Physical activity and academic performance in adolescents: a systematic review of intensity, frequency, and mediating factors

Article information

J Exerc Rehabil Vol. 22, No. 1, 3-8, February, 2026

Publication date (electronic) : 2026 February 23

doi : https://doi.org/10.12965/jer.2550854.427

Haemi Jee

Department of Physical Therapy, Namseoul University, Cheonan, Korea

^*Corresponding author: Haemi Jee, https://orcid.org/0000-0002-9066-1472, Department of Physical Therapy, Namseoul University, 91 Daehak-ro, Seonghwan-eup, Sebuk-gu, Cheonan 31020, Korea, Email: wlgpal9@gmail.com

Received 2025 December 20; Revised 2026 January 3; Accepted 2026 January 6.

Abstract

INTRODUCTION

Adolescence is widely recognized as a critical period of human development, encompassing rapid biological, psychological, and behavioral changes (Mastorci et al., 2024). Prior research suggests that physical activity (PA) may enhance cognitive processes relevant to learning, such as attention, executive function, and processing speed, potentially leading to better academic outcomes in school-aged youth. Systematic evidence indicates that chronic PA interventions can yield small-to-medium positive effects on academic achievement, though results are heterogeneous and moderated by factors including activity type, duration, and intensity (Barbosa et al., 2020).

Meta-analytic evidence further demonstrates that both acute and chronic PA interventions are associated with preferable cognitive outcomes such as attention, working memory, and language skills in adolescents and young adults, which are implicated in school performance (Haverkamp et al., 2020). School-based PA programs, particularly those of moderate intensity and sufficient duration, have been shown to benefit overall academic achievement and specific subjects like mathematics (He et al., 2025).

However, associations between PA and academic performance (AP) are not uniformly positive, and evidence suggests that optimal academic benefits may depend on balanced dose and context, with higher volumes or intensities not always conferring greater effects (Barbosa et al., 2020). Additionally, physiological mechanisms, such as neurotrophic modulation, may underlie the influence of PA on adolescent brain development; however, evidence in this population remains limited (Haverkamp et al., 2020).

Despite strong evidence supporting public health recommendations that advocate daily moderate-to-vigorous PA (MVPA) for optimal physical and mental development, a decline in PA levels across adolescence has been consistently reported (Husøy et al., 2024). To promote PA in adulthood, it is essential to understand PA patterns across different stages of growth and development (Bélanger et al., 2025).

Given these mixed findings and the increased focus on adolescent populations in recent research (2024–2025), a systematic synthesis of current empirical studies is needed to clarify the nature of the PA–AP relationship, including activity characteristics, developmental moderators, and potential mediating pathways. Therefore, this study aims to systematically synthesize recent empirical evidence on the relationship between PA and AP among adolescents, with particular attention to activity intensity and frequency, developmental stage, and cognitive and psychosocial mediating mechanisms.

MATERIALS AND METHODS

This systematic review was conducted and reported in accordance with the Preferred Reporting Items For Systemic Review and Meta-analyses (PRISMA) 2020 statement, which is the current guideline for reporting systematic reviews and meta-analyses in health and rehabilitation sciences (Forero et al., 2025).

Eligibility criteria were defined as priori. Studies were included if they met all the following: (1) peer-reviewed original research; (2) examined the association PA and AP; (3) included healthy adolescent participants; and (4) employed quantitative study designs such as randomized controlled trials, quasi-experimental designs, longitudinal cohort studies, or cross-sectional studies. Exclusion criteria were: (a) nonoriginal research (reviews, conference abstracts, editorials), (b) studies primarily involving clinical populations with medical conditions, (c) studies not focusing on PA or AP as principal measures, and (d) studies requiring external devices or apps not directly related to standardized PA assessment.

A systematic literature search was conducted in PubMed, ScienceDirect, and Google Scholar to identify potentially relevant studies published from January 2024 through December 2025. The search strategy combined controlled vocabulary and free-text terms related to PA, AP, adolescents, and similar constructs using appropriate Boolean operators. The final search was completed on December 15th, 2025.

All records retrieved were imported and duplicates were removed. Two independent reviewers screened titles and abstracts against the eligibility criteria. Articles deemed potentially eligible were retrieved in full text and assessed independently by the same reviewers. Discrepancies were resolved by discussion and consensus. The study selection process, including numbers of articles at each screening stage and reasons for exclusion, is presented in the PRISMA 2020 flow diagram (Fig. 1). Data were also extracted independently and extracted information included: authors, year of publication, country, study design, sample size and characteristics, PA measurement methods, AP outcomes, analytic approach, and key results. When studies included subjects beyond adolescent age ranges, data specific to the adolescent subgroup were extracted.

