonomous and meaningful learning, in which the nature of knowledge is based on the creation of a constructivist curriculum (Tobón, 2013).

The implementation of active methodologies such as the STEAM approach (Science, Technology, Engineering, Arts, and Mathematics) has emerged as an innovative proposal that integrates different disciplines to promote interdisciplinary, meaningful, and contextualized learning (Wu et al., 2025). Its aim is to improve learning, competencies, and students’ perceptions, as well as to strengthen positive emotions and attitudes toward learning (Mateos-Núñez et al., 2020). Unlike traditional approaches, STEAM connects areas of knowledge to solve real-world problems creatively and collaboratively (Campina et al., 2025), enhancing the development of critical thinking (Chen, 2019), problem-solving skills (Mantei & Kervin, 2025), creativity (Kim et al., 2023), participation (Avendaño-Uribe et al., 2022), and teamwork (Leal, 2025). Furthermore, the inclusion of art reinforces creativity in scientific learning, making content more useful and applicable (VV, 2017).

Other methodologies, such as project-based learning or game-based learning, also foster motivation and collaboration (Brecl et al., 2024), promoting students’ behavioral, cognitive, and emotional development (Y. Chang et al., 2024). Additionally, methodological strategies such as gamification, augmented reality, and the flipped classroom are key for learning Natural Sciences, as they increase motivation, participation, and critical thinking (S. C. Chang & Hwang, 2018; Q. Liu et al., 2024; Lotter & Ramnarain, 2025; Soriano-Sánchez & Jiménez-Vázquez, 2025). Hattie and Yates (2014) further note that practices that reinforce self-esteem, respect individual learning paces, and promote active learning contribute to more positive and enriching educational experiences.

1.4. Current State of the Issue

Among the factors that affect low-quality learning outcomes in science subjects are lack of motivation, negative attitudes toward teachers, low self-esteem, and lack of confidence in one’s own abilities (Dwi & Putu, 2020; Setianingsih et al., 2019; Handayani et al., 2017; Wira Bayu et al., 2023). These aspects not only impact academic performance but can also influence students’ future professional aspirations (Vinni-Laakso et al., 2022). Another significant factor contributing to poor outcomes is the teacher’s approach to structuring instruction when it focuses solely on content coverage without attending to the learning process (Suryani et al., 2019). Pramana and Suarjana (2019) identify low student attention during traditional classes, attributed to a weak connection between the content and the students’ context. In the teaching-learning process, teachers often limit themselves to presenting information related to concepts, failing to capture students’ interest.

Learning Natural Science content should involve the development of critical thinking skills, scientific curiosity, and understanding of the natural world (Ruiz & Guete, 2023). Consequently, academic performance is influenced by multiple factors, with motivation occupying a central role (Banda & Nzabahimana, 2023; Kavčič et al., 2022; Sigmund et al., 2013). An adequate level of student motivation promotes greater engagement, enhances knowledge retention, and supports the development of scientific competencies (Tinungki et al., 2024). However, despite growing attention to the role of motivation in educational contexts, there is still no clear and updated synthesis that integrates the available evidence on its specific impact on academic performance in Natural Sciences across different educational stages and in relation to the methodological strategies employed.

Despite increasing interest in this topic, a gap persists in the literature regarding the relationship between motivation and academic performance in Natural Sciences, highlighting the need to identify, analyze, and organize existing empirical findings. This will guide future research, support the design of more effective pedagogical strategies, and provide a comprehensive understanding of how fostering motivation relates to academic performance in learning these subjects. In particular, through a systematic review of the literature, this study seeks to offer a solid theoretical foundation to identify effective instructional strategies that enhance motivation and increase student academic performance, moving toward inclusive and quality education. Therefore, the general objective of this study is to analyze the influence of motivation on academic performance in the learning of Natural Sciences across different educational stages. From this, the following specific objectives are derived:

(a)

To identify methodological strategies that promote motivation in learning Natural Sciences.

(b)

To examine the main themes and connections influencing learning in Natural Sciences through concurrency analysis.

(c)

To determine, via meta-analytic results, the degree of significance of interventions on motivation in fostering learning.

(d)

To identify instruments used to measure motivation and academic performance in the learning of Natural Sciences.

2. Materials and Methods

2.1. Procedure and Search Strategies

This study is based on a systematic review of the scientific literature. The guidelines proposed by the PRISMA statement were followed, a guide for conducting and reporting systematic reviews and meta-analyses (Page et al., 2021) (Figure 1). The search was conducted in the Web of Science (WoS) and Scopus databases, both part of Elsevier. These are the most prestigious scientific databases, encompassing the well-known JCR and SJR impact indices, respectively. The search was carried out without any temporal restrictions. The search formula used was as follows: ((motivation) AND (academic achievement) AND (Natural Sciences)). A total of 157 documents were retrieved. This formula was applied to the fields “title,” “abstract,” and “keywords” and was adapted to the syntax of each database to ensure compatibility.

