PSYCHOLOGICAL ISSUES (V DRAPEAU AND V IVEZAJ, SECTION EDITORS)

The Impact of Physical Activity on Food Reward: Reviewand Conceptual Synthesis of Evidence from Observational, Acute,and Chronic Exercise Training Studies

Kristine Beaulieu1& Pauline Oustric1 & Graham Finlayson1

# The Author(s) 2020

AbstractPurpose of Review This review brings together current evidence from observational, acute, and chronic exercise training studiesto inform public debate on the impact of physical activity and exercise on food reward.Recent Findings Low levels of physical activity are associated with higher liking and wanting for high-energy food. Acute boutsof exercise tend to reduce behavioral indices of reward for high-energy food in inactive individuals. A dissociation in liking(increase) and wanting (decrease) may occur during chronic exercise training associated with loss of body fat. Habitual moderate-to-vigorous physical activity is associated with lower liking and wanting for high-fat food, and higher liking for low-fat food.Summary Food reward does not counteract the benefit of increasing physical activity levels for obesity management. Exercisetraining appears to be accompanied by positive changes in food preferences in line with an overall improvement in appetitecontrol.

Keywords Physical activity . Exercise . Food reward . Appetite . Liking and wanting . Obesity management

Introduction

Among the reasons that people with obesity cite for avoidingexercise is a lack of enjoyment and perceived failure to loseweight [1, 2]. Moreover, there is a misconception that persistsamong some individuals that exercise is counter-productivefor weight management. This common assertion is reinforcedby occasional reports in the media about exercise and foodreward [3, 4]. Biological explanations, reliant on soft evi-dence, have been put forward suggesting that glycogen deple-tion, reduced blood glucose levels, endorphin release or othersignals generated during exercise can increase appetite orcause specific cravings for foods. Alternatively, psychologicalaccounts propose that high fat or sugary foods are sought outpost-exercise to counteract negative affect or reward virtuousbehavior. Research over the past 10 years has shown that

physical activity and eating behavior are loosely coupled,but the physiological and neurocognitive mechanisms thatcontribute to this relationship are complex [5]. Moreover, theevidence for the impact of physical activity on food reward isdifficult to assess due to the absence of randomized controlledtrials and differences between study designs—encompassingobservational, acute, and chronic interventions. Differencesalso exist in the modality and intensity of physical activityexamined and the variety of methodologies used to measurereward responses to food. In a review of longitudinal weightmanagement interventions that measured food reward out-comes at baseline and follow-up, Oustric and colleagues[6••] identified 17 studies consisting of dietary, pharmacolog-ical, cognitive, and exercise-based intervention types. Overall,a post-intervention reduction in food reward was found acrossall treatment types—except exercise, where no consistentchanges were reported. While it is interesting to speculate thatexercise training may have a moderating influence on therelationship between weight management and food reward,only three exercise studies were eligible for inclusion in thissystematic review. Moreover, interpretation was limited bysmall sample sizes, lack of a control condition in one study,and inconsistent use of food reward measures. Further re-search is needed to update and summarize the available

This article is part of the Topical Collection on Psychological Issues

* Graham FinlaysonG.S.Finlayson@leeds.ac.uk

1 Appetite Control & Energy Balance Research Group, School ofPsychology, Faculty of Medicine & Health, University of Leeds,Leeds, West Yorkshire, UK

https://doi.org/10.1007/s13679-020-00372-3Current Obesity Reports (2020) 9:63–80

Published online: 15 April 2020

evidence from observational, acute, and chronic study designsto gain an overview of the influence of physical activity andexercise on food reward. With this objective in mind, thespecific aims of this review were to describe the role of phys-ical activity and food reward in appetite control and weightmanagement; to tabulate and synthesize the different types ofevidence that have addressed the impact of physical activityon food reward; and to discuss, and where possible harmo-nize, the findings in the light of relevant moderators and meth-odological issues. To our knowledge, there is currently nocomprehensive review of the literature on physical activityand food reward. A critical examination of the evidence mayhelp to clarify some of the perceived barriers for engaging inphysical activity and exercise for obesity management.

Current Thinking on the Role of PhysicalActivity in Appetite Control

In the last 5 years, evidence has accrued showing that physicalactivity affects both episodic (meal-to-meal) and tonic (basal)homeostatic mechanisms that influence appetite control [7].Acute exercise influences gastric emptying [8], and attenuatesthe release of ghrelin and increases the secretion of satietypeptides, e.g., peptide YY, glucagon-like peptide-1, and pan-creatic polypeptide [9]. Chronic exercise improves body com-position [10–12], and leptin and insulin sensitivity [13–15].

Our research has shown that habitual physical activity has asmall positive association with daily energy intake (account-ing for around ~ 3% of the variance) [16], but this is onlylogical when considering the increase in energy requirementsand the longer-term indirect effects from increased restingmetabolic rate after changes in fat-free mass. It has been pro-posed that chronic exercise influences appetite control throughan increase in hunger but also a strengthening of post-mealsatiety [17]. Indeed, exercise and physical activity appear tointeract with nutritional factors to enhance satiety signaling,with studies showing that people who engaged in more exer-cise and physical activity were better able to compensate fordifferences in the energy content of food (achieved by increas-ing the ratio of fat to carbohydrate) at a subsequent meal thantheir less active counterparts [18]. While more active individ-uals are driven to eat more due to their greater energy require-ments, their stronger satiety response to food appears to allowfor a better matching between energy intake and energy ex-penditure [19••]. This relationship between physical activitylevel and daily energy intake is best represented by a J-shapedcurve, whereby individuals with low levels of physical activ-ity on the left of the curve have dysregulated appetite withgreater intake than expenditure, and individuals with greaterlevels of physical activity towards the right of the curve have aproportional increase in intake with increasing expenditure.These findings suggest that concerns about exercise or

increased levels of physical activity being counter-productive for obesity management should be concerned withnon-homeostatic and food reward-related mechanisms thatmay be driving unhealthy food choices.

Defining Food Reward and Its Importancefor Weight Management

Food reward is important for understanding appetite controland has the logical status of an intervening variable that guideseating behavior. Food reward is encoded by distinct neuralpathways in the brain and can be modulated by metabolicsignaling, sensory stimuli from the food environment and cog-nitive processes such as attention, learning, and memory [20].Food reward is often conceptualized to consist of two distinctsub-components—“liking” and “wanting.” These have beenbroadly studied [21], following extensive work demonstratingtheir dissociation in the brain and behavior of many speciesincluding humans [22]. Liking is the sensory pleasure exertedwhile eating a food and wanting is rather the, often implicit,drive to eat triggered by a food cue [23]. Liking and implicitwanting for energy-dense foods are related to excess energyintake in free-living and laboratory settings [23, 24]. However,liking accounts only for a small proportion of the variance infood intake [25, 26], and unconscious processes such as im-plicit wanting may play a larger role in driving overeating [27,28]. Food reward is an important factor in weight managementthrough its intervening status between the nutritional require-ments of the body and stimulation from the food environment[23], but it is likely that liking and implicit wanting sometimesact separately to influence appetite control [6••].

While several techniques to measure food reward havebeen developed [21] (e.g., reinforcing value tasks, willingnessto pay), the Leeds Food Preference Questionnaire (LFPQ) hasbeen designed to measure both liking and implicit wanting fordistinct dimensions (e.g., fat and sweet taste) of foods com-mon in the diet. Food reward measured by the LFPQ can beinterpreted as both a state- and a trait-dependent measure de-pending on the timing and condition of measurement. Indeed,liking and wanting have been shown to be partially dissocia-ble pre to post food consumption according to sensory-specific satiety [29]. On the contrary, food cravings, definedhere as spontaneous instances of strong explicit wanting for aspecific food, tend to be measured as a trait reflecting thefrequency, intensity or quality of cravings experienced overa specified time period [30–32]. Neural activation to foodsmeasured through functional magnetic resonance imaging(fMRI) are also used as an inference of food reward and thistechnique allows the analysis of different regions of interestand their functional connectivity [33]. However, fMRI mea-sures of reward should be used in conjunction with behavioralmeasures or actual food intake to support their interpretation.