Fig. 1

Flow diagram of evaluated studies.

Although a formal risk of bias tool was not applied, bias considerations were incorporated into the interpretation of findings. Key potential biases assessed qualitatively included measurement approach for PA (objective vs. self-report), selection bias, confounding control, and analytic rigor. These considerations are discussed in the Discussion section. Given heterogeneity in outcome measures, PA assessment methods, and study designs, quantitative meta-analysis was not appropriate. Findings were synthesized narratively. Studies were grouped by PA characteristics and AP outcome types, and patterns in associations were summarized descriptively (Table 1).

Table 1

Summary of studies on physical activity and academic performance

RESULTS

The systematic search across PubMed, ScienceDirect, and Google Scholar identified 825 articles, of which 816 were excluded after screening, resulting in 9 eligible studies published between 2024 and 2025. Most excluded articles focused on clinical populations or outcomes, such as depression, anxiety, loneliness, sleep, weight, suicide, attention deficit hyperactive disorder, or autism, while others addressed application usability, validation, or technical development rather than PA outcomes. Studies involving younger children or adults were excluded to focus on healthy adolescents. The included articles were systematically organized and summarized in Table 1. This table presents key study characteristics, including first author, country/region of the study conducted, sample size, study design, PA exposure, academic outcome, statistical analysis, and a summary of key findings.

Alfaro-Chaverri et al. (2025) examined students from elementary to high school, and only adolescent data were included in this review. MVPA and vigorous PA (VPA) decreased with grade level, while higher levels of in-school and total PA were associated with greater academic success. Grade point average (GPA) increased by 0.44 points per 30 min of MVPA and 0.61 points per 10 min of VPA, indicating a modest-to-moderate effect of PA intensity on academic outcomes (Alfaro-Chaverri et al., 2025). Ali et al. (2025) investigated associations among PA, motor skills (vertical jump, handgrip strength, sit-and-reach flexibility, and 60-sec push-ups), and AP by grade level. PA was significantly associated with motor skills, particularly vertical jump and handgrip strength, which in turn correlated with academic achievement.

Chiluwal et al. (2024) compared the previous term’s average grades with PA types (running, walking, team sports, and yoga). Running and walking were positively associated with AP, whereas team sports and yoga showed inconsistent associations. PA intensity was negatively correlated with AP (r=−0.392, P=0.029), suggesting that excessive intensity may impair academic outcomes. Ilić et al. (2024) used the IPAQ to assess walking, moderate, and vigorous PA among 836 high school students, and measured AP with a 5-point Likert scale. Vigorous (r=0.127), moderate (r=0.098), and light (r=0.080) PA, as well as total METs (r=0.181), were significantly correlated with AP (P≤0.005).

Jee (2025) analyzed the Korea Youth Risk Behavior Survey, examining perceived PA frequency and AP across grades 6–12. Associations were stronger among middle school students, with higher PA linked to better AP, while high school students showed inconsistency, potentially reflecting increased academic demands. Martín-Bozas et al. (2025) assessed PA using the PAQ-A and evaluated executive function (attention, inhibition, emotional regulation) using the SENA questionnaire. AP was measured via last-term grades in core subjects. The attention-deficit model explained 15% and the inhibitory-deficit model 26.1% of AP variance, with both significantly mediated by PA (P<0.01). Attention deficit (β= −0.65, P<0.01) and inhibitory deficit (β=−0.42, P<0.01) negatively predicted AP (Martín-Bozas et al., 2025).

Pan et al. (2025) examined PA intensity, duration, and frequency using the Physical Activity Rating Scale-3 scale and assessed AP in Chinese, mathematics, and English. PA had a significant direct effect on AP (β=0.24), while indirect effects via self-concept and physical/mental health accounted for 14.86% and 8.86% of the total effect, respectively, highlighting multiple mediating pathways. Roig-Hierro et al. (2025) evaluated PA frequency and intensity and their effects on AP and perceived competence. Moderate PA (2–5 days/wk) was associated with preferable AP, whereas >5 days/wk was linked to reduced performance. Moderate-intensity PA correlated significantly with outcomes in mathematics, science, and language, while high-intensity PA showed only partial associations (Roig-Hierro et al., 2025).

