High-Intensity Interval Training for Cognitive

and Mental Health in Adolescents

SARAH A. COSTIGAN1, NARELLE EATHER1, RONALD C. PLOTNIKOFF1, CHARLES H. HILLMAN2,
and DAVID R. LUBANS1
1
Priority Research Centre for Physical Activity and Nutrition, School of Education, University of Newcastle, Callaghan,
AUSTRALIA; and 2Department of Kinesiology and Community Health, University of Illinois, Urbana, IL

ABSTRACT
Downloaded from http://journals.lww.com/acsm-msse by BhDMf5ePHKav1zEoum1tQfN4a+kJLhEZgbsIHo4XMi0hCywCX1AWnYQp/IlQrHD3YiN3qTGLcvIZ/6VAa2fMHVdcmiSJwjN+y9WA9ZmGXzC4nZcNXxiyFA== on 09/14/2020

COSTIGAN, S. A., N. EATHER, R. C. PLOTNIKOFF, C. H. HILLMAN, and D. R. LUBANS. High-Intensity Interval Training for
Cognitive and Mental Health in Adolescents. Med. Sci. Sports Exerc., Vol. 48, No. 10, pp. 1985–1993, 2016. Purpose: Emerging
literature suggests that physical activity and fitness may have a positive effect on cognitive and mental health in adolescents. The purpose
of the current study was to evaluate the efficacy of two high-intensity interval training (HIIT) protocols for improving cognitive and
mental health outcomes (executive function, psychological well-being, psychological distress, and physical self-concept) in adolescents.
Methods: Participants (n = 65; mean age = 15.8 T 0.6 yr) were randomized to three conditions: aerobic exercise program (AEP; n = 21),
resistance and aerobic program (RAP; n = 22), and control (n = 22). HIIT sessions (8–10 min per session) were delivered during physical
education lessons or at lunchtime three times per week for 8 wk. Assessments were conducted at baseline and immediately
postintervention to detect changes in executive function (trail making test), psychological well-being, psychological distress, and
physical self-description by researchers blinded to treatment allocation. Intervention effects were examined using linear mixed models.
Cohen_s d effect sizes and clinical inference were also calculated. Results: While results were not significant, small improvements in
executive function (mean change (95% CI) j6.69 (j22.03, 8.64), d = j0.32) and psychological well-being (mean change (95% CI) 2.81
(j2.06, 7.68), d = 0.34) were evident in the AEP group; and moderate improvements in executive function (mean change (95% CI) j10.73
(j26.22, 4.76), d = j0.51), and small improvements in well-being (mean change (95% CI) 2.96 (j1.82, 7.75), d = 0.36) and perceived
appearance (mean change (95% CI) 0.32 (j0.25, 0.86), d = 0.35), were observed for the RAP group. Mean feeling state scores improved
from preworkout to postworkout in both HIIT conditions, with significant results for the AEP (P = 0.001). Conclusions: This study
highlights the potential of embedding HIIT within the school day for improving cognitive and mental health among adolescents.
Key Words: HIGH-INTENSITY INTERVAL TRAINING, COGNITIVE HEALTH, MENTAL HEALTH, ADOLESCENTS

R
egular participation in physical activity is associated achieve the international physical activity recommendations
with a wide range of physical health benefits for of 60 minIdj1 (16), and trends in this generation show a
young people, including improvements in body secular decline in health-related physical fitness (especially
composition, physical capacity, and overall health-related cardiorespiratory fitness) (5). These findings, combined with
indicators (e.g., blood pressure, insulin resistance, lipid pro- the dramatic decline in physical activity (33) typically observed
file) (20). Emerging literature also suggests that physical ac- during adolescence (7% per year from age 12 to 19) (14),
tivity and fitness may have a positive effect on mental health highlight the need for effective solutions to the inactivity pan-
outcomes for youth (e.g., depression and anxiety) (34). Fur- demic. However, physical activity and fitness interventions
thermore, it has been suggested that participation in physical targeting adolescents have been largely unsuccessful (13), and
activity and the attainment of high levels of physical fitness developing innovative and time efficient strategies that provide
are linked to enhanced brain structure and function, cognition, potent health benefits for young people are urgently needed.
and academic performance, via direct and indirect physio- High-intensity interval training (HIIT) has emerged as a
logical, cognitive, emotional, and learning mechanisms (17). feasible and efficacious strategy for increasing physical health
Despite the extensive benefits of an active lifestyle, ap- outcomes in young people (9,25). HIIT involves either a)
APPLIED SCIENCES
proximately 80% of young people across the globe do not short or long intervals (from e45 s to 2–4 min) of high-
intensity exercise (e.g., 985% max heart rate) interspersed by
short rest periods or b) reoccurring short or long (G10 s to
Address for correspondence: David Lubans, Ph.D., School of Education,
Faculty of Education and Arts, The University of Newcastle, University Drive, 20–30 s) bouts of maximal sprints interspersed by a pro-
Callaghan NSW 2308, Australia; E-mail: [email protected]. longed rest period between exercises (8). The main appeal
Submitted for publication February 2016. of HIIT is that it can be completed in a short period while
Accepted for publication May 2016. resulting in equivalent physiological adaptations to longer
0195-9131/16/4810-1985/0 sessions of traditional aerobic training (8). There is strong
MEDICINE & SCIENCE IN SPORTS & EXERCISEÒ evidence indicating that HIIT can improve physical health
Copyright Ó 2016 by the American College of Sports Medicine (9,22), with additional evidence demonstrating a positive ef-
DOI: 10.1249/MSS.0000000000000993 fect on depression (41), sleep quality (41), and emotional