Curr Obes Rep (2020) 9:63–8064

Review of Literature

Methods

The inclusion and exclusion criteria for this non-systematicreview were established prospectively. We included studieswith a general, healthy population of adults (≥ 18 years) orchildren (< 18 years), including those with overweight or obe-sity. Studies that included adults or children with overweightor obesity who were specifically trying to lose weight werealso included. We excluded studies in populations with dis-eases (including substance related and addictive disorders),in vitro and animal studies. The interventions and compari-sons of interest included any type of structured exercise orcomparisons between different physical activity levels. Allchronic training studies had to have a minimum interventionduration of 4 weeks. The outcomes of interest included allpsychometric measures of food reward obtained either directly(e.g., ratings or pleasantness or desire to eat) or indirectly (e.g.,measure of the willingness to work to obtain a food, reactiontime), as well as neuronal response to food cues measured byfMRI. We included unpublished and ongoing studies whererelevant.

The search strategy for this review combined electronicsearches and hand searching. To identify ongoing or complet-ed, but unpublished trials, ClinicalTrials.gov was searched on22 November 2019 and researchers known to be active in thetopic were contacted. Limits were set to include all paperspublished in English or French after 2009, in healthy humansamples. Authors were contacted for additional information ifunclear or not reported in the manuscript.

Observational Studies of Physical Activityand Exercise

As shown in Table 1, only three studies have examined foodreward differences in defined active and inactive groups [34,35, 38]. Horner et al. [38] found that in the fed state, overallliking, and liking for high-fat savory, high-fat sweet, and low-fat sweet food measured by the LFPQ was lower in activecompared with inactive men differing in BMI, and differencesfor foods overall and low-fat sweet foods remained afteradjusting for differences in percentage body fat. In both fedand hungry states, active men had a greater implicit wantingfor low-fat savory foods comparedwith inactivemen, but onlythe differences in the fed state remained after adjusting forpercentage body fat. Faster gastric emptying was found to beassociated with greater liking for savory food and lower im-plicit wanting for high-fat food. Two studies in men and wom-en differing in physical activity levels but matched for BMI (~23 kg/m2) found no differences in liking or wanting for high-fat relative to low-fat food in the hungry or fed states [34, 35].

In terms of correlational studies, objectively measuredMVPAwas found to be inversely associated with liking (r =− 0.25, p < 0.001) and wanting (r = − 0.27, p = 0.001) forhigh-fat relative to low-fat food measured by the LFPQ in156 women across a range of BMI [41]. This is in line witha study showing that self-reported physical activity was asso-ciated with reduced fMRI responses to high-energy relative tolow-energy food [39]. Another fMRI study in participantsranging in BMI found an inverse association between self-reported moderate-to-vigorous physical activity and brain re-sponse to food cues [40•]. This relationship appeared to bemore prominent in the participants with obesity than the leanparticipants. In individuals with overweight/obesity and im-paired fasting glucose and/or glucose tolerance, food com-pared with non-food brain activation was negatively associat-ed with leisure-time physical activity [36•]. Interestingly, therewas a positive association between brain activation and work-related physical activity, which lost statistical significance af-ter adjusting for BMI and age, and there was no associationwith sport-related physical activity.

Overall, these studies suggest a reduction in food reward(both liking and wanting) with increasing levels of habitualphysical activity, particularly at higher levels of adiposity.Those who accumulate more time in moderate-to-vigorousphysical activity tend to prefer low-fat food and those whoengage in more sedentary activity prefer high-fat food.These findings do not seem to be restricted to those whoperform structured exercise. The inter-relationships betweenfood reward, physical activity, and adiposity remain to be fullydisentangled, as well as the mechanisms underlying these ob-served effects. One simple explanation could be that avoid-ance of high-fat food perceived to be unhealthy, preference forlow-fat food and greater physical activity levels are all drivenby a dispositional desire to be healthy and engage in a range ofpositive health behaviors [64, 65]. Alternatively, it has beenproposed that physical activity may act as a reward “buffer”against liking and wanting for high-fat foods, whereas lowlevels of physical activity may render people more susceptibleto hedonic eating [66, 67]. The possibility of stronger inverseassociations between physical activity and food reward in leangroups versus groups with obesity suggests that exercise andleisure-time physical activity may be an effective strategy forcontrolling hedonic eating in people with obesity.

Acute Exercise Studies

At least eight studies have investigated the effect of acuteexercise on food reward measured by the LFPQ which allowsdirect comparison of outcomes (Table 1). All these studies hada no-exercise control group except for Alkahtani et al. [42],but the exercise varied in intensity (low to high), modality(aerobic vs resistance; cycling vs swimming) or in musclecontraction type (eccentric vs concentric). Two studies

Curr Obes Rep (2020) 9:63–80 65

Table1

Descriptio

nof

theobservational,acute,andchronictraining

studiesexam

iningtheim

pactof

physicalactiv

ityon

food

reward

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

Observationalstudies

Beaulieuetal.

2017

[34]

UK

Sex:

females

andmales

BMIstatus:<

29.9

kg/m

2

Active:n=20

(50%

males)

Age:3

0(10)

years

BMI:22.6(1.9)kg/m

2

Inactive:n=19

(42%

males)

Age:3

0(9)years

BMI:23.1(2.7)kg/m

2

PAlevelassessm

ent:IPAQandPA

monito

r(SenseWear)

Active:≥4exercise

sessions/week

(MVPA

=182(67)

min/day)

Inactive:≤1exercise

session/week

(MVPA

=103(37)

min/day)

Exercisesession:

≥40

min

MVPA

-LFP

Q:likingandim

plicitwantin

gbias

forfat

-Setting:laboratory

-State:preandposthigh-fator

high-carbohydratelunch(adlibitu

m)

-Nodifference

infood

reward(likingand

wantin

g)betweengroups.

-Differencebetweenlik

ing/wantin

g:none

-Fo

odintake:energyintake

greaterin

high-fatrelativ

eto

high-carbohydrate

inboth

groups.

-Eatingbehavior

traits:tendencyfor

restrainttobe

higher

inHiPA

comparedwith

LoPA

-Bodycomposition:

HiPAhadlower

body

fatp

ercentagethan

LoPA

Beaulieuetal.

2017

[35]

UK

Sex:

females

andmales

BMIstatus:<

29.9

kg/m

2

HiM

VPA

:n=12

(33%

males)

Age:2

9(10)

years

BMI:22.4(2.1)kg/m

2

ModMVPA

:n=11

(27%

males)

Age:2

6(3)years

BMI:22.7(2.2)kg/m

2

LoM

VPA

:n=11

(27%

males)

Age:3

0(11)

years

BMI:23.1(2.9)kg/m

2

PAlevelassessm

ent:Tertilesof

daily

MVPA

measuredby

PAmonito

r(SenseWear)

HiM

VPA

:174

(39)

min/day

ModMVPA

:121

(15)

min/day

LoM

VPA

:83(16)

min/day

-LFP

Q:likingandim

plicitwantin

gbias

forfat

-Setting:laboratory

-State:preandpostpreloads—HE

(~700kcal)andLE(~

260kcal)

relativ

eto

watercontrol(0kcal)

-Nodifference

infood

reward(likingand

wantin

g)betweengroups

-Differencebetweenlik

ing/wantin

g:none

-Fo

odintake:g

reaterreductionin

liking

andwantingafterHEpreloadrelative

toLEpreloadin

allg

roups.

ModMVPA

andHiM

VPA

reducedad

libitumenergy

intake

atthelunchmeal

follo

wingconsum

ptionof

theHE

comparedwith

theLEpreload,while

theLoM

VPA

groupdidnot.Noeffect

ofMVPA

groupon

energy

intake

atdinner

oreveningsnackbox.