Finally, Shin et al. (2024) conducted a longitudinal study assessing MVPA during physical education classes over three years. Direct and attention-mediated indirect effects of MVPA were significant among female students (β=0.06 and 0.18, P<0.05), emphasizing the role of executive function in long-term academic outcomes (Shin et al., 2024).

DISCUSSION

This systematic review synthesized evidence from 9 studies published between 2024 and 2025 examining the relationship between PA and AP among healthy adolescents. Overall, the findings indicate a consistent, albeit nuanced, association between PA and AP, with effects varying according to activity intensity, frequency, developmental stage, and mediating cognitive and psychosocial factors.

While MVPA and VPA are generally associated with better academic outcomes, effect sizes were modest, indicating that PA is one of multiple contributors to AP. Dose-response for both MVPA and VPA were positively associated with GPA, with stronger effects observed for VPA (Alfaro-Chaverri et al., 2025; Donnelly et al., 2016; Renke et al., 2022; Singh et al., 2025). However, the magnitude of these effects remained modest, indicating that PA is one of multiple contributors to academic success rather than a sole determinant.

One of the studies highlighted the importance of motor competence and physical fitness as correlates of AP. Ali et al. (2025) found significant associations between PA, motor skills such as vertical jump and handgrip strength, and academic outcomes. These findings support contemporary frameworks emphasizing physical literacy and motor competence as foundational elements linking PA to cognitive and academic development (Moon et al., 2024). Better motor skills may reflect more frequent engagement in structured or unstructured PA, which in turn supports attention, classroom behavior, and learning readiness (Domínguez-Martín et al., 2024; Moon et al., 2024).

The association between PA and AP was not uniformly positive. Several studies reported that excessively high PA frequency or intensity was linked to weaker or even negative academic outcomes (Chiluwal et al., 2024; Roig-Hierro et al., 2025). These findings indicate a curvilinear relationship, whereby moderate PA optimizes academic outcomes, while excessive demands may compete with study time, increase fatigue, or elevate stress. Longitudinal and meta-analytic evidence likewise cautions against assuming linear academic benefits with increasing PA volume (Muntaner-Mas et al., 2021; Owen et al., 2022).

Clearer positive associations were observed in middle school students, whereas findings were inconsistent among high school students, likely reflecting heightened academic pressure and reduced discretionary time in late adolescence (Jee, 2023, 2025). This aligns with prior evidence suggesting that PA benefits are more pronounced during early and middle adolescence, when executive function development is particularly sensitive to behavioral inputs (Teng et al., 2025).

Mechanistic studies indicate that PA supports AP primarily through indirect pathways. Attention and inhibitory control significantly mediated the PA–AP relationship (Martín-Bozas et al., 2025), while self-concept and physical and mental health also contributed to indirect effects (Pan et al., 2025). These results are consistent with models proposing that PA is related to favorable academic outcomes by improving executive function, mental health, and classroom engagement (Diamond and Ling, 2019; Fernandes et al., 2024), with attention emerging as a particularly important mediator, especially among female students (Shin et al., 2024).

Several limitations of this review should be acknowledged. First, interpretation of the findings should consider key methodological limitations, including reliance on self-reported PA measures, the predominance of cross-sectional study designs, and heterogeneity in AP outcomes. Although a formal risk of bias assessment tool was not applied, qualitative consideration of selection bias, confounding, and analytic rigor indicates that the results should be interpreted with caution. In addition, many studies employed ambiguous or inconsistent PA assessment methods, which may have contributed to variability in reported associations.

Effect sizes across studies were generally modest, suggesting that PA represents only one of multiple contributors to AP. Few studies explicitly examined cognitive or psychosocial mediating mechanisms, and several analyses relied on previously collected datasets, potentially limiting their contemporary relevance. In the present review, mediating factors were examined as a secondary focus; consequently, many studies primarily centered on mediators rather than the direct PA–AP relationship were excluded, resulting in a final sample of nine studies.

Finally, this review was limited to studies published between 2024 and 2025 to minimize the influence of coronavirus disease 2019-related disruptions; however, some included analyses were based on data collected prior to this period. Future research should prioritize longitudinal or experimental designs using objective and standardized PA measures, while accounting for developmental stage and contextual moderators.

In conclusion, the results of the studies suggest promoting regular, moderate-intensity PA within adolescents’ daily routines, particularly through accessible school-based programs. However, inconsistent findings for high-frequency and high-intensity PA highlight the importance of balanced approaches that support both health and academic demands without overburdening students.

Article information Continued