Breaking barriers: challenging the notion of a strong link between physical fitness and executive functions in 10-year-olds

Abstract

Background

This study utilized a replicated measurement approach to comprehensively explore the connections between various aspects of physical fitness and executive functions in prepubescent children.

Methods

The sample consisted of 62 students (32 boys, 30 girls, aged 10.44 ± 0.33 years) with assessments of physical fitness and executive functions conducted at two time points 4 months apart. Physical fitness assessment involved evaluating body composition (body-mass index, fat mass, fat-free mass) and measures of motor coordination (Obstacle Course Backward test), strength (Long Standing Jump test), speed of movement (Plate Tapping test), and aerobic fitness (20 m Shuttle Run test). Executive functions, specifically inhibition and cognitive flexibility, were evaluated using the Modified Stroop task and Trail Making task, respectively.

Results

Initial measurements showed only low correlations (r = 0.12–0.20) between the Plate Tapping and Shuttle Run with executive function tasks, which did not reach statistical significance, while other connections were consistently trivial. In the follow-up measurement, the Plate Tapping test showed a moderate correlation (r = 0.39) with the Stroop task, while remaining correlations were either negligible or low and not significant. The findings suggest two important conclusions: (i) body composition shows limited association with executive functions in children; (ii) among motor variables, only the speed of limb movement may have some relevance for inhibition, but this association is relatively modest and inconsistent.

Conclusions

Overall, the morphological status and motor competence of prepubertal children seem to have minimal effects on cognitive tasks involving inhibition and cognitive flexibility, and vice versa.

Introduction

Over the past two decades, the notion that physical fitness and cognition are linked has gained wider attention, sparking interest in exploring the relationships between different aspects of these two complex domains and the potential benefits of physical activity on cognitive functioning, particularly in school-aged children and adolescents [1,2,3] and in the aging population [4]. The former, physical fitness, encompasses “the ability to function efficiently and effectively, to enjoy leisure, to be healthy, to resist disease, and to cope with emergency situations”, with two subdomains – health-related components (body-composition, aerobic fitness, flexibility, and strength), and skill-related components, or motor fitness (agility, balance, coordination, and speed) [5, 6]. The latter, cognition, includes “all forms of knowing and awareness, such as perceiving, conceiving, remembering, reasoning, judging, imagining, and problem solving” [7], though executive functions (EF), which are in the focus of this study, along with working memory and IQ measures (primarily fluid intelligence) being of particular interest to researchers focusing on the relationship between cognition and PF.

EF encompass “higher level cognitive processes of planning, decision making, problem solving, action sequencing, task assignment and organization, effortful and persistent goal pursuit, inhibition of competing impulses, flexibility in goal selection, and goal-conflict resolution” [8]. In general, EF can be divided into three relatively autonomous categories: inhibitory control, working memory, and cognitive flexibility [9, 10]. Inhibitory control involves the ability to resist distractions, disregard irrelevant information, and suppress automatic responses in order to concentrate on a particular task or objective. Working memory involves the ability to temporarily store and process information, which enables the individual to perform complicated cognitive activities such as problem-solving or language comprehension. Cognitive flexibility, on the other hand, involves adapting and transitioning between various tasks, rules or modes of thinking essentially adjusting cognitive processes to accommodate new information or shifting demands [9,10,11,12].

The notion that physical fitness and EF are linked is based on several perspectives. First, comparable brain structures, in particular the prefrontal cortex and the primary motor cortex, play a role in both motor performance and EF [3, 10]. Furthermore, both motor and cognitive abilities develop in a structured, stage-like manner, progressing from basic to more complex skills, with their developmental trajectories showing notable similarities between the ages of 5 and 12 years [13]. In this age period, children experience significant growth in both domains, with motor skills such as balance, coordination, and fine motor control serving as a foundation for the more complex cognitive tasks required in school and daily life [3]. For instance, the ability to write and manipulate objects requires the development of fine motor skills, which in turn support cognitive tasks such as reading comprehension and problem-solving [14]. In addition, beyond its well-documented benefits for motor development, physical activity has been shown to promptly enhance cerebral blood flow, neurogenesis, and synaptic plasticity, particularly in the prefrontal cortex and hippocampus - key regions involved in EF and learning [10, 15, 16]. Previous meta-analysis [16] has shown that the most significant impact of physical activity on cognition is particularly pronounced in preadolescents (9–12 years) and the elderly (> 65 years). Additionally, physical fitness contributes to cognitive function not only through direct physiological effects on the brain but also by fostering improved attention and problem-solving skills [17]. The bidirectional nature of this relationship suggests that improvements in motor competence can support cognitive development, while enhanced cognitive control facilitates motor learning and execution [17, 18]. Consequently, there is a logical basis for considering the relationship between physical fitness and EF as interdependent, especially in preadolescence. However, given the broadness of the two domains, research on their relationship is highly diverse, with both physical fitness and EF being represented by different functions, measured by different sets of variables, leading to complex patterns of contradictory results that are difficult to interpret and summarize [1, 11, 16]. This diversity of findings may be attributed to studies selectively focusing on only one or two PF variables or investigating the relationship between physical fitness and EF at a composite level, thus obscuring an understanding of the multilevel nature of this relationship in children [11].

