Energy intake and expenditure during sedentary screen time and
motion-controlled video gaming1–3
Elizabeth J Lyons, Deborah F Tate, Dianne S Ward, and Xiaoshan Wang
Background: Television watching and playing of video games
(VGs) are associated with higher energy intakes. Motion-controlled
video games (MC) may be a healthier alternative to sedentary
screen-based activities because of higher energy expenditures, but
little is known about the effects of these games on energy intakes.
Objective: Energy intake, expenditure, and surplus (intake 2 ex-
penditure) were compared during sedentary (television and VG) and
active (MC) screen-time use.
Design: Young adults (n = 120; 60 women) were randomly assigned to
the following 3 groups: television watching, playing traditional VGs, or
playing MCs for 1 h while snacks and beverages were provided. En-
ergy intakes, energy expenditures, and appetites were measured.
Results: Intakes across these 3 groups showed a trend toward a sig-
nificant difference (P = 0.065). The energy expenditure (P , 0.001)
was higher, and the energy surplus (P = 0.038) was lower, in MC
than in television or VG groups. All conditions produced a mean
(6SD) energy surplus as follows: 638 6 408 kcal in television,
655 6 533 kcal in VG, and 376 6 487 kcal in MC groups. The
OR for consuming $500 kcal in the television compared with the
MC group was 3.2 (95% CI: 1.2, 8.4). Secondary analyses, in which
the 2 sedentary conditions were collapsed, showed an intake that
was 178 kcal (95% CI: 8, 349 kcal) lower in the MC condition than
in the sedentary groups (television and VG).
Conclusion: MCs may be a healthier alternative to sedentary screen
time because of a lower energy surplus, but the playing of these
games still resulted in a positive energy balance. This trial was reg-
istered at clinicaltrials.gov as NCT01523795.
Am J Clin Nutr
of video games (VGs)4, are highly prevalent (1) and have wide-
ranging public health consequences including obesity and related
negative health outcomes (2, 3). Preliminary cross-sectional evi-
(4–6). Laboratory studies of eating behaviors have consistently
shown an increased energy intake during television watching
compared with no stimulus (7–9), even in the absence of food-
related advertising (9, 10).
Although television watching remains the most prevalent
sedentary behavior (1), video gaming is also quite popular. There
is evidence that gaming may ameliorate some of the negative
health effects of television watching by replacing sedentary
behavior with light (11, 12) or moderate-vigorous physical ac-
tivity(13, 14). However,increases in energyexpenditures may be
offset by increased energy intakes. Studies that compared gaming
with no distraction haveshown similar effects as television-based
distraction (15–17), but to our knowledge, no studies have
compared television and VGs to one another to determine effects
on energy intakes.
The recent popularity of motion-based controllers raises the
question of whether VGs may differ from television and also
between types of game control. It is possible that the effects of
motion-controlled video games (MCs) on energy intakes and
expenditure are different from those of typically controlled
games (ie, button-based gamepad controllers). MCs could affect
energy intakes because of increased energy expenditures or
because MCs occupy hands that would otherwise be used to
consume snacks. A comparison of snacking during traditional
seated gamingorgaming whilewalkingonatreadmill showedno
difference in intakes (18), which suggested that differences in
energy expenditures during game play may not affect intakes. We
are not aware of any published studies that have compared
motion-controlledgaming to traditionalgaming toinvestigatethe
effects of greater hand occupation on snacking behavior.
There is a need for rigorously controlled investigations of the
effects of different forms of screen time on energy intake and
expenditure. The purpose of this study was to investigate po-
tential differences in energy intake, expenditure, and surplus (the
difference between intake and expenditure) associated with 2
sedentary screen behaviors (ie, television watching and tradi-
tional VG playing) and one potentially active screen behavior
(the playing of MCs). This investigation excluded advertisements
and offered a choice of television or VG content, food, and
beverages to minimize potential confounding because of in-
dividual preferences. It was hypothesized that MCs would be
1Fromthe Departments of Nutrition (EJL, DFT, and DSW), Health Be-
havior and Health Education (DFT), and Biostatistics (XW) and the Line-
berger Comprehensive Cancer Center (EJL and DFT), The University of
North Carolina at Chapel Hill, Chapel Hill, NC.