The results obtained from each database were as follows: 113 in WoS and 11 in Scopus. In addition, manual checks of the references of the included studies were conducted to identify potentially eligible articles that had not been captured in the initial database searches. Finally, the search was carried out in July 2025.

2.2. Eligibility Criteria

First, the following inclusion criteria were defined: (a) Empirical studies; (b) Research presenting statistical analyses (e.g., correlation) to measure motivation and academic performance in the learning of Natural Sciences; (c) Studies covering different educational stages, from Early Childhood Education to the university level; (d) Higher Education studies conducted within initial teacher education programs (Degrees in Early Childhood Education and Primary Education); and (e) Studies published in English or Spanish.

Conversely, the following exclusion criteria were established:

(a) Studies with restricted access to the publication; (b) Duplicate studies; (c) Conference proceedings, book chapters, or books; (d) Theoretical studies, reviews, or case studies; (e) Research published in languages other than English or Spanish; (f) Studies conducted in university degrees other than initial teacher education; (g) Bibliometric analyses; (h) Studies irrelevant to the research topic (on other subjects related to education or other educational areas); and (i) Research that does not provide instruments for analyzing motivation and academic performance.

2.3. Results Extraction

The selection of studies was carried out jointly by the three authors. To minimize selection bias, rigorous strategies were implemented, including the clear definition of inclusion criteria, the systematic application of search terms across multiple databases, and manual review of references. Additionally, an expert in the field was consulted and supervised the process, ensuring precision and consistency, in accordance with PRISMA guidelines, which recommend the participation of at least two reviewers. The selection process was conducted in several phases:

• Initial selection by title and abstract: All identified records were evaluated by the three authors.
• Full-text assessment: Preselected articles were read in their entirety to apply the inclusion and exclusion criteria.
• Duplicate detection: An Excel spreadsheet was used to organize studies and identify duplicates.

In summary, titles and abstracts of the identified records were reviewed. Preselected records were obtained and read in full. In cases of controversy, a thorough reading of the full text was conducted to apply the remaining conceptual and methodological criteria. Information from records meeting the eligibility criteria was extracted into an Excel file.

The search formula yielded a total of 157 documents, distributed as 113 in WoS and 44 in Scopus. Regarding exclusion criteria, a total of 127 documents were discarded as follows: 4 studies were excluded due to restricted access (criterion a); 21 studies were excluded as duplicates (criterion b); 5 studies were excluded for being conference proceedings, book chapters, or books (criterion c); 13 studies were excluded for being theoretical studies, reviews, or case studies (criterion d); 4 studies were excluded because they were published in languages other than English or Spanish (criterion e), specifically 2 in Chinese and 2 in Russian; 1 study was excluded for being conducted in a university degree other than initial teacher education (criterion f); 1 study was excluded for being a bibliometric analysis (criterion g); 78 studies were excluded for being irrelevant to the research topic (criterion h); and 1 study was excluded based on evaluation criteria (criterion i), which, although analyzing student academic performance and discussing motivation, did not employ any instrument to measure this psychological construct.

Finally, 4 studies met the eligibility criteria and were included in the review. The selection process is summarized in the PRISMA flow diagram (Figure 2).

2.4. Data Extraction

For data extraction, an Excel spreadsheet was designed to include information based on the established inclusion criteria. These were coded following the process described below: author(s); (2) year of publication; (3) study objective; (4) educational stage; (5) sample size (N); (6) country (continent) where the study was conducted; (7) age–mean age (Medad); (8) study design; (9) variables analyzed; (10) statistical analyses used. Using the Excel spreadsheet, the most relevant quantitative and qualitative data from each study were extracted. This process was carried out thoroughly to ensure the highest reliability in data collection.

The procedure was initially performed by two researchers (J.G.S.-S. and R.Q.-L.) and subsequently verified independently by the third researcher (M.S.S.R.) to ensure data accuracy. The results were summarized in a summary table presenting the key characteristics of the included studies, in accordance with the PRISMA statement.

2.5. Data Analysis: Meta-Analysis

First, the “Intervention Review” option was selected to assess the effectiveness of interventions, following a fixed-effect model because it was assumed that the included studies share a common underlying effect and that observed differences among them are due solely to sampling error. The meta-analysis was conducted using the Cochrane Review Manager (RevMan), version 5.4 (Cochrane, London, UK), to evaluate study heterogeneity, effect size, data quality, etc. (Sánchez-Meca & Ato, 1989), as it provides a high level of evidence on the effectiveness of interventions.

To analyze the data, the standardized mean difference statistical method for fixed effects (Fixed Effect Model) was used, assuming that all studies estimate the same true effect. The direction of effect sizes is considered favorable if the results indicate an improvement across the set of interventions, with significance observed through the p-value, where the effectiveness of the overall interventions is considered significant if p < 0.05. Regarding heterogeneity, it is considered high if I² ≥ 75%, moderate between 50% and 75%, and low when I² ≤ 25% (Bölek et al., 2022; Higgins et al., 2003). Due to the limited number of studies, subgroup or sensitivity analyses were not possible.

