Journal of Applied Research in Memory and Cognition 7 (2018) 261–269

Contents lists available at ScienceDirect

Journal of Applied Research in Memory and Cognition

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

Sweat So You Don’t Forget: Exercise Breaks During a University

Lecture Increase On-Task Attention and Learning

Barbara Fenesi, Kristen Lucibello, Joseph A. Kim, Jennifer J. Heisz∗

McMaster University, Canada

We examined the impact of taking exercise breaks, non-exercise breaks, or no breaks on learning among
first year Introductory Psychology students. Three 5-minute breaks were equally distributed throughout a 50-
minute computer-based video lecture. The exercise breaks group performed a series of callisthenic exercises; the
non-exercise breaks group played a computer game; the no breaks group watched the lecture without breaks.
Mind-wandering questions measured attention during the lecture. Exercise breaks promoted attention through-
out the lecture compared to no breaks and non-exercise breaks, and resulted in superior learning when assessed
on immediate and delayed tests. The exercise breaks group also endorsed higher ratings for narrator clarity and
perceived understanding than the other two groups. This is the first study to show that exercise breaks promote
attention during lecture and improve learning in university students.

General Audience Summary

We tested the use of exercise breaks as a tool to focus student attention and promote learning. In children,
exercise breaks reduce off-task physical behaviors like fidgeting and shifting, while also improving academic
performance. However, it is unclear whether the benefits of exercise breaks extend to older students who may
manifest off-task behaviors in less overt ways, such as through mind wandering away from the primary task.
Incorporating exercise breaks into a university lecture improved on-task attention and memory for lecture
material compared to non-exercise breaks or no breaks. Learners who received exercise breaks found the
lecturer and lecture material easier to understand than did learners who received no breaks. Engaging minds
through classroom learning and practice is a direct route to promoting academic performance, and physical
exercise breaks may augment these processes.

Keywords: Physical activity, Mind wandering, Memory, Academic performance, Cognition

We remember what we attend to but our attention has a limited on a lecture declines as the lecture proceeds (Bunce, Flens, &
capacity (Engle, 2002). The appropriate allocation of attention Neiles, 2010). The net effect of reduced on-task attention dur-
during an effortful task, such as listening to a lecture, is criti- ing a lecture is reduced memory performance, which can impair
cal for durable learning to occur (Shiffrin & Schneider, 1977). academic performance (Wammes, Seli, Cheyne, Boucher, &
However, as time on task increases, minds naturally wander Smilek, 2016). As lectures remain the primary point of con-
from a selected activity; not surprisingly, the tendency to focus tent delivery in most university courses, determining strategies

Author Note
Barbara Fenesi, Kristen Lucibello and Jennifer J. Heisz, Department of from McMaster University’s Science Media Lab for help with experimental
Kinesiology, McMaster University, Hamilton, Canada; Joseph A. Kim, Depart- materials. Funding for this project was provided by Social Sciences and Human-
ment of Psychology, Neuroscience & Behavior, McMaster University, Hamilton, ities Research Council of Canada (SSHRC #430-2017-00616) to JJH.
Canada. ∗ Correspondence concerning this article should be addressed to Jennifer J.

The authors would like to thank Michelle Ogrodnik, Kaaran Gupta and Amy Heisz, Department of Kinesiology, McMaster University, 1280 Main Street
Pachai for invaluable contributions during data collection, as well as the Intro- West, Hamilton, ON, Canada.
ductory Psychology course coordinator, Michelle Cadieux, and Greg Atkinson
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 262