2Supportedby a National Institute of Mental Health National Research
Service Award postdoctoral fellowship (grant 5-T32MH75854-05).
3Addresscorrespondence and reprint requests to EJ Lyons, The University
of Texas Medical Branch, 301 University Boulevard, Galveston, TX 77555-
0342. E-mail: firstname.lastname@example.org.
4Abbreviationsused: MC, motion-controlled video game; MET, meta-
bolic equivalent; VG, video game.
ReceivedOctober 10, 2011. Accepted for publication May 9, 2012.
Firstpublished online July 3, 2012; doi: 10.3945/ajcn.111.028423.
Am J Clin Nutr 2012;96:234–9. Printed in USA. ? 2012 American Society for Nutrition
associatedwith alower energy intake, higher energy expenditure,
playing of traditional VGs with a handheld gamepad controller.
SUBJECTS AND METHODS
One hundred twenty 18–35-y-old adults (n ¼ 60 men and 60
women) were recruited by using a university mailing list and
television advertisements on a local cable news channel. To be
eligible for the study, participants were required to weigh ,300
pounds (a requirement for one of the game controllers), have
played VGs $3 times over the previous year, have transpor-
tation to the study location, and be willing to fast 2 h before their
appointment and be videotaped during their appointment. Of
324 individuals who requested information and eligibility cri-
teria, 199 subjects completed eligibility information; of those
199 subjects, 132 potential participants were scheduled, and 120
participants completed the study protocol. All data were col-
lected between October 2010 and February 2011. Appointments
were scheduled to occur near traditional mealtimes for lunch
(eg, 1200 or 1400) or dinner (eg, 1600 or 1800). The time of day
did not differ across the 3 groups [x2
2= 1.88, P = 0.382].
This study was conducted in a dedicated media laboratory in
a university-owned building. All media were played on a 58-inch
high-definition plasma screen television. Participants provided
written informed consent, after which anthropometric and ques-
tionnaire measureswere taken. Fasting for a period of $2 h before
the time of the appointment was confirmed. All participants wore
an activity monitor on the upper right arm. Participants were
randomly assigned to watch television, play traditional VGs, or
play MCs. In all 3 groups, shows or games could be changed at
the discretion of participants during the 1-h study period.
Participants (n = 40) watched television programs by using
Netflix instant-streaming software (Netflix Inc) via an Internet-
connected Xbox 360 console (Microsoft). A wide variety of
television shows were available. Participants watched a range of
programming types, including action (eg, “24” and “Avatar: The
Last Airbender”), comedy (eg, “The Office” and “30 Rock”),
drama (eg, “Weeds” and “Dexter”), nature (eg, “Blue Planet”
and “Life of Mammals”), and reality (eg, “Say Yes to the Dress”
and “Flavor of Love”). This streaming service offers selections
from commercially available DVD releases and, thus, does not
include advertisements of any kind.
Traditional VG group
Participants (n = 40) played one or more of 10 available VGs
on a Playstation 3 console (Sony) (single-player only and not
connected to the Internet). Games were chosen to represent as
many genres as possible, with #2 games from any one genre. All
games received ratings of .75 from an online critical ranking
aggregator with scores that ranged from 0 (worst) to 100 (best).
Scores .75 indicated generally positive reviews from critical and
gaming-enthusiast publications. All of the games used a standard
Playstation 3 Dualshock controller (Sony), which included 2 an-
alog sticks, a directional pad, 4 primary buttons, and 4 secondary
buttons on the controller shoulder. No games included motion
control of any kind; all game inputs were based on pressing
buttons or moving the analog sticks. Five of the games were rated
M for mature, and 5 of the games were rated from E (everyone)
to T (teen). The games were Assassin’s Creed 2 (Ubisoft Enter-
tainment SA), Call of Duty: Modern Warfare 2 (Activision Inc),
Dead Rising 2 (Capcom Entertainment Inc), Dead Space (Elec-
tronic Arts Inc), Final Fantasy XIII (Square Enix Co Ltd), Little
Big Planet (Sony Computer Entertainment America LLC),
Ratchet & Clank Future: Tools of Destruction (Sony Computer
Entertainment America LLC), Red Dead Redemption (Take-Two
Interactive Software Inc), Street Fighter IV (Capcom Entertain-
ment Inc), and 3D Dot Game Heroes (Atlus Co Ltd).