-Eatingbehavior

traits:n

odifferences

betweengroups

-Bodycomposition:

nodifferences

betweengroups

Drummen

etal.2019

[36•]

Netherlands

Sex:

females

andmales

BMIstatus:>

25kg/m

2

n=39

(56%

males)

Age:5

3(11)

years

BMI:32.3(3.7)kg/m

2

Participantshadim

paired

fasting

glucoseand/or

glucose

tolerance

PAlevelassessm

ent:Baecke

questio

nnaire

(work,sport,and

leisure-tim

ephysicalactiv

ity)

[37]

Work=2.6(0.8)range1.3–4.3

Sport=

2.6(0.7)range1.0–4.0

Leisure=2.9(0.7)range1.5–4.4

Scores

rangingfrom

1(low

levelo

fPA

)and5(highlevelo

fPA

)

-fM

RI:BOLDsignalsto

high-energy

food,low

-energyfood

andnon-food

images

-Setting:laboratory

-State:afterovernightfast

-Inverseassociationbetween

leisure-tim

ePA

andfood

compared

with

non-food

brainactiv

ationin

the

rightthalamus,leftm

iddlecingulate

gyrus,rightp

recuneus,leftp

utam

en,

andleftangulargyrus.

-Po

sitiv

eassociationbetween

work-relatedphysicalactiv

ityand

brainactiv

ation(disappeared

after

adjustingforBMIandage)

-Noassociationwith

sport-related

physicalactiv

ity.

-Differencebetweenliking/wanting:

NR

-Fo

odintake:N

R-Eatingbehavior

traits:p

ositive

associationbetweenbrainactivation

anddisinhibition.N

oassociationwith

restrainto

rsusceptibility

tohunger.

-Bodycomposition:

noassociation

Horneretal.

2016

[38]

Australia

Sex:

males

BMIstatus:1

8–40

kg/m

2

Active:n=22

Age:2

9(8)years

BMI:24.5(2.6)kg/m

2

Inactive:n=22

Age:3

1(9)years

BMI:27.4(4.2)kg/m

2

PAlevelassessm

ent:Self-reportand

PAmonitor(A

ctiGraph)

Active:≥4exercise

sessions/week

(PA=709(239)kcal/day)

Inactive:≤1exercise

session/week

(PA=525(185)kcal/day)

Exercisesession:

≥40

min

MVPA

-LFP

Q:likingandim

plicitwantin

gfor

HFS

W,L

FSW,H

FSA,L

FSA

-Setting:laboratory

-State:postfixedbreakfast(400kcal)

andprelunch5hlater

-Fed:

Activehadlower

likingfor

high-fatfoods,low-fatsw

eetfoods

andforfoodsoverallcom

paredwith

inactiv

e.Activehadgreaterwanting

forlow-fatsavory

foods.

-Hungry:

Nodifferencesin

liking

betweenactiv

eandinactiv

e.Active

hadgreaterwantin

gforlow-fatsavory

foods.

-Gastricem

ptying:Inverse

association

with

post-prandialchanges

inlik

ing

low-fatsavory

foods.Po

sitive

associationwith

likingtastebias

inhungry

state(i.e.,fastergastric

emptying

associated

with

greater

likingforsavory

foods).N

oassociationwith

likingfatb

iasnor

implicitwantingtastebias.P

ositive

associationwith

implicitwantin

gfat

Curr Obes Rep (2020) 9:63–8066

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

-Fedto

hungry:A

ctivehadgreater

increase

inlik

ingforallfood

categories

combinedthan

inactive.

-Differencebetweenlik

ing/wantin

g:Fastergastricem

ptying

associated

with

likingforsavory

foodsand

slow

ergastricem

ptying

associated

with

greaterim

plicitwantingfor

high-fatfoods.

bias

infedandhungry

states.E

ffects

independento

fbody

fat.

Killgore

etal.

2013

[39]

USA

Sex:

females

andmales

BMIstatus:1

9.8–34.8

kg/m

2

n=37

(59%

males)

Age:3

0(8)years

BMI:24.5(3.7)kg/m

2

PAlevelassessm

ent:Self-reported

habitualPA

inmin/week(typical

days/week×duration/day)

PA=151(160)min/week(range

0–540)

-fM

RI:response

tohigh-energyfood,

low-energyfood

andnon-food

images

-Su

bjectivefood

preferences:desire

toeatd

epictedfood

item

atthatmom

ent

(VAS)

-Setting:laboratory

-State:1hfasted

(pre

scan

food

intake

323(245)kcal)

-InverseassociationbetweenhabitualPA

andfM

RIresponses(m

edial

orbitofrontalcortexandleftinsula)to

high-energyfoodsrelativeto

low-energyfoods

-InverseassociationbetweenhabitualPA

andpreference

forhigh-energysavory

foods

-Su

bjectiv

efood

preference:fMRI

responsespositiv

elyassociated

with

preference

forhigh-energysavory

foods(not

forhigh-energysw

eet

foods).N

oassociationbetweenfM

RI

responsesto

high-energyrelativ

eto

low-energyfoodsandpreference

for

low-energyfoods.

Luo

etal.

2018

[40•]

USA

Sex:

females

andmales

BMIstatus:N

Rn=40

(48%

males)

Leanindividuals:n=22

(46%

males)

Age:2

1(2)years

BMI:22.6(1.9)kg/m

2

Individualswith

obesity:n

=18

(50%

males)

Age:2

2(2)years

BMI:35.2(4.0)kg/m

2

PAlevelassessm

ent:Self-reported

from

3to

524-h

recalls

over

2months(m

eandaily

minutes

ofMVPA

i.e.,activities

≥3METs)

Leanindividuals:MVPA

=125(84)

min/day

Individualswith

obesity:

MVPA

=134(114)min/day

-fM

RI:responsesto

high-energyfood

andnon-food

images

-Setting:laboratory

-State:9–11

amaftero

vernightfast,task

performed

20–30min

after75

gglucoseingestion

-Inverse

associationbetweenMVPA

and

brainresponse

tofood

cues

inmiddle

insulaandleftpostcentralg

yrus.

-Individualswith

obesity:inverse

associationbetweenMVPA

andbrain

responses.

-Leanindividuals:non-significant

inverseassociationbetweenMVPA

andbrainresponses.

-Associatio

nbetweenMVPA

andbrain

responsesstronger

inmales

than

females.

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Oustricetal.

2018

[41]

UK

Sex:

females

BMIstatus:1

8.5–45.0

kg/m

2

n=156

Age:5

3(11)

years

BMI:32.3(3.7)kg/m

2

Pooled

datafrom

6studies

PAlevelassessm

ent:Quintilesof

daily

MVPA

measuredby

PAmonito

r(SenseWear)

Q1:

25(8)min/day

Q2:

53(9)min/day

Q3:

83(9)min/day

Q4:

120(12)

min/day

Q5:

197(62)

min/day

-LFP

Q:likingandim

plicitwantin

gbias

forfat/taste

-Setting:laboratory

-State:hungry

(3–10hfast)

-MVPA

inverselyassociated

with

liking

andim

plicitwantingfatb

ias.

-MVPA

positiv

elyassociated

with

liking

tastebias.

-Q5greaterlik

ingandwantin

gfor

low-fatfoods,whileQ1-Q3greater

likingandwantin

gforhigh-fatfoods.

-Differencebetweenlik

ing/wantin

g:none

-Fo

odintake:N

R-Eatingbehavior

traits:N

oassociation

betweenPA

andfood

cravings.

Craving

forsw

eetfood(Control

ofEatingQuestionnaire;C

oEQ)

positivelyassociated

with

explicit

likingandim

plicitwantin

gforsw

eet

foodson

theLFP

Q.C

raving

for

savory

foods(CoE

Q)associated

with

LFP

Qexplicitwantin

gforsavory

foods.Craving

control(CoE

Q)

negativ

elyassociated

with

implicit

wantin

gforhigh-fatfoods.

-Bodycomposition:

Fatm

assindex

(FMI)butn

otwaistcircum

ference

(WC)was

inverselyassociated

with

explicitlikingandim

plicitwantin

gfor

Curr Obes Rep (2020) 9:63–80 67

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

sweetrelativeto

savory

foods.WC

was

positiv

elyassociated

with

all

likingandim

plicitwantin

gforhigh-fat

relativ

etolow-fatfoods.FM

Iwas

also

associated

with

implicitwantin

gfor

high-fatfoods.

