Does Learning Through Movement Improve Academic Performance in Primary Schoolchildren? A Systematic Review

Physically active children have greater motor competence and a faster maturation compared with their sedentary peers. Recent research also suggests that physical activity during childhood may also promote cognitive development and therefore improve academic performance. The aim of this study was to understand if physically active academic lessons may improve academic achievement in primary schoolchildren. A systematic review following the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines was conducted. The search was performed on the following database: PubMed, Web of Science, Scopus, Education Resources Information Center (ERIC), and PsycINFO (APA). Studies evaluating schoolchildren aged between 3 and 11 years taking part in educational contexts that include physical activity and natural environments evaluating physical fitness and/or educational outcomes were included. A total of 54 studies (for a total sample of 29,460 schoolchildren) were considered eligible and included in the qualitative synthesis. The Effective Public Health Practice Project risk-of-bias assessment revealed a moderate quality of the included studies with only two considered weeks. Despite differences in the retrieved protocols, physically active academic lessons improve the total time engaged in physical activity, motor skills, and/or academic performance. The results of this review suggest that learning through movement is an effective, low-cost, and enjoyable strategy for elementary schoolchildren.


INTRODUCTION
Children spend an ever-increasing time in sedentary behaviors such as the ∼2 h (children aged 3 years) or 3 h (children aged between 3 and 5 years) per day in television view (1). Sedentary behaviors are also in school settings during which ∼80% of the time children are seated (2), and only 5% of the time is spent in moderate to vigorous activities in European schoolchildren (3). Consequently, considering the classroom as a place where students spend the majority of their waking time, the school setting can be considered as an ideal setting to improve physical activity and academic achievement and also because it has positive results inside and outside the school (4,5).
The key point to limit sedentary behaviors is to propose health promotion programs to promote physical activity since early childhood (6). Physical activity interventions for children should improve physical fitness, promote health-related behaviors, and facilitate mental development (7,8). Physical activity, especially during development, has positive effects on the measures of adiposity, motor skill, bone and skeletal health, psychosocial health, cardiometabolic health indicators, and cognitive development (9)(10)(11). Gross and motor skill practice has also positive effects on cognitive development (12) and functions (such as perceptual skills, intelligence quotient, academic achievement and readiness, verbal and mathematics tests, developmental level) (13,14), non-executive cognitive functions, core executive functions, and higher-level executive functions (15). It seems that aerobic training has the largest effects (16). Its performance during early childhood could become a lifelong habit, improving cognitive and physical health (17), making the physical movement even more important in this phase of life. Consequently, states need to monitor and evaluate strategies to increase physical activity during school time, adopting a policy specific to prevent potential loopholes (18). On the other side, elementary schools could be a platform for early intervention to improve daily physical activity, but further investigations are required to secure the successful assimilation of movement integration into routine practices (19). Especially in children, physical activity practice during school days can be incorporated, and it increases moderate to vigorous physical activity levels (20) and improve aerobic fitness (4) and also has positive learning outcomes and consequently academic achievement (21). It can integrate physical activity in the academic curriculum and consequently propose a classroombased physical activity program, increase children's cognition (15) and energy expenditure (22), develop social skills, improve mental health, and reduce risk-taking behaviors, but it also has short-term cognitive benefits (23). A physical education program could be a decisive education strategy to enhance motor and cognitive learning in preschool children and to achieve successful academic outcomes (24). Physically active lessons can be proposed with different contents such as math, language, arts, and social sciences, and this has also positive effects on physical activity level and learning and attention (25). The inclusion of physical activity in the curriculum to improve learning outcomes is feasible, and it is suggested in elementary schoolchildren (21).
Schools and teachers are culturally changing, adopting active learning and other kinds of learning methods, but further improvement is required (26). Unfortunately, individuals and schools limit the application of these kinds of programs (27). In 2012, Erwin et al. (21) suggested that more research is required to study integrated physical activity interventions, both on the learning outcome and physical activity levels. Consequently, the objective of this systematic review was to analyze the protocols adopted and the effects of outdoor learning on schoolchildren.

MATERIALS AND METHODS
The systematic review was conducted following the principles outlined by PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines (28).