Participants (n = 40) played $1 of 10 available VGs on either
a Wii (Nintendo) or Xbox 360 (Microsoft) console (single-player
only and not connected to the Internet). Games were included that
required at least arm motions to play (at least throwing or hitting
motions). Because this study was specifically focused on games, it
did not include fitness-themed games that were based on exercises
with no game (eg, strategic or narrative) content. The games were
Boom Blox Bash Party (Electronic Arts Inc; throwing motions),
Dance Dance Revolution: Universe 2 (Konami Digital Entertain-
ment Inc; jumping and dancing motions), NHL Slapshot (Elec-
tronic Arts Inc; swinging motions), Punch-Out!! (Nintendo of
America Inc; punching motions), Rayman Raving Rabbids: TV
Party (Ubisoft Entertainment SA; dancing, swinging, and throwing
motions), Rock Band 2 (Electronic Arts Inc; drumming motions),
Wario Ware: Smooth Moves (Nintendo of America Inc; throwing
motions), We Cheer 2 (Namco Bandai Games America Inc;
dancing motions), Wii Fit Plus (Nintendo of America Inc; jogging,
throwing, and leaning motions), and Wii Sports Resort (Nintendo
of America Inc; throwing, punching, and swinging motions). A
Wii Remote and nunchuk (Nintendo), dance mat, hockey stick,
balance board, and drum set were the controllers available for use.
Snacks and beverages
During the study period, snacks and beverages were made
available. Snacks were placed on a table alongside the chair of the
participant, which was placed w6 feet from the television. Four
clear plastic containers (16-cup/3.8-L capacity) were placed on the
table with small plastic bowls in front of each. The snacks available
were M & M’s candy (Mars Inc; regular), Baked Lay’s and Doritos
chips (Frito-Lay North America Inc), and Trader Joe’s Simply the
Best Trek Mix (Trader Joe’s Co; which consisted of cashews, al-
monds, dried pineapples, cranberries, and cherries). Snacks were
chosen on the basis of pilot data on snack and beverage preferences
of young adult gamers (Lyons, unpublished data, 2009) and to
include 2 savory and 2 sweet options. Three bottles or cans each of
water, Coca-Cola, Diet Coke (both Coca-Cola Co), and Mountain
Dew (Pepsi-Co) were available in a refrigerator placed next to the
chair of the participant.
Energy intakes were measured by weighing food and beverage
containers to the nearest gram before and after each session by
using a digital food scale (Tanita). Energy expenditures were
ENERGY BALANCE DURING SCREEN TIME
measured by using a SenseWear Pro armband (BodyMedia Inc)
on the upper right arm. The SenseWear Pro armband uses
accelerometry as well as a galvanic skin response to estimate
energy expenditures. The armband has shown an adequate val-
idity in physical activity measurement (19) and has been used in
laboratory-based VG studies (20). Heights and weights were
measured with subjects wearing light street clothes without shoes
by using a wall-mounted stadiometer (Perspective Enterprises
Inc) and calibrated scale (Tanita).
Participants responded to 4 questions on hunger, fullness, desire
to eat, and prospective food consumption by using a 200-mm
visual analog scale with anchors at either end (an example item
was “How strong is your desire to eat?” with anchors “very weak”
and “very strong”). The appetite score was calculated by adding
hunger, desire to eat, prospective food consumption, and 200
minus fullness and then dividing by 4 to produce an appetite score
between 1 and 200 [hunger 1 desire to eat 1 prospective food
consumption 1 (200 2 fullness)/4].
The total weight for each food and beverage consumed was
measured by subtracting the weight of food or beverage con-
tainers at the conclusion of the study period from the weight
measured before the study period. Kilocalories per gram were
calculated by using reported nutrition information from food and
beverage packaging. The energy expenditure in kilocalories was
measured over the entire 1-h study period. Kilocalories per ki-
logram of body weight per hour, which was equal to metabolic
intensity (sedentary behavior: ,1.5 METs; light-intensity physi-
cal activity: 1.5–2.9 METs; and moderate-intensity physical ac-
tivity: 3–5.9 METs) (21). The energy surplus was calculated as
the energy expenditure subtracted from the energy intake.