to sustain attention throughout a lecture period may promote bet-

ter learning. Exercise breaks incorporated into teaching periods
may be one such promising intervention.
In elementary school classrooms, exercise breaks reduce off-
task physical behaviors like fidgeting and shifting (Janssen et al.,
2014; Ma, Mare, & Gurd, 2014) while improving math flu-
ency (Howie, Schatz, & Pate, 2015; Vazou & Skrade, 2016) and Figure 1. Experimental design. The lecture was divided into three parts. Partic-
language skills (Martin & Murtagh, 2015; Mullender-Wijnsma ipants watched each part of the lecture, followed by a mind-wandering question
et al., 2016). These off-task behaviors are presumed to be a phys- (MW), a break (unless part of the no breaks group) and a measure of heart rate
(HR). This process was repeated until the lecture was complete, followed by
ical manifestation of waning attention and a reduction in such
an immediate learning test. After 48 h participants returned and completed a
behaviors following exercise represents the refocusing of atten- delayed learning test.
tion on classroom instruction (Carriere, Seli, & Smilek, 2013).
Although these studies position exercise breaks as a promising would thus be less effective at supporting attention and memory
intervention to promote learning, research making causal con- for the lecture material.
nections has been limited. First, there is currently no evidence The present study examined the attention and learning impact
that directly associates a reduction in off-task behaviors follow- of three 5-minute breaks distributed throughout a 50-minute
ing exercise breaks to improved learning of instructional content; computer-based video lecture taken from a post-secondary
thus, the link between refocused attention and improved learning Introductory Psychology course. The exercise breaks group per-
has not been fully investigated. Second, previous studies have formed a series of callisthenic exercises; the non-exercise breaks
evaluated learning gains from exercise breaks using experiment- group played a simple visual matching computer game (Bejew-
based tests such mathematics speed tests, spelling tests, and eled), which acted as a sham break to control for the level
reading tests (Ma et al., 2014; Mullender-Wijnsma et al., 2016). of arousal. The no breaks group watched the lecture contin-
Although these measures provide insight into the ability of exer- uously without breaks. Figure 1 provides a schematic of the
cise breaks to foster general academic skills and knowledge, they experimental design. The lecture was divided into three parts.
do not directly evaluate how exercise breaks promote learning For each part, participants watched the lecture and answered a
of authentic instructional material. Third, research on exercise mind-wandering question to gauge on-task attention. Then they
breaks has been limited to school-age children; it is unclear completed the experimental break (unless part of the no breaks
whether the benefits of exercise breaks extend to older students group) and provided a heart rate measure as an index of phys-
who may manifest off-task behaviors in less overt ways, such as iological arousal. We hypothesized that exercise breaks during
through mind wandering (i.e., the shifting of attention away from learning of university-level educational content would heighten
the primary task and toward internal cognitions; Smallwood, physiological arousal to promote on-task attention and improve
Fishman, & Schooler, 2007; Wammes, Seli, et al., 2016). learning compared to non-exercise breaks and no breaks.
Converging lines of evidence further support a role for
exercise in promoting general cognitive function and aca- Method
demic learning. An acute bout of exercise prior to completing
a task promotes executive functioning and working memory Participants
(Chang, Labban, Gapin, & Etnier, 2012), along with imme- The a priori recommended sample size was 75 participants
diate and long-term memory for basic motor skills (Roig, based on .90 power estimate for a small-to-medium effect
Skriver, Lundbye-Jensen, Kiens, & Nielsen, 2012), word lists size (Subramaniapillai, Tremblay, Grassmann, Remington, &
(Coles & Tomporowski, 2008), and picture–location associ- Faulkner, 2016); two additional participants were recruited (for a
ations (van Dongen, Kersten, Wagner, Morris, & Fernández, total of 77) due to missing cells. Thus, 77 undergraduate students
2016). With respect to academic learning, students of all ages at McMaster University enrolled in an Introductory Psychology
who have a higher level of physical fitness also have higher aca- course (78% females; age M ± SD = 18.7 ± 1.4 years; 18–22yrs
demic achievement both in the classroom and on standardized old) were recruited from an online recruitment portal and were
tests (Carlson et al., 2008; Grissom, 2005; Singh, Uijtdewil- pre-screened for prior exposure to lecture content. Only partic-
ligen, Twisk, Van Mechelen, & Chinapaw, 2012). Although ipants without prior exposure to the lecture content could enlist
controlled studies that test memory for complex educational in the study. The study was conducted at McMaster Univer-
material are lacking, an acute bout of exercise does have sity. All participants provided informed consent (study approved
direct effects on the brain by increasing arousal via the release by McMaster Research Ethics Board, # 2014 131) and were
of stress hormones known to improve attention and memory compensated with course credit for their participation.
(McEwen, 2007). Exercise-induced increases in epinephrine
may improve on-task attention during the lecture, while activa-
Materials
tion of the hypothalamic-pituitary-adrenal (HPA) axis leading to
increases in cortisol may improve memory for the lecture mate- Lecture. Participants watched a condensed 50-min
rial (Brisswalter, Collardeau, & René, 2002). This cascade of computer-based lecture about form perception, corresponding
arousal responses to physical exercise is expected to be absent to about one week’s worth of content and taken directly from
in non-exercise breaks (such as playing a computer game) which the Introductory Psychology course. The course curriculum
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 263