2.6. Protocol Registration

The protocol for this review was registered in PROSPERO (2025 CRD420251091156), ensuring transparency and traceability of the review process in accordance with best practices recommended by PRISMA. This registration allowed the objectives, methods, and planned analyses to be publicly documented prior to conducting the study.

2.7. Keyword Co-Occurrence Network Analysis

To analyze the co-occurrence of keywords obtained from the literature search, an automated analysis was conducted using VOSviewer, version 1.6.20, a software tool specialized in the construction and visualization of bibliometric networks (Van Eck & Waltman, 2010).

2.8. Risk of Bias Assessment and Methodological Quality of Included Studies

The risk of bias in the studies included in the meta-analysis was assessed by examining the distribution of points in funnel plots, following the guidelines suggested by Higgins et al. (2003). Each included study was evaluated independently, and the results were tabulated in a risk-of-bias matrix. This process was carried out by two researchers, with disagreements resolved by a third researcher.

The Grading of Recommendations Assessment, Development and Evaluation (GRADE) system was used in this study to evaluate the quality or certainty of the evidence obtained for all studies included in the systematic review. This standardized tool allowed for an assessment of the reliability and robustness of the conclusions drawn from the analyzed studies, facilitating an informed interpretation of the results. GRADE classifies the certainty of evidence into four main levels: high, moderate, low, and very low. These categorizations are based on the degree of confidence that the results will remain stable in future research.

This ensured a rigorous evaluation of the robustness of the evidence, contributing to better data-driven decision-making. For the methodological assessment, the criteria comprising the checklist were framed with either a general or a specific focus, in the latter case adapted to the quantitative approach used. Open-ended questions were operationally defined to provide a yes/no response, using the following eight items (Eadie et al., 2018):

• Item 1: Did the study address a clearly focused topic?
• Item 2: Was the cohort recruited with acceptable precision?
• Item 3: Was the outcome measured accurately to minimize bias?
• Item 4: Did the authors identify all important confounding factors?
• Item 5: Did the authors account for confounding factors in the study design and/or analysis?
• Item 6: Was participant follow-up sufficiently complete?
• Item 7: Was participant follow-up sufficiently long?
• Item 8: Were the results precise (e.g., reporting confidence intervals, standard errors, or standard deviations)?

Additionally, Items 9 and 10 were included, established for this type of study by Soriano-Sánchez and Jiménez-Vázquez (2023):

• Item 9: Is there a relationship between the data and the conclusion?
• Item 10: What is the quality of the study design?

The risk-of-bias assessment was conducted by two researchers (J.G.S.-S. and R.Q.-L.), with disagreements resolved in a consensus meeting with a third researcher (M.S.S.R.).

3. Results

3.1. Descriptive Analysis of the Selected Studies

First, interest in this research topic has increased in recent years, as it was not until 2022 that investigations began in this scientific line. Regarding the study objective, all four studies shared the aim of analyzing the level of motivation and academic performance of students in the area of Natural Sciences (

Table 1. Summary of the studies included in the systematic review.

Author(s)	Year of Publication	Objective	Educational Stage	N	Country (Continent)	Age- Mage (SD)	Study Design	Analyzed Variables	Analysis Used
(Banda & Nzabahimana, 2023)	2023	To investigate the impact of using PhET simulations on students’ motivation and academic performance in the learning of oscillations and waves	Secondary Education	280 (44.60% girls)	Malawi (Africa)	17.50 (1.42)	Quasi-experimental	a and b	Mean differences
(Kavčič et al., 2022)	2022	To investigate the use of self-regulation strategies in individual learning of scientific hypertexts and their relationship with academic performance	Secondary Education	443 (224 girls and 219 boys)	Slovenia (Europe)	14.38 (0.40)	Correlational study with a cross-sectional approach	a, b, c and e	Correlations
(R. Liu et al., 2022)	2022	To develop a series of lessons in immersive virtual reality (IVR) and examine the effects of these lessons on learning outcomes	Primary Education	362 (183 boys and 179 girls)	China (Asia)	9.5	Mixed methods design	a, b and d	Mean differences
(Xie et al., 2023)	2023	To investigate whether learning science through immersive virtual reality improves learning outcomes compared to watching videos, and whether corrective feedback is more effective than explanatory feedback in virtual reality environments	Secondary Education	47 (G1 = 17; G2 = 15; G3 = 15)	China (Asia)	13.8	Randomized controlled pilot trial	a, f, g and h	Mean differences and correlations

Note. PhET = Physics Education Technology; G1 = Learning with instructional videos; G2 = Intervention group with VR learning and corrective feedback; G3 = Intervention group with VR learning and explanatory feedback; a = Motivation; b = Academic performance; c = Learning strategies; d = Cognitive load; e = Cognitive learning strategies; f = Academic self-efficacy; g = Learning satisfaction; h = Cognitive load.