All statistical analyses were performed with SAS 9.2 software
(SAS Inc). Nonparametric ANCOVAs were performed to test
hypotheses (22, 23). This method needed only minimal assump-
tions beyond randomization in the study design. The sample size
was sufficient to apply a large sample approximation. The Markov
chain Monte Carlo method was used to impute missing data (10
copies) (24). A Markov chain is a sequence of random variables in
thepreviousone.Markov chain MonteCarlosimulation constructs
a long Markov chain to establish a stationary distribution, which is
the distribution of interest. By repeatedly simulating steps of the
chain, the method draws imputed estimates from the distribution.
proposed by Rubin (25). Correlations were performed to test as-
as the global assessment. If a global test was significant, pairwise
comparisons were examined with Bonferroni adjustment for man-
aging multiplicity. Logistic regression was used to investigate the
likelihood of consuming .500 kcal.
As a secondary analysis, the 2 sedentary conditions (television
watching and traditional VG; METs ,1.5) were collapsed. The
collapsed group was used as a reference in the comparison with
the MC group.
Means (6SDs) for participant characteristics are displayed in
Table 1. Sixty-two percent of the sample was white, 17% of the
samplewas black, 14%of the samplewasAsian,7%ofthesample
was classified as other, and 8% of the sample reported Hispanic
ethnicity. Most participants were normal weight (63%), 26% of
participants were overweight, and 11% of participants were obese,
4.4 y. Participants reported w2.5 h/d of combined television
watching and VG playing. No significant differences were shown
differed by sex (P , 0.001), but BMI did not differ by sex (P =
the study (P = 0.325).
Energy intake, expenditure, and surplus
a trend toward a significant difference was shown (P = 0.065;
Table 2). Overall, participants consumed a mean of 672 6 488
kcal from foods and beverages during the 1-h study period. Lo-
gistic regression analysis showed that consuming .500 kcal (n =
65) was predicted by male sex (OR: 3.2; 95% CI: 1.3, 7.9) and
watching television (compared with in the MC group, OR: 3.2;
95% CI: 1.2, 8.4). No difference was shown between VG and MC
groups (OR: 1.6; 95% CI: 0.6, 4.0). Not consuming any calories
(n = 9; 7 women; 5 overweight subjects) was rare.
In the secondary analysis, we collapsed the 2 sedentary
conditions to compare them to motion-controlled active gaming.
The comparison of sedentary to active screen time showed a
significant difference (P , 0.001). Participants who played MCs
consumed 179 kcal (95% CI: 8, 349 kcal) less than did subjects
who engaged in sedentary screen time. Men consumed 353 kcal
(95% CI: 193, 514 kcal) more than did women (P , 0.001).
Adjustment of the model for BMI or height did not alter findings
(P = 0.033). For the 2-group logistic regression, the comparison
of sedentary screen time to motion-controlled gaming showed
a trend toward significance for the consumption of $500 kcal
(OR: 2.21; 95% CI: 0.97, 5.02; P = 0.058).
A comparison of energy expenditures showed a significant
difference between the 3 groups (P , 0.001). When multiplicity
adjustment was considered, the MC group produced a signifi-
cantly higher energy expenditure than that of both the television
group (mean difference: 1.42 kcal $ kg21$ h21; 95% CI: 1.18,
1.66 kcal $ kg21$ h21; P , 0.001) and VG group (mean dif-
ference: 1.20 kcal $ kg21$ h21; 95% CI: 0.95, 1.45 kcal $ kg21$
h21; P , 0.001). A higher energy expenditure in the VG group
than in the television group was also shown (mean difference:
0.22 kcal $ kg21$ h21; 95% CI: 0.11, 0.33 kcal $ kg21$ h21; P =
0.001), although traditional VGs were still sedentary (1.30
METs). After adjustment for group and BMI, men produced 0.23
kcal $ kg21$ h21(95% CI: 0.04, 0.41 kcal $ kg21$ h21) more
than women did (P = 0.015). The secondary analysis also showed
a significant difference in energy expenditure (mean difference:
1.31 kcal $ kg21$ h21; 95% CI: 1.07, 1.55 kcal $ kg21$ h21;
P , 0.001) between sedentary screen time and MCs.