Eligibility Criteria
The selection criteria of this review were of the PICO-S (Population, Intervention, Comparison, Outcomes, and Study) design.
The population was composed of young children aged between 3 and 11 years of primary/elementary schools. Studies that investigated only a special population such as people with disabilities were excluded because of the possible disabilityspecific outcomes.
The intervention of interest had to be the use of movement and natural environment with educational elements integrated to improve physical fitness and/or educational outcomes. Curriculum physical education, physical activity breaks without educational elements, recess, and after-school interventions were excluded.
The comparison and the outcomes of interest comprised physical fitness parameters and education outcomes.
About the study design, only English-written original and peer-reviewed studies were considered because of the limitations of the authors with the languages. Intervention, cross-sectional, longitudinal, correlational (randomized and non-randomized controlled, and quasi-randomized studies) studies were also included. Reviews, meta-analyses, abstracts and scientific conference abstracts, citations, opinion articles, books and book reviews, letters, editorials, statements, and commentaries were excluded.

Study Record
The selected articles were included in EndNote software (EndNote version X8; Thompson Reuters, NY, USA). In the first step, duplicates were detected. After this step, two investigators, who worked independently, performed a selection process based on the inclusion and exclusion criteria on the title, abstract, and full-length articles. If the two investigators were in disagreement in categorizing an article, the coordinator of the study was involved and, independently, provided the tie-breaking decision. All investigators were not blinded to the authors or associated institutions of the articles during the selection process.
Information related to the sample (age, gender, and sample size) and intervention (type, duration, frequency) characteristics, and on physical fitness and educational outcomes was collected. The data were discussed narratively and represented through tables.

Risk-of-Bias Assessment
To detect the risk of bias and the quality of the study, the Effective Public Health Practice Project tool (29) was adopted. This tool is composed of three scores (weak, moderate, or strong) that were assigned to the following: (1) selection bias assessment, (2) study design evaluation, (3) confounder factors, (4) blinding, (5) data collection methods, (6) withdrawals, and (7) dropouts, to provide an overall rating. A "strong" scoring was provided to a study if at least four strong ratings and no weak rating were provided to each sub-domain. A "moderate" scoring was provided to a study if it had less than four strong ratings and one weak rating provided to the subdomains. A "weak" scoring was provided to a study if two or more weak ratings were provided to the subdomains. In order to numerically quantify the subdomains, a score of 3 was attributed to a strong evaluation, a score of 2 was attributed to a moderate evaluation, and a score of 1 was attributed to a weak evaluation.

RESULTS
A total of 17,862 studies were found in the electronic databases searched, and 6,820 of the articles were immediately removed because they were duplicates. The final number of included studies after the eligibility criteria screening has been of 54 (three studies were included in a second moment after the reference checking of the included studies). A summary of the search process is provided in Figure 1.

Risk of Bias
The quality of the studies was overall moderate, with only two studies deemed weak. The mean score for selection bias was 3/3, for the study design 2.5/3, for the confounder 2.4/3, for the blinding of 1.3/3, for the data collection of 2.2/3, and for the dropout of 2.9/3, reaching an overall mean total score of 1.8 out of 3.

Study Characteristics
A summary of the study's characteristics is proposed in Table 1.
The number of participants included in the studies was 29,460; one study did not specify the number of students, but the number of classrooms included was 4. A total of 11,392 were composed of girls (39%), 11,021 were boys (38%), whereas in 6,486, the gender was not specified (23%). The mean age (standard deviation) of the included participants was 8.2 (0.7) years, and it ranged from 3.9 to 11.2 years.
The studies were performed in different countries. The majority of the studies were performed in the United States (n = 22). In Australia, a total of nine studies were conducted. Five studies were conducted in the United Kingdom and Norway. More than one study was conducted in Denmark (n = 4), in the Netherlands (n = 3), in Greece (n = 2), and in Ireland (n = 2). Only one study was conducted in Italy, New Zealand, and Vietnam.
The majority of the studies were randomized controlled trials (n = 25). They were followed by quasi-experimental design (n = 13), observational studies (n = 5), intervention studies (n = 4), and pilot studies (n = 3). Other study designs such as mixed factorial, mixed experimental, within subject, and pedagogical experiments were adopted only one time.
Seven interventions provided negative feedback on the effect of integrated lessons on physical activity and/or academic outcomes, and there are no aspects between the studies that could suggest excluding some aspects of the intervention such as the duration of the program or session, or the kind of intervention, or the subject considered.