1985

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 well-being (1) in a range of adult population groups (e.g., circumference (AEP: mean change = j1.5, 95% confidence
older adults, cancer patients, and cancer survivors). However, interval [CI] = j3.4 to 0.4, d = j0.5; RAP: mean change =
little is known regarding the effect of HIIT on cognitive and j2.1, 95% CI = j4.0 to j0.3, d = j0.7), body mass index
mental health outcomes in adolescent populations. (AEP: mean change = j0.27, 95% CI = j0.57 to 0.04, d =
Mental health is not merely the absence of a mental illness j0.5; RAP: mean change j0.28, 95% CI = j0.57 to 0.02,
but a state of positive well-being and effective functioning in d = j0.5), and body mass index z (AEP: mean change =
which an individual realizes his or her potential and is able j0.10, 95% CI = j0.20 to j0.01, d = j0.6; RAP: mean
to make a positive contribution to his or her community change = j0.08, 95% CI = j0.17 to 0.01, d = j0.5) in
(44). Therefore, indicators of mental health can be classified comparison with the control group. Heart rate targets were
into three broad categories: cognitive function (e.g., atten- met, with a higher average heart rate evident for the RAP
tion, perception, and memory), well-being (e.g., self-concept (AEP: 74.04% of max, 148.09 bpm; RAP: 77.58% of max,
and eudemonic well-being), and ill-being (e.g., depression 155.15 bpm).
and anxiety). Relative to cognition, the strongest relation- Power calculations were based on change in the primary
ship for physical activity appears for tasks or task compo- outcome (cardiorespiratory fitness, assessed using the mul-
nents that require extensive amounts of executive function tistage shuttle test [24]). A between-group difference of
(i.e., the intentional component of environmental interaction 10 laps was considered achievable, assuming an SD of nine
entailing processes such as inhibition, working memory, and laps, 80% power with alpha levels set at 0.05. It was deter-
cognitive flexibility) (18). Considering the global inactivity mined that 20 participants per group would provide adequate
pandemic and the large numbers of adolescents who have power to detect statistically significant effects.
been diagnosed as having a mental illness (35), the current After baseline measures had been assessed by research
study was designed to evaluate the efficacy of two HIIT assistants blinded to treatment allocation, participants were
protocols (aerobic exercise program [AEP] and resistance randomized using a random number-producing algorithm. A
and aerobic program [RAP]) for improving cognitive and stratified random sampling procedure was conducted to en-
mental health outcomes (executive function, psychological sure that equal numbers of boys and girls were allocated
well-being, psychological distress, and physical self- between the three groups.
concept) in a sample of adolescents from one secondary Participants randomized to the intervention conditions
school in New South Wales, Australia. (AEP and RAP groups) participated in three HIIT sessions
per week for 8 wk (24 sessions in total). Two HIIT sessions
per week were delivered in scheduled PE lessons, and a third
METHODS session was delivered at lunch time. The focus of each of the
three programs included the following:
Study Design and Participants
The study methods have been described in detail previ- i. AEP: Participants completed HIIT sessions primarily in-
ously (10). Briefly, ethical approval to conduct the study volving gross motor cardiorespiratory exercises (e.g.,
was gained from the University of Newcastle Human Re- shuttle runs, jumping jacks, and skipping).
search Ethics Committee (H-2014-0083). The study proto- ii. RAP: Participants completed HIIT sessions that included
col was registered with the Australian and New Zealand a combination of cardiorespiratory and body weight resis-
Clinical Trials Registry (ACTRN12614000729628). The tance training exercises (e.g., shuttle runs, jumping
school principal, parents, and study participants provided jacks, skipping, combined with body weight squats, and
written informed consent to participate in the study. Study push-ups). For example, one RAP work phase included
participants (n = 65) were students in grades 9 and 10 (ages the following sequence of cardiorespiratory and resis-
14–16 yr) attending the study school. The design, conduct, tance exercises (four walking lunges, 10-m sprint, and
and reporting for this randomized controlled trial (RCT) three push-ups) repeated as many times as possible in a
adhered to the guidelines of the Consolidated Standards of 30-s period. The RAP treatment did not include a sep-
Reporting Trials (32). arate resistance training component with a prespecified
APPLIED SCIENCES

A three-arm school-based RCT was conducted with ado- number of sets and repetitions.
lescents from one secondary school. The HIIT sessions iii. Control: Participants continued with their programmed
ranged from 8 to 10 min in duration (weeks 1–3: 8 min; PE and usual lunchtime activities for the 8-wk inter-
weeks 4–6: 9 min; weeks 7–8: 10 min), with a work to rest vention period (Fig. 1).
ratio of 30:30 s. The AEP and the RAP sessions were de-
livered by the research team at the study school. The AEP and the RAP groups engaged in their HIIT ses-
Our first study examined changes in health-related fitness sions (inclusive of a short warm-up activity including dynamic
outcomes associated with the two HIIT protocols (10). stretching, 8–10 min of HIIT and cooldown), whereas the
Briefly, a small intervention effect was evident for cardio- control group did their typical warm-up and stretching and
respiratory fitness in the RAP group; participants in the AEP completed one activity with their PE teacher. Following the
and RAP groups had moderate intervention effects for waist HIIT session, the groups were combined to complete the

1986 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
FIGURE 1—Flow of participants through the study. K10, Kessler Psychological Distress Scale; FL SCALE, Flourishing Scale; MOT, physical activity
motivation; PSC, physical self-concept.