Acuteexercise

studies

Alkahtani

etal.2014

[42]

Australia

Sex:

males

BMIstatus:>

25kg/m

2

PAlevel:sedentary(criteriaNR)

n=12

Age:2

9(4)years

BMI:29.1(2.4)kg/m

2

-Intensity:m

oderate-intensity

intervaltraining

(MIIT;

alternatingbetween±20%

FATmax)vs.high-intensity

intervaltraining

(HIIT;

85%

VO2peak)

-Ty

pe:cycleergometer

-Duration:

MIIT5-min

stages

at±2

0%FA

Tmaxfor30

min,H

IIT

15-sintervalsand15-srecovery

(workloadmatched

toMIIT;

~18

min)

-Tim

ing:

morning

-Control

condition:n

o;MIITvs.

HIIT

-LFP

Q:likingandim

plicitwantin

gfor

HFS

W,L

FSW,H

FSA,L

FSA

-Setting:laboratory

-State:beforeandafterexercise

(afteran

overnightfast)

-Decreasein

wantingandincrease

inlik

ingforallthe

food

categories

independento

ftheintensity

-Differencebetweenlik

ing/wantin

g:lik

ingincreasedwhilewantin

gdecreasedbutw

ithouta

controlthis

response

might

bedueto

theeffectof

timerather

than

exercise

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Alkahtani

etal.2019

[43]

SaudiA

rabia

Sex:

males

BMIstatus:N

RPA

level:moderatelyactive

(2–5

hstructured

aerobic

exercise/week)

n=14

(8forfood

rewarddata)

Age:2

4(6)years

BMI:23.4(3.3)kg/m

2

-Intensity

:moderate(60%

VO2m

ax)

interspersed

with

low(30%

VO2m

ax)

-Ty

pe:contractio

ntype

eccentric

(dow

nhill

runningat−12%

inclination)

vs.concentric(flat

running)

-Duration:

5stages

of8min

at60%VO2max/2

min

at30%VO2max

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-LFP

Q:likingandim

plicitwantin

gbias

forfat/taste

-Setting:laboratory

-State:before,after

exercise

and24

hafterexercise

(beforean

adlibitu

mlunch)

-Nochange

infood

rewardafterexercise

-Differencebetweenlik

ing/wantin

g:greaterlik

ingof

savory

foodsover

sweetfoods

indownhill

runningthan

frontrunning

-Fo

odintake:n

ochange

-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Crabtreeetal.

2014

[44]

UK

Sex:

males

BMIstatus:2

1.8–26.6

kg/m

2

PAlevel:moderatelyactive

(2h/week)

n=16

Age:2

3(3)years

BMI:24.2(2.4)kg/m

2

-Intensity:h

igh(70%

VO2max)

-Ty

pe:treadmill

run

-Duration:

60min

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-fM

RI:BOLDsignalsto

high-and

low-energyfood

cues

comparedwith

non-food

pictures

-Setting:laboratory

-State:fasted

-Decreased

activationinthepallidumfor

high-energyfood

andincrease

for

low-energyfood

afterexercise

-Differencebetweenliking/wanting:

NR

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Evero

etal.

2012

[45]

USA

Sex:

females

andmales

BMIstatus:<

25kg/m

2

PAlevel:habitually

activ

e(>

3h/week)

n=30

(57%

males)

-Intensity:h

igh(83%

HRmax)

-Ty

pe:cycleergometer

-Duration:

60min

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-fM

RI:BOLDsignalsto

high-and

low-energyfood

cues

comparedwith

neutralcontrol

-Setting:laboratory

-Exercisereducedtheneuronalresponse

tofood

cues

inbrainregionsrelated

with

food

reward(i.e.,insula,

putamen,rolandicoperculum)

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Curr Obes Rep (2020) 9:63–8068

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

Age:2

2(4)years

BMI:23.6(2.2)kg/m

2-State:fM

RIwas

performed

after

exercise

afteran

8–12

hovernightfast-Differencebetweenlik

ing/wantin

g:decrease

inregionsrelatedto

liking

andwantingbutn

obehavioral

measures

Farahetal.

2012

[46]

UK

Sex:

females

andmales

BMIstatus:n

olim

itsPA

level:NR

n=27

(52%

males)

Female:n=13

Age:2

6(3)years

BMI:BMI:22.8(3.1)kg/m

2

Male:n=14

Age:3

6(10)

years

BMI:26.1(3.3)kg/m

2

-Intensity:m

oderate(6

METs)

-Ty

pe:treadmill

walk

-Duration:

60min

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-VAS:

liking

-Setting:laboratory

-State:im

mediately,60,120,and

180min

afterexercise

(overnight

fasted)

-Nochange

inlik

ing

-Differencebetweenlik

ing/wantin

g:wantin

gnotm

easured

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Finlayson

etal.2009

[47]

UK

Sex:

females

BMIstatus:<

25kg/m

2

PAlevel:2.4(1.2)

engagements/week

n=24

Age:2

4(6)years

BMI:22.3(2.9)kg/m

2

-Intensity:m

oderate(70%

HRmax)

-Ty

pe:cycleergometer

-Duration:

50min

-Tim

ing:

morning

-Controlcondition:yes,noexercise

-LFP

Q:relativepreference

(food

choice),lik

ingandim

plicitwantin

gforHFS

W,L

FSW,H

FSA,L

FSA

-Setting:laboratory

-State:beforeandafterexercise(2

hafter

afixedbreakfast,kcalNR)and

afteran

adlib

itum

lunch30-m

inpostexercise

-Increase

inim

plicitwantin

gafter

exercise

inthosewho

compensated

oratemoreatthead

libitu

mlunchin

response

toexercise

-Differencebetweenlik

ing/wantin

g:Changes

inim

plicitwantin

gbutn

otlik

ing

-Fo

odintake:A

fter

exercise

some

individualsincreasedtheirenergy

intake

(com

pensators)andhad

enhanced

implicitwantin

g-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Martin

setal.

2015

[48]

Norway

Sex:

females

andmales

BMIstatus:>

25kg/m

2

PAlevel:sedentary(criteriaNR)

n=12

(42%

males)

Age:3

3(10)

years

BMI:32.3(2.7)kg/m

2

-Intensity:H

IITand½HIIT(allout;

average~85%

HRmax),

continuous

(70%

HRmax)

-Ty

pe:H

IITvs.½

HIITvs.

continuous

cycling

-Duration:

HIIT(8-sintervalsand

12-srecovery

for250kcal;

~18

min),½

HIIT(8-sintervals

and12-srecovery

for125kcal;

~9min),continuous

exercise

(250

kcal;~

27min)

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-LFP

Q:relativepreference

(food

choice),lik

ingandim

plicitwantin

gbias

forfat

-Setting:laboratory

-State:before

andafterexercise/b

efore

anad

libitum

lunch

(standardizedbreakfasto

f600kcal

consum

ed1hbefore

exercise/rest)

-Nochange

infood

reward

-Differencebetweenlik

ing/wantin

g:no

differences

-Fo

odintake:n

ochange

-Eatingbehavior

traits:N

R-Bodycomposition:

NR

McN

eiletal.

2015

[49]

Canada

Sex:

females

andmales

BMIstatus:<

25kg/m

2

PAlevel:inactive

(<150min/week)

n=16

(50%

males)

Age:2

2(3)years

BMI:22.8(1.8)kg/m

2

-Intensity:h

igh(aerobic70%

VO2peak,resistance70%

1-repetitionmaxim

um)

-Ty

pe:aerobicvs.resistance

-Duration:

aerobic~24

min,

resistance

~86

min

(matched

for

energy

expenditu

reat4kcal/kg;

~275kcal)

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-LFP

Q:relativepreference

(food

choice),lik

ingandim

plicitwantin

gbias

forfat/taste

-Setting:laboratory

-State:preandpostad

libitum

lunch

30min

afterexercise

(standardized

breakfasto

f534kcalconsum

ed~1.5hbefore

exercise)

-Decreasein

therelativepreference

for

high-fatrelativ

eto

low-fatfoodsafter

both

exercise

-Decreasein

likingforhigh-fatfoods

follo

wingresistance,but

notaerobic

-Differencebetweenlik

ing/wantin

g:change

infood

choice

andliking,but

notimplicitwantin

g

-Fo

odintake:n

ochange

-Eatingbehavior

traits:N

R-Bodycomposition:

nodifference

inbodyweight

Miguetetal.