). In terms of educational stage, one study was conducted in Primary Education (R. Liu et al., 2022) and three in Secondary Education (Banda & Nzabahimana, 2023; Kavčič et al., 2022; Xie et al., 2023). Regarding sample size, the number of participants ranged from N = 47 (Xie et al., 2023) to N = 443 (Kavčič et al., 2022), with all samples including participants of both male and female genders. The studies were carried out in countries such as Malawi, located in Africa (Banda & Nzabahimana, 2023); Slovenia, in Europe (Kavčič et al., 2022); and China, in Asia (R. Liu et al., 2022; Xie et al., 2023). Participant ages ranged from a mean of M = 9.5 years (R. Liu et al., 2022) to M = 17.5 years (Banda & Nzabahimana, 2023).

Regarding study design, one study employed a quasi-experimental design (Banda & Nzabahimana, 2023), a second study used a correlational design with a cross-sectional approach (Kavčič et al., 2022), a third study adopted a mixed-methods design (R. Liu et al., 2022), and a fourth study conducted a pilot randomized controlled trial (Xie et al., 2023). Finally, to analyze motivation and academic performance, two studies performed statistical analyses of mean differences (Banda & Nzabahimana, 2023; R. Liu et al., 2022), another conducted correlations (Kavčič et al., 2022), and a final study employed both mean differences and correlations (Xie et al., 2023).

3.2. Measures Used to Assess Motivation and Academic Performance

First, to assess motivation, R. Liu et al. (2022) used the translated version of the Motivation to Learn Science Questionnaire by Barak et al. (2011), consisting of 20 items rated on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree). In contrast, Banda and Nzabahimana (2023) used the Motivation Questionnaire adapted from Tuan et al. (2005). This version was modified by replacing the word “science” with “oscillations and waves” to measure students’ motivation in that specific content area. It consisted of 38 items measured on a 5-point Likert scale (1 = strongly disagree, 5 = strongly agree) and included subscales to assess self-efficacy, active learning strategies, value of learning physics, performance goals, achievement goals, stimulation of the learning environment, and attitudes toward using PhET simulations (Physics Education Technology) in teaching and learning. The Cronbach’s alpha was α = 0.85.

Other authors, such as Kavčič et al. (2022), developed a measurement scale introducing questions to assess motivation on a 7-point scale (1 = not at all, 7 = very much). Example items included “Total time spent watching the content from start to finish” and “Time spent watching the video.” Cronbach’s alpha values for both tests exceeded 0.8, indicating good internal consistency. When designing the scale, the authors paid special attention to factors affecting data quality, such as scale length, clarity of questions, and reference points. Xie et al. (2023), on the other hand, evaluated motivation in learning using the Self-Regulation Questionnaire developed by Deci et al. (1994).

Regarding the assessment of academic performance, R. Liu et al. (2022) developed a content test to evaluate academic achievement across three science lessons. The test included 25 items (17 multiple-choice and 8 true/false) assessing fundamental knowledge of the scientific content covered in the curriculum prior to the study, in order to determine the level of equivalence in general science knowledge before participation (maximum score = 50). For example, one true/false question was: “Earthworms can live in water for a long time.” Cronbach’s alpha values for both tests exceeded 0.8, indicating good internal consistency.

Similarly, Banda and Nzabahimana (2023) adapted questions from previous exams. The tests consisted of 16 structured items (open-ended questions) on oscillations and waves, with a Cronbach’s alpha of α = 0.78, within the recommended range (Cohen, 1988). In contrast, Kavčič et al. (2022) assessed academic performance through the average grades obtained in science subjects, including Biology, Physics, and Chemistry. Finally, Xie et al. (2023) evaluated academic performance through learning self-efficacy, using the Motivated Strategies for Learning Questionnaire by Lee et al. (2010).

3.3. Synthesis of the Evidence Found

The studies agree that motivation is a key factor for academic performance in students learning Natural Sciences. In Primary and Secondary Education, different studies employed immersive virtual reality (IVR). This is a specific type of virtual reality designed to make the user feel completely “inside” the virtual environment. To achieve this, devices such as VR headsets or goggles (e.g., Oculus Quest, HTC Vive), headphones, haptic gloves, motion sensors, and others are used. The idea is to isolate the user from the real world and fully immerse them in the virtual world, creating a strong sense of presence (Mills & Brown, 2021). In Primary Education, R. Liu et al. (2022) showed that an IVR-based classroom, conducted over 5 weeks with three 45-min science lessons per week, significantly improved students’ academic performance and motivation while reducing cognitive load. They used an IVR system with a virtual assistant named Degree, an artificial robot that guided students with immediate feedback through explanations, formative questions, and real-time corrections. Through interactive 3D scenes, students explored topics such as the climate, and the respiratory and digestive systems. Degree supported learning by providing corrections and reinforcing key knowledge in real time.