remainder of the scheduled PE lesson. Session duration and awarded to the pairs in each study arm receiving the most
intensity were the same for both intervention groups. Partici- certificates at the end of the program. To promote autonomy,
pants wore heart rate monitors (Polar H7) to encourage participants were also given the opportunity a) to choose
maintenance of the appropriate exercise intensity, which were music (student playlists used weeks 2–8), b) to select specific
connected to a central iPad application (Polar Team). Heart exercises to be completed during a workout (weeks 4–6), and
rates were displayed on a projector screen during sessions. c) to choose a workout (between two workouts previously
Given that both adolescent girls and boys have reported completed; weeks 7 and 8) once exercises were mastered.
difficulty starting and adhering to regular exercise, several
approaches (based on self-determination theory [39]) were
Outcomes
undertaken to promote adherence to the program. First, ses-
sions were designed to be enjoyable by including a fun warm- All assessments were conducted by trained researchers
APPLIED SCIENCES
up and cooldown activity or game, and participants worked blinded to group allocation. A measurement training session
with a partner of their choice (one participant undertook the and protocol manual, including specific instructions for
‘‘work’’ phase of the sessions, while their partner completed conducting all assessments, was provided for all research
the ‘‘rest’’ phase). To create a supportive environment, a fo- staff to ensure accuracy and consistency. All physical as-
cus of all sessions was to promote and reward students for sessments were conducted discretely, and questionnaires
providing verbal encouragement and support to peers and for were completed under exam-like conditions.
working hard during the HIIT sessions. A ‘‘Trainer of the
Day’’ certificate was presented to one pair per session for
Mental Health Outcomes
providing positive feedback and motivation for their partner
and for demonstrating outstanding effort and dedication dur- Executive function. The trail making test (TMT) is a
ing the workout. Prizes (e.g., a gift voucher) were also measure of visual attention, speed, scanning, speed of

EFFECT OF HIIT ON COGNITIVE AND MENTAL HEALTH Medicine & Science in Sports & Exercised 1987

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 processing, and mental flexibility and has been validated in Process Evaluation
youth (43). The TMT involves a two-part visual task in
Feelings state. A one-item Feelings State questionnaire
which participants are required to, first (trail a), draw a line
was administered before and after each HIIT session for the
from one point to the next as quickly as possible to connect
duration of the intervention (total 24 sessions) (38). Partici-
numbers in ascending order (e.g., 1-2-3-4, etc.) and, second
pants were asked to respond on an 11-point scale (j5 = very
(trail b), draw a line from one point to the next as quickly as
bad to +5 = very good) to the question How are you feeling
possible to connect both numbers and letters in an ascending
right now? Mean pre- and postworkout scores were calcu-
and alternating order (e.g., 1-a-2-b-3-c-4-d, etc.) (37). Lower
lated for each session according to intervention condition.
scores indicate greater cognitive performance, and in the
Heart rate data. To monitor exercise intensity (target:
literature, various methods have been used to obtain an
efforts Q85% of heart rate maximum), participants were fitted
overall measure of cognitive flexibility. For instance, in one
with Polar H7 heart rate monitors, which were connected to a
method, the time to complete trail A is subtracted from the
central iPad application (Polar Team). The mean heart rate for
time to complete trail b (B-A) (40), whereas another method
the entire session and the mean maximum heart rate were
includes time to complete trail b divided by the time to
tracked over the study period.
complete trail A (B/A) (40). As TMT B is a more complex
test compared with TMT A, some studies have only con-
Statistical Analyses
sidered the time taken to complete trail b (30). Therefore,
each of these methods are calculated and reported. Statistical analyses of the primary and secondary out-
Psychological well-being. The Flourishing Scale is a comes were conducted with linear mixed models using IBM
brief eight-item summary measure of the respondent_s self- SPSS Statistics for Windows, Version 20.0 (2010 SPSS Inc.,
perceived success in areas such as relationships, self-esteem, IBM Company Armonk, NY). Intervention effects for the
purpose, and optimism. Students responded on a seven-point primary and secondary outcomes were examined by using
scale (1 = strongly disagree to 7 = strongly agree) to how linear mixed models. Because of the small sample size and
much they agreed with each statement relating to indicators the potential issues associated with interpreting P values
of social well-being (e.g., I lead a purposeful and mean- (15), Cohen_s d was included to provide a measure of effect
ingful life). The scale provides a single psychological well- size (adjusted difference between HIIT and control groups
being score. A composite score was created by summing the over time divided by the pooled SD of change), and 95% CI
scores for each item (possible range, 8–56). Higher scores values were also determined. Moderators of HIIT effects were
indicate greater well-being. Cronbach_s > was used as a explored using linear mixed models with interaction terms for
measure of scale reliability for psychological well-being the following: i) sex (boys vs girls), ii) weight status (healthy
(baseline: > = 0.77; posttest: > = 0.92). weight vs overweight/obese), and iii) baseline fitness level
Psychological distress. The Kessler Psychological (i.e., healthy fitness zone vs needs improvement). Subgroup
Distress Scale (K10) involves 10 questions about a person_s analyses were only conducted if significant interaction effects
emotional state (2). The K10 questionnaire is intended were observed (P = 0.10).
to measure distress based on questions about anxiety and In addition to Cohen_s d effect sizes (mean difference
depressive symptoms that a person has experienced in the last [posttest – baseline] between groups divided by the pooled
4 wk. Each question is scored from 0 (none of the time) to SD of change for the whole group), and based on a previous
5 (all of the time). Scores for the 10 questions are then literature, the clinical inference of the true value of change
summed, producing a minimum score between 0 and 50. Low scores was derived using a custom made spreadsheet de-
scores indicate low levels of psychological distress, and high veloped by Hopkins (19). A clinical inference was based on
scores indicate high levels of psychological distress (2). the probabilities of harm and benefit for each outcome and is
Cronbach_s > was used as a measure of scale reliability for presented as the chance that the true value of the change
the K10 (baseline: > = 0.93; posttest: > = 0.91). scores was beneficial, trivial, or harmful (19). Our study
Physical self-concept. The global physical self- used the default probabilities (%) and associated descriptors
concept and perceived appearance subscales from the of 0 ‘‘most unlikely,’’ 0.5 ‘‘very unlikely,’’ 5 ‘‘unlikely,’’ 25
APPLIED SCIENCES

Physical Self-Description Questionnaire (validated in adoles- ‘‘possibly’’ 75 ‘‘likely,’’ 95 ‘‘very likely,’’ and 99.5 ‘‘most
cence [28]) were used. Participants were asked to respond on likely’’ (19).
a six-point scale (1 = ‘‘false,’’ to 6 = ‘‘true’’) how true each
statement was for them (e.g., ‘‘I am attractive for my age’’ and
RESULTS
‘‘I feel good about who I am and what I can do physically’’).
Total scores were divided by number of items to provide a The number of participants involved at each phase of the
mean value for the subscales of global physical self-concept study is reported in Figure 1. One secondary school was
and perceived appearance. Cronbach_s > was used as a successfully recruited, and 65 adolescents from three classes
measure of scale reliability for perceived appearance (base- (45 males, 20 females, mean T SD age = 15.8 T 0.6 yr) from
line: > = 0.94; posttest: > = 0.94) and global physical self- grades 9 and 10 completed baseline testing (see Table 1). The
concept (baseline: > = 0.88; posttest: > = 0.95). intervention groups were similar for baseline characteristics.

