Journal of Science and Medicine in Sport 21 (2018) 501–507

Contents lists available at ScienceDirect

Journal of Science and Medicine in Sport

journal homepage: www.elsevier.com/locate/jsams

Review

Effects of physical activity on executive functions, attention and

academic performance in preadolescent children: a meta-analysis
Johannes W. de Greeff a,∗ , Roel J. Bosker b,c , Jaap Oosterlaan d , Chris Visscher a ,
E. Hartman a
a
University of Groningen, University Medical Center Groningen, Center for Human Movement Sciences, The Netherlands
b
University of Groningen, Faculty of Behavioral and Social Sciences, Department of Educational Sciences, The Netherlands
c
University of Groningen, Groningen Institute for Educational Research, The Netherlands
d
VU University Amsterdam, Faculty of Psychology and Education, Clinical Neuropsychology, The Netherlands

a r t i c l e i n f o a b s t r a c t

Article history: Objectives: The aim of this meta-analysis was to provide a systematic review of intervention studies that
Received 13 February 2017 investigated the effects of physical activity on multiple domains of executive functions, attention and aca-
Received in revised form demic performance in preadolescent children (6–12 years of age). In addition, a systematic quantification
18 September 2017
of the effects of physical activity on these domains is provided.
Accepted 26 September 2017
Design: Systematic review and meta-analysis.
Available online 10 October 2017
Methods: Searches of electronic databases and examining relevant reviews between 2000 and April 2017
resulted in 31 intervention studies meeting the inclusion criteria. Four subdomains of executive functions
Keywords:
Primary school children
(inhibition, working memory, cognitive flexibility and planning), three subdomains of attention (selec-
Cognition tive, divided and sustained) and three subdomains of academic performance (mathematics, spelling and
Academic achievement reading) were distinguished. Effects for different study designs (acute physical activity or longitudinal
Exercise physical activity programs), type of physical activity (aerobic or cognitively engaging) and duration of
Review intervention were examined separately.
Results: Acute physical activity has a positive effect on attention (g = 0.43; 95% CI = 0.09, 0.77; 6 studies),
while longitudinal physical activity programs has a positive effect on executive functions (g = 0.24; 95%
CI = 0.09, 0.39; 12 studies), attention (g = 0.90; 95% CI = 0.56, 1.24; 1 study) and academic performance
(g = 0.26; 95% CI = 0.02, 0.49; 3 studies). The effects did depend on the subdomain.
Conclusions: Positive effects were found for physical activity on executive functions, attention and aca-
demic performance in preadolescent children. Largest effects are expected for interventions that aim for
continuous regular physical activity over several weeks.
© 2017 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.

1. Introduction increased activity in selective parts of the brain structural network

and especially improves executive functions.5 Executive functions
European preadolescent children (aged 6–12 years) spend are higher order cognitive functions that are responsible for initiat-
209 min/day (64%) of their school time in sedentary activities, while ing, adapting, regulating, monitoring, and controlling information
spending only 16 min/day (5%) in moderate to vigorous physi- processes and behavior.6,7 These functions are often thought of as
cal activity (MVPA).1 This is concerning because apart from the an important prerequisite for successful learning in preadolescent
clear physical health benefits of MVPA in children,2,3 an expanding children.7 Other researchers have focused on more lower-order
body of literature shows that MVPA is positively associated with cognitive tasks, with particular interest in attention.8 Attention is
key cognitive functions that are important for success in school.4 defined as a cognitive state in which a child focuses on a selection
Recently, many researchers have been focusing on the executive of available perceptual information.9 Although attention is closely
functions hypothesis.5 This hypothesis states that MVPA leads to related to executive functions,10 it can be seen as a lower order
cognitive function and it is mostly measured with performance on
reaction time or other simple decisional tasks.11 Improvements in
these cognitive functions as a result of increased physical activity
∗ Corresponding author. may, in turn, improve children’s academic performance.
E-mail address: [email protected] (J.W. de Greeff).

https://doi.org/10.1016/j.jsams.2017.09.595
1440-2440/© 2017 Sports Medicine Australia. Published by Elsevier Ltd. All rights reserved.
502 J.W. de Greeff et al. / Journal of Science and Medicine in Sport 21 (2018) 501–507