Similarly, in Secondary Education, Xie et al. (2023) also used IVR, creating an IVR environment with VeryEngine and Unity 3D. This enabled interactive simulations in a virtual laboratory on series electrical circuits. Before the experiment, participants in the experimental group received 30 min of training on using virtual reality devices, including how to collect objectives, operate with rays, and enter numbers using virtual keyboards. Participants were also encouraged to ask questions and clarify doubts during the allotted time. The first 2 min were dedicated to an introduction to the learning topic, followed by guided task completion with feedback provided both in audio and text for specific operations. The intervention groups experienced IVR for 25 min, while control group participants watched a video for the same duration.

In Secondary Education, Banda and Nzabahimana (2023) evaluated the impact of PhET simulations on students’ motivation and academic performance in the teaching of oscillations and waves over six weeks. Learning with these simulations, which are based on interactive environments, facilitated the understanding of complex scientific phenomena through virtual exploration, experimentation, and active learning. In the experimental group, students manipulated variables, formulated and tested hypotheses, and visualized abstract concepts, which promoted higher motivation and better academic performance compared to the control group, which used traditional active methods without simulations. Thus, the study confirmed the effectiveness of the PhET approach in enhancing science learning through direct interaction with virtual models, immediate feedback, and active knowledge construction (Banda & Nzabahimana, 2023).

On the other hand, Kavčič et al. (2022) developed an online learning environment that recorded students’ activities while navigating a hypertext online. This study was conducted in two sessions separated by a two-month interval. Interactions such as clicks on the glossary to obtain definitions and the use of controls to play videos were recorded, with timestamps for all events, including session start and end. The learning unit, “Eyes and Color Perception,” from the Slovenian digital book Chemistry in Life, covered concepts such as the structure of the eye, visual perception, the role of rods and cones, related chemical processes, as well as topics like light, photons, and animal mimicry. The hypertext contained 1073 words plus 226 in the glossary to clarify concepts, divided into six chapters with text, 20 images, five chemical diagrams, and a video clip on β-carotene solubility. Students progressed chapter by chapter and could review the entire content at the end, with a link to the glossary always available in the top menu bar for resolving questions.

3.4. Summary of Meta-Analytic Results

The meta-analytic findings on the promotion of motivation in academic performance in Natural Sciences learning, based on a total of k = 2 studies, included 314 participants in the experimental group and 328 in the control group. The results indicated moderate heterogeneity (I² = 49%), with an effect size of SMD = 9.90 and a 95% confidence interval (CI) [8.20, 11.60]. Regarding significance, the results showed a significant effect (p < 0.0001) (Figure 3).

For the interpretation of the meta-analysis, the symbols have the following meanings:

Figure 3. Forest plot of effect size for the set of interventions. 🟩 Green squares: represent the effect size (mean difference) of each individual study. The size of the square reflects the weight of the study in the meta-analysis. — Black horizontal lines crossing the squares: represent the 95% confidence interval (95% CI) for each study. ◆ Black diamond at the end of the plot: represents the overall combined effect of all included studies. The width of the diamond shows the 95% confidence interval for the combined effect. Note. Tau² = estimate of effect size variance; Chi² = test for heterogeneity; df = degrees of freedom; p = statistically significant heterogeneity; green dots = effect size of each study; black diamond on each variable = effect size of the set of studies; final black diamond = effect size of the subgroup. Meta-analysis conducted with the studies presented by Banda and Nzabahimana (2023) and R. Liu et al. (2022). In this case, since all the squares and the diamond were observed to the right of the line of no effect (0), the results favored the experimental group.

Figure 3. Forest plot of effect size for the set of interventions. 🟩 Green squares: represent the effect size (mean difference) of each individual study. The size of the square reflects the weight of the study in the meta-analysis. — Black horizontal lines crossing the squares: represent the 95% confidence interval (95% CI) for each study. ◆ Black diamond at the end of the plot: represents the overall combined effect of all included studies. The width of the diamond shows the 95% confidence interval for the combined effect. Note. Tau² = estimate of effect size variance; Chi² = test for heterogeneity; df = degrees of freedom; p = statistically significant heterogeneity; green dots = effect size of each study; black diamond on each variable = effect size of the set of studies; final black diamond = effect size of the subgroup. Meta-analysis conducted with the studies presented by Banda and Nzabahimana (2023) and R. Liu et al. (2022). In this case, since all the squares and the diamond were observed to the right of the line of no effect (0), the results favored the experimental group.

3.5. Risk of Bias and Quality Criteria of the Included Studies

Once the studies for the meta-analysis were selected, the reliability of the results was evaluated through a risk of bias assessment, using the Cochrane risk of bias tool and the Funnel Plot (a graph used to check for the presence of publication bias). This allowed us to observe the risk of bias in the psychological parameters studied. Consequently, the risk of bias was low in the two included studies, as both studies fell within the triangular area provided by this tool (Figure 4).