2018

[50]

Sex:

females

andmales

(adolescents)

-Intensity:h

igh(70%

,75%

,80%

,85%,and

90%

HRmax)

-LFP

Q:relativepreference,likingand

implicitwantingbias

forfat/taste

-Fo

odintake:d

ecreasein

energy

intake

atlunchanddinner

Curr Obes Rep (2020) 9:63–80 69

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

France

BMIstatus:>

29.9

kg/m

2

PAlevel:inactive(<2h/week)

n=33

(36%

males)

Age:1

3(1)years

BMI:35.0(4.3)kg/m

2

-Ty

pe:h

igh-intensity

interval

training

cycling

-Duration:

5×2-min

increasing

intensity

intervalsfollowed

by30-srecovery

(15min)

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-Setting:laboratory

-State:preandpostad

libitum

lunch

30min

afterexercise

(standardized

breakfasto

f500kcalconsum

ed2.5h

before

exercise)

-Decreasein

implicitwantin

gforsw

eet

intheexercisecondition

vs.increasein

thecontrol

-Differencebetweenlik

ing/wantin

g:im

plicitwantingbutn

otlik

ingis

decreasing

-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Thackray

etal.

unpublished

UK

Sex:

females

andmales

BMIstatus:1

8.5–29.9

kg/m

2

PAlevel:habitually

activ

en=32

Age:2

3(2)years

BMI:23.9(2.6)kg/m

2

-Intensity:self-determ

ined

moderate-to-highintensity

(RPE

of15

“hard”)

-Ty

pe:swim

mingvs.cyclin

g-Duration:

6×8-min

intervalswith

2-min

recovery

-Tim

ing:

morning

-Controlcondition:yes,noexercise

-LFP

Q:relativepreference

(food

choice),lik

ingandim

plicitwantin

gbias

forfat/taste

-Setting:laboratory

-State:post-exercise(3

hafterfixed

breakfasto

f650kcalformales,

525kcalforfemales,and

before

adlibitu

mlunchmeal)

-Tendency

foramaineffectof

trialfor

implicitwantingfatb

ias(posth

oc:

cycling<control,cycling<

swim

ming).

-Noim

pactof

swim

mingor

cyclingon

otherfood

rewardparameters.

-Differencebetweenlik

ing/wantin

g:Changes

inim

plicitwantin

gbutn

otlik

ing

-Fo

odintake:increasein

adlib

itum

energy

intake

aftersw

immingbutn

otaftercycling

-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Thiveletal.

2019

[51]

France

Sex:

females

andmales

BMIstatus:<

25kg/m

2

PAlevel:moderatelyactive

(150–240

min/week)

n=19

(52%

males)

Age:2

1(1)years

BMI:22.3(2.9)kg/m

2

-Intensity:low

50%

VO2max,high

75%

VO2max

-Ty

pe:cyclin

g-Duration:

lowintensity

45min,

high

intensity

30min

-Tim

ing:

morning

-Controlcondition:yes;noexercise

-LFP

Q:relativepreference,likingand

implicitwantingbias

forfat/taste

-Setting:laboratory

-State:preandpostfixedlunch30

min

afterexercise

(fem

ales

750kcaland

males

900kcal;standardized

breakfasto

f500kcalconsum

ed3h

before

exercise).

-Nochange

infood

reward

-Differencebetweenlik

ing/wantin

g:no

differences

-Fo

odintake:(self-reported)

nochange

-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Saanijo

kietal.2018

[52]

Finland

Sex:

males

BMIstatus:1

9.9–26.9

kg/m

2

PAlevel:NR

n=24

Age:2

7(5)years

BMI:23.5(1.6)kg/m

2

-Intensity:m

oderate(74%

HRmax)

-Ty

pe:aerobiccycling

-Duration:

60min

-Tim

ing:

NR

-Controlcondition:yes;noexercise

-fM

RI:BOLDsignalsto

palatableand

non-palatablefoodscomparedwith

neutralcontrol

(cars)

-Setting:laboratory

-State:post-exercise

fasted

for3

hbefore

thescans

-Noeffectsof

exercise

vsreston

neuronalresponses.

-Individualvariability

intheBOLD

signalsafterexercise

might

beexplainedby

changesin

thebrain

opioid

system

.Participantswho

show

edmostincreases

inendogenous

opioid

releasealso

hadhighest

anticipatoryfM

RIrewardresponses

follo

wingtheexercise

-Differencebetweenlik

ing/wanting:

not

assessed

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Chronicexercise

training

studies

Alkahtani

etal.2014

[53]

Australia

Sex:

males

BMIstatus:≥

25kg/m

2

PAlevel:inactive(criteriaNR)

n=10

Age:2

9(4)years

Moderate-intensity

interval

training

(MIIT):

BMIbaselin

e=30.7(3.5)kg/m

2

BMIpost=30.8(3.5)kg/m

2

-Frequency:

3days/weekfor4

week(cross-overwith

6-week

washout)

-Intensity:M

IIT±20%

workloadat

45%

VO2peak,HIIT90%

VO2peak

-Ty

pe:M

IITvs.H

IIT

-Duration:30–45min(M

IIT5-min

stages

alternatingbetween±20%

-LFP

Q:likingandim

plicitwantin

gfor

HFS

W,L

FSW,H

FSA,L

FSA

-Setting:laboratory

-State:preandpost45-m

incyclingat

45%

VO2m

ax

-Measurementtim

epoints:w

eek0and

week4in

each

interventio

n

-Tendency

forexplicitlik

ingforhigh-fat

non-sw

eetfoods

afteracuteexercise

toincrease

with

MIITanddecrease

with

HIIT.

-Nochangesin

wanting.

-Differencebetweenliking/wantin

g:yes;

changesin

likingbutn

otwantin

g

-Fo

odintake:N

oeffectsof

training

onfood

intake

orenergy

intake.

Tendency

forfatintake(g)and%

energy

from

fattoincrease

afterMIIT.

-Eatingbehavior

traits:N

R-Bodycomposition:

NR

Curr Obes Rep (2020) 9:63–8070

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

High-intensity

intervaltraining

(HIIT):

BMIbaselin

e=30.9(3.2)kg/m

2

BMIpost=30.9(3.2)kg/m

2

workload,HIIT30-sintervals

and30-srecovery)

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

no;M

IITvs.H

IIT

Beaulieuetal.

2019

[54]

UK

Sex:

females

andmales

BMIstatus:2

6.0–38.0

kg/m

2

PAlevel:inactive(≤

2h/week)

Exercisers:n=46

(35%

males)

Age:4

3(8)years

BMIbaselin

e=30.5(3.8)kg/m

2

BMIpost=29.9(4.0)kg/m

2

Controls:n=15

(40%

males)

Age:4

1(11)

years

BMIbaselin

e=31.4(3.7)kg/m

2

BMIpost=31.8(3.9)kg/m

2

-Frequency:

5days/weekfor12

weeks

-Intensity:7

0%HRmax

-Ty

pe:aerobic(treadmill,row

er,

cycleergometer,and

ellip

tical)

-Duration:

500kcal(m

ales

~40–45min,fem

ales

~60

min)

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

yes;no

exercise

-LFP

Q:likingandim

plicitwantin

gbias

forfat

-Setting:laboratory

-State:preandpostfixedlunch(high-fat

orhigh-CHO;8

00kcal)

-Measurementtim

epoints:w

eek0and

week12

-Decreasein

wantin

gaftertraining

-Nochange

inlik

ing

-Differencebetweenliking/wantin

g:yes;

changesin

wantin

gbutn

otlik

ing

-Fo

odintake:reductio

nin

high

fatad

libitu

mdinnerintake

butnochange

indaily

high-fatenergy

intake

[55]

-Eatingbehavior

traits:d

ecreasein

disinhibition

andbingeeatingscore

-Bodycomposition:

reductionin

body

weightand

percentage

body

fat,but

notassociatedwith

changesinwantin

g

Cornier

etal.