Within the preadolescent age range, previous meta-analyses tigated the effects of physical activity on multiple domains of
have shown that enhanced cognitive functioning as a result of executive functions, attention and academic performance in pread-
physical activity is most clearly seen in executive functions12 and olescent children. In addition, a systematic quantification of the
attention.8 These cognitive functions are indispensable for suc- effects of physical activity on these domains is provided.
cess throughout life and are often thought of as an important
prerequisite for successful learning.7 Several underlying mecha-
nisms might explain the effects of physical activity on cognitive 2. Methods
functions. First, a single bout of physical activity (acute physi-
cal activity) is thought to immediately elevate the child’s level The electronic databases PubMed, Web of Science, MEDLINE
of physiological arousal, which in turn facilitates the cognitive and ERIC were searched for studies that investigated the effects
performance by an increased allocation of attention.13,14 From of physical activity on attention, executive functions and/or aca-
a psycho-physiological perspective, acute physical activity trig- demic performance. Key search terms included the words physical
gers an increase of neurotransmitters (e.g. epinephrine, dopamine, activity, physical fitness, executive functions, cognition, academic
brain-derived neurotrophic factors), which are thought to enhance performance and children. Medical Subject Headings (MeSH)
cognitive processes.15,16 Secondly, according to the cardiovascular terms, free text words and all possible equivalents were used
fitness hypothesis, an intervention program that contains contin- (Table 1S, see Supplementary material). The current meta-analysis
uous aerobic physical activity over several weeks (longitudinal included all studies that: (a) investigated the effects of physical
physical activity program) is thought to improve aerobic fitness activity on executive functions, attention and/or academic perfor-
and consequently improve cognitive performance.17 This hypoth- mance, (b) were written in the English language and published
esis is supported by the argue that physical activity enhances between 2000 and April 2017, (c) focused on primary school
the angiogenesis18 and neurogenesis16 in areas of the brain that children between the age of 6–12 years, (d) included a random
support memory and learning, subsequently enhancing cogni- assignment or matching with appropriate adjustments for any
tive performance.19 More recently, other researchers argue that pre-test differences28 and (e) included outcome variables mea-
instead of ‘simple’ aerobic physical activity (i.e. physical activity suring executive functions, attention or academic performance
that is intended to improve cardiovascular performance), cog- on interval- or ratio-level scale. Exclusion criteria for the current
nitively engaging physical activity (i.e. physical activity that is meta-analysis were: (a) studies targeting special populations (e.g.
cognitively challenging) is more beneficial for cognition.20–22 Cog- children with mental or cognition disorders, nervous system dis-
nitive engagement is the amount of both the allocation of attention eases or brain injuries), (b) studies without appropriate control
and the cognitive effort that are needed for a certain activity.20,23 conditions or groups29 and (c) studies of which the intervention
Physical activities with a relatively high cognitive engagement (e.g. consisted of more than just specific physical activity (e.g. inter-
tennis, where children have to plan strategically, focus attention, ventions that included physical active and cognitive tasks). The
and so forth) are suggested to have more effect on executive func- PRISMA-statement for reporting systematic reviews and meta-
tions, compared to physical activities with a relatively low cognitive analysis was used as a guideline to conduct the review.30 A trained
engagement (e.g. long distance running, which involves more auto- research assistant screened the titles of all studies retrieved from
mated movements).20,22 These different underlying mechanisms the electronic databases for potentially suitable studies, after which
suggest that the effects for physical activity to improve attention, the trained research assistant and the first author screened the
executive functions and academic performance in children might abstracts of the selected studies. If there was a doubt about the suit-
depend on the duration or type (aerobic vs cognitively engaging) ability of the study based on the abstract, the authors assessed the
of physical activity is chosen. eligibility based on the full text of the article. The reference list of
Previously, studies have mainly focused on the association relevant reviews were searched for additional studies. Lead authors
between physical activity and overall cognitive functioning in from studies without details that allowed for the calculation of ESs
children. The results from a previous meta-analysis showed a were contacted by email to retrieve missing details.
positive association between physical activity and overall cogni- After removing duplicates and adding 7 studies from previous
tive functioning in children (effect size [ES] = 0.21; 8–10 years).24 reviews, our initial electronic search yielded 3032 studies that were
Meta-analyses on studies allowing the investigation of causal rela- reviewed based on their title (Fig. 1). Seventy five (n = 75) full-text
tionships showed significant positive effects of physical activity articles were reviewed, after which 41 were excluded (Table 2S,
on children’s executive functions (ES = 0.57; 6–12 years)12 and see Supplementary material). Common exclusion reasons were:
academic performance (ES = 0.27; 3–18 years).4 In these previ- studies without appropriate control conditions or groups (n = 9),
ous meta-analyses only a few intervention studies investigated cognitive tests that did not explicitly assess executive functions
the causal effects in preadolescent children, especially those inter- (n = 7), ages of participants were (partly) outside of target age range
vention studies that implemented a longitudinal physical activity (n = 7) or interventions that combined physical activity with aca-
program.12 More recently, several randomized controlled trials demic assignments (n = 6). Finally, one study was excluded because
have become available, aimed at investigating the effects of acute additional details were needed to calculate an ES and we received
physical activity and longitudinal physical activity programs on no response on our email nor on our reminder.31 Thus, a total of
cognitive functioning in preadolescent children.20,25,26 31 studies (4593 children) were included in the meta-analysis. The
The current meta-analysis updates and expands previous meta- characteristics of the included studies can be found in Table 3S (see
analyses by including only studies that investigate the effects of Supplementary material).
acute physical activity or longitudinal physical activity programs The study quality of the selected studies was assessed indepen-
with an appropriate control group. All correlational studies were dently by a research assistant and the first author according to the
excluded from the present meta-analysis, as these designs do not Physiotherapy Evidence Database (PEDro) scale. The PEDro scale is
allow investigation of causal effects. In addition, uncontrolled stud- an 11-item scale which has been used extensively in meta-analyses
ies were excluded, because these designs do not allow conclusions and reliably assesses randomization, blinding, intention-to-treat,
on whether the possible improvements would also have been found between-group comparison and measures of variability.32 Scores
if the participants had been exposed to another intervention not on the PEDro scale range between 0 and 10 (one item pertains
involving physical activity.27 The aim of the present meta-analysis external validity and is not used to calculate the score). An ade-
is to provide a systematic review of all available studies that inves- quate quality is defined as a study having an adequate generation
 J.W. de Greeff et al. / Journal of Science and Medicine in Sport 21 (2018) 501–507 503