In the context of physical fitness, previous research has predominantly focused on aerobic fitness and/or composite motor scores (i.e., summarized motor variables) to explore their associations with EF. Despite these associations being sometimes significant, their strength tends to be weak, with correlation coefficients (r) typically falling below 0.25, as evidenced by several cross-sectional studies [19,20,21,22,23,24,25]. Beyond these broad measures, only a limited number of studies [22,23,24,25] have delved into specific fitness variables such as strength, speed or coordination. These studies suggest that certain aspects of EF, particularly inhibition and cognitive flexibility, may intricately intertwine with individual fitness components. For example, De Bruijn and colleagues [23] used four motor tests (i.e., long jump, plate taping, endurance shuttle run, and 10 × 5 shuttle run) to assess strength, speed, aerobic fitness, and agility in 9-year-old children and examine potential correlations with EF such as inhibition, working memory, and cognitive flexibility. Their findings indicated low correlations (r ≤ 0.18) between cognitive flexibility and the motor tasks assessed, as well as between working memory and both speed and strength. Notably, inhibition demonstrated a significant correlation (r = 0.21) exclusively with the plate tapping task, which was used to measure the speed of limb movement. In another study with overweight prepubertal children, Mora-Gonzalez et al. [22] examined strength, aerobic fitness and agility and found that only agility displayed significant correlations with both inhibition (r = 0.23) and cognitive flexibility (r = 0.30), while aerobic fitness only showed a significant relationship with cognitive flexibility (r = 0.25). On the other hand, Schmidt et al. [24] found that of three fitness components (coordination, strength, aerobic fitness), only coordination served as a significant predictor of composite EF in 10-12-year-old children.

The present study was designed to provide a comprehensive examination of the relationships between physical fitness and EF within a cohort of prepubertal children. To thoroughly explore the correlation between physical fitness and EF, we utilized a replicated measurement approach in which the relationship between the two variables was examined at two different time points. Consequently, we believe that this research design offers several advantages, such as increasing the confidence and credibility of the results by reducing the likelihood of false positives, sampling bias, or measurement error [26]. By incorporating repeated measurements, we aimed to capture more stable and representative data, minimizing intra-individual fluctuations and increasing the precision of our estimates. Given the limitations of previous research, which often focused on a single fitness variable or composite motor score, we incorporated multiple motor variables, including strength, aerobic fitness, coordination, and speed of limb movement. The main focus was to explore potential correlations with EF, with particular emphasis on inhibitory control and cognitive flexibility. The choice of these two EF domains was guided by their significant correlations with specific fitness variables documented in earlier studies [19, 20, 27]. To increase the depth of our study, we also included a thorough body composition assessment. This decision is particularly relevant in light of previous research findings suggesting that children who are obese or overweight may exhibit deficits in EF compared to their normal-weight peers [28,29,30]. Indeed, studies have shown that obese and overweight children tend to perform worse on tasks requiring inhibitory control, cognitive flexibility, and working memory compared to their normal-weight peers [29, 30]. These impairments may stem from both physiological and behavioral factors, including reduced physical activity levels, systemic inflammation, altered brain structure and function, and metabolic dysregulation [28, 30]. Given the gaps in the existing literature, the present study aims to provide a comprehensive investigation of the relationships between various motor components, body composition, and specific EF (inhibition and cognitive flexibility) in 10-year-old children.