2012

[56]

USA

Sex:

females

andmales

BMIstatus:>

25kg/m

2

PAlevel:NR

n=12

(42%

males)

Age:3

8(10)

years

BMIbaselin

e=33.3(4.3)kg/m

2

BMIpost=NR

-Frequency:

5days/weekfor24

weeks

-Intensity:u

pto

75%

VO2m

ax

-Ty

pe:treadmill

-Duration:

upto

500kcal/day

(40–60

min/day)

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

no

-fM

RI:responsesto

food

vsnon-food

cues

-Setting:laboratory

-State:fasted

withoutexercisefor24

h(chronicexercise)andfasted

within

30min

ofacuteexercise

(chronic+

acute;500kcal60–75%

VO2m

axfor

40-60min)

-Measurementtim

epoints:w

eek0and

week24

-Chronicexercise:reductio

nin

neuronal

responsesobserved

inthebilateral

parietalcortices,leftinsulaandvisual

cortex.

-Chronic+acuteexercise:intermediate

attenuationof

theresponse

tovisual

food

cues

inbrainregion

importantin

food

regulatio

ncomparedwith

chronicexercise

andbaselin

e-Differencebetweenliking/wanting:

NR

-Foodintake:self-reported

energy

intake

lower

aftertraining

comparedwith

baselin

ebutn

ochange

inmacronutrient

intake.N

oassociation

with

changesin

neuronalresponses.

-Eatingbehavior

traits:n

ochange

indietaryrestrainto

rdisinhibition

-Bodycomposition:

reductionin

body

fatp

ercentage.Changes

inanterior

insularesponsespositivelyassociated

with

changesin

body

massandfat

mass.

Finlayson

etal.2011

[57]

UK

Sex:

females

andmales

BMIstatus:2

6.0–38.0

kg/m

2

PAlevel:inactive(≤2h/week)

Responders(non-com

pensators):

n=20

(43%

males)

Age:4

1(9)years

BMIbaselin

e=32.3(4.3)kg/m

2

BMIpost=30.9(4.3)kg/m

2

Non-Responders(com

pensators):

n=14

(50%

males)

Age:3

7(12)

years

BMIbaselin

e=29.7(2.2)kg/m

2

BMIpost=29.3(2.5)kg/m

2

-Frequency:

5days/weekfor12

weeks

-Intensity:7

0%HRmax

-Ty

pe:aerobic(treadmill,row

er,

cycleergometer,and

ellip

tical)

-Duration:

500kcal(m

ales

~40–45min,fem

ales

~60

min)

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

no

-LFP

Q:relativepreference

(food

choice),liking,explicitwantin

gfor

HFS

W,L

FSW,H

FSA,L

FSA

-Setting:laboratory

-State:preandpostacuteexercise

(scheduled

session)

-Measurementtim

epoints:w

eek0and

week12

-Increase

inlikingafterexercise

innon-responderscomparedwith

respondersatbaselineandweek12.

-Increase

inexplicitwantin

gforhigh-fat

sweetfoods

innon-responders.

-Increase

inrelativ

epreference

for

high-fatsw

eetfoodinnon-responders.

-Differencebetweenlik

ing/wantin

g:im

plicitwantingNR

-Fo

odintake:N

R-Eatingbehavior

traits:N

R-B

odycompositio

n:greaterfatmassloss

inresponders

Finlayson

etal.

unpublished

UK

Sex:

females

andmales

BMIstatus:2

6.0–38.0

kg/m

2

PAlevel:inactive(≤

2h/week)

Non-com

pensators:n=15

(33%

males)

-Frequency:

5days/weekfor12

weeks

-Intensity:7

0%HRmax

-Ty

pe:aerobic(treadmill,row

er,

cycleergometer,and

ellip

tical)

-LFP

Q:likingandim

plicitwantin

gbias

forfat

-Setting:laboratory

-State:preandpostfixedlunch(high-fat

orhigh-CHO;8

00kcal)

-Non-com

pensatorsshow

edasm

aller

likingandim

plicitwantin

gforhigh-fat

food.

-Atb

aseline,compensatorsshow

eda

strong

likingandwantin

gforhigh-fat

-Fo

odintake:tendencyfor

non-compensatorsto

decrease

adlib

itum

dinnermealsizefrom

baseline

topost-interventionthatwas

notseen

incompensators.

Curr Obes Rep (2020) 9:63–80 71

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

Age:4

2(8)years

BMIbaselin

e=30.7(4.9)kg/m

2

BMIpost=29.1(5.0)kg/m

2

Com

pensators:n=15

(33%

males)

Age:4

1(9)years

BMIbaselin

e=31.8(3.7)kg/m

2

BMIpost=32.1(3.9)kg/m

2

Controls:n=15

(33%

males)

Age:4

1(11)

years

BMIbaselin

e=31.4(3.7)kg/m

2

BMIpost=31.8(3.9)kg/m

2

-Duration:

500kcal(m

ales

~40–45min,fem

ales

~60

min)

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

yes,no

exercise

-Measurementtim

epoints:w

eek0and

week12

food

whereas

non-compensators

show

edno

difference

betweenhigh-fat

andlow-fatfood.

-Greater

baselinerewardforhigh-fat

food

incompensatorsreduced

follo

wingtheexercise

interventio

n.-Inthenon-compensators,sm

allincrease

inlik

ingforhigh-fatfood

after

exercise

training,but

asimultaneous

decrease

inwantin

gforhigh-fatfood.

-Differencebetweenliking/wanting:yes,

non-compensatorsincreasedlik

ingbut

decreasedwantingforhigh-fatfood

post-intervention.

-Eatingbehavior

traits:N

R-B

odycompositio

n:significantreductio

nin

BMI,body

mass,fatm

assandWC

innon-compensators,whereas

nochangesincompensators,andincrease

inbody

massandWC.

Martin

etal.

2019

[58]

USA

Sex:

females

andmales

BMIstatus:2

5–45

kg/m

2

PAlevel:inactive(<

20min

<3

days/week)

8kcal/kgbody

weight/w

eek

(KKW):n=59

(27%

males)

Age:4

8(11)

years

BMIbaselin

e=31.4(4.6)kg/m

2

BMIpost=NR(~

31.3

kg/m

2)

20KKW:n

=51

(29%

males)

Age:4

9(12)

years

BMIbaselin

e=30.6(4.4)kg/m

2

BMIpost=NR(~

30.0

kg/m

2)

Control:n

=61

(26%

males)

Age:5

0(1)years

BMIbaselin

e=32.3(4.8)kg/m

2

BMIpost=NR(~

32.2

kg/m

2)

Pooled

exercisers(n=110)

divided(m

ediansplit)into

compensators/non--

compensatorsbasedon

actual

andpredictedweightloss.

-Frequency:

3–5days/week

(self-selected)for24

weeks

-Intensity:6

5–85%

VO2peak

(self-selected)

-Ty

pe:treadmill

-Duration:

8KKW

~35

min/session

(~700kcal/week)

vs.20KKW

~55

min/session

(~1760

kcal/week)

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

yes;no

exercise

-Fo

odPreference

Questionnaire

[59]

preferencesforfood

classified

asalongside2components:fat(2factors:

HighFatand

Low

Fat)and

carbohydrate(3

factors:HighSimple

Sugar,HighCom

plex

CHO,and

Low

CHO/HighProtein)

-Setting:laboratory

-State:NR

-Measurementtim

epoints:w

eek0and

week24

-8KKW

groupincreasedpreference

for

high

fat/highCHOfoodswhereas

20KKW

groupdecreased.

-Com

pensatorsdecreasedpreferences

forhigh

CHO,low

fat,andlow

fat/highCHOfoodswhereas

non-compensatorsincreased.