1988 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
TABLE 1. Participant baseline demographics (Australia, July 2014).

0.891 j0.04 (j1.83 Very unlikely harmful

Control AEP RAP

Possibly beneficial

Possibly beneficial

Possibly beneficial

Possibly beneficial
Inferenceb

0.935 j0.03 (j0.28 Unlikely harmful

Clinical
Total (n = 22) Total (n = 21) Total (n = 22)
Mean T SD Mean T SD Mean T SD

RAP j Control

0.35 (j0.74 Unclear

Age (yr) 15.6 T 0.6 15.7 T 0.7 15.5 T 0.6
Weight (kg) 66.0 T 15.8 64.7 T 9.8 67.0 T 12.9

j7.76 (j21.79 to 6.27) 0.273 j0.40 (j8.24

0.222 j0.37 (j0.29

j10.73 (j26.22 to 4.76) 0.171 j0.51 (j8.92

0.36 (j3.86
Cohen_s d
Height (cm) 171.3 T 10.6 172.3 T 8.6 173.8 T 7.1

(95% CI)

to 8.85)

to 1.04)

to 9.73)

to 3.13)

to 1.96)

to 0.01)

to 0.35)
BMI (kgImj2) 22.29 T 3.53 21.72 T 2.10 22.08 T 3.56
BMI-z 0.51 T 0.94 0.43 T 0.60 0.45 T 1.05
BMI = body mass index.

2.96 (j1.82 to 7.75) 0.219

0.249
P
Changes in Executive Function (TMT)

Adjusted Difference between Groupsa

j0.56 (j1.47 to 0.35)

j0.19 (j2.97 to 2.59)

0.32 (j0.25 to 0.86)

0.02 (j0.45 to 0.48)

Small-to-moderate intervention effects for executive func-

(95% CI)
Change
tion were found for the RAP condition for all methods of

Mean
calculation used (B/A: j0.56, 95% CI = j1.47 to 0.35; d =
j0.37, 95% CI = j0.29 to 1.04; B-A: j7.76, 95% CI = j21.79
to 6.27; d = j0.40, 95% CI = j8.24 to 8.85; TMT B: j10.73,

0.00 (j0.32 Most unlikely harmful

Very unlikely harmful

Very unlikely harmful

0.737 j0.11 (j1.76 Very unlikely harmful

0.10 (j0.48 Very unlikely harmful

95% CI = j26.22 to 4.76; d = j0.51, 95% CI = j8.92 to 9.73),

Possibly beneficial
Inferenceb
Clinical
which were all classified as ‘‘possibly beneficial.’’ For the AEP
condition, however, a small intervention effect for executive

Unclear
AEP j Control

Adjusted difference between groups and 95% CI between intervention and control groups after the 8-wk intervention (AEP minus control; RAP minus control).
function was only evident when considering TMT B scores
(j6.69, 95% CI = j22.03 to 8.64; d = j0.32, 95% CI = j9.12

0.01 (j0.69

0.34 (j3.84
0.571 j0.21 (j8.44

0.386 j0.32 (j9.12

Cohen_s d
(95% CI)
to 9.77), which was also classified as ‘‘possibly beneficial.’’

to 8.87)

to 0.66)

to 9.77)

to 3.32)

to 2.13)

to 0.29)

to 0.33)
Changes in Psychological Well-being

0.968

0.252

0.753

0.957
(Flourishing Scale)

P
j3.95 (j17.85

j6.69 (j22.03
Small intervention effects for well-being were found for

0.02 (j0.90

2.81 (j2.06

j0.49 (j3.37

0.89 (j0.47

j0.01 (j0.49
(95% CI)
Change

to 9.95)

to 0.94)

to 8.64)

to 7.68)

to 2.40)

to 0.65)

to 0.46)
Mean
both HIIT conditions (AEP: 2.81, 95% CI = j2.06 to 7.68;
d = 0.34, 95% CI = j3.84 to 3.32; RAP: 2.96, 95% CI =
j1.82 to 7.75; d = 0.36, 95% CI = j3.86 to 3.13), which was
Postintervention

‘‘possibly beneficial.’’
32.81 (24.55

53.79 (44.50

48.28 (45.36

17.55 (14.31
to 41.07)

to 63.08)

to 51.20)

to 20.78)
to 2.55)

to 4.83)

to 4.91)
(95% CI)

2.01 (1.47

4.35 (3.88

4.41 (3.91
Mean
RAP Group

Changes in Psychological Distress (K10)

There were no intervention effects for psychological dis-
39.88 (32.33

65.07 (56.24

46.59 (44.48

17.68 (14.50
to 47.43)

to 73.91)

to 48.70)

to 20.86)
(95% CI)

2.40 (1.92

3.52 (2.95

4.35 (3.88
Baseline

to 2.88)

to 4.09)

to 4.83)
Mean

tress for either HIIT groups, in comparison with the control

condition. A clinical inference of ‘‘very unlikely harmful’’
was apparent for both conditions.
Postintervention