Methods

Experimental design

Participants underwent a comprehensive physical fitness assessment, which included a body composition analysis and four field-based tests to measure motor fitness. In addition, EF were assessed using two tasks. All measurements were administered twice to the participants, allowing us to have a single database that includes both measurement points – an initial session and a follow-up replication session to further confirm the results of the initial session. The two sessions took place 4 months apart.

In both sessions, the experimental procedure took part on two consecutive days. On the first day, body composition and motor fitness were assessed. On the following day, EF were assessed. On both days, the data was collected in the morning hours (8–11 am) at a constant room temperature (20–25° Celsius). To ensure consistency, all participants were familiarized with the motor and EF tests during a visit prior to the actual data collection. In addition, participants were instructed to not to engage in any physical activity and eat solid food for two hours before body composition testing.

Subjects

Sixty-two children (32 boys and 29 girls, aged 10.44 ± 0.33 years), students from a school in Belgrade, Serbia, participated in the initial session. Considering this sample size, the resulting statistical power is 1 − β = 0.639. All participants were healthy and had no history of musculoskeletal injuries or cardiovascular issues. According to the school psychologists, none of the participants had an individualized learning plan, and there were no participants with ADHD. Of the initial sample, a total of 58 participants (32 boys and 26 girls) were available to participate in the replication session. Participants and their parents were fully informed about the experimental procedures and potential risks and signed a written informed consent form prior to participation in the study. The study was approved by the Institutional Ethics Committee (ID 451-03-1/2023-01/4) and conducted in accordance with the Declaration of Helsinki.

Assessment of physical fitness

The anthropometric measurements included body height (BH) and body mass (BM), whereas body mass index (BMI) was also calculated. The Martin’s portable anthropometer (Siber-Hegner, Switzerland), with an accuracy of 0.1 cm, was used for BH measurement. The BMI was calculated according to the standardized formula [31]. The Tanita MC-780 MA device (Tanita Corporation, Tokyo, Japan) was used to determine BM and body composition variables, especially body fat mass (BF) and fat-free mass (FFM). This entailed the application of the Direct Segmental Multi-frequency–Bioelectrical Impedance Analysis (DSM–BIA) method. The BIA device was manually programmed with data such as height, age and gender. Before the analysis took place, participants were instructed to abstain from eating in the morning, avoid engaging in any form of exercise within 24 h prior to measurement, and address any physiological needs. They were requested to stand for at least 5 min before the measurement to allow for the redistribution of bodily fluids. During the measurement, all participants wore light sportswear and took care to remove all metallic accessories.

The motor test battery comprised a total of 4 items and was administered according to a standardized protocol [32, 33]: for accessing speed (or frequency) of movement – Plate Taping Test (freq); for assessing explosive strength - Standing Long Jump Test (m); for assessing body coordination – Obstacle Course Backwards Test (0.1s); for accessing aerobic fitness – Shuttle Run Test (s). All testing sessions were supervised by two experienced physical education teachers. Attention was paid to correct form throughout the testing.

The Plate Tapping Test (PTT) was conducted on a wooden table featuring two 20 cm diameter circles, with their centers 60 cm apart. Seated participants were instructed to place their non-dominant hand between the circles in the center of the table, while their dominant hand crossed over the weaker hand on one circle. The task involved rapidly tapping both circles with the fingers of the dominant hand for a duration of 20 s. For the Standing Long Jump Test (SLJ), participants stood with their feet shoulder-width apart and jumped forward to land as far as possible on both feet. The measurement from the starting line to the back of the heel nearest to the starting line determined the score. In the Obstacle Course Backwards Test (OCB), participants moved backwards on all fours over a 10-meter distance. Two obstacles, a vaulting box lid and a frame, were placed 3 and 6 m from the starting line. The participants climbed over the first obstacle and crawled under the second. In the Shuttle Run Test (STR) the participants had to run between two lines that were 20 m apart. The pace, controlled by an audio signal, started at 8.5 km/h and increased by 0.5 km/h every minute. The test ended when the participants stopped due to fatigue or failed to reach the end line in sync with the audio signal on two consecutive occasions. The final level achieved was recorded to assess aerobic fitness.

Executive functions assessment

EF of the participants were assessed using two tasks: [1] Stroop task and [2] Trail-Making Test.