includes both in-class and online components and thus the Perceptions. Immediately after viewing the lecture, partici-
lecture used in this study represents how the course material pants provided subjective ratings of the lecture with respect to
is authentically delivered. Participants were accustomed to their perceptions of difficulty, engagement, interest, understand-
viewing such online lectures; however, they had not previously ing of the content, and narrator clarity using a 7-item Likert scale
viewed this particular lecture, which was scheduled to appear (see Appendix 1 for description of subjective measures). Two
later in the course. questions were used to assess each construct, and the average rat-
Exercise and non-exercise breaks. The exercise breaks ing was used in the analysis. Prior research examining exercise
consisted of five 50-s experimenter-led callisthenic exercises breaks during classroom instruction reported that children per-
(jumping jacks, heeltaps, high knees, split jumps, and hamstring ceived the classroom time as being more enjoyable, and teachers
kickers) separated by a 10-s rest. Since the experimenter had to perceived greater participation and engagement from their stu-
lead the exercise sessions, this precluded the study from being dents following the exercise breaks (Camahalan & Ipock, 2015;
blinded. This was done to ensure consistency in data collection Mulrine, Prater, & Jenkins, 2008). The novelty of the present
across conditions. study is that we evaluated both perceptions and performance
The non-exercise breaks consisted of playing the computer indicators to provide a more comprehensive assessment of learn-
game, Bejeweled, on the same computer as the video lecture ing (Moullin, 2004). Exercise breaks were expected to produce
for an equal time period as the exercise breaks. In pilot test- a better learning experience as evidenced by higher perception
ing (N = 15), Bejeweled was rated as the most enjoyable among ratings.
four computer game options (Bejeweled = 39%, Pacman = 25%, Physical activity level. Participants completed the Physi-
Tetris = 19%, Pong = 17%). Other non-exercise breaks, such as cal Activity and Sedentary Behaviour Questionnaire (Tremblay
web browsing, viewing social media, or listening to music were et al., 2007) to verify no group differences in physical activity
disqualified because of the potentially arousing effects of the prior to the study. Physical activity was derived from partici-
emotional content (Scherer & Coutinho, 2013) and variance pants’ response to the following question: “How many minutes
in enjoyment across participants. Sitting silently was also dis- per week do you engage in moderate-to-vigorous physical activ-
qualified as a non-exercise break because of the potential for ity?”
rumination on the lecture material or life events (Schwarzer,
1996). Procedure
Arousal. A pilot protocol was conducted to assess the accu-
racy of carotid artery monitoring by pairing it with a heart rate Participants were randomly assigned to exercise breaks
(HR) monitor and comparing the two values. There was no (N = 26), non-exercise breaks (N = 23), or no breaks (N = 28).
significant difference between the values provided by the two We randomized using the free online randomizer tool (random-
methods, t(5) = 0.16, p = .88. Given that HR monitors require an ization.com), which applies the method of randomly permuted
under-the-clothes chest band and a properly synced wrist moni- blocks. Figure 1 provides a schematic of the experimental
tor, the carotid artery method was chosen to minimize participant design. During each experimental session, approximately four
burden and optimize data collection efficiency. participants, who were assigned to the same experimental group,
watched the computer-based form perception lecture using
Attention. We used a validated mind-wandering question
individual laptops with attached headsets. For exercise or non-
for classroom research (Wammes, Boucher, Seli, Cheyne, &
exercise breaks groups, three 5-minute breaks were distributed
Smilek, 2016). The question read, “Which of the following
approximately every 17 min. This time interval structure is sim-
responses best characterizes your mental state just before this
ilar to work done by Ma et al. (2014), where 4-min exercise
screen appeared?” The answer options were (1) on task, (2)
breaks were introduced after 20 min of classroom instruction.
intentionally mind wandering, or (3) unintentionally mind wan-
Participants in the no breaks group watched the lecture continu-
dering. There were a total of three mind-wandering questions:
ously without a break. A mind wandering question was presented
baseline (after the first part of the lecture), Time 1 (after the sec-
before each break or at a similar time point in the lecture for the
ond part), and Time 2 (after the final part). Prior work by Szpunar,
no breaks group.
Khan, and Schacter (2013) relied on a single mind-wandering
Immediately following each break and at the very beginning
question at the end of the experimental session. In the present
and end of the initial experimental session, participants mea-
study, the three mind-wandering questions were distributed
sured their heart rate (HR) by applying their middle and index
equally throughout the lecture to more accurately capture mind
fingers to their carotid artery for 15 s (note that participants were
wandering without relying on retrospective judgment. To min-
trained to use this method prior to the experiment). We multiplied
imize interference during learning, mind-wandering questions
the recorded value by four to determine the number of beats per
occurred prior to each break (except for controls, who did not
minute. The control group provided HR measures at similar time
receive a break but were administered the mind-wandering ques-
points as the exercise and non-exercise breaks groups, and this
tion at a similar time point during the lecture).
was done immediately after responding to each mind-wandering
Learning. Learning was evaluated using a 30-question question.
multiple-choice test that was answered immediately following Learning was assessed by a comprehension test that was
the lecture. Participants returned 48 h later to complete a delayed administered immediately after the lecture. The test questions
test with different questions, matched for content and difficulty. were drawn from the Introductory Psychology course test bank.
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 264