Intervention Characteristics
Different studies were based on national or international intervention programs. The Comprehensive School Physical Activity Programs was the intervention program adopted majority of times (n = 4). Adopted in three different studies is the Active Smarter Kids intervention. Less adopted assessment methods are provided in Table 2.
Most of the interventions wanted to improve mathematics (n = 30) and language (n = 14) learning. Language arts was proposed as integrated lessons in five studies, social studies in only three studies, and two times for geography and history. Other subjects such as reading, handicrafts, science, general health, statistics, biology, and religion were studied only one time.
In different studies, no information related to the curriculum subjects studied has been provided (n = 22).
The mean length of the intervention was 153.5 days, with a range from 5 to 1,095 days. The mean duration of the integrated physical activity was of 28.5 min, with interventions that were of 10 and others arrived to 60 min. Different studies proposed three interventions a week (n = 10), but other studies proposed only 2 days a week of curriculum-integrated physical activity (n = 7). Five studies proposed more than 3 days a week of intervention (n = 5). Unfortunately, the majority of the studies (n = 32) have not provided this information.
The majority of the 29 studies ( Table 2) that included data related to the physical activity level collected with accelerometers or pedometers had positive results, with a percentage range of improvement from 1.8 to 96.2. Only one study reported no improvement with the integrated movement program. Unfortunately, the data are not heterogeneous; indeed, studies compared different groups or the same group before and after the intervention. Studies reported the time in which the children were engaged in moderate to the vigorous physical activity or the number of steps. Studies collected data during the school hours or during the week or the day.
Academic achievements or cognitive functions were assessed majority of times through the academic outcomes and the time on task (n = 3). In three studies, the authors evaluated them through the "on-task" behavior. Less adopted assessment methods are provided in Table 2.
Related to physical activity assessment, 29 studies evaluated it through an accelerometer or a pedometer. Some studies evaluated health-related physical fitness characteristics through physical tests such as the Test of Gross Motor Development 2 (n = 4), Andersen test (n = 2), 20-m shuttle run test (n = 3), and Progressive Aerobic Cardiovascular Endurance Run (n = 2). Less adopted evaluation methods are provided in Table 2. Skill-related physical fitness was evaluated through test to evaluate executive functions and motor skills (n = 1). Studies adopted also batteries to evaluate physical fitness such as the FITNESSGRAM (n = 2). The most interesting subjective physical activity evaluation methods were the System for Observing Student Movement in Academic Routines and Transitions (n = 3), Physical Activity Questionnaire for Older Children (n = 1), interviews, and observations.

Intervention Proposal
Some studies reported the intervention in detail or examples of intervention, and the following are proposals of the included studies. Some studies proposed outdoor structured nature-based play (33,71) or adopted the outdoor environment to learn math, language, history, or religion (70). Games-centered interventions (34,35) or games related the pedometer with mathematics (36) or free play or semistructured physical activity have been proposed (39). Always through play was the intervention  of Pham and colleagues, which adopted balls with numbers, letters, and mathematical symbols on the surface (60). Complex, independent, and symbolic play (72) and playing with mathrelated materials to examine children's verbal and non-verbal mathematics exploration without adult guidance (82) were also proposed. Other proposals that comprised cooperative activities integrate health education into several school subjects (41). An intervention proposed as language activity "Scrabble relay, " where children worked in groups, or "Bingo" to improve mathematics (43). In another study, one teacher read a story while students perform the movements in the story (45).  different subjects. To learn language, children enacted the actions indicated by the words to be learned by physically exercising (i.e., for the word "fly, " children ran and moved their hands as if they were flying) (51). To learn geography, children "traveled" from one continent to the other, imitating the movements of the animal representing the continent (52). To learn math, foam blocks of numbers were placed on the floor, shaping a straight line, and the children ran, jumped, and stepped each time on one number while counting or walked or ran backward, sideward, or forward (53). A similar intervention to learn geography was the one proposed by Oliver et al. (59). Norris et al. (57) proposed in their intervention presentation sessions known as Virtual Field Trips, designed to be delivered using existing classroom interactive whiteboards. Similarly, children autonomously navigated through two skill stations with at least three levels of difficulty at each station (67). Other language and mathematics interventions consisted in the performance of a spell by jumping in place for every mentioned letter or to jump to solve multiplications. Similar academic tasks with different words or sums were exercised during one lesson (55,56). Other interventions consisted in building two-digit numbers by making and simultaneously verbalizing out loud different-sized steps (68). Students stand on their self-space and jump the answer to a problem the teacher provided and the second by moving around the classroom, picking a card with a problem working as a group or with a partner (73). Locomotor skills of running, skipping, hopping, and galloping (75) integrating structured movement and motor skill practice with preschool learning concepts and integrating auditory, visual, and kinesthetic learning methods (80) were also proposed. Use and modification of movement elements, development of creative thinking during movement activities through exploration, use of movement for experienced learning of concepts of different teaching thematic areas such as mathematics, and development of critical thinking during movement activities were also adopted (81).