In the Trail Making Test, participants had two tasks (Trail Making task-A and Trail Making task-B). In the Trail Making task-A, participants were given a piece of paper with the numbers 1–25 placed in circles scattered on it. Their task was to connect the consecutive numbers, starting with 1, and they had unlimited time to complete it. In Trail Making task-B, participants were asked to connect consecutive numbers and letters interchangeably (1-A-2-B…), again with unlimited time. If they made a mistake, the experimenter pointed out to them that they had made a mistake, so they corrected themselves and continued. The experimenter measured the total time taken for each task. The ratio of the two times (Trail Making task-B/Trail Making task-A) was calculated to better capture EF by reducing the contribution of the simpler processes also involved in the Trail Making task-A test [34]. Note that in our version of the Trail Making Test, the letters of Serbian Cyrillic alphabet were used, rather than Serbian Latin or English Latin alphabet, even though the original version of the Trail Making Test is often used with adults. In the Serbian language, both alphabets are used, but Cyrillic is the one that students learn first, so schoolchildren are likely the most familiar with the Cyrillic orthography and the Cyrillic traditional order of letters, out of the three options.

In the Stroop task, executive function was assessed using an analogous procedure. The experimenter provided the participants with four lists of 30 colour names printed on paper, arranged in six rows, one after the other. The word lists were taken from the Stroop task made available by the Genetics Science Learning Center at Utah University [35]. In the first list (List A), all colour names were printed in black. In the second list (List B), each colour name was printed in the congruent colour. The third list (List C) contained meaningless strings printed in different colours. The final list (List D) contained colour names printed in an incongruent colour. For the first two lists, participants were asked to read the colour names aloud. For the third list, the task was to name colours. Participants read the final list twice, once to name the colours and the second time to read the words. If a participant made a mistake, the experimenter corrected the participants, unless they corrected themselves first, which they did in most cases, and then the participant said the correct answer. In each case, the experimenter measured the time it took a participant to read each of the four lists to the end. For the assessment of inhibition, the score for the Stroop task was calculated as the interference score according to Eq. 1, where t_D−naming is the time, it took readers to complete the naming task on list D, and t_A and t_B are the times it took readers to complete tasks A and B, respectively.

$$\:I={t}_{D-naming}-\frac{{t}_{A}+{t}_{B}}{2}$$

(1)

Analogous to the Trail Making Test, the procedure, which includes corrections and self-corrections, allows the measure of total time to include both speed and accuracy in the same index. On the other hand, calculating the interference index helps to separate the executive function component from other factors that contribute to task success, such as reading ability [36]. This is particularly relevant in school children, whose reading skills are still developing.

Statistical analysis

The descriptive statistics are presented using means, standard deviations and value ranges. The Shapiro-Wilk test was used to assess the normality of the distributions, and the values for skewness and kurtosis were calculated. The Pearson correlation coefficient (r), for which 95% confidence intervals are given in addition to statistical significance, was used to examine the relationship between physical fitness and EF. The strength of the Pearson correlation was categorized according to Hopkins et al. [37] as follows: trivial (> 0.01), small (> 0.1), moderate (> 0.3), large (> 0.5), very large (> 0.7), and extremely large (> 0.9). If the distribution of any of the variables was not normal, Spearman’s rho coefficient (ρ) was also used. The correlation analysis was also performed between the same variables in the initial and the replication sessions. A second- and third-order polynomial regression analysis was performed to examine the nonlinear relationship between physical fitness and EF. Paired-sample t-tests were used to assess differences between baseline and follow-up measurements for the tested variables. When the assumption of normality was violated, the non-parametric Wilcoxon Signed-Rank Test was applied instead. The threshold for statistical significance was p ≤ 0.05. All calculations were performed using JASP 0.18.1.

Results

Initial measurement session

Concerning fitness variables, distribution of body height, SLJ, PTT, and OCB did not deviate significantly from the normal distribution (W = 0.97–0.99, p > 0.05), while the distributions of body mass, BMI, body fat, fat-free mass, and SRT were positively skewed (W = 0.88–0.94, p < 0.05). Concerning EF variables, both Stroop and Trail Making tasks showed normal distribution (W = 0.96, p > 0.05).