The lecture perceptions questionnaire was administered imme- Table 1

diately after the learning assessment, following which the last Arousal During Lecture
HR measure was collected. Delayed learning was assessed 48 h Exercise breaks Non-exercise breaks No breaks
later. The completion time for the control condition was approx- M (SD) M (SD) M (SD)
imately seventy minutes (50 min lecture; 20 min for the learning
HR Pre 84 (34) 77 (18) 85 (12)
test and questionnaires). The completion time for the exercise HR Break 1 136 (39)*** 75 (16) 83 (12)
and non-exercise breaks conditions was approximately ninety HR Break 2 145 (42)*** 73 (16) 80 (10)
minutes (50 min lecture; 15 min [3× 5 min] of breaks; 3–5 min HR Break 3 145 (41)*** 73 (12) 79 (11)
of total transition time between breaks and lecture; 20 min for HR Post 93 (34)* 76 (13) 78 (10)
the learning test and questionnaires). Note. Heart rate (HR) in beats per minute (bpm) was higher following each exer-
cise break compared to the other groups, and these group differences remained
until the end of the experimental session (HR post; main effects of group and
a group × time interaction: all p < .001). Parentheses denote standard deviation.
Statistical Analyses Independent samples t test revealed significant differences between the exercise
breaks group compared to the other groups, which were not significantly differ-
Software package SPSS Version 23 was used for analyses. ent from each other at any time point.
Arousal and learning were assessed by separate mixed-model *p < .05; ***p < .001.
ANOVAs with a between-subject factor of Group (exercise
breaks, non-exercise breaks, no breaks) and a within-subject Arousal
factor of Time. To measure off-task attention, unintentional and
Exercise breaks significantly increased arousal compared to
intentional mind wandering were collapsed for the analyses.
non-exercise breaks and no breaks (Table 1). The HR for the
Analyses focused on attention on-task (coded as 0) versus off-
exercise breaks increased to approximately 70% of the age-
task (both unintentional and intentional mind wandering, coded
predicted HR max, indicating a moderate-to-vigorous level of
as 1). The initial mind-wandering question, done prior to the
intensity for the exercise group. The HR of the exercise-breaks
experimental manipulation, was added as a covariate in a binary
group following each break was higher than the HR of the other
logistic regression model as a baseline measure of attention,
two groups. These group differences remained until the end of
with simple linear contrasts comparing groups. Both immedi-
the experimental session (i.e., after the learning test and ques-
ate and delayed learning tests had high internal consistency
tionnaires; see Table 1). Higher HR following each break in the
reliability scores (Immediate ωh = .96; Delayed ωh = .96). We
exercise breaks group was supported by main effects of Group,
also conducted a 3 × 2 repeated-measures ANCOVA to deter-
F(2, 72) = 28.60 p < .001, η2p = .44, and Time, F(4, 288) = 65.73,
mine whether the benefit of exercise breaks emerged at different
stages in the lecture (with a between-subject factor of Group and p < .001, η2p = .48, as well as a Group × Time interaction, F(8,
a within-subject factor of Lecture Time; the outcome measure 288) = 87.84, p < .001, η2p = .71. Post hoc independent-samples
was performance on the immediate test questions that evaluated t tests confirmed that the exercise breaks group had higher HRs
the material that was presented after each exercise break). We compared to the other two groups following the first exercise
controlled for any baseline differences in learning across groups break until the end of experimental session (see Table 1). In con-
by adding performance from the first part of the lecture as a trast, the non-exercise breaks group and no breaks group were
covariate. not significantly different form each other at any time point, all
To directly test our hypothesis that heightened physiologi- p > .05.
cal arousal from exercise would promote on-task attention and
improve learning, we conducted Spearman’s correlations to test Attention
the strength of the association of arousal following the exercise Table 2 and Figure 2A depict the attention results. As the
breaks with the measures of attention and learning. Subjec- lecture progressed, exercise breaks promoted greater on-task
tive perception ratings were assessed with separate one-way attention than non-exercise breaks or no breaks (Figure 2A).
ANOVAs with a between-subject factor of Group. When time points were analyzed by separate binary logistic
regressions, there were no group differences following the first
Results mind-wandering question (Time 1; p > .05). Group differences
were observed at the second mind-wandering question (Time
Manipulation Check 2), with the exercise breaks group engaging in more on-task
Spearman correlation confirmed that the pairs of perception attention than the non-exercise breaks group, OR = 6.59, 95%
questions were significantly correlated, all p < .004; all rs > .36; CI = 1.77, 24.50, p < .01, and the no breaks group, OR = 3.98,
all r2 > .13, and the average score between perception questions 95% CI = 1.22, 12.96, p < .05.
was computed. There were no group differences in physical
activity among the groups, F(2, 73) = 0.82, p = .44, confirm- Learning
ing that individuals of similar levels of physical activity were Table 3 and Figure 2B depict the learning results. Exercise
assigned to each group. All analyses met assumptions of nor- breaks resulted in better test performance on both immediate
mality. and delayed assessments than non-exercise breaks or no breaks
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 265

Table 2
Descriptive Statistics for Mind Wandering Questions Across Groups

Exercise breaks Non-exercise breaks Control

Baseline T1 T2 Baseline T1 T2 Baseline T1 T2

N = 25 N = 25 N = 25 N = 22 N = 22 N = 22 N = 28 N = 28 N = 27

On task 19 18 18 12 16 7 19 16 11
Off task 6 7 7 10 6 15 9 12 16
Unintentional 6 6 5 7 6 10 9 11 13
Intentional 0 1 2 3 0 5 0 1 3

Note. Collapsed mind wandering is represented by number of participants endorsing on-task versus off-task attention at each mind-wandering question: Baseline
(covariate), Time 1 (T1), and Time 2 (T2). Pre-collapsed mind wandering is represented by number of people endorsing each mind-wandering options.