Energy surplus, which was operationalized as energy intake
minus energy expenditure, showed a similar pattern to energy
LYONS ET AL
expenditure. When the 3 groups were compared, energy surplus
differed significantly (P = 0.038) such that the energy surplus was
lower in the MC group than in the television group (mean dif-
ference: 2263 kcal; 95% CI: 2452, 273 kcal; P = 0.006) and
VG group (mean difference: 2279 kcal; 95% CI: 2480, 278
kcal; P = 0.007), which were significant after multiplicity ad-
justment by using the Bonferroni method. The 2-group compar-
ison in the secondary analysis was also significant (P = 0.002), in
which the MC group produced 271-kcal lower energy surplus
(95% CI: 101, 441 kcal) than did the sedentary screen time group.
The energy surplus was 317 kcal greater (95% CI: 155, 479 kcal)
in men than in women (P , 0.001). Most (n = 108) of the 120
participants were in a positive energy balance at the conclusion of
the 1-h study period (range: 2304 to 1923 kcal).
Energy intake and energy expenditure were unrelated when
analyzed by using the full sample (r = 20.04, P = 0.666) and
when analyzed separately for each group (television: r = 0.10,
P = 0.53; VG: r = 0.00, P = 0.989; and MC: r = 0.27, P = 0.089).
BMI was not related to intake (r = 20.02, P = 0.824) but
showed a trend toward an association with body mass–corrected
expenditure (r = 20.17, P = 0.070). The association between
height-adjusted energy intake and weight was not significant
(r = 20.31, P = 0.735). Age was not related to intake (r =
20.14, P = 0.115) or expenditure (r = 20.11, P = 0.237). Race
did not predict intake (P = 0.092) or expenditure (P = 0.983).
There was no difference between groups (P = 0.883) for
poststudy appetite, and the change in appetite from before to after
the study period also showed no difference by group (P = 0.291).
Appetite before the study period was positively associated with
energy intake (r = 0.467, P , 0.001), whereas poststudy appetite
(r = 20.384, P , 0.001) and change in appetite (r = 20.571, P ,
0.001) were negatively associated with energy intake. Prestudy,
poststudy, and change in appetite were not associated with energy
expenditure (P . 0.50), and the results did not differ by group.
No significant differences in the 3 groups were shown for
energy intake. However, a trend toward a significant difference
was shown (P = 0.065), and the energy expenditure was sig-
nificantly higher and the energy surplus significantly lower in
the MC group than in the television and traditional video-
gaming groups. In addition, in secondary analyses, when motion-
controlled gaming was contrasted with sedentary screen time
(television watching and traditional video gaming), differences in
energy intakes were significant. We hypothesize that this dif-
ference in results was due to issues related to statistical power.
Greater variability reduces power, and we showed greater vari-
ability than has been reported in previous studies of snacking
during television watching (8, 26).
Participants in this study showed a mean energy surplus of 555
kcal, which meant that they consumed, on average, more energy
during the 1-h study period than they expended; however, sub-
jects who played MCs had an energy surplus that was 273 kcal
lower than in the sedentary screen groups. Participants who
MCs to consume $500 kcal during the 1-h study period. Results
from the current study suggested that MCs may reduce the
positive energy balance associated with sedentary screen time.
Television (n = 40)
Video games (n = 40)
Motion-controlled video games (n = 40)
Total (n = 120)
26.1 6 4.7
23.2 6 4.3
24.6 6 4.7
22.8 6 4.8
24.4 6 3.8
23.6 6 4.2
23.8 6 4.8
24.3 6 4.2
24.0 6 4.4
24.1 6 4.4
177.2 6 7.7
165.0 6 5.2
171.1 6 9.0
179.9 6 8.8
163.7 6 6.3
171.8 6 11.1
163.4 6 6.4
179.9 6 8.8
170.1 6 9.2
171.0 6 9.8*
82.3 6 17.4
62.9 6 9.8
72.6 6 17.1
79.4 6 13.1
64.6 6 15.5
72.0 6 16.0
74.4 6 10.7
65.6 6 9.5
70.0 6 10.9
71.5 6 14.8*
26.3 6 5.0
23.1 6 3.6
24.7 6 4.6
24.6 6 3.1
24.0 6 4.9
24.3 6 4.0
23.9 6 3.1
24.7 6 4.5
24.3 6 4.2
24.4 6 4.1
111.4 6 36.2
134.0 6 20.4
122.7 6 31.2
124.8 6 33.6
115.7 6 29.1
120.3 6 31.3
120.2 6 31.1
105.1 6 33.8
112.7 6 33.0
118.5 6 31.9
79.1 6 37.3
68.5 6 35.1
73.8 6 36.1
79.8 6 31.0
69.2 6 39.4
74.6 6 35.3
75.7 6 32.1
77.6 6 37.2
76.7 6 34.3
75.0 6 35.0
1All values are means ± SDs. Differences were tested by using 2-way (sex and group assignment) ANCOVA. * Significant difference between sexes, P < 0.05.