32.80 (24.69

55.97 (46.80

47.92 (44.92

18.17 (14.74
to 40.91)

to 65.14)

to 50.91)

to 21.59)
to 2.73)

to 5.19)

to 5.24)
(95% CI)

2.18 (1.62

4.69 (4.19

4.71 (4.19
Mean

Changes in Physical Self-Concept

TABLE 2. Changes in cognitive and mental health outcomes postintervention.
AEP Group

Changes for all outcomes are reported in Table 2. Analyses

of efficacy (adjusted difference between groups and Cohen_s
63.21 (54.10

46.38 (44.17

18.60 (15.27
36.05 (28.28

to 72.33)

to 48.59)

to 21.94)
to 43.83)
(95% CI)

1.99 (1.49

4.09 (3.49

4.69 (4.19
Baseline

to 2.48)

to 4.68)

to 5.19)
Mean

d effect sizes reported) identified a small intervention effect

for the RAP condition for perceived appearance (0.32, 95%
Likelihood of intervention being beneficial/trivial/harmful.

CI = j0.25 to 0.86; d = 0.35, 95% CI = j0.74 to 0.01;

Postintervention

34.95 (26.41

57.32 (47.43

47.00 (44.22

22.10 (18.83
to 43.49)

to 67.21)

to 49.78)

to 25.36)

clinical inference: ‘‘unclear’’). However, no intervention ef-

to 2.17)

to 5.48)

to 5.54)
(95% CI)

1.63 (1.09

5.00 (4.52

5.04 (4.54

APPLIED SCIENCES
Mean

fects were apparent for global physical self-concept in either

Control Group

HIIT group (clinical inference: AEP ‘‘most unlikely harmful’’;

RAP ‘‘unlikely harmful’’).
34.25 (25.07

48.27 (46.16

22.05 (18.87
to 43.45)

to 68.54)

to 50.38)

to 25.23)
(95% CI)

1.46 (0.98

57.87 (47.2

4.49 (3.92

5.00 (4.52
Baseline

to 1.94)

to 5.06)

to 5.48)
Mean

Process Outcomes
Changes in feelings state. Mean pre- and postworkout
self-description score
score (appearance)
Psychological distress
(trails B j trail A)

scores were calculated for each session according to inter-

(global physical)
(trail B / trail A)

self-description
Executive function

Executive function

Executive function

(psychological
Flourishing Scale

vention condition (see Figs. 2 and 3). For the AEP group,
well-being)

Mean physical

Mean physical
(TMT B)

mean feeling state scores increased from pre = 1.57 T 1.13

Variable

*P G 0.05.

to post = 2.54 T 1.00, which was statistically significant

(P = 0.001), representing an average improvement of 0.97 T
b
a

EFFECT OF HIIT ON COGNITIVE AND MENTAL HEALTH Medicine & Science in Sports & Exercised 1989

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 FIGURE 2—Mean feeling state scores recorded pre- and postsessions for the AEP condition.

1.08. The improvement in mean feeling state scores among and mental health outcomes in a sample of adolescents.
participants in RAP approached statistical significance (pre = Overall, small improvements in executive function (TMT B)
1.85 T 1.54 to post = 2.19 T 1.54; P = 0.06), an improvement and psychological well-being were evident in the AEP
of 0.34 T 0.80. group, and small improvements in executive function (B-A;
Heart rate data. Mean heart rate scores were calculated B/A), well-being, and perceived appearance were observed
for each session according to intervention conditions. Higher for the RAP group. However, when considering TMT B
average heart rates (AEP: 74.0%, 148.1 bpm; RAP: 77.6%, only, a moderate effect was apparent for the RAP condition.
155.2 bpm) were evident in the RAP group, in comparison Mean feeling state scores improved from preworkout to
with the AEP (note: session average heart rate included the postworkout in both HIIT groups; however, significant results
warm-up, work periods, rest periods, and cooldown). By were observed only among participants in the AEP group.
contrast, mean maximum heart rate was higher for the AEP Executive function comprises several cognitive processes,
group (AEP: 92.4%, 184.8 bpm; RAP: 91.8%, 182.2 bpm) in which contribute to organizing and controlling goal-directed
comparison with the RAP group. behavior, and includes inhibition, working memory, and
Moderators. As there were no significant (P G 0.10) cognitive flexibility (31). Although the evidence is still
interaction effects for any of the potential moderators, sub- emerging, regular participation in physical activity has been
group analyses were not conducted. linked to enhanced brain function and cognition and im-
proved academic performance in adolescents (17). Evidence
also suggests that incorporating physical activity into the
DISCUSSION
school day is associated with improvements in attention,
The aim of the current study was to evaluate the effect of concentration, and time on task in the classroom (36). In our
two HIIT protocols (AEP and RAP) on a range of cognitive study, small and moderate improvements in executive
APPLIED SCIENCES

FIGURE 3—Mean feeling state scores recorded pre- and postsessions for the RAP condition.