Fig. 1. PRISMA flow diagram of the selection of studies.

of random sequence, concealment of allocation and blinding of out- acute physical activity (in minutes) and for the longitudinal physi-
come assessors.33 This is represented by a PEDro summary score cal activity program (in weeks).
of at least 5 points.33 Discrepancies regarding study quality were No statistical comparisons were made between the domains and
discussed until consensus was reached. subdomains because data from multiple domains and subdomains
Statistical analysis was performed using Comprehensive Meta- came from the same children making it unlikely that the assump-
Analysis (version 2.2). First, studies were coded in terms of study tion of independence was true. To control for dependency between
design, distinguishing between studies investigating acute physi- multiple outcome variables (e.g. accuracy and reaction time) within
cal activity or a longitudinal physical activity program. Secondly, a subdomain (e.g. inhibition), a mean ES of the outcome variables
primarily for the purpose of the aim, three separate meta-analyses was used. In case of multiple independent subgroups within a study
were conducted for the domains executive functions, attention and (e.g. multiple intervention groups), subgroup was used as the unit
academic performance. Third, a subgroup analysis was conducted of analysis. To adjust for respective sample sizes, Hedges’ adjusted g
for several subdomains. Four subdomains of executive functions was used and studies were weighted by the study inverse variance.
were distinguished: inhibition, working memory, cognitive flexi- The magnitude of Hedges’ g was interpreted using Cohen’s guide-
bility and planning. These subdomains were chosen because they lines, distinguishing between small (<0.2), moderate (0.5), and large
are relatively easy to operationalize and are likely to be impli- (>0.8) ESs.37 We used a random effects approach to compute over-
cated in the performance of the tests used in most of the included all ESs.38 Several studies collected data at multiple points. In such
studies.6,34,35 Three subdomains of attention were distinguished: cases, only the baseline scores and the scores closest to the end of
selective, divided and sustained.36 In addition, three subdomains of the intervention were used to calculate the ES. Heterogeneity of
academic performance were distinguished: mathematics, spelling each effect was assessed using the I2 and Q statistics.38 The pres-
and reading. Fourth, two additional meta-analysis were conducted ence of a publication bias for executive functions, attention and
for aerobic and cognitively engaging physical activity. Last, a meta- academic performance was assessed by using a funnel plot, calcu-
regression was conducted to investigate the effects of duration for lating the Rosenthal’s fail-safe N and performing the Egger’s linear
regression method. The moderation effect of duration was inves-
504 J.W. de Greeff et al. / Journal of Science and Medicine in Sport 21 (2018) 501–507

Table 1
Individual effect sizes for each subgroup within the included studies.

tigated with a fixed effects meta-regression analysis. Statistical ences between the ESs [I2 = 76%; Q = 41.31; p < 0.001]. No studies
significance was adopted for all tests when p < 0.05. focused on sustained or divided attention. Overall, acute physical
activity resulted in a small to moderate improvement of selective
attention [Hedges’ g = 0.43; 95% CI = 0.09, 0.77; p = 0.013].
3. Results No overall significant effect was found for acute physical activ-
ity on academic performance (s = 4, k = 5), with minor differences
The ESs of the individual studies can be found in Table 1. The between the studies [I2 = 26%; Q = 5.39; p = 0.250].42,45,46,52 Investi-
funnel plot resembled a funnel and the Egger’s linear regression gating each distinguished subdomain of academic performance, a
method was non-significant for executive functions, attention and small to moderate effect was found for spelling [Hedges’ g = 0.25;
academic performance indication that there was no evidence of 95% CI = 0.03, 0.48; p = 0.030] (s = 2, k = 3).42,45 No significant effects
publication bias. There were however considerable differences were found for mathematics (s = 4, k = 5)42,45,46,52 and reading (s = 3,
between the ESs derived from the individual studies [I2 = 84%; k = 4).42,45,52
Q = 286.87; p < 0.001]. The mean PEDro score was 4.3 (below Of the studies that investigated the effects of acute physical
adequate) and had a range between 2 (below adequate) and 7 activity, 15 studies (k = 24) focused on the effects of aerobic physi-
(adequate) (Table 3S, see Supplementary material). cal activity, resulting in a small to moderate positive effect [Hedges’
Seventeen studies (s = 17) were included that examined the g = 0.28; 95% CI = 0.09, 0.46; p = 0.004].25,39–46,48,49,51–54 Five stud-
effects of acute physical activity on cognitive functions [I2 = 87%; ies (k = 5) focused on the effects of cognitively engaging physical
Q = 217.18; p < 0.001], which resulted in a total of 29 comparisons activity which showed no significant effect.39,43,47,50,51 To include
(k = 29).25,39–54 Of these studies, 10 (59%) reported positive findings all studies in the meta-regression, a correction on the standard
on at least one of the outcome measures,25,40–42,44–48,53 six stud- errors (dividing the standard error by the square root of the num-
ies (35%) reported no significant findings39,49–52,54 and one study ber of comparisons) was applied for one study that used different
(6%) reported a negative finding.43 Overall, acute physical activity durations of acute physical activity.46 Duration (in minutes) had
resulted in a small to moderate improvement of cognitive functions no significant influence on the effects of acute physical activity
[Hedges’ g = 0.24; 95% CI = 0.08, 0.40; p = 0.004] (Table 2). (␤ < 0.001; p = 907).
No significant effect was found for acute physical activity on Fourteen included studies (k = 18) implemented a longitudinal
executive functions (s = 9, k = 15). When looking at the subdomains physical activity program and examined the effects on cognitive
of executive functions, a small to moderate positive effect of acute functions [I2 = 65%; Q = 48.46; p < 0.001].20,26,55–66 Nine of these
physical activity was found for inhibition [Hedges’ g = 0.28; 95% studies (64%) reported positive findings on at least one of the
CI = 0.01, 0.56; p = 0.042]. No significant effect was found for work- outcome measures,20,26,56,58,59,61–63,66 five studies (36%) reported
ing memory (s = 4, k = 6) or cognitive flexibility (s = 5, k = 8). No no significant finding55,57,60,64,65 and no studies reported a neg-
studies investigated the effects of acute physical activity on plan- ative finding. Longitudinal physical activity programs resulted in
ning. an overall small to moderate improvement of cognitive functions
Six studies focused on the effects of acute physical activity on [Hedges’ g = 0.37; 95% CI = 0.20, 0.55; p ≤ 0.001].
selective attention (k = 11),25,43,48,50,51,53 with considerable differ-
 J.W. de Greeff et al. / Journal of Science and Medicine in Sport 21 (2018) 501–507 505