After evaluating the risk of bias according to the established items, it was observed that the different studies presented a low risk of bias. The four quantitative investigations met all 8/8 criteria proposed by Eadie et al. (2018) for the quality of cohort study designs, such as subject recruitment, minimization of bias, control of confounding factors, and follow-up. Furthermore, regarding items 9 and 10, introduced by Soriano-Sánchez and Jiménez-Vázquez (2023), most of the studies demonstrated high quality in terms of study design (

Table 2. Risk of bias in the included studies.

Study	Item 1	Item 2	Item 3	Item 4	Item 5	Item 6	Item 7	Item 8	Item 9	Item 10
Banda and Nzabahimana (2023)	✓	✓	✓	✓	✓	✓	✓	✓	✓	L
Kavčič et al. (2022)	✓	✓	✓	✓	✓	✓	✓	✓	✓	L
R. Liu et al. (2022)	✓	✓	✓	✓	✓	✓	✓	✓	✓	L
Xie et al. (2023)	✓	✓	✓	✓	✓	✓	✓	✓	✓	L

Note. L = Low; ✓ = Explicitly present.

).

3.6. Results of the Keyword Co-Occurrence Network Analysis

The results of the thematic co-occurrence analysis allowed for the visualization of a rich and interconnected conceptual network (Figure 5). Through semantic mapping, three main clusters or communities were identified, each representing distinct research focuses related to learning, educational interventions, and assessment in educational contexts. In particular, the following clusters were identified:

Red Cluster: “Intervention Dimension and Academic Performance”: This group, centered around terms such as student, group, academic performance, intervention, effect, change, and knowledge, reflected a strong focus on pedagogical interventions and their impact on students. A high interconnectivity was observed between the student node and concepts such as group, effect, and academic performance, reinforcing the emphasis on the student as the central figure in educational improvement processes. Additionally, terms like intervention and change stood out, highlighting the importance of analyzing how changes introduced in the classroom affect academic performance and knowledge acquisition.

Green Cluster: “Motivational and Disciplinary Dimension”: This cluster grouped terms such as motivation, development, impact, achievement, type, and natural science. Here, the network focused on how motivation and academic achievement are deeply interconnected, both with each other and with variables such as the type of intervention or the disciplinary context, highlighting their essential role in sustained learning.

Blue Cluster: “Methodological and Assessment Dimension”: Formed by terms such as importance, implication, relationship and measure. The prominent presence of nodes like measure and relationship indicated a focus on grounding the findings and ensuring the relevance and significance of the results.

Overall, the visual analysis demonstrates how research on learning and academic performance is organized around three main axes: interventions targeting the student, motivational and disciplinary foundations, and methodological rigor in the assessment of results.

The student node, located at the center of the network and interconnected with all clusters, illustrates its central role in designing effective educational strategies.

4. Discussion

The results obtained have allowed the main objective of this study to be achieved, highlighting the role that motivation plays in students’ academic performance in the learning of Natural Sciences. In this regard, Trigueros et al. (2022) indicate that higher motivation helps reduce emotional exhaustion and disinterest in academic tasks, thereby generating a more positive educational experience. This finding reinforces the idea that fostering motivation not only improves students’ disposition toward learning but also contributes to their emotional well-being, a particularly relevant aspect in the context of Natural Sciences, where a high degree of engagement and active curiosity is required.

Zhang et al. (2021) introduce an innovative perspective by considering temporal orientation as a key variable for understanding student motivation. Their findings show that when there is coherence between a student’s dispositional temporal orientation (i.e., their natural way of relating to time) and the orientation promoted by the educational environment (focused on long-term goals), motivation is strengthened, positively affecting academic performance. This highlights the need to adapt educational strategies not only in terms of content but also considering students’ temporal and organizational profiles, fostering learning environments that enhance planning, self-regulation, and personal projection.

Other authors, such as Chan and Lam (2023), support the effectiveness of active learning methodologies, such as the Jigsaw method, a cooperative learning strategy designed for students to work together, teach each other, and actively learn in groups. This methodological strategy, by promoting active participation, collaboration, and shared responsibility, stimulates deeper and sustained motivation, which translates into better content comprehension and positive development of social skills. This suggests that learning is enhanced when students become active protagonists in constructing knowledge.

In response to the first specific research objective, the results obtained (Figure 6) indicate that the application of immersive virtual reality (IVR) leads to increased motivation and academic performance, as well as reduced cognitive load in Natural Sciences lessons, providing a framework for integrating this methodological strategy into existing classrooms (R. Liu et al., 2022). In this regard, Hsu (2024) suggests that such strategies stimulate students’ curiosity and enthusiasm for learning, especially in those who may feel bored or disengaged in traditional classroom settings. IVR educational materials can increase student engagement through interactivity and immersion, enabling experiences with live scientific experiments or recreations of distant scenes.