DISCUSSION
The findings of the review highlighted that different interventions were proposed to teach different curriculum subjects through movement with a lack of standardization in the protocols adopted by the authors ( Table 2). Similar to the findings of Erwin and colleagues, physical activity integrated in the academic curriculum is proposed with other interventions (such as breaks), and the details on the effect on children's learning and physical fitness are not always provided (21). It is important to propose a structured intervention; only in this way that it is possible to contextualize and generalize the finding and make the procedure safer (83), and the teachers have a crucial role in following the procedures proposed (22). Differences were also detected in the length, duration, and week frequency, making impossible a comparison among the studies. These findings are similar to the study by Daly-Smith and colleagues, where differences in the design, interventions, duration and intensity, and outcomes were detected (84). The intervention duration in this study started from 5 days arriving to more 1,000 days, differently from other studies in which the intervention ranged from 13 to 300 days (21). Even if the literature suggests that the length of the intervention did not influence the effect of the intervention (21), a short-duration program is not useful to have a long-term improvement on academic performance (85). Furthermore, it is important to propose the integrated programs in daily or weekly schedule because it increases also the physical activity during the school day, and it is feasible (86). Ideally, the physical activity interventions should be three times per week to obtain the best results on children's cognitive and achievement outcomes (16).
Differences were detected also in the interventions. The movement integration program wants to teach students through the movement. It is well-known that physical activity interventions have a positive effect on cognitive performance and academic performance in children (87,88). Integrated physical activity in the classroom can increase children's academic intrinsic motivation, perceived competence, and effort without influencing academic lessons (89).
For those studies that proposed play as an intervention, positive outcomes have been detected. Learning through play forces children to make choices and assuming responsibility having fun at the same time, working on the internal cognitive transactions and intrinsic motivation, determine life habits (90). Play should have to be enjoyable, freely chosen, non-literal safe, and actively engaged; only in this way that learning is through intrinsic motivation (90).
Other interventions, instead, were based outdoors. This way of learning can be incorporated within conventional teaching methods (91); it increases physical activity and reduces sedentary behaviors (92). Open learning environments want to educate the students with own initiative, planning, experimentation, elaboration, and self-evaluation, which is an interesting way (26).
The interventions showed improvements in the academic outcomes, motor skills, or amount of physical activity (through step count), but an important point is that they are costeffective, teachers are not required to prepare them, and they are enjoyable both for teachers and children (5), making them ideal for primary schools. The advantage of a classroom-based physical activity program integrated in the school curriculum is that it takes time from other subjects, but improves physical activity and on-task behavior without sacrificing or influencing academic performance (93,94). Furthermore, physical education in elementary school children has no negative effects on standardized academic achievement test scores (95). Indeed, physical activity improves mathematics-related skills, reading, and composite scores such as the classroom behaviors, suggesting physical exercise lessons in the curriculum and physical activity integration in classroom lessons (96). Even brief bouts (1 h long) of outdoor active play can improve on-task behavior (97). The level of physical activity enjoyed outdoors on the playground is higher, and the increase in on-task classroom behavior is greater; simple play outdoors seems to be not sufficient (97). Physical activity incorporated into the school day improves attention to task (98). Physical, active academic lessons have several benefits for schools and students; indeed, they are cost-effective. Children and teachers enjoy them. They do not require additional teacher preparation time and improve academic achievement scores (5).

Limitations and Future Studies
Data obtained from accelerometers were not analyzed because of the limitation of this tool in detecting activities performed with the upper body (30). Furthermore, the studies included in the review present a wide variety of testing conditions and interventions, making the performance of a meta-analysis impossible. The sample background (physical activity participation outside the school, social status, or other influencing factors) was not detected, making the comparison even harder.
The study has been focused only on a specific population. It has been suggested by the literature (16) that children with learning disabilities also present improvements in academic abilities when physical activity interventions are adopted, making the study of these interventions also in this population even more important. There is a lack of heterogeneity among the study interventions, with differences not only in the length of the program, duration of the session, and frequency but also in the intervention methodology and in the subjects included in the programs. Differences were also in the outcome studies, both for physical activity and academic performance evaluation. Future studies should focus their attention on review of the literature about physical activity breaks during classroom time to improve physical fitness and academic performance. Attention should be focused also on interventions performed in nature, for two reasons: first, the intervention moves the children outside, and second, this intervention can help the children to understand the importance of nature.

CONCLUSION
All the interventions, despite differences in the protocols, have a common aspect: they improve physical activity and/or academic performance, making this kind of approach ideal in elementary schools.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding author.