Table 2
Meta-analytic results with potential moderators affecting the effects of physical activity on cognitive performance.

Sample size k Meta-analytic effect size Heterogeneity

Hedges’ g 95% CI p-value I2 Q p-value

Acute physical activity 2827 29 0.24 [0.08, 0.40] 0.004 87.11 217.18 <0.001
Domains
Executive functions 1384 15 0.20 [−0.04, 0.42] 0.096 91.03 156.02 <0.001
Inhibition 349 10 0.28 [0.01, 0.56] 0.042 77.64 40.25 <0.001
Working memory 1084 6 0.27 [−0.12, 0.66] 0.176 95.88 121.38 <0.001
Cognitive flexibility 458 8 0.30 [−0.14, 0.73] 0.182 88.29 59.80 <0.001
Planning – – – – – – – –
Attention 616 11 0.43 [0.09, 0.77] 0.013 75.79 41.31 <0.001
Academic performance 941 5 0.09 [−0.05, 0.22] 0.196 25.76 5.39 0.250
Mathematics 941 5 −0.18 [−0.48, 0.13] 0.254 84.27 25.42 <0.001
Reading 847 4 0.17 [−0.08, 0.41] 0.185 57.75 7.10 0.069
Spelling 56 3 0.25 [0.03, 0.48] 0.030 <0.01 0.52 0.772
Type of physical activity
Aerobic 2472 24 0.28 [0.09, 0.46] 0.004 89.31 215.21 <0.001
Cognitively engaging 355 5 0.07 [−0.11, 0.25] 0.424 <0.01 1.95 0.745

Longitudinal physical activity 1766 18 0.37 [0.20, 0.55] <0.001 64.92 48.46 <0.001
Domains
Executive functions 1179 15 0.24 [0.09, 0.39] 0.001 34.00 21.21 0.096
Inhibition 688 7 0.19 [−0.04, 0.42] 0.097 49.72 11.93 0.063
Working memory 579 8 0.36 [0.10, 0.62] 0.007 56.79 16.20 0.023
Cognitive flexibility 556 4 0.18 [0.01, 0.35] 0.040 4.79 3.15 0.369
Planning 394 4 0.12 [−0.08, 0.32] 0.224 <0.01 0.78 0.855
Attention 299 2 0.90 [0.56, 1.24] <0.001 51.49 2.06 0.151
Academic performance 565 4 0.26 [0.02, 0.49] 0.032 38.84 4.91 0.179
Mathematics 231 2 0.09 [−0.17, 0.35] 0.490 <0.01 <0.01 0.983
Reading 277 3 0.15 [−0.15, 0.46] 0.316 35.31 3.10 0.213
Spelling 46 1 0.34 [−0.23, 0.92] 0.243 – – –
Type of physical activity
Aerobic 1231 12 0.29 [0.13, 0.45] 0.001 43.23 19.38 0.055
Cognitively engaging 535 6 0.53 [0.14, 0.92] 0.008 78.87 23.67 <0.001

Note: k = number of comparisons.