ENERGY BALANCE DURING SCREEN TIME
Compared with eating without distraction, television watching
and gaming haveconsistently been shown to produce increases in
energy intakes (15–17). Comparisons of different media have
produced various results. In 2 studies, television watching and
an audio recording of a detective story were shown to produce
equivalent energy intakes (27, 28), and a comparison of seated
traditional gaming to traditional gaming while walking on a
treadmill also showed equivalent intakes (18). However, a com-
parison of television watching to listening to a symphony showed
a greater intake in the television group (26), and a comparison of
continuous television watching with watching a television seg-
ment repeatedly showed a greater intake in the continuous group
(10). The hypothesized mechanism by which these media in-
fluence intake is distraction from satiety cues (10).
It is possible that continuous television watching produces
greater narrative transportation (29) than watching a repeated
segment or listening to classical music, which may have led to
increased distraction from satiety cues. Narrative transportation
is the feeling of being immersed and absorbed in a story, which
study, MCs may have produced lower feelings of transportation
than did television or traditional VGs. None of the included MCs
included an involved storyline, whereas many of the traditional
games and television shows emphasized their narrative contents.
It is also possible that motion controllers could discourage
snacking because of the possibility that thegame could interpret
eating movements as intended game inputs, which could result
in poorer performance. The lack of an association between energy
expenditure and energy intake suggests that it was not the in-
tensity of the activity that produced a lower intake. Additional
of different screen-based media on eating behaviors. The use of
camera-based motion control, such as Xbox 360 Kinect games
(Microsoft), may produce differing results than those shown in
the current study from handheld motion control.
It has been reported that 36% of eating occasions in young
adults occur while multitasking (ie, while distracted) (30). It does
not appear that the substitution of traditional video gaming with
television watching affects energy intakes, but the substitution of
motion-controlled gaming may have a beneficial effect on both
sides of the energy-balance equation. Because of the widespread
use of television and VG technology as well as the prevalence of
eating while distracted by these media, the replacement of
television watching or traditional video gaming with motion-
controlled gaming has the potential for a broad public health
impact. However, such replacement would be a harm-reduction
kcal–positive energy balance.
Despite the many strengths of this study, results should be
interpreted in light of several limitations. Laboratory conditions
cannot accurately replicate in-home media consumption and
snacking behaviors. Results from this study should not be taken
as representative of the amount consumed during typical screen-
based leisure activities. The presence of highly palatable snack
foods and beverages offered at no cost likely produced greater
intakes than might be expected typically. An observational study
in the home environment or studies of screen-based media use
while participants eat self-purchased snacks or meals may pro-
effects of VGs on subsequent eating instances (15, 16); in this
Outcomes across groups and sexes1
Television (n = 40)
Video games (n = 40)
Motion-controlled video games (n = 40)
Food intake (kcal)
691 6 403
615 6 367
653 6 382
899 6 535
452 6 418
675 6 525
697 6 558
307 6 254
502 6 472
Soda intake (kcal)
73 6 89
54 6 76
63 6 82
114 6 107
28 6 67
71 6 98
73 6 93
29 6 68
51 6 84
Total intake (kcal)
764 6 424
669 6 396
716 6 407
1013 6 504
481 6 442
747 6 540
769 6 571
336 6 290
553 6 498
Energy expenditure (kcal $ kg21$ h21)
1.08 6 0.13
1.06 6 0.11
1.07 6 0.12*
1.36 6 0.22
1.24 6 0.35
1.30 6 0.29*
2.72 6 0.76
2.29 6 0.73
2.50 6 0.77
Energy surplus (kcal/h)
676 6 431
603 6 393
638 6 408*
906 6 505
404 6 442
655 6 533*
565 6 564
187 6 305
376 6 487
1All values are means 6 SDs. Differences were tested by using nonparametric ANCOVA adjusted for sex and pairwise comparisons with Bonferroni adjustment as necessary. Significant differences
between sexes are shown for all variables, P , 0.01 *Significantly less than for motion-controlled video games, P , 0.05.