Similarly, PhET simulation-based learning has been shown to improve both academic performance and motivation, such as in the study of oscillations and waves. The visualizations and instructional resources provided by these simulations facilitate the understanding of complex content, promoting more effective learning and greater student engagement (Banda & Nzabahimana, 2023). Other research also emphasizes the importance of integrating pedagogical strategies based on experimentation and technological tools to optimize the learning of Natural Sciences and promote a more dynamic approach (Fuenmayor et al., 2025).

On the other hand, the use of hypertext in the acquisition of scientific content is associated with higher student motivation and academic performance, as well as increased use of cognitive and metacognitive strategies in learning, such as: (a) silently repeating content while learning new material; (b) thinking about how to apply the content to daily life; (c) reflecting on whether attention has been paid to the topic during learning; (d) looking up new or unfamiliar words; and (e) checking whether learning has been fully achieved (Kavčič et al., 2022).

This suggests that hypertext, through videos and term glossaries, promotes more meaningful learning compared to traditional listening or reading of content. Moreover, it could serve as an innovative educational strategy, allowing the personalization of learning and fostering student autonomy by giving learners the ability to choose their own learning paths. It also encourages active knowledge construction, as students integrate diverse sources and perspectives, and promotes the development of critical thinking through comparison and validation of information.

From the perspective of Self-Determination Theory proposed by Deci and Ryan (1985), it can be inferred that students feel more motivated, engaged, and learn better when they have some degree of choice and control (autonomy), feel capable of achieving their goals (competence), and perceive a supportive environment and sense of belonging (relatedness). According to Pintrich and De Groot (1990), academic motivation is closely related to self-efficacy and the perceived relevance of content—factors that can be enhanced by demonstrating the real-life and practical applicability of science. Similarly, Schunk et al. (2008) highlight that exploration- and experimentation-based teaching fosters curiosity and intrinsic interest, which are key to sustaining long-term motivation. Thus, to achieve truly competency-based learning in Natural Sciences, an effective approach could be designing activities that promote choice, provide positive feedback, and stimulate collaborative work.

Science learning provides an ideal framework for developing critical thinking and a deep understanding of the environment, essential elements to awaken student motivation and improve academic performance (Ospankulova et al., 2024). Therefore, integrating teaching strategies that connect scientific content with students’ prior experiences in meaningful contexts could contribute not only to improved motivation but also to better performance. Within this framework, initial teacher training becomes crucial, as it should foster a critical and sustainability-oriented citizenship through meaningful educational experiences. An example is interventions in eco-didactic gardens, which enhance environmental awareness, attitudes toward Education for Sustainability, and students’ connection with nature in Early Childhood and Primary Education degrees (Rico et al., 2025).

Regarding the second specific objective, the co-occurrence analysis results reflect that Natural Sciences learning revolves around three clearly differentiated thematic cores, as shown in the semantic network. The red cluster emphasizes the central role of the student and the importance of pedagogical interventions aimed at improving academic performance through collaborative and meaningful experiences. The green cluster highlights the relationship between motivation, achievement, and skill development within the specific context of Natural Sciences, reinforcing this context as particularly favorable for active and innovative practices. Finally, the blue cluster shows a transversal concern for rigorous evaluation and the relevance of educational practices. Consequently, student participation is indirectly represented through the nodes student and group, which sit at the center of the network and establish multiple connections with motivation and achievement, suggesting that motivation and academic achievement are strongly linked to student engagement and group work. This interlinked structure confirms that beyond content transmission, current pedagogical practices should focus on strengthening autonomy and active engagement, integrating critical reflection, innovative methodological strategies, and technological resources that enhance meaningful learning in scientific literacy.

Concerning the third specific objective, meta-analytic results indicate a favorable effect for experimental groups, demonstrating that the application of active methodological strategies—such as immersive virtual reality (IVR) and hypertext—enhances motivation and academic performance in Natural Sciences learning. Regarding risk of bias, the funnel plot shows its absence. Additionally, based on responses to the evaluation items for the four studies included in the systematic review, the absence of bias is further confirmed.

For the fourth specific objective, among the instruments used to assess motivation are the Motivation to Learn Science Questionnaire (Barak et al., 2011) and the adapted Motivation Questionnaire (Tuan et al., 2005). To evaluate academic performance, R. Liu et al. (2022) and Banda and Nzabahimana (2023) designed content-based science tests. In contrast, Kavčič et al. (2022) measured academic performance using average grades in Biology, Physics, and Chemistry. Xie et al. (2023) assessed academic performance through learning self-efficacy using the Motivated Strategies for Learning Questionnaire (Lee et al., 2010). This underscores the importance of using robust evaluation procedures and instruments to assess competency acquisition and reflect on teaching practice to ensure quality education.