A small to moderate positive effect was found for longitudi- (in weeks) had no significant influence on the effects of longitudinal
nal physical activity programs on executive functions [Hedges’ physical activity programs (␤ < 0.001; p = 0.952).
g = 0.24; 95% CI = 0.09, 0.39; p = 0.001] (s = 12, k = 15).20,26,56–60,62–66
Investigating the subdomains of executive functions, a small to
4. Discussion
moderate positive effect was found for working memory [Hedges’
g = 0.36; 95% CI = 0.10, 0.62; p = 0.007] (s = 6, k = 8)20,57,58,62,63,66 and
The current meta-analysis investigated the effects of physical
a small positive effect was found for cognitive flexibility [Hedges’
activity on executive functions, attention and academic perfor-
g = 0.18; 95% CI = 0.01, 0.35; p = 0.040] (s = 3, k = 4).20,26,66 No sig-
mance in preadolescent children based on the outcome data of 31
nificant effects of longitudinal physical activity programs were
studies. The meta-analysis showed that acute physical activity has
found for inhibition (s = 6, k = 7)20,26,56,57,65,66 and planning (s = 4,
a positive small to moderate effect on attention (ES = 0.43), while
k = 4).59,60,64,66
longitudinal physical activity programs has a positive small to mod-
One study (k = 2) found a large positive effect of a longitudinal
erate effect on executive functions (ES = 0.24), a positive large effect
physical activity program on selective attention [Hedges’ g = 0.90;
on attention (ES = 0.90) and a positive small to moderate effect on
95% CI = 0.56, 1.24; p = < 0.001].61 No studies focused on sustained
academic performance (ES = 0.26).
or divided attention.
This meta-analysis expanded the evidence by examining the
Three studies (k = 4) focused on the effects of a longitudinal
results of 9 recent studies that aimed to investigate the effects
physical activity program on academic performance, with minor
of acute physical activity on executive functions in preadolescent
differences between the studies [I2 = 39%; Q = 4.91; p 0.179].55,58,59
children. For acute physical activity (i.e. a single bout of physical
An overall small to moderate positive effect was found on aca-
activity), no effect was found in the current meta-analysis, while
demic performance [Hedges’ g = 0.26; 95% CI = 0.02, 0.49; p = 0.032].
a moderate to large positive effect (ES = 0.57; s = 2) was found in
However, investigating each distinguished subdomain of academic
Verburgh et al.12 The reason for this difference is likely due to the
performance, no significant effects of longitudinal physical activity
inclusion of a larger number of studies in the current meta-analysis
were found for mathematics (s = 1, k = 2),59 reading (s = 2, k = 3)58,59
(s = 9) which may reflect a more reliable result. Despite the inclu-
and spelling (s = 1, k = 1).58
sion of a large number of studies, still large ES differences between
Investigating the type of physical activity within studies that
studies were found. The effects largely depend on different modera-
implemented a longitudinal physical activity program, 11 stud-
tors, such as outcome measure,40,41,45,49 duration46 or effects were
ies (k = 12) focused on aerobic physical activity20,26,55,56,59–65
only found in a subgroup.39,41,44 The positive small to moderate
and five studies (k = 6) focused on cognitively engaging physical
effect of acute physical activity on attention (ES = 0.43) is compa-
activity.57,58,61,63,66 A small to moderate positive effect was found
rable with the small to moderate effect found in the meta-analysis
for aerobic [Hedges’ g = 0.29; 95% CI = 0.13, 0.45; p = 0.001] and a
by Chang et al.8 (ES = 0.42). These findings are in accordance with
moderate to large positive effect for cognitively engaging physical
the hypothesis that acute physical activity immediately facili-
activity [Hedges’ g = 0.53; 95% CI = 0.14, 0.92; p = 0.008]. Duration
tates the cognitive performance by increasing the allocation of
attention.13,14 The combined finding that acute physical activity
506 J.W. de Greeff et al. / Journal of Science and Medicine in Sport 21 (2018) 501–507