Relationship between physical fitness and executive functions

As can be seen from Tables 1 and 2, Pearson’s r (or Spearman’s ρ) coefficients were mostly in the range of trivial correlations (< 0.1), with even their 95% confidence interval margins in the range of small effects. Only three correlations exceeded the 0.1 threshold, but they were also in the range of small correlations, and neither was statistically significant (PTT and Stroop, r = -0.2, p > 0.05, SRT and Stroop, r = -0.12, p > 0.05, Trail Making task B/A ratio and SRT, r = -0.14, p > 0.05).

Table 1 Cross-correlations between anthropometric and EF domains during initial session

Table 2 Cross-correlations between motor and EF domains during initial session

Replication session

As in the initial session, the body height, SLJ, PTT and OCB distributions did not deviate from the normal distribution to a statistically significant degree (W = 0.96–0.98, p > 0.05), while the distributions of the body mass, BMI, body fat, fat-free mass and SRT were positively skewed (W = 0.87–0.94, p < 0.05). The Trail Making task ratio showed significant positive skewness (W = 0.90, p < 0.01), as did the Stroop task results (W = 0.90, p < 0.01).

Compared to the first measurement, the means and standard deviations indicate slightly better results for most tasks and anthropometric measurements. Significant differences were found for all anthropometric variables (p < 0.05), except for BMI (p = 0.833). Similarly, all motor variables showed significant improvements (p < 0.05), except for OCB (p = 0.631). The Trail task also showed significant improvement (p < 0.05), whereas differences in Stroop task performance did not reach statistical significance (p = 0.296) (Tables 3 and 4).

Table 3 Baseline and follow-up values for anthropometric variables

Table 4 Baseline and follow-up values for motor and EF variables

Relationship between physical fitness and executive functions

As shown in Table 5, in this session, one correlation was of moderate magnitude (r > 0.3) and statistically significant – the correlation between the PTT and the Stroop test (r = -0.39, 95% CI = [-0.12, 0.0.61], p < 0.01; ρ = -0.32, 95% CI = [-0.03, 0.55], p < 0.05). The remaining correlations were either trivial or low and not significant (Tables 5 and 6).

Table 5 Cross-correlations between motor and EF variables during replication session

Table 6 Cross-correlations between anthropometric and EF variables during replication session

A second- and third-order polynomial regression analysis was performed to examine the nonlinear relationship between the observed variables. For most variables, there were no significant differences between the linear and polynomial regressions, indicating that the relationship between the variables was largely linear.

To evaluate the consistency of measurements between sessions, we calculated Pearson or Spearman correlation coefficients for all physical fitness and EF variables. The results showed strong correlations for anthropometric and motor fitness measures: body height (r = 0.989, p < 0.01), body mass (ρ = 0.984, p < 0.01), BMI (ρ = 0.962, p < 0.01), body fat (ρ = 0.930, p < 0.01), fat-free mass (ρ = 0.981, p < 0.01), SLJ (r = 0.910, p < 0.01), PTT (r = 0.858, p < 0.01), OCB (r = 0.748, p < 0.01), and SRT (ρ = 0.821, p < 0.01). EF tasks also demonstrated moderate consistency: Stroop task (ρ = 0.622, p < 0.01) and Trail Making Task ratio (ρ = 0.607, p < 0.01). These findings support the stability of the measurements across the four-month interval.

Discussion

In this study, a replicated measurement approach was used to comprehensively investigate the relationship between physical fitness and EF in 10-year-old children. According to the current findings, it can be concluded that body composition shows a limited association with of EF in children. As far as motor variables are concerned, of strength, aerobic fitness, body coordination and speed of movement, only the speed component might have some relevance for inhibition, but this association is relatively modest and inconsistent. Overall, morphological status and motor competence of children appear to have minimal effects on cognitive tasks involving inhibition and cognitive flexibility, and vice-versa.

The current findings indicate that a definitive relationship between physical fitness and EF isn’t readily apparent in typically developing 10-year-old children. Conversely, several studies [19, 21, 22, 38,39,40] pointed out that motor competence plays a significant role in EF. However, it should be noted that while the aforementioned studies often highlight the importance of this relationship, their strength is rather weak, with r coefficients typically below 0.25, which is partially consistent with the current findings. Apart from the fact that researchers often only present significant correlations, which in this case are usually low, we consider it very important to also show the non-significant results in order to avoid bias and obtain a realistic overall picture of a phenomenon. Unfortunately, the literature reveals numerous instances of misreporting or misrepresenting results, with non-significant findings often omitted [41]. The failure to report negative results introduces bias in meta-analyses and leads to the unnecessary replication of studies, ultimately wasting resources [42, 43]. Therefore, we emphasize the importance of transparently reporting all findings, regardless of statistical significance, to ensure a more reliable scientific process.