-Differencebetweenliking/wanting:

NR

-Fo

odintake:adjusteddoubly-labeled

water

energy

intake

increasedin

both

exercise

groups

(relativeto

control

group).N

ochangesintestmealenergy

intake.

-Eatingbehavior

traits:greaterreduction

indisinhibition

innon-compensators

relativ

eto

compensators.

-Bodycomposition:

difference

inbody

massandbody

fatp

ercentageloss

between20

KKW

andcontrol.

Martin

setal.

2017

[60]

Norway

Sex:

females

andmales

BMIstatus:≥

30kg/m

2

PAlevel:inactive(<1day/week

MVPA

,<20

min/day

<3days/weeklight

PA)

HIIT:n

=16

(13completers;40%

males)

Age:3

4(8)years

BMIbaselin

e=33.2(3.5)kg/m

2

BMIpost=NR

-Frequency:

3days/weekfor12

weeks

-Intensity:H

IITand½

HIIT

(85–90%

HRmax),continuous

(70%

HRmax)

-Ty

pe:H

IITvs.½

HIITvs.

continuous

cycling

-Duration:

HIIT(8-sintervalsand

12-srecovery

for250kcal;

~20

min),½

HIIT(8-sintervals

-LFP

Q:relativepreference

(food

choice),lik

ingandim

plicitwantin

gbias

forfat/taste

-Setting:laboratory

-State:preandpostfixedbreakfast

(600

kcal)

-Measurementtim

epoints:w

eek0and

week12

-Noeffectof

exercise

onlikingor

wantin

g-Differencebetweenlik

ing/wantin

g:no

-Fo

odintake:n

ochange

[61]

-Eatingbehavior

traits:N

R-Bodycomposition:

reductionin

body

weightand

trunkandleg%

fatm

ass

[61]

Curr Obes Rep (2020) 9:63–8072

Tab

le1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

½HIIT:n

=16

(9completers;

80%

males)

Age:3

4(7)years

BMIbaselin

e=32.4(2.9)kg/m

2

BMIpost=NR

Contin

uous:n

=14

(13

completers;60%

males)

Age:3

3(10)

years

BMIbaselin

e=33.3(2.4)kg/m

2

BMIpost=NR

and12-srecovery

for125kcal;

~10

min),continuous

exercise

(250

kcal;~

32min).

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

no;H

IITvs.½

HIITvs.contin

uous

Miguetetal.

under

review

France

Sex:

females

andmales

(adolescents)

BMIstatus:≥

95th

percentilefor

sexandage

PAlevel:inactive(<2h/week)

n=30

(23%

males)

Age:1

3(1)years

BMIbaselin

e=35.7(4.5)kg/m

2

BMIpost=30.9(5.0)kg/m

2

-Frequency:

4days/weekfor

10months

-Intensity:N

R-Ty

pe:v

arious

(aerobic,strength,

aquatic

andleisure-tim

eactiv

ities)

-Duration:

60min

-Tim

ing:

NR

-Su

pervision:

yes

-Control

group:

noAspartof

aninpatient

multidisciplinaryweightloss

program

-LFP

Q:likingandim

plicitwantin

gfor

HFS

W,L

FSW,H

FSA,L

FSA

-Setting:laboratory

-State:preandpostlunch(adlib

itum)

-Measurementtim

epoints:b

aseline,

5months,10

months

-Hungry:

Increase

inlik

ingat5months

follo

wed

bydecrease

at10

months

(sim

ilartobaselin

evalues).Nochange

inwanting.

-Fed:

Decreasein

likingat5and

10months.Nochange

inwantin

g.-Hungryto

fed:

Decreasein

likingat5

and10

months,butn

otbaselin

e.No

change

inwantin

g.-D

ifferencebetweenliking/wantin

g:yes;

decrease

inlik

ing,no

change

inwantin

g

-Fo

odintake:d

ecreasein

lunchEIat

10months

-Eatingbehavior

traits:d

ecreasein

uncontrolledeatingandem

otional

eatin

gat5and10

months

-Bodycomposition:

decrease

inpercentage

fatm

assat5and

10months

Riouetal.

2019

[62]

Canada

Sex:

females

(premenopausal)

BMIstatus:>

27kg/m

2

PAlevel:inactive

(<150min/week)

Low

intensity

:n=11

week

Moderateintensity

:n=10

Age:3

1(11)

years

BMIbaselin

e=35.1(6.2)kg/m

2

BMIpost=NR(~

35.5

kg/m

2)

-Frequency:

5days/weekfor

12–14weeks

-Intensity:L

OW

(40%

VO2reserve)

vs.m

oderate(M

OD;6

0%VO2reserve)

-Ty

pe:aerobic(treadmill

orcycle

ergometer)

-Duration:

to300kcal(LOW

~62

min,M

OD~46

min)

-Tim

ing:

NR

-Su

pervision:

3/5days

-Controlgroup:no;L

OW

vs.M

OD

-LFP

Q:likingandim

plicitwantin

gbias

forfat/taste

-Setting:laboratory

-State:postbreakfast/p

reexercise

(ad

libitu

mbreakfasto

nfirstsession

and

quantitiesreplicated

onsubsequent

sessions;L

OW

~648kcal,M

OD

~746kcal)andpostrest(w

eek4)

orexercise

(scheduled

session;

weeks

1and12–14)

-Measurementtim

epoints:w

eek4,week

1,andweek12–14(inlin

ewith

menstrualcycle)

-Decreasein

wantin

gforfatafter

training.

-Increaseinlikingforsavoryfoodspreto

post-exerciseatweek1butdecreaseat

week12–14.

-Differencebetweenliking/wantin

g:yes;

changesin

wantin

gbutn

otlik

ingfor

fatw

ithtraining

-Fo

odintake:n

ochange

-Eatingbehavior

traits:increasein

susceptib

ility

tohunger

-Bodycomposition:

groupby

time

interactionforbody

weightand

fat

massshow

ingincrease

inMODand

decrease

inLOW

Thiveletal.

2019

[63]

France

Sex:

females

andmales

(adolescents)

BMIstatus:>

90th

percentilefor

sexandage

PAlevel:NR

Eccentric:n

=12

(50%

males)

Age:1

4(1)years

BMIbaselin

e=34.8(5.5)kg/m

2

BMIpost=29.0(4.5)kg/m

2

Concentric:n=12

(50%

males)

-Frequency:

3days/weekfor12

weeks

(+2h/weekphysical

education)

-Intensity:5

0%up

to70%

VO2peak

-Ty

pe:eccentricvs.concentric

cycling

-Duration:

30up

to45

min

-Tim

ing:

NR

-Su

pervision:

yes

-LFP

Q:relativepreference

(food

choice),lik

ingandim

plicitwantin

gbias

forfat/sweet

-Setting:laboratory

-State:fasted

-Measurementtim

epoints:b

aseline,

week12

andweek24

-Eccentricgroup:

increase

inpreference

forhigh-fatfoodsandsavory

foods

(decreasein

sweetb

ias).Increasein

implicitwantingforsavory

foods

(decreasein

sweetb

ias).

-Group

bytim

einteractionshow

edconcentricgroupincreasedpreference

andim

plicitwantin

gforsw

eetfoods

whileeccentricdecreased.

-Fo

odintake:totaldaily

energy

intake

increasedinbothgroups

from

baselin

eto

week24,but

only

increasedin

concentricgroupfrom

week12

to24.