1990 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
function (TMT B) were evident in the AEP and RAP groups, an educational component in addition to physical activity
respectively. Similarly, a 4-month cluster RCT of Spanish (7). Conversely, a review of five studies investigating the
adolescents (n = 67) examined the effect of four PE lessons effect of exercise for the prevention and treatment of anxiety
(embedded with high-intensity activities) per week versus and depression in youth (1983–2005) reported that physical
four regular lessons per week versus two regular lessons per activity interventions had a statistically significant effect on
week (control; regular lesson duration = 55 min) on cogni- reducing depression (standard mean difference effect size =
tive performance and academic achievement (3). Overall, no j0.66, 95% CI = j1.25 to j0.08) (23). Given the incon-
differences in cognitive performance or academic achieve- sistent findings and limited number of current studies
ment were evident between having two or four regular PE available, future research is needed to clarify the relation-
lessons per week; however, students randomized to four ship between physical activity and psychological distress
sessions per week embedded with high-intensity activities among adolescents and to determine the optimal intensity
achieved improvements in all cognitive performance vari- and duration of physical activity intervention to achieve the
ables (excluding verbal reasoning) and for average school greatest improvements in psychological distress. Moreover,
grades. Our study builds on these findings by using very subgroup analyses are required to determine whether the
short duration HIIT to improve cognitive function in ado- effects of physical activity on depressive symptoms are
lescents. Further research involving larger sample sizes and stronger among individuals who are ‘‘at risk’’ or have de-
long-term follow-up is needed to investigate the effect that pression. Because of the small sample size, this was not
the intensity of PE lessons can have on cognition and aca- possible for the current study.
demics for adolescents. Although session attendance plays a Physical self-concept refers to an individual_s beliefs
role for adolescent cognition and academics (18), the effect about their physical characteristics, and adolescence repre-
of intensity does not appear to have been examined. sents a significant period for the development of physical
Psychological well-being refers to an individual_s per- self-concept (29). Evidence suggests that in comparison with
ception of their success in relationships, self-esteem, pur- adolescents who engage in low levels or no physical activ-
pose, and optimism (12). In our study, there was a small ity, adolescents participating in higher levels of physical
intervention effect for psychological well-being suggestive activity have greater self-concept (4). Adolescents_ percep-
of a ‘‘possibly beneficial’’ effect for both HIIT intervention tions of physical appearance are typically formed by com-
protocols. Each HIIT session was designed to meet partici- paring themselves with peers (29). In our study, the RAP
pants_ basic psychological needs for autonomy (e.g., choice condition achieved a small positive intervention effect for
of music, exercise choices during a workout, and choice of the perceived appearance subscale. Similarly in a previous
workout), competence (e.g., provision of challenging yet 8-wk study examining the effects of resistance training on
achievable workouts and sense of accomplishment), and physical self-perception in a sample of adolescents (n = 108)
relatedness (e.g., working in pairs, sessions focused on (26), significant changes in perceived body attractiveness
promoting encouragement, and support to peers), which may were evident among girls randomized to the free weights re-
have also affected well-being. Similarly, in a recent 8-month sistance training condition (P G 0.01, d = 0.76); however, no
RCT examining psychological well-being in adolescent significant changes were found for boys. A recent meta-
boys (mean T SD age = 12.7 T 0.5) participating in the Ac- analysis examining physical activity and domains of physical
tive Teen Leaders Avoiding Screen time study (27), the in- self-concept in youth reported a weak association between
tervention resulted in a small yet statistically significant perceived appearance and physical activity (r = 0.14, 95%
effect on well-being. Interestingly, the intervention effect on CI = 0.09–0.18, P 9 0.001) (4). Given that physical self-concept
well-being was mediated by improvements in muscular fit- (including perceived competence and appearance) is an im-
ness and the provision of autonomy within lessons, indi- portant component of global self-esteem, the small improve-
cating that the inclusion of resistance training within the ments in perceived appearance demonstrated in our study (RAP
intervention facilitated improvements in well-being. Al- condition) may have important implications for improving
though resistance training was included in the sessions mental health outcomes. However, longer-term studies may be
programmed for the RAP group, there do not appear to be necessary (i.e., 98 wk) to improve perceived appearance, par-
APPLIED SCIENCES
any substantial differences in well-being effects between ticularly among boys, and further investigations are needed to
groups in the current study. As the effect of HIIT on well- determine whether improvements in perceived appearance
being appears promising, further research spanning beyond contribute to improvements in global self-esteem and serve as a
one school term with a focus on resistance training is needed protective factor against mental illness during adolescence.
to investigate the ongoing effect of this approach to exercise. Process outcomes. The high retention rate suggests
The HIIT intervention effect on psychological distress that intervention strategies used in this program appealed to
was marginal in the current study. Similarly, a meta-analysis participants and resulted in their continued involvement in
of nine studies (1982–2010) examining the effect of physical the program. Although the maximum heart rate data suggest
activity on depression in youth (7) reported a small yet sig- that our target heart rates were achieved (i.e., Q85% maxi-
nificant treatment effect. Of note, the greatest effects on mum heart rate), this cannot be confirmed with the available
depression were evident in RCT of G3 months, which used data. It should be noted that heart rate monitors were worn

EFFECT OF HIIT ON COGNITIVE AND MENTAL HEALTH Medicine & Science in Sports & Exercised 1991