has an effect on attention and inhibition (ES = 0.28), while no effect erate to vigorous intensities.21 Secondly, the current meta-analysis
on other domains of executive functions (e.g. working memory, only included studies that were published after 2000 and that con-
cognitive flexibility and planning) supports the hypothesis that tained a control condition with no or a lower amount of physical
attentional problems in children are related with specific domains activity. Although this resulted in excluding only a small amount of
of executive functions, particularly with inhibition.10 In addition, studies, the inclusion of these studies would limit the generalizabil-
the positive effects on attention are expected to increase the time ity of our findings. Including different types of control conditions
children are engaged in academic subjects,67 with better academic would require additional comparisons, for which we would like to
performance in the long term. refer to a recent meta-analysis by Vazou et al.22
For longitudinal physical activity programs (i.e. continuous
regular physical activity over several weeks), a positive small 5. Conclusions
to moderate effect on executive functions (ES = 0.24) was found,
which is slightly lower than the positive small to moderate effect Based on the results of the current meta-analysis positive effects
on executive functions (ES = 0.46) found in a previous meta-analysis were found for both acute physical activity as well as for longi-
including both children and adolescents.22 Verburgh et al.12 found tudinal physical activity programs on cognitive functions (in the
no significant effect of longitudinal physical activity programs current study a combined effect of the domains executive functions,
(s = 3), which might be explained by the inclusion of a larger num- attention and academic performance) in preadolescent children.
ber of studies in the current meta-analysis (s = 12). With regard to The positive effects of acute physical activity were only found for
academic performance, a small to moderate effect of longitudinal attention, while the positive effects of longitudinal physical activity
physical activity programs was found (ES = 0.26). These effects are programs were consistent for all domains. The results indicate that
similar with the small to moderate effects found by Fedewa and benefits are largest for continuous cognitively engaging physical
Ahn4 (ES = 0.27). There is an increasing pressure to reduce time activity over several weeks.
in the school curriculum for physical education, in order to make
room for learning academic skills such as spelling, mathematics Practical implications
and reading. It is therefore important to acknowledge the positive
effect of physical activity on academic performance.68 Interest- - Single bout of physical activity can be a successful strategy to
ingly, the meta-analysis by Fedewa and Ahn4 found the highest stimulate attention, but not executive functions or academic per-
effect on mathematics and closely followed by reading, while no formance in children between the age of 6 and 12 years.
effect was found on any of the subdomains (mathematics, read- - Intervention programs that implement continuous regular phys-
ing and spelling) in the current meta-analysis. It seems unlikely ical activity over several weeks are more likely to improve
that this difference is explained because of an age difference, since executive functions and academic performance than a single bout
Fedewa and Ahn4 found the highest ESs on academic performance of physical activity.
for children at the elementary school level, compared with the - Interventions programs over several weeks that include cogni-
middle and high school levels. By focusing only on preadolescent tively challenging physical activities seem to be more effective in
children, the number of studies per subdomain of academic perfor- improving cognitive performance than aerobic physical activity.
mance in the current meta-analysis was low. Therefore, the results
of the subdomains of academic performance should be interpreted Acknowledgments
with caution and possibly explaining the differences between the
current results and the ones found by Fedewa and Ahn.4 This study was supported by a grant from the Netherlands Ini-
Interestingly, the longitudinal physical activity programs that tiative for Education Research (405-15-410) and the Dutch Brain
focused on cognitively engaging physical activity (s = 11) showed Foundation. The authors would like to thank all the researchers
a moderate to large positive effect (ES = 0.53), while aerobic phys- who provided additional data of their studies and Elsje de Vries for
ical activity showed a small to moderate positive effect (ES = 0.29). her work during the search process.
For cognitively engaging physical activity, a child has to allocate a
high amount of attention and cognitive effort.23 It is thought that by Appendix A. Supplementary data
specifically selecting physical activities with a relatively high cogni-
tive engagement, performance on cognitive functions will increase. Supplementary data associated with this article can be found, in
Although both types of physical activity showed beneficial effects, the online version, at https://doi.org/10.1016/j.jsams.2017.09.595.
the current findings suggests that physical activity with a higher
cognitive engagement has more effect on executive functions com-
pared to aerobic physical activity. This finding is supported by the References
1
recent meta-analysis that compared the effects of different types of
longitudinal physical activity programs on cognitive functioning in 1. van Stralen MM, Yıldırım M, Wulp A et al. Measured sedentary time and phys-
children and adolescents.22 This makes cognitively engaging phys- ical activity during the school day of European 10-to 12-year-old children: the
ENERGY project. J Sci Med Sport 2014; 17(2):201–206.
ical activity an interesting focus for future intervention programs.
2. Bailey DP, Boddy LM, Savory LA et al. Associations between cardiorespiratory
Some limitations should be considered before drawing con- fitness, physical activity and clustered cardiometabolic risk in children and ado-
clusions. First, although different types of moderators that might lescents: the HAPPY study. Eur J Pediatr 2012; 171(9):1317–1323.
3. Boddy LM, Murphy MH, Cunningham C et al. Physical activity, cardiorespiratory
have influenced the size of the effects of physical activity were
fitness, and clustered cardiometabolic risk in 10- to 12-year-old school children:
investigated, there are additional moderators not included in the the REACH Y6 study. Am J Human Biol 2014; 26(4):446–451.
current meta-analysis that might have an effect on the physiologi- 4. Fedewa AL, Ahn S. The effects of physical activity and physical fitness on chil-
cal responses of physical activity. For example, it is recommended dren’s achievement and cognitive outcomes: A meta-analysis. Res Q Exerc Sport
2011; 82(3):521–535.
that future studies test different exercise intensities and initial fit- 5. Colcombe SJ, Kramer AF. Fitness effects on the cognitive function of older adults:
ness levels. It has been suggested that there is an inverted-U effect a meta-analytic study. Psychol Sci 2003; 14(2):125–130.
with MVPA showing the largest effect on cognition, whereas light
and vigorous physical activity showing smaller effects.69 Neverthe-
less, almost all studies that focused on the effects of physical activity 1
Asterisks have been added to those references referring to studies included in
on cognitive performance in children were conducted with mod- the current meta-analysis.
 J.W. de Greeff et al. / Journal of Science and Medicine in Sport 21 (2018) 501–507 507