Figure 2. Attention during lecture and learning of the lecture material. (A) As the lecture progressed, exercise breaks prevented a reduction in on-task attention
compared to non-exercise breaks or no breaks. Proportion of participants reporting on-task attention controlling for baseline attention was used to illustrate attention
differences across groups, although statistical analyses were conducted on dichotomous mind-wandering values (0 = on-task, 1 = off-task). Error bars represent ± SE.
(B) Exercise breaks resulted in better test performance on both immediate and delayed assessments than non-exercise breaks or no breaks. For illustrative purposes,
immediate and delayed test performances were normalized into a z score using the mean (46.1%) and standard deviation (10.5%) for the immediate test. This was
done to normalize performance to the average so that it would improve generalizability and also to anchor the delayed performance to the immediate test. Error bars
represent ± 1 SE.

(Figure 2B). This was supported by a main effect of Group, F(2, points of the lecture across groups (controlling for baseline dif-
73) = 9.34, p < .001, η2p = .20. Post hoc independent-samples t ferences in learning from Lecture Part 1). There was a main
tests confirmed that group differences were driven by higher effect of Group, F(2, 73) = 4.82, p = .011, η2p = .12, as previously
test scores for exercise breaks compared to non-exercise breaks, revealed by the omnibus ANOVA reported above. Importantly,
Cohen’s d = 1.1; t(47) = 4.07, p < .001, and no breaks, Cohen’s there was no effect of Time, F(1, 73) = 0.93, p = .34, η2p = .13,
d = 1.2; t(51) = 3.90, p < .001, with a large magnitude of effect. or interaction between Group and Time, F(2, 73) = 0.98, p = .38,
Test scores for the non-exercise breaks and no breaks groups η2p = .26.
were not significantly different from each other, t(48) = 0.25, Spearman’s correlations (Table 4) revealed that heightened
p = .81. physiological arousal from exercise was associated with learn-
An ANCOVA was conducted to determining whether differ- ing as assessed by both immediate and delayed tests but was only
ences in learning on the immediate test emerged at different associated with on-task attention near the end of the lecture as
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 266

Table 3 and the no breaks group, t(46) = 2.49, p = .017. There were no
Descriptive Statistics for Untransformed Learning Outcomes group differences with respect to perceived difficulty, interest, or
Exercise breaks Non-exercise breaks Control engagement.

Immediate learning .51 (.05) .44 (.10) .43 (.13)