LYONS ET AL
study, we evaluated intakes only during exposure to the media Download full-text
studied. Although evidence on the effects of exercise on appetite
later compensatory eating in some individuals, which would
attenuate the benefits of MCs on energy balance.
The instrument used to measure energy expenditure (ie, the
SenseWear Pro armband; BodyMedia Inc) is a valid measure but
not the gold standard of expenditure measurement. Thus, energy-
expenditure estimates may have been less precise than if they
had been measured by indirect calorimetry. Exercise- and fitness-
themed games were not included, and these games may produce
different results than those of other MCs. In addition, MCs vary in
their content and the extent to which motions are integrated into
programs used in this study were streamed rather than watched
live; the watching of network or cable television, which includes
commercial breaks, may also produce different results.
In conclusion, MCs hold promise as a healthier alternative to
sedentary screen time such as television watching and traditional
VG playing because of a lower energy surplus. MCs may also be
capable of producing health benefits associated with reduced
sedentary behavior and increased physical activity. However,
participants in the MC group, like those in the sedentary screen
time groups, were in a positive energy balance after 1 h of play.
Although less harmful than sedentary screen time, playing MCs
should not be assumed to produce a net benefit on weight or
health. Snacking on energy-dense sweet and savory foods and
sodas during screen time was consistently associated with pos-
itive energy balances across the 3 conditions. There is a need for
more research into the potential benefits as well as possible
drawbacks of integrating motion control into VGs.
We thank Phillip Carr and Stephanie Komoski for their assistance in data
acquisition and cleaning, Kristen Polzien for her assistance with energy-
expenditure measurements and analyses, and Karen Erickson for her assis-
tance with study administration.
The authors’ responsibilities were as follows—EJL, DFT, and DSW: de-
signed the research and wrote the manuscript; EJL and XW: analyzed data;
DFT, DSW, and XW: edited the manuscript; EJL: conducted the research
and had primary responsibility for the final content of the manuscript; and all
authors: read and approved the final manuscript. None of the authors had
a conflict of interest.
1. US Department of Labor. American Time Use Survey—2008 Results.
Available from: http://www.bls.gov/news.release/pdf/atus.pdf (cited 1
2. Dunstan DW, Barr EL, Healy GN, Salmon J, Shaw JE, Balkau B,
Magliano DJ, Cameron AJ, Zimmet PZ, Owen N. Television viewing
time and mortality: the Australian Diabetes, Obesity and Lifestyle
Study (AusDiab). Circulation 2010;121:384–91.
3. Ekelund U, Brage S, Froberg K, Harro M, Anderssen SA, Sardinha LB,
Riddoch C, Andersen LB. TV viewing and physical activity are in-
dependently associated with metabolic risk in children: the European
Youth Heart Study. PLoS Med 2006;3:e488.
4. Van den Bulck J, Van Mierlo J. Energy intake associated with television
viewing in adolescents, a cross sectional study. Appetite 2004;43:181–4.
5. Bowman SA. Television-viewing characteristics of adults: correlations
to eating practices and overweight and health status. Prev Chronic Dis
6. Thomson M, Spence JC, Raine K, Laing L. The association of tele-
vision viewing with snacking behavior and body weight of young
adults. Am J Health Promot 2008;22:329–35.
7. Chaput JP, Klingenberg L, Astrup A, Sjodin AM. Modern sedentary
activities promote overconsumption of food in our current obesogenic
environment. Obes Rev 2011;12:e12–20.
8. Bellissimo N, Pencharz PB, Thomas SG, Anderson GH. Effect of
televisionviewing at mealtime on food intake after a glucose preload in
boys. Pediatr Res 2007;61:745–9.
9. Temple JL, Giacomelli AM, Kent KM, Roemmich JN, Epstein LH.
Television watching increases motivated responding for food and en-
ergy intake in children. Am J Clin Nutr 2007;85:355–61.