6. Miyake A, Friedman NP, Emerson MJ et al. The unity and diversity of executive 39. *Best JR. Exergaming immediately enhances children’s executive function. Dev
functions and their contributions to complex frontal lobe tasks: a latent variable Psychol 2012; 48(5):1501–1510.
analysis. Cogn Psychol 2000; 41(1):49–100. 40. *Chen A, Yan J, Yin H et al. Effects of acute aerobic exercise on multiple
7. Diamond A. Executive functions. Annu Rev Psychol 2013; 64:135–168. aspects of executive function in preadolescent children. Psychol Sport Exerc 2014;
8. Chang Y, Labban J, Gapin J et al. The effects of acute exercise on cognitive per- 15(6):627–636.
formance: a meta-analysis. Brain Res 2012; 1453:87–101. 41. *Drollette ES, Scudder MR, Raine LB et al. Acute exercise facilitates brain func-
9. Gerrig RJ, Zimbardo PG. Psychology and life, Boston, USA, Pearson Education, tion and cognition in children who need it most: an ERP study of individual
2010. differences in inhibitory control capacity. Dev Cogn Neurosci 2014; 7:53–64.
10. Friedman NP, Haberstick BC, Willcutt EG et al. Greater attention problems during 42. *Duncan M, Johnson A. The effect of differing intensities of acute cycling on
childhood predict poorer executive functioning in late adolescence. Psychol Sci preadolescent academic achievement. Eur J Sport Sci 2014; 14(3):279–286.
2007; 18(10):893–900. 43. *Gallotta MC, Emerenziani GP, Franciosi E et al. Acute physical activity and
11. Purdy MH. Executive functions: theory, assessment, and treatment, in Cognitive delayed attention in primary school students. Scand J Med Sci Sports 2015;
communication disorders, 1st ed., Kimbarow ML, editor, San Diego, CA, Plural 25(3):e331–e338.
Publishing cop, 2011, p. 77–93. 44. *Hill L, Williams JH, Aucott L et al. Exercising attention within the classroom.
12. Verburgh L, Königs M, Scherder EJ et al. Physical exercise and executive functions Dev Med Child Neurol 2010; 52(10):929–934.
in preadolescent children, adolescents and young adults: a meta-analysis. Br J 45. *Hillman CH, Pontifex MB, Raine LB et al. The effect of acute treadmill walk-
Sports Med 2014; 48(12):973–979. ing on cognitive control and academic achievement in preadolescent children.
13. Tomporowski PD. Effects of acute bouts of exercise on cognition. Acta Psychol Neuroscience 2009; 159(3):1044–1054.
2003; 112(3):297–324. 46. *Howie EK, Schatz J, Pate RR. Acute effects of classroom exercise breaks on exec-
14. Audiffren M. Acute exercise and psychological functions: a cognitive-energetic utive function and math performance: a dose-response study. Res Q Exerc Sport
approach, in Exercise and cognitive function, 1st ed., McMorris T, Tomporowski 2015; 86(3):217–224.
PD, Audiffren M, editors, Oxford, Wiley Online Library, 2009, p. 3–39. 47. *Jäger K, Schmidt M, Conzelmann A et al. Cognitive and physiological effects
15. Roig M, Nordbrandt S, Geertsen SS et al. The effects of cardiovascular exercise of an acute physical activity intervention in elementary school children. Front
on human memory: a review with meta-analysis. Neurosci Biobehav Rev 2013; Psychol 2014; 5:1473.
37(8):1645–1666. 48. *Niemann C, Wegner M, Voelcker-Rehage C et al. Influence of acute and chronic
16. Dishman RK, Berthoud H, Booth FW et al. Neurobiology of exercise. Obesity 2006; physical activity on cognitive performance and saliva testosterone in preado-
14(3):345–356. lescent school children. Ment Health Phys Act 2013; 6(3):197–204.
17. Etnier JL, Salazar W, Landers DM et al. The influence of physical fitness and 49. *Pirrie AM, Lodewyk KR. Investigating links between moderate-to-vigorous
exercise upon cognitive functioning: a meta-analysis. J Sport Exerc Psychol 1997; physical activity and cognitive performance in elementary school students. Ment
19:249–277. Health Phys Act 2012; 5(1):93–98.
18. Isaacs KR, Anderson BJ, Alcantara AA et al. Exercise and the brain: angiogenesis in 50. *Schmidt M, Egger F, Conzelmann A. Delayed positive effects of an acute
the adult rat cerebellum after vigorous physical activity and motor skill learning. bout of coordinative exercise on children’s attention. Percept Mot Skills 2015;
J Celeb Blood Flow Metab 1992; 12(1):110–119. 121(2):431–446.
19. Hillman CH, Buck SM, Themanson JR et al. Aerobic fitness and cognitive develop- 51. *Schmidt M, Benzing V, Kamer M. Classroom-based physical activity breaks and
ment: event-related brain potential and task performance indices of executive children’s attention: cognitive engagement works! Front Psychol 2016; 7:1474.
control in preadolescent children. Dev Psychol 2009; 45(1):114–129. 52. *Thompson HR, Duvall J, Padrez R et al. The impact of moderate-vigorous
20. *Schmidt M, Jäger K, Egger F et al. Cognitively engaging chronic physical activity, intensity physical education class immediately prior to standardized testing on
but not aerobic exercise, affects executive functions in primary school children: student test-taking behaviors. Ment Health Phys Act 2016; 11:7–12.
a group-randomized controlled trial. J Sport Exerc Psychol 2015; 37(6):575–591. 53. *Tine MT, Butler AG. Acute aerobic exercise impacts selective attention: an
21. Pesce C. Shifting the focus from quantitative to qualitative exercise char- exceptional boost in lower-income children. Educ Psychol 2012; 32(7):821–834.
acteristics in exercise and cognition research. J Sport Exerc Psychol 2012; 54. *Tomporowski PD, Davis CL, Lambourne K et al. Task switching in over-
34(6):766–786. weight children: effects of acute exercise and age. J Sport Exerc Psychol 2008;