Lecture Part 1 .51 (.13) .49 (.16) .42 (.16) Discussion
Lecture Part 2 .55 (.19) .42 (.17) .49 (.19)
Lecture Part 3 .49 (.12) .40 (.13) .39 (.16) The present study demonstrates the benefits of exercise
Delayed learning .50 (.10) .42 (.12) .42 (.10) breaks on attention and learning using authentic course mate-
Note. Table 3 provides a breakdown of learning outcomes by lecture part for
rials with university students. Incorporating exercise breaks
immediate learning. Delayed learning was not assessed by lecture part because improved on-task attention and memory for lecture material
of the time delay between lecture exposure and testing, which would have compared to non-exercise breaks or no breaks. Furthermore,
confounded meaningful interpretations of the association between on-task atten- subjective perception ratings suggest that learners who received
tion and lecture learning. Learning is represented by proportion correct on the exercise breaks found the lecturer and lecture material easier to
learning assessment. Parentheses denote standard deviation.
understand than learners who received no breaks.
This is the first study to demonstrate that the benefits of
exercise breaks on learning extend to older students. Although
Table 4 consistent with prior findings with elementary school children
Spearman’s Correlations Following Exercise Breaks With Measures of Attention
and Learning
(Janssen et al., 2014), an important difference is that children’s
off-task behavior is typically more overt through actions such as
Attention Learning fidgeting (Ma et al., 2014; Mullender-Wijnsma et al., 2016) com-
pared to that of a trained adult learner who is more accustomed to
Baseline T1 T2 Immediate Delayed
proper classroom etiquette. The mind-wandering questions used
Arousal here provided a window into the covert off-task behavior of uni-
HR1 .09 −.03 .22 .28* .39** versity students (Wammes, Boucher, et al., 2016). Despite age
HR2 .03 −.003 .27* .33** .39**
HR3 −.05 −.04 .32** .30** .44**
differences in the manifestation of off-task attention, the results
suggest that exercise breaks improved attention and memory for
Note. Time 1 (T1), Time 2 (T2). these older learners as well.
*p < .05; **p < .01. On-task attention declined as the lecture progressed, but
only for the non-exercise and no breaks groups. The exercise
breaks buffered against declines in attention resulting in supe-
Table 5 rior engagement during the latter part of the lecture compared
Subjective Perception Ratings
to the other two groups. Although this suggests that exercise
Exercise breaks Non-exercise breaks No breaks breaks may not necessarily result in a direct boost for on-
task attention, the exercise breaks did seem to be beneficial for
Interest 5.2 (1.1) 4.8 (1.3) 4.4 (1.2)
learning by mitigating otherwise natural reductions in attention
Engagement 4.6 (1.0) 4.1 (1.2) 4.1 (1.2)
Narrator clarity 5.7 (1.0) 5.1 (1.1) 4.8 (1.2) as lecture length increased. This finding has important impli-
Difficulty 4.4 (1.1) 4.7 (1.3) 4.2 (1.2) cations for the integration of exercise breaks into classroom,
Understanding 4.6 (0.9) 4.0 (1.0) 3.8 (1.3) workplace, and training contexts suggesting that the greatest
benefit of exercise breaks is for long-duration learning or work
Note: Mean perceived interest, engagement, narrator clarity, difficulty and under-
standing ratings across conditions using Likert scale 1–7. Higher values indicate sessions during which attention becomes increasingly depleted
greater endorsement. Parentheses denote standard deviation. (Wammes, Boucher, et al., 2016).
This study underscores how connections between brain and
body physiology affects complex cognitive function. Spear-
man’s correlations showed that physiological arousal was
revealed by a significant correlation with mind-wandering ques-
associated with both on-task attention at the end of the lecture
tion at Time 2. There were no significant correlations between
and learning as assessed by both immediate and delayed tests.
attention and learning, all p > .3.
This finding lends support for the HPA activation hypothesis,
with exercise-induced arousal facilitating the release of stress
Subjective Perception hormones known to improve attention and memory (McEwen,
2007). While engaging minds through classroom learning,
Main effect of group was found for narrator clarity, F(2, study, and practice is a direct route to promoting academic
64) = 4.01, p = .02, η2p = .11, and perceived understanding, F(2, performance, physical exercise may augment these processes.
64) = 3.49, p = .037, η2p = .10 (Table 5). Post hoc indepen- Collectively engaging both the body and the mind during class-
dent-samples t tests revealed that the exercise breaks group room learning may have a direct positive impact on attentional
perceived the narrator to be clearer than the no breaks group, fortitude and educational outcomes.
t(46) = 2.77, p = .008, and their own understanding to be bet- An interesting observation was that non-exercise breaks did
ter than both non-exercise breaks group, t(37) = 2.19, p = .035, not provide additional improvement in attention or memory over