In line with these considerations, we undertook a replication study to bolster the reliability of the findings by mitigating potential sources of error such as measurement inaccuracies, sampling biases, or false positives [26]. Additionally, what sets our study apart is its emphasis on motor aspects, revealing a notable association solely between the speed of limb movement and EF, particularly inhibitory control. It is noteworthy that the correlation between PTT and the Stroop task was only significant in the second assessment. Although this correlation was more pronounced in the first assessment compared to other motor aspects assessed (r = 0.20 and r = 0.03–0.12, respectively), it didn’t reach statistical significance. Unfortunately, the scarcity of studies that include limb movement speed as a motor variable in their research design makes it difficult to directly compare our findings with previous reports. To our knowledge, only De Bruijn et al. [23] have examined limb movement speed using PTT and investigated possible associations with EF in prepubertal children. Among four motor variables, their results underlined PTT as the most significant predictor (r = 0.21) of the Stroop test, which is consistent with our results. Similarly, studies [39, 44, 45] that focused on academic performance and intelligence also found a low significant correlation with PTT, but not with other fitness components (strength, agility, aerobic fitness). Conversely, studies [46, 47] on maximal sprint speed and EF in children show a trivial correlation, suggesting that not all speed-related factors are relevant predictors of cognitive function. This disparity could be due to the fact that cognitive functions are more directly involved in tasks that require precise and coordinated limb movements, such as in PTT [48], whereas sprint speed is influenced by physiological factors that determine the body’s ability to generate force and propel itself efficiently [49]. Specifically, the modest correlation observed between PTT and Stroop task could be explained by similar task demands. In particular, tasks involving limb motor speed, such as rapid tapping or reaction time tests like the PTT, may require inhibitory control to maintain accuracy and avoid errors [50]. Consequently, inhibition tasks, as seen in the Stroop task, involve motor responses that could be influenced by motor speed abilities [51].

In contrast to PTT, other motor variables (i.e., coordination, aerobic fitness and strength) did not show significant associations with EF tasks. The current findings are consistent with previous reports [27, 52], where SLJ (i.e., strength) showed no significant correlation with EF. Regarding aerobic fitness, our results are consistent with some studies [46, 52], but contradict other studies [19, 21, 53], in which aerobic fitness was found to be a significant predictor of EF, particularly cognitive flexibility. However, it should be noted that regardless of whether the relationship between aerobic fitness and cognitive flexibility was significant or not in previous literature, the strength of the correlation was consistently low (r ≤ 0.21). Although statistical significance was not reached in the present study, the Trail Making test showed a low correlation (r = 0.14 and r = 0.18) with STR, while the association with other motor variables was always trivial (r ≤ 0.08). Given our study’s sample size of approximately 60 children, a larger sample could increase the likelihood of establishing statistical significance in the relationship between STR and the Trail Making task. Nevertheless, in light of previous and current findings, aerobic fitness appears to have at most only a small effect on the cognitive flexibility of 10-year-old boys and girls.

In addition to strength and aerobic fitness, the lack of significant correlations between motor coordination and EF is somewhat unexpected. Given that coordination is a multifaceted fitness component that is highly reliant on the integrity of neural networks such as the prefrontal cortex, basal ganglia, and cerebellum, areas that also play a role in EF, this would suggest a potential link [3]. However, previous research has yielded conflicting findings regarding the relationship between coordination and EF, ranging from trivial to moderate correlations [39, 45, 54, 55]. Two possible factors could account for this inconsistency. First, it is important to recognize that motor coordination is a multifaceted concept that encompasses various dimensions (e.g., bilateral coordination, gross motor coordination, etc.), each of which can be assessed by different motor tasks [56]. Second, the relationship between coordination and EF may vary with age and is particularly pronounced in preschool children compared to older age groups [20, 57]. This is likely attributed to the rapid development of both motor and cognitive domains in early childhood, a critical period for establishing neural connections and brain development, fostering significant overlap and interdependence between motor and cognitive functions [57]. However, as children mature and grow older, these connections tend to become more specialized, with cognitive functions increasingly independent of motor development [20].