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
 for the entire HIIT session, which included a warm-up, rest participating in the study may limit the generalizability of
periods, and cooldown. Therefore, it is not surprising that our findings. In addition, the intervention period was rela-
participants_ mean heart rates were less than 85% maximum tively short, with no long-term follow-up conducted. Finally,
heart rate. Unfortunately, we did not specifically ask par- the heart rate monitoring application did not allow for heart
ticipants to reflect on the potential benefits of incorporating rate to be recorded during the work interval only (i.e., maximum
resistance exercise into the HIIT protocols. However, con- heart rate and mean heart rate for the entire session were
sidering the benefits of muscular fitness for health among recorded). Our heart rate monitoring protocol provided a sum-
adolescents (42), there is clear advantage to including re- mary of the entire session and did not isolate participants_ heart
sistance exercise in future interventions targeting adoles- rates during the work interval only (i.e., maximum heart rate
cents. Encouragingly, mean feeling state scores improved and mean heart rate for the entire session were recorded). In
from pre- to postworkout, suggesting that high-intensity future studies, researchers are encouraged to use the lap
exercise is likely to result in improved affective responses function on the heart rate monitors to isolate HIIT work pe-
for this population group (38). However, results were only riods from rest periods and if possible record the length of time
statistically significant for the AEP group, which could be (number of minutes and percentage of session) that each par-
explained by the slightly lower average heart rate reached ticipant met heart rate targets. This protocol will provide a
across AEP sessions in comparison with the RAP sessions. better indication of HIIT session intensity.
It may be that AEP sessions were somewhat less strenuous
than the RAP sessions and perceived as more achievable and
CONCLUSION
enjoyable for participants. Although a recent commentary
(6) has proposed that prescribing intense exercise to the The outcomes of this research contribute to understanding
general/sedentary population may lead to feelings of in- how short bouts of intense exercise influence cognitive and
competence and failure resulting in reduced physical activity mental health outcomes in adolescent populations. Although
motivation and participation, this was not the case in our evidence from this study highlights the potential of embed-
study. Interestingly, in a study conducted by Crisp et al. (11), ding HIIT within the school day for improving executive
the male participants did not consider sprint interval cycling function, physical self-concept (especially appearance), and
to be more strenuous than moderate-intensity continuous cy- well-being among adolescents, no significant (P G 0.05)
cling, and in a study conducted by Jung et al. (21), examining findings emerged. Therefore, further longitudinal research
the affective response to high-intensity exercise compared with longer follow-up periods, investigating a larger sample
with continuous moderate- or vigorous-intensity exercise of adolescents from a range of year levels and schools,
conducted on cycle ergometers, participants reported greater should be conducted.
enjoyment and a preference to participate in high-intensity
intervals in comparison with continuous moderate-intensity This project was supported by a Hunter Medical Research Institute
exercise and continuous vigorous-intensity exercise. This project grant. D. R. L. is supported by an Australian Research Council
Future Fellowship. R. C. P. is supported by a National Health and
highlights the importance of the type of exercises included in Medical Research Council Senior Research Fellowship. The authors
HIIT sessions being appealing for participants to maintain/ are grateful for the support and cooperation of the participating
improve enjoyment and motivation for physical activity. school and students. The authors thank Dan Beckett, Tara Finn,
Emma Pollock, and Sarah Kennedy for their assistance in the de-
Strengths and limitations. This novel study has sev- livery of HIIT sessions and for data collection.
eral strengths, including the randomized design, assessor The authors declare no conflict of interest. The results of this study
blinding, and high retention rates. However, some limita- do not constitute endorsement by the American College of Sports
Medicine.
tions should also be acknowledged. The small sample from Trial Registration: Australian and New Zealand Clinical Trials Reg-
one school and the uneven distribution of girls and boys istry (ACTRN12614000729628).

REFERENCES
1. Adamsen L, Quist M, Andersen C, et al. Effect of a multimodal States: baseline results from the NFL PLAY 60 FITNESSGRAM
APPLIED SCIENCES

high intensity exercise intervention in cancer patients undergoing Partnership Project. J Pediatr. 2015;167(3):662–8.
chemotherapy: randomised controlled trial. BMJ. 2009;339:b3410. 6. Biddle SJ, Batterham AM. High-intensity interval exercise training
2. Andrews G, Slade T. Interpreting scores on the Kessler psycho- for public health: a big HIT or shall we HIT it on the head? Int J
logical distress scale (K10). Aust N Z J Public Health. 2001; Behav Nutr Phys Act. 2015;12(1):95.
25(6):494–7. 7. Brown HE, Pearson N, Braithwaite RE, Brown WJ, Biddle SJ. Physical
3. Ardoy D, Fernández-Rodrı́guez J, Jiménez-Pavón D, Castillo R, activity interventions and depression in children and adolescents: a
Ruiz J, Ortega F. A physical education trial improves adolescents_ systematic review and meta-analysis. Sports Med. 2013;43(3):195–206.
cognitive performance and academic achievement: the EDUFIT 8. Buchheit M, Laursen PB. High-intensity interval training, solu-
study. Scand J Med Sci Sports. 2014;24(1):e52–61. tions to the programming puzzle. Part I: cardiopulmonary empha-
4. Babic MJ, Morgan PJ, Plotnikoff RC, Lonsdale C, White RL, Lubans sis. Sports Med. 2013;43(5):313–38.
DR. Physical activity and physical self-concept in youth: systematic 9. Costigan S, Eather N, Plotnikoff R, Taaffe DR, Lubans DR. High-
review and meta-analysis. Sports Med. 2014;44(11):1589–601. intensity interval training for improving health-related fitness in
5. Bai Y, Saint-Maurice PF, Welk GJ, Allums-Featherston K, adolescents: a systematic review and meta-analysis. Br J Sports
Candelaria N, Anderson K. Prevalence of youth fitness in the United Med. 2015;49(19):1253–61.