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 267

having no breaks at all. Although this is in line with our hypothe- In conclusion, this is the first study to demonstrate the ben-
sis that heightening physiological arousal from exercise is a key efits of exercise breaks on learning extend to older students
component in the initiation of refocused attention and learning, by inducing a state of physiological arousal promoting on-task
it is possible that other non-exercise breaks may be equally bene- attention and memory. Engaging students in physical exercise
ficial through alternative mechanisms. One possibility is that the breaks during classroom instruction may promote their ability
computer game played during the non-exercise break may have to optimally engage their attentional resources and thus become
acted as a second cognitive task as opposed to a cognitive break. better learners.
Switching between two cognitive tasks can deplete attention and
impair performance for both tasks (Rogers & Monsell, 1995; Conflict of Interest Statement
Wylie & Allport, 2000); thus, alternating between the lecture and
The authors declare no conflict of interest.
the cognitive game may not be equivalent to other non-exercise
breaks that are not as cognitively demanding, such as mind-
Author Contributions
fulness meditation (Lutz, Slagter, Dunne, & Davidson, 2008;
Semple, 2010). There is also some evidence that including low- All authors contributed equally to the generation of ideas,
stakes quizzes during a lecture can reduce mind wandering and experimental design, analysis, and reporting. KL led data col-
boost performance (Szpunar et al., 2013), thus offering an alter- lection.
native method for promoting attention and learning distinct from
exercise breaks. Appendix 1.
Another alternative explanation is that the change in context
Subjective Perception Measures
that was specific to the exercise breaks group (i.e., completely
moving away from the computer screen) may be a key con- Please rate your agreement with the following statements
tributor to helping learners sustain their attention and improve based on the scale indicated:
their comprehension rather than physiological changes due to
exercise. A non-exercise breaks condition that alters the break 1 = absolutely disagree
context so that attention is removed from the learning envi- 2 = mostly disagree
ronment (i.e., computer screen) but is also non-exercise in 3 = somewhat disagree
nature would evaluate this alternative hypothesis. Identifying 4 = neutral
non-exercise breaks that could produce similar attentional and 5 = somewhat agree
6 = mostly agree
learning gains would have a practical benefit for students or 7 = absolutely agree
learning environments under which physical exercise may be
restricted. Furthermore, the study’s findings have the poten-
tial to inform the structure of massive open online courses Perceived Interest
(MOOCs) that require students to view extended video lectures
on a computer screen. Perhaps breaking up content delivery with 1. I found the material presented in this lecture to be interesting
prompted short exercise breaks will foster learning in both the 2. The material captured my interest
classroom and online environments.
While this study provides an important step toward under- Perceived Understanding
standing how exercise breaks improve attention and learning in
1. I found that the presentation style helped me to understand
an academic setting, there are limitations which warrant fur-
the lecture material
ther study. It may not be feasible to perform these exercises in
2. I found that I had a meaningful understanding of the material
all university lectures, especially classrooms with space restric-
overall
tions. Future research should evaluate whether exercises that
require less physical space can generate similar attention and
Perceived Narrator Clarity
learning benefits. Also, determining the optimal dose of exer-
cise break (e.g., 1, 2, 3 breaks) and the interaction with lecture 1. The narrator clearly communicated the material
duration (e.g., 1-hour class, 3-hour class) will be essential for 2. I understood what the narrator was saying with minimal effort
incorporating exercise break prescriptions into the educational
curriculum. Although mind-wandering questions are the princi- Perceived Difficulty
ple research tool for capturing state attention, the limitation of
this introspective measure must be acknowledged. Furthermore, 1. The lecture material has a high level of difficulty
this sample consisted of more females than males. Although 2. The information was difficulty to understand
sex may impact the effect of exercise on physical outcomes
(Hands, Parker, Larkin, Cantell, & Rose, 2016), there is no Perceived Engagement
evidence to suggest that sex would alter the impact of exer-
cise on these cognitive outcomes in young adults (Hülya Aşçı, 1. I found the lecture engaging
2009). 2. The lecture captured my attention
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 268