22. Vazou S, Pesce C, Lakes K et al. More than one road leads to Rome: a narrative 30(5):497–511.
review and meta-analysis of physical activity intervention effects on cognition 55. *Ahamed Y, Macdonald H, Reed K et al. School-based physical activity does
in youth. Int J Sport Exerc Psychol 2016:1–26. not compromise children’s academic performance. Med Sci Sports Exerc 2007;
23. Tomporowski PD, McCullick B, Pendleton DM et al. Exercise and children’s cog- 39(2):371–376.
nition: the role of exercise characteristics and a place for metacognition. J Sport 56. *Chaddock-Heyman L, Erickson KI, Voss MW et al. The effects of physical activity
Health Sci 2015; 4(1):47–55. on functional MRI activation associated with cognitive control in children: a
24. Sibley BA, Etnier JL. The relationship between physical activity and cognition in randomized controlled intervention. Front Hum Neurosci 2013; 7(72):1–13.
children: a meta-analysis. Pediatr Exerc Sci 2003; 15(3):243–256. 57. *Crova C, Struzzolino I, Marchetti R et al. Cognitively challenging physical
25. *Altenburg TM, Chinapaw MJ, Singh AS. Effects of one versus two bouts of mod- activity benefits executive function in overweight children. J Sports Sci 2014;
erate intensity physical activity on selective attention during a school morning 32(3):201–211.
in Dutch primary schoolchildren: a randomized controlled trial. J Sci Med Sport 58. *Dalziell A, Boyle J, Mutrie N. Better movers and thinkers (BMT): an exploratory
2016; 19(10):820–824. study of an innovative approach to physical education. Eur J Psychol 2015;
26. *Hillman CH, Pontifex MB, Castelli DM et al. Effects of the FITKids random- 11(4):722–741.
ized controlled trial on executive control and brain function. Pediatr 2014; 59. *Davis CL, Tomporowski PD, McDowell JE et al. Exercise improves executive
134(4):e1063–e1071. function and achievement and alters brain activation in overweight children: a
27. Diamond AB. The cognitive benefits of exercise in youth. Curr Sports Med Rep randomized, controlled trial. Health Psychol 2011; 30(1):91–98.
2015; 14(4):320–326. 60. *Fisher A, Boyle JM, Paton JY et al. Effects of a physical education intervention on
28. Slavin RE, Lake C, Groff C. Effective programs in middle and high school mathe- cognitive function in young children: randomized controlled pilot study. BMC
matics: a best-evidence synthesis. Rev Educ Res 2009; 79(2):839–911. Pediatr 2011; 11:97.
29. Harris KC, Kuramoto LK, Schulzer M et al. Effect of school-based physical activ- 61. *Gallotta MC, Emerenziani GP, Iazzoni S et al. Impacts of coordinative training on
ity interventions on body mass index in children: a meta-analysis. CMAJ 2009; normal weight and overweight/obese children’s attentional performance. Front
180(7):719–726. Hum Neurosci 2015; 9:577.
30. Moher D, Liberati A, Tetzlaff J et al. Preferred reporting items for system- 62. *Kamijo K, Pontifex MB, O’Leary KC et al. The effects of an afterschool physical
atic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; activity program on working memory in preadolescent children. Dev Sci 2011;
6(7):e1000097. 14(5):1046–1058.
31. Ellemberg D, St-Louis-Deschênes M. The effect of acute physical exercise on cog- 63. *Koutsandreou F, Wegner M, Niemann C et al. Effects of motor versus cardio-
nitive function during development. Psychol Sport Exerc 2010; 11(2):122–126. vascular exercise training on children’s working memory. Med Sci Sports Exerc
32. Maher CG, Sherrington C, Herbert RD et al. Reliability of the PEDro scale for 2016; 48(6):1144–1152.
rating quality of randomized controlled trials. Phys Ther 2003; 83(8):713–721. 64. *Krafft CE, Schwarz NF, Chi L et al. An 8-month randomized controlled exercise
33. Armijo-Olivo S, da Costa BR, Cummings GG et al. PEDro or Cochrane to assess trial alters brain activation during cognitive tasks in overweight children. Obesity
the quality of clinical trials? A meta-epidemiological study. PLoS One 2015; 2014; 22(1):232–242.
10(7):e0132634. 65. *Krafft CE, Pierce JE, Schwarz NF et al. An eight month randomized controlled
34. Diamond A, Barnett WS, Thomas J et al. Preschool program improves cognitive exercise intervention alters resting state synchrony in overweight children. Neu-
control. Science 2007; 318(5855):1387–1388. roscience 2014; 256:445–455.
35. Miller EK, Cohen JD. An integrative theory of prefrontal cortex function. Annu 66. *van der Niet AG, Smith J, Oosterlaan J et al. Effects of a cognitively demanding
Rev Neurosci 2001; 24(1):167–202. aerobic intervention during recess on children’s physical fitness and executive
36. Janssen M, Toussaint HM, van Mechelen W et al. Effects of acute bouts of physical functioning. Pediatr Exerc Sci 2016; 28(1):64–70.
activity on children’s attention: a systematic review of the literature. Springer- 67. Duncan GJ, Dowsett CJ, Claessens A et al. School readiness and later achievement.
Plus 2014; 3(1):410. Dev Psychol 2007; 43(6):1428.
37. Cohen J. Statistical power analysis for the behavioral sciences, 2nd ed. Hillsdale, 68. Wilkins J, Graham G, Parker S et al. Time in the arts and physical education and
NJ, Lawrence Erlbaum Associates, 1988. school achievement. J Curriculum Stud 2003; 35(6):721–734.
38. Higgins J, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat Med 69. McMorris T, Hale BJ. Differential effects of differing intensities of acute exercise
2002; 21(11):1539–1558. on speed and accuracy of cognition: a meta-analytical investigation. Brain Cogn
2012; 80(3):338–351.