On the other hand, although the present study highlights the role of motivation in learning Natural Sciences, it is also essential to consider the social factors that influence these aspects. Research has shown that family support is closely related to students’ motivation; for example, a recent study found that students who receive parental support are influenced by adolescents’ motivational beliefs (Starr et al., 2025). Moreover, Bandura’s Social Learning Theory (Bandura, 1977) suggests that learning occurs through observation, imitation, and modeling, influenced by factors such as attention, motivation, attitudes, and emotions. This perspective emphasizes the importance of the social contexts in which students interact and learn, implying that social experiences can significantly affect motivation to learn.

Overall, this study has consolidated a robust theoretical framework that facilitates the identification of teaching practices aimed at fostering student motivation, a key factor in enhancing learning and academic achievement. Findings highlight the importance of inclusive and meaningful educational approaches, which individualize and personalize instruction to ensure quality. Moreover, results show that motivation not only acts as a catalyst for academic performance but also influences students’ willingness to actively engage in learning. Therefore, it is crucial for teachers across educational stages to incorporate the identified methodological strategies wherever possible to promote intrinsic motivation and create a stimulating learning environment. In conclusion, based on the integrated analysis of the reviewed studies, academic performance depends not only on cognitive factors but is profoundly influenced by motivational, emotional, and contextual dimensions.

4.1. Limitations of the Present Study

However, despite the results obtained, this study is not without limitations. The main limitation may lie in the databases consulted, which could have inadvertently led to the exclusion of relevant studies published in other electronic resources. Additionally, there is a scarcity of research reporting pretest–posttest mean differences, which has prevented conducting a meta-analysis with a larger number of studies. Another limitation is the lack of studies conducted in Early Childhood Education and at the university level.

4.2. Future Research Directions

The relationship between motivation and academic performance could be further explored through studies that integrate different methodological and contextual approaches across the various educational stages (from Early Childhood Education to Higher Education). This could not only enrich theoretical knowledge but also provide practical tools to improve the quality of education in the learning of Natural Sciences content. Additionally, the findings of this study highlight the need for future intervention research to clearly report results from pretest and posttest phases, in order to facilitate the conduct of meta-analyses that provide more robust support for the accumulated empirical evidence.

4.3. Practical Implications of the Results

The results obtained offer relevant practical implications. Based on the various studies integrated into this work, the importance of implementing innovative methodological strategies in teaching practice can be inferred, particularly through the use of digital tools. These strategies should address the content, attitudes, skills, and learning strategies that foster competency-based learning in Natural Sciences, aiming to face the challenges of the 21st century. For this purpose, it is essential that both in-service teachers and those in initial teacher education are knowledgeable about the use of these tools, and that educational institutions commit to continuous professional development, the provision of adequate technological resources, and the creation of collaborative environments that facilitate the effective integration of these technologies into the teaching–learning process (Soriano-Sánchez & Jiménez-Vázquez, 2025).

5. Conclusions

This study highlights the decisive influence of motivation on students’ academic performance in Natural Sciences, demonstrating that a high level of motivation not only enhances achievement but also positively impacts emotional well-being. This deep connection between learning and well-being invites a rethinking of teaching practice from a more holistic and human-centered perspective. In this context, the work provides relevant implications for initial teacher education, emphasizing the urgency of transforming traditional approaches toward ones that acknowledge the emotional dimension of learning and promote active methodologies capable of stimulating students’ interest, curiosity, and engagement in their own learning process.

The implementation of methodological strategies such as Hypertext, PhET simulations, and Immersive Virtual Reality (IVR) has proven effective in promoting deep and sustained motivation as well as academic performance, while fostering participation, collaboration, and social skill development. This underscores the need for student-centered learning that enhances autonomy and agency to achieve meaningful, competency-based learning.

The co-occurrence analysis shows that learning in Natural Sciences is organized around three key axes: the student’s central role, the relationship between motivation and achievement, and the importance of rigorous evaluation. The semantic network reflects how student engagement and participation, especially in collaborative contexts, are crucial for fostering meaningful learning. These findings reinforce the need to adopt innovative pedagogical approaches that promote autonomy, critical reflection, and the use of digital tools in scientific literacy.

The meta-analysis shows a consistent positive effect of the interventions, with low heterogeneity and low risk of bias. Additionally, the co-occurrence analysis highlights that motivation plays a central role in learning Natural Sciences, evidencing a significant integration between theory and practice and establishing academic performance as a cross-cutting concept. Among the instruments used to measure motivation is the Motivation to Learn Science Questionnaire, while academic performance was assessed using the Motivated Strategies for Learning Questionnaire.

Ultimately, it is essential to understand that academic performance is deeply influenced by motivational, emotional, and contextual factors, which calls for a reconsideration of traditional teaching methods. Training teachers to integrate innovative, student-centered methodological strategies not only enhances engagement and autonomy in learning but also prepares educators to more effectively support the competency development of their students. Therefore, promoting a more human, active, and reflective education is an indispensable step toward achieving a more meaningful and transformative teaching and learning process.