10. Epstein LH, Saad FG, Giacomelli AM, Roemmich JN. Effects of al-
location of attention on habituation to olfactory and visual food stimuli
in children. Physiol Behav 2005;84:313–9.
11. Wang X, Perry AC. Metabolic and physiologic responses to video
game play in 7- to 10-year-old boys. Arch Pediatr Adolesc Med 2006;
12. Lyons EJ, Tate DF, Ward DS, Bowling JM, Ribisl KM, Kalyararaman
S. Energy expenditure and enjoyment during video game play: dif-
ferences by game type. Med Sci Sports Exerc 2011;43:1987–93.
13. Miyachi M, Yamamoto K, Ohkawara K, Tanaka S. METs in adults
while playing active video games: a metabolic chamber study. Med Sci
Sports Exerc 2010;42:1149–53.
14. Lanningham-Foster L, Foster RC, McCrady SK, Jensen TB, Mitre N,
Levine JA. Activity-promoting video games and increased energy ex-
penditure. J Pediatr 2009;154:819–23.
15. Chaput JP, Visby T, Nyby S, Klingenberg L, Gregersen NT, Tremblay
A, Astrup A, Sjodin A. Video game playing increases food intake in
adolescents: a randomized crossover study. Am J Clin Nutr 2011;93:
16. Oldham-Cooper RE, Hardman CA, Nicoll CE, Rogers PJ, Brunstrom
JM. Playing a computer game during lunch affects fullness, memory
for lunch, and later snack intake. Am J Clin Nutr 2011;93:308–13.
17. Brunstrom JM, Mitchell GL. Effects of distraction on the development
of satiety. Br J Nutr 2006;96:761–9.
18. Mellecker RR, Lanningham-Foster L, Levine JA, McManus AM. En-
ergy intake during activity enhanced video game play. Appetite 2010;
19. King GA, Torres N, Potter C, Brooks TJ, Coleman KJ. Comparison of
activity monitors to estimate energy cost of treadmill exercise. Med Sci
Sports Exerc 2004;36:1244–51.
20. Leatherdale ST, Woodruff SJ, Manske SR. Energy expenditure while
playing active and inactive video games. Am J Health Behav 2010;34:
21. Pate RR, O’Neill JR, Lobelo F. The evolving definition of “sedentary”.
Exerc Sport Sci Rev 2008;36:173–8.
22. Tangen CM, Koch GG. Non-parametric analysis of covariance for
confirmatory randomized clinical trials to evaluate dose-response re-
lationships. Stat Med 2001;20:2585–607.
23. Koch GG, Tangen CM, Jung JW, Amara IA. Issues for covariance
analysis of dichotomous and ordered categorical data from randomized
clinical trials and non-parametric strategies for addressing them. Stat
24. Schafer JL. Analysis of incomplete multivariate data. New York, NY:
Chapman and Hall, 1997.
25. Rubin DB. Multiple imputation for nonresponse in surveys. New York,
NY: John Wiley & Sons Inc, 1987.
26. Blass EM, Anderson DR, Kirkorian HL, Pempek TA, Price I, Koleini
MF. On the road to obesity: Television viewing increases intake of
high-density foods. Physiol Behav 2006;88:597–604.
27. Bellisle F, Dalix AM, Airinei G, Hercberg S, Peneau S. Influence of
dietary restraint and environmental factors on meal size in normal-
weight women. A laboratory study. Appetite 2009;53:309–13.
28. Bellisle F, Dalix AM, Slama G. Non food-related environmental
stimuli induce increased meal intake in healthy women: comparison of
television viewing versus listening to a recorded story in laboratory
settings. Appetite 2004;43:175–80.
29. Green MC, Brock TC. The role of transportation in the persuasiveness
of public narratives. J Pers Soc Psychol 2000;79:701–21.
30. Laska MN, Graham D, Moe SG, Lytle L, Fulkerson J. Situational
characteristics of young adults’ eating occasions: a real-time data
collection using Personal Digital Assistants. Public Health Nutr (Epub
ahead of print 8 December 2010).
31. Hopkins M, King NA, Blundell JE. Acute and long-term effects of
exercise on appetite control: is there any benefit for weight control?
Curr Opin Clin Nutr Metab Care 2010;13:635–40.
ENERGY BALANCE DURING SCREEN TIME