Another important aspect of the current study is that it addresses the relationship between body composition and EF, an area that has been overlooked in previous research. In this way, our study provides a more comprehensive understanding of the relationship between physical fitness and EF in typically developing prepubertal children. Recent theories suggest that weight status may be associated with EF, whereby obese or overweight children are expected to have deficits in cognitive functioning compared to their normal-weight peers [28, 58, 59]. However, our findings contradict this assumption, as we found no such association in normal-weight children. Moreover, our results are in line with recent reports [59], which also found no significant association between body composition variables (BF, FFM, and BMI) and the ability to delay gratification and affective decision-making in a sample of 6-12-year-old Polish children. Hence, it is quite possible that there is a relationship between body composition and EF in the overweight and obese population, but this relationship is not evident in normal-weight and underweight children, as suggested by our results. One possible explanation for this difference could be attributed to the level of physical activity, as it significantly influences both weight status and cognitive ability [60]. In this regard, physical activity level could act as a mediator in the relationship between weight status and cognitive function [24].

Although our study was not primarily designed to assess test-retest reliability, we examined the consistency between the two measurement sessions. Despite the four-month interval, most variables demonstrated moderate to strong correlations, with particularly high stability observed for anthropometric and motor fitness measures. Executive function scores also showed moderate consistency (ρ = 0.622 for Stroop; ρ = 0.607 for TMT), indicating that variability in our results is unlikely to stem from measurement error. These findings further support the reliability of our assessments.

Strengths and limitations

The present study provides a comprehensive analysis of the relationships between physical fitness and EF in prepubertal children, including the relationship between body composition and EF, which has been overlooked in previous studies. Our study also overcame the limitations of previous research, which typically focused on a single physical fitness variable or composite motor score by including multiple motor variables such as strength, aerobic fitness, coordination, and movement speed. To enhance the previously employed methodology and validate the findings of this study, we conducted a replication study, following best practices outlined in prior research [61,62,63]. While meta-analytic studies have generally found weak associations between physical fitness and EF, these results often come from single-session studies, which are prone to measurement error and intra-individual variability. By including a second measurement session, we aimed to determine whether the observed associations were stable over time, reducing the likelihood that any effects (or lack thereof) were caused by random fluctuations, sampling bias, or false positives. This methodological decision provides stronger empirical support for assessing the true nature of the relationship between physical fitness and executive function in prepubertal children. Additionally, this approach strengthens the internal validity of present study by ensuring that observed effects are not due to transient factors or measurement inconsistencies. Moreover, we deem that a special strength of our study is emphasizing a lack of significant correlations between the abovementioned variables [64]. In this way, we have broken the unwritten rule in the present study and emphasized the importance of reporting non-significant results and not overlooking them. We believe that this is also a very special and important contribution to science.

Despite the strengths evident in the study, it is important to acknowledge several limitations. First, the cross-sectional design of the study limits the ability to establish direct causality. Second, we did not assess the children’s maturity level, a factor that potentially affects both physical fitness and EF. Another potential limitation is the relative age effect, where children born earlier in the academic year may exhibit more advanced physical and cognitive development than their younger peers. This may have contributed to variability in the observed outcomes, although birth month was not specifically analyzed in this study. The fourth limitation concerns the lack of measurement of certain unaccounted social variables, such as family income. Finally, the study used only one test to assess motor coordination (i.e., the OCB test), thus overlooking different dimensions of coordination. In light of these considerations, further research is needed to elucidate whether diverse facets of motor coordination could be associated with EF in children. Additionally, as daily physical activity levels and food intake were not assessed, future studies should consider tracking these variables to further refine our understanding of these relationships. Future research should also consider using exact birth dates to enhance the accuracy of age-related analyses.

Conclusion

In this study, we used a replicated measurement approach to thoroughly investigate the relationship between different components of physical fitness (such as body composition and four motor aspects) and EF (specifically inhibitory control and cognitive flexibility) in prepubertal children. The findings of our study indicate that body composition variables, including fat and fat-free mass, have trivial associations with EF, suggesting that they should not be considered significant contributors of cognitive functions. Among motor variables, only limb movement speed appears to be of interest in relation to inhibitory control; however, this association is likely due to similar task demands. Overall, our current results suggest that the relationship between fitness and EF is not readily apparent in typically developing prepubertal children.