1992 Official Journal of the American College of Sports Medicine http://www.acsm-msse.org

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.
10. Costigan SA, Eather N, Plotnikoff RC, et al. Preliminary efficacy and 29. Marsh HW, Barnes J, Cairns L, Tidman M. Self-Description
feasibility of embedding high intensity interval training into the school Questionnaire: age and sex effects in the structure and level of
day: a pilot randomized controlled trial. Prev Med Rep. 2015;2:973–9. self-concept for preadolescent children. J Educ Psychol. 1984;
11. Crisp NA, Fournier PA, Licari MK, Braham R, Guelfi KJ. Adding 76(5):940.
sprints to continuous exercise at the intensity that maximises fat 30. Martı́n-Martı́nez I, Chirosa LJ, Reigal RE, Hernández-Mendo A,
oxidation: implications for acute energy balance and enjoyment. de Mier RJ-R, Guisado R. Effects of physical activity on executive
Metabolism. 2012;61(9):1280–8. function in a sample of adolescents. Anales de Psicologia. 2015;31(2):
12. Diener E, Wirtz D, Tov W, et al. New well-being measures: short 962–71.
scales to assess flourishing and positive and negative feelings. 31. Miyake A, Friedman NP, Emerson MJ, Witzki AH, Howerter A,
Social Indicators Research. 2010;97(2):143–56. Wager TD. The unity and diversity of executive functions and their
13. Dobbins M, De Corby K, Robeson P, Husson H, Tirilis D. School- contributions to complex ‘‘frontal lobe’’ tasks: a latent variable
based physical activity programs for promoting physical activity and analysis. Cogn Psychol. 2000;41(1):49–100.
fitness in children and adolescents aged 6–18. Cochrane Database 32. Moher D, Hopewell S, Schulz KF, et al. CONSORT 2010 expla-
Syst Rev. 2009;(1):CD007651. nation and elaboration: updated guidelines for reporting parallel
14. Dumith SC, Gigante DP, Domingues MR, Kohl HW 3rd. Physical group randomised trials. BMJ. 2010;340:c869.
activity change during adolescence: a systematic review and a 33. Ortega FB, Konstabel K, Pasquali E, et al. Objectively measured
pooled analysis. Int J Epidemiol. 2011;40(3):685–98. physical activity and sedentary time during childhood, adoles-
15. George BJ, Beasley TM, Brown AW, et al. Common scientific cence and young adulthood: a cohort study. PLoS One. 2013;8(4):
and statistical errors in obesity research. Obesity. 2016;24(4):781–90. e60871.
16. Hallal PC, Andersen LB, Bull FC, et al. Global physical activity 34. Parfitt G, Eston RG. The relationship between children_s habitual
levels: surveillance progress, pitfalls, and prospects. Lancet. 2012; activity level and psychological well-being. Acta Paediatr. 2005;
380(9838):247–57. 94(12):1791–7.
17. Hillman CH, Erickson KI, Kramer AF. Be smart, exercise your 35. Patton GC, Coffey C, Cappa C, et al. Health of the world_s ado-
heart: exercise effects on brain and cognition. Nat Rev Neurosci. lescents: a synthesis of internationally comparable data. Lancet.
2008;9(1):58–65. 2012;379(9826):1665–75.
18. Hillman CH, Pontifex MB, Castelli DM, et al. Effects of the 36. Rasberry CN, Lee SM, Robin L, et al. The association between
FITKids randomized controlled trial on executive control and brain school-based physical activity, including physical education, and
function. Pediatrics. 2014;134(4):e1063–71. academic performance: a systematic review of the literature. Prev
19. Hopkins WG. A spreadsheet for deriving a confidence interval, Med. 2011;52:S10–20.
mechanistic inference and clinical inference from a P value. 37. Reitan RM, Wolfson D. The trail making test as an initial screening
Sportscience. 2007;11:16–21. procedure for neuropsychological impairment in older children. Arch
20. Janssen I, LeBlanc AG. Systematic review of the health benefits of Clin Neuropsychol. 2004;19(2):281–8.
physical activity and fitness in school-aged children and youth. 38. Rejeski WJ, Best DL, Griffith P, Kenney E. Sex-role orientation
IJBNPA. 2010;7(40):1–16. and the responses of men to exercise stress. Res Q Exerc Sport.
21. Jung ME, Bourne JE, Little JP. Where does HIT fit? An exami- 1987;58(3):260–4.
nation of the affective response to high-intensity intervals in 39. Ryan RM, Deci EL. Self-determination theory and the facilitation
comparison to continuous moderate- and continuous vigorous- of intrinsic motivation, social development, and well-being. Am
intensity exercise in the exercise intensity-affect continuum. PLoS Psychol. 2000;55(1):68–78.
One. 2014;9(12):e114541. 40. Salthouse TA. What cognitive abilities are involved in trail-making
22. Kilpatrick MW, Jung ME, Little JP. High-intensity interval train- performance? Intelligence. 2011;39(4):222–32.
ing: a review of physiological and psychological responses. 41. Singh NA, Stavrinos TM, Scarbek Y, Galambos G, Liber C,
ACSMs Health Fit J. 2014;18(5):11–6. Fiatarone Singh MA. A randomized controlled trial of high versus
23. Larun L, Nordheim LV, Ekeland E, Hagen KB, Heian F. Exercise low intensity weight training versus general practitioner care for
in prevention and treatment of anxiety and depression among clinical depression in older adults. J Gerontol A Biol Sci Med Sci.
children and young people. Cochrane Database Syst Rev. 2006;(3): 2005;60(6):768–76.
CD004691. 42. Smith JJ, Eather N, Morgan PJ, Plotnikoff RC, Faigenbaum AD,
24. Léger LA, Mercier D, Gadoury C, Lambert J. The multistage 20 metre Lubans DR. The health benefits of muscular fitness for children and
shuttle run test for aerobic fitness. J Sports Sci. 1988;6(2):93–101. adolescents: a systematic review and meta-analysis. Sports Med.
25. Logan GR, Harris N, Duncan S, Schofield G. A review of adolescent 2014;44(9):1209–23.
high-intensity interval training. Sports Med. 2014;44(8):1071–85. 43. Strauss E, Sherman EM, Spreen O. A Compendium of Neuropsy-
26. Lubans DR, Aguiar EJ, Callister R. The effects of free weights and chological Tests: Administration, Norms, and Commentary. Oxford
elastic tubing resistance training on physical self-perception in University Press; 2006.
adolescents. Psych Sport Exerc. 2010;11(6):497–504. 44. World Health Organisation. Promoting Mental Health: Concepts, APPLIED SCIENCES
27. Lubans DR, Smith JJ, Morgan PJ, et al. Mediators of psychological Emerging evidence, Practice: A report of the World Health Or-
well-being in adolescent boys. J Adolesc Health. 2015;58(2):230–6. ganization, Department of Mental Health and Substance Abuse in
28. Marsh HW. Construct validity of Physical Self-Description Ques- collaboration with the Victorian Health Promotion Foundation
tionnaire responses: relations to external criteria. J Sport Exerc and the University of Melbourne. Geneva: World Health Organi-
Psychol. 1996;18:111–31. zation; 2005. Available from: World Health Organization.

EFFECT OF HIIT ON COGNITIVE AND MENTAL HEALTH Medicine & Science in Sports & Exercised 1993

Copyright © 2016 by the American College of Sports Medicine. Unauthorized reproduction of this article is prohibited.