References Mullender-Wijnsma, M. J., Hartman, E., de Greeff, J. W.,

Doolaard, S., Bosker, R. J., & Visscher, C. (2016). Physically
Brisswalter, J., Collardeau, M., & René, A. (2002). Effects of acute active math and language lessons improve academic achieve-
physical exercise characteristics on cognitive performance. Sports ment: A cluster randomized controlled trial. Pediatrics, 137(3),
Medicine, 32(9), 555–566. 1–9.
Bunce, D. M., Flens, E. A., & Neiles, K. Y. (2010). How long can Mulrine, C. F., Prater, M. A., & Jenkins, A. (2008). The active
students pay attention in class? A study of student attention decline classroom: Supporting students with attention deficit hyperactivity
using clickers. Journal of Chemical Education, 87(12), 1438–1443. disorder through exercise. Teaching Exceptional Children, 40(5),
Camahalan, F. M. G., & Ipock, A. R. (2015). Physical activity breaks 16–22.
and student learning: A teacher-research project. Education, 135(3), Rogers, R. D., & Monsell, S. (1995). Costs of a predictable switch
291–298. between simple cognitive tasks. Journal of Experimental Psychol-
Carlson, S. A., Fulton, J. E., Lee, S. M., Maynard, L. M., Brown, D. ogy: General, 124(2), 207–231.
R., Kohl, H. W., III, & Dietz, W. H. (2008). Physical education and Roig, M., Skriver, K., Lundbye-Jensen, J., Kiens, B., & Nielsen, J. B.
academic achievement in elementary school: Data from the early (2012). A single bout of exercise improves motor memory. PLoS
childhood longitudinal study. American Journal of Public Health, ONE, 7(9), e44594.
98(4), 721–727. Scherer, K. R., & Coutinho, E. (2013). How music creates emotion: A
Carriere, J. S., Seli, P., & Smilek, D. (2013). Wandering in both mind and multifactorial process approach. In The emotional power of music.
body: Individual differences in mind wandering and inattention pre- pp. 121–145.
dict fidgeting. Canadian Journal of Experimental Psychology/Revue Schwarzer, R. (1996). Thought control of action: Interfering self-
canadienne de psychologie expérimentale, 67(1), 19–31. doubts. In Cognitive interference: Theories, methods, and findings.
Chang, Y.-K., Labban, J., Gapin, J., & Etnier, J. L. (2012). The effects pp. 99–115.
of acute exercise on cognitive performance: A meta-analysis. Brain Semple, R. J. (2010). Does mindfulness meditation enhance
Research, 1453, 87–101. attention? A randomized controlled trial. Mindfulness, 1(2),
Coles, K., & Tomporowski, P. D. (2008). Effects of acute exercise on 121–130.
executive processing, short-term and long-term memory. Journal of Shiffrin, R. M., & Schneider, W. (1977). Controlled and automatic
Sports Sciences, 26(3), 333–344. human information processing: II. Perceptual learning, automatic
Engle, R. W. (2002). Working memory capacity as executive attention. attending and a general theory. Psychological Review, 84(2),
Current Directions in Psychological Science, 11(1), 19–23. 127–190.
Grissom, J. (2005). A study of the relationship between physical fitness Singh, A., Uijtdewilligen, L., Twisk, J. W., Van Mechelen, W., & Chi-
and academic achievement in California using 2004 test results. napaw, M. J. (2012). Physical activity and performance at school:
California Department of Education. A systematic review of the literature including a methodological
Hands, B., Parker, H., Larkin, D., Cantell, M., & Rose, E. (2016). quality assessment. Archives of Pediatrics & Adolescent Medicine,
Male and female differences in health benefits derived from physical 166(1), 49–55.
activity: Implications for exercise prescription. Journal of Women’s Smallwood, J., Fishman, D. J., & Schooler, J. W. (2007). Counting the
Health, Issues and Care, 5(4). cost of an absent mind: Mind wandering as an underrecognized influ-
Howie, E. K., Schatz, J., & Pate, R. R. (2015). Acute effects of class- ence on educational performance. Psychonomic Bulletin & Review,
room exercise breaks on executive function and math performance: 14(2), 230–236.
A dose–response study. Research Quarterly for Exercise and Sport, Subramaniapillai, M., Tremblay, L., Grassmann, V., Remington, G., &
86(3), 217–224. Faulkner, G. (2016). The effect of an acute bout of exercise on exec-
Hülya Aşçı, F. (2009). Sex differences in psychological effects of exer- utive function among individuals with schizophrenia. Psychiatry
cise. International Journal of Psychology, 44(4), 313–320. Research, 246, 637–643.
Janssen, M., Chinapaw, M., Rauh, S., Toussaint, H., van Mechelen, W., Szpunar, K. K., Khan, N. Y., & Schacter, D. L. (2013). Interpolated
& Verhagen, E. (2014). A short physical activity break from cog- memory tests reduce mind wandering and improve learning of online
nitive tasks increases selective attention in primary school children lectures. Proceedings of the National Academy of Sciences of the
aged 10–11. Mental Health and Physical Activity, 7(3), 129–134. United States of America, 110(16), 6313–6317.
Lutz, A., Slagter, H. A., Dunne, J. D., & Davidson, R. J. (2008). Atten- Tremblay, M. S., Shephard, R. J., Brawley, L. R., Cameron, C., Craig, C.
tion regulation and monitoring in meditation. Trends in Cognitive L., Duggan, M., . . . & Janssen, I. (2007). Physical activity guidelines
Sciences, 12(4), 163–169. and guides for Canadians: Facts and future. Applied Physiology,
Ma, J. K., Mare, L. L., & Gurd, B. J. (2014). Classroom-based high- Nutrition, and Metabolism, 32(S2E), S218–S224.
intensity interval activity improves off-task behaviour in primary van Dongen, E. V., Kersten, I. H., Wagner, I. C., Morris, R. G.,
school students. Applied Physiology, Nutrition, and Metabolism, & Fernández, G. (2016). Physical exercise performed four hours
39(12), 1332–1337. after learning improves memory retention and increases hippocam-
Martin, R., & Murtagh, E. M. (2015). Preliminary findings of active pal pattern similarity during retrieval. Current Biology, 26(13),
classrooms: An intervention to increase physical activity levels of 1722–1727.
primary school children during class time. Teaching and Teacher Vazou, S., & Skrade, M. A. (2016). Intervention integrating physi-
Education, 52, 113–127. cal activity with math: Math performance, perceived competence,
McEwen, B. S. (2007). Physiology and neurobiology of stress and and need satisfaction. International Journal of Sport and Exercise
adaptation: Central role of the brain. Physiological Reviews, 87(3), Psychology, 1–15.
873–904. Wammes, J. D., Boucher, P. O., Seli, P., Cheyne, J. A., & Smilek,
Moullin, M. (2004). Eight essentials of performance measurement. D. (2016). Mind wandering during lectures I: Changes in rates
International Journal of Health Care Quality Assurance, 17(3), across an entire semester. Scholarship of Teaching and Learning
110–112. in Psychology, 2(1), 13–32.
 EXERCISE BREAKS DURING A LECTURE INCREASE ATTENTION AND LEARNING 269

Wammes, J. D., Seli, P., Cheyne, J. A., Boucher, P. O., & Smilek, D. Received 25 October 2016;
(2016). Mind wandering during lectures II: Relation to academic
received in revised form 11 October 2017;
performance. Scholarship of Teaching and Learning in Psychology,
2(1), 33–48.
accepted 30 January 2018
Wylie, G., & Allport, A. (2000). Task switching and the measurement Available online 5 April 2018
of “switch costs”. Psychological Research, 63(3–4), 212–233.