-Eatingbehavior

traits:N

R-Bodycomposition:

greaterpercentage

body

fatlossaftereccentricvs.

concentricexercise

from

week12

to24

Curr Obes Rep (2020) 9:63–80 73

showed a decrease in food reward after acute exercise com-pared with the sedentary control. McNeil and colleagues re-vealed a decrease in the preference for high-fat relative to low-fat food independently of the modality of exercise (aerobic orresistance) in inactive adults within the normal range of BMI[49]. Miguet and colleagues showed a decrease in implicitwanting for sweet relative to savory food in response to anad libitum meal after a session of high intensity interval exer-cise in inactive adolescents with obesity [50]. Interestingly, nochange in food intake was observed in the McNeil studywhereas a decrease in total energy intake (both lunch anddinner) was noted in the Miguet study. This might be relatedto the fact that changes in implicit wanting are a greater driverof food intake than changes in liking. Also, the McNeil studymight have been underpowered to detect a change in implicitwanting. Four studies showed no changes in food reward (fator taste bias for liking, relative preference or implicit wanting).One study compared eccentric vs concentric exercise on mod-erately active men (BMI 23.4 ± 3.3 kg/m2) and showed noeffect on either appetite sensations, appetite-related hormones,or food reward [43]. Three studies compared the intensity ofexercise (low versus high [51] or high- or moderate-intensityintermittent cycling [42, 48]). Two reported no effects on foodreward or food intake [48, 51] whereas one found a decreasein wanting and increase in liking after both high- andmoderate-intensity exercise [42]. Of note, these studies wereconducted in moderately active normal-weight adults [51] andinactive adults with overweight/obesity [42, 48]. A recentlycompleted study compared bouts of swimming or cycling to ano-exercise control in a within-subjects design (Thackrayet al. unpublished). While they found that energy intake wasincreased after swimming but not cycling compared with con-trol, no differences in food reward were detected except for atendency for a main effect of trial for implicit wanting fat biaswith wanting being smaller after cycling relative to swimmingand control. Lastly, Finlayson and colleagues demonstratedthat implicit wanting was increased in response to moderate-intensity exercise only in those individuals that increased theirenergy intake relative to no exercise (i.e., compensators) [47].Consequently, even with the samemethodology to assess foodreward, the response to acute exercise seems to be equivocaland subject to individual variability. This could be explainedby methodological issues; even though the same tool is used,the studies were conducted in different countries (UK, SaudiArabia, Canada, France, and Norway) and the food imagesused may not have been cross-culturally validated [68]. Thesample sizes were relatively small for most of the studies(ranging from 12 to 33), including mainly both genders andwith different ranges of BMI. However, it can be noticed thatexercise seems to affect food reward more clearly in inactiveindividuals compared with active ones in both adolescents andadults. One tentative hypothesis could be that inactive indi-viduals (within the non-regulated zone of the J-shape curve)T

able1

(contin

ued)

Reference

Participantcharacteristics

Exercise/physicalactiv

itydetails

Food

rewardmethod

Food

rewardresults

Associatio

nswith

appetiteoutcom

es

Age:1

3(1)years

BMIbaselin

e=31.8(3.8)kg/m

2

BMIpost=27.6(4.0)kg/m

2

-Control

group:

no;eccentricvs.

concentric

Phase2(w

eek12–24)

ofa24-w

eek

inpatient

multidisciplinary

weightlossprogram

-Differencebetweenliking/wantin

g:yes;

changesin

implicitwantingforsw

eet

foods,no

change

inlik

ing.

Dataaremeans

(SD).CoE

Q,C

ontrol

ofEatingQuestionnaire;F

MI,fatm

assindex;

fMRI,functio

nalm

agnetic

resonanceim

aging;

LFPQ,L

eeds

Food

Preference

Questionnaire;L

FSA

,low

-fatsavory;

LFSW

,low

-fatsw

eet;HFSA

,high-fatsavory;H

FSW

,high-fatsw

eet;HIIT,high-intensity

intervaltraining;MET,metabolicequivalent

ofatask;MVPA

,moderate-to-vigorousphysicalactiv

ity;N

R,n

otreported;PA,physicalactivity

;VAS,visualanalogue

scales;W

C,w

aistcircum

ference

Curr Obes Rep (2020) 9:63–8074

would benefit more from acute exercise than active individ-uals for whom the appetite control system is more sensitive.

Farah and colleagues used a computer-based task to mea-sure the effect of acute exercise on liking (visual analoguescale) and other non-homeostatic indicators (ideal portionsize, food utility, hunger) [46]. They found that a 60-minmoderate-intensity exercise bout reduced hunger and idealportion size but not liking. This is in concordance with previ-ous results showing that implicit wanting rather than liking

might be influenced by exercise. However, the physical activ-ity level of the participants was not reported, and implicitwanting was not measured.

Acute exercise has also been shown to have an effect on brainreward measured by BOLD response to food cues with fMRI.Evero and colleagues showed that a 60-min high-intensity exer-cise bout decreased the neuronal response to food cues in brainregions related to food reward, visual attention, and inhibitorycontrol [45]. Interestingly, regions related to both liking (i.e.,

Fig. 2 Conceptual model of the impact of habitual physical activity andexercise on appetite control. The model builds upon the relationshipbetween physical activity level, energy intake, and body fat recentlydemonstrated by Beaulieu et al. [19••]. Higher levels of physicalactivity are associated with enhanced satiety signaling—within a“regulated zone” of appetite control—resulting in a better matchingbetween energy intake and energy expenditure. Lower levels ofphysical activity are associated with higher body fat, weaker satietysignaling, and greater responsiveness to hedonic inputs from the foodenvironment—within a “non-regulated zone” of appetite control—allowing for overconsumption to occur and body fat to increase further.This review adds to this model by proposing effects of physical activity

on liking and wanting as processes of food reward. Specifically, lowerlevels of physical activity are associated with greater liking and wantingfor high-fat/high-energy food, with wanting as the stronger driver ofexcess energy intake. Acute exercise leads to a reduction in liking andwanting, especially in inactive individuals. As habitual levels of physicalactivity increase (including during chronic exercise interventions), thereis a small increase in liking and decrease in wanting that accompanyweight loss and improvement in appetite control. Finally, higher levelsof habitual physical activity (e.g., regular bouts of moderate-to-vigorousphysical activity) are associated with greater liking for low-fat/low-energy food and lower wanting for high-fat food

Fig. 1 The impact of a 12-weekaerobic exercise intervention(70% HRmax, 500 kcal/day,5 days/week) on liking andimplicit wanting for high-fat foodin weight loss non-compensators(n = 15), compensators (n = 15),and non-exercising controls (n =15) (Finlayson et al.unpublished). *p < 0.05,**p < 0.01, #p = 0.08, §p = 0.10

Curr Obes Rep (2020) 9:63–80 75

insula, orbitofrontal cortex) and wanting (i.e., putamen) werereduced after exercise. This is in line with another study thatfound that exercise increases neural responses in reward-relatedregions in response to images of low-calorie foods and sup-presses activation during the viewing of high-calorie foods[44]. These central responses were associated with exercise-induced changes in peripheral signals related to appetite controland hydration status. However, liking and implicit wanting werenot measured directly in these studies (i.e., behavioral measures)nor was food intake. Lastly, Saanikoji et al. found no effect of anacutemoderate-intensity aerobic exercise on brain food reward inlean men [52]. However, they showed that individuals who in-creased the most in endogenous opioid release had the highestbrain reward response after the exercise compared with the con-trol. Consequently, the opioid systemmight contribute to explainsome individual variability in the food reward responses toexercise.

Chronic Exercise Training Studies

As mentioned above, our recent systematic review on weightmanagement interventions found limited evidence on the im-pact of exercise interventions on food reward [6••]. Amongthe included studies, two investigated the impact of high-intensity interval exercise compared with either moderate-intensity interval training [53] or continuous training [60].Using a cross-over design, Alkahtani et al. [53] found that inresponse to acute exercise after a 4-week exercise interven-tion, liking for high-fat savory food seemed to increase afterMIIT and decrease after HIIT in men with overweight or obe-sity (interaction p = 0.09). Another study in individuals withobesity found no changes in food reward in response to abreakfast meal (hungry and fed states) after a 12-week inter-vention of either high-intensity interval exercise (125 or250 kcal, 3 days/week) or moderate-intensity continuous ex-ercise (250 kcal, 3 days/week) [60]. In another study, nochanges in liking or implicit wanting for high-fat relative tolow-fat food in the hungry state were observed after a 12-weekaerobic exercise intervention (500 kcal, 5 days/week), al-though a trend towards a reduction in wanting was noted[69]. However, more recent analyses with a non-exercisingcontrol group revealed that in response to a high-fat andhigh-carbohydrate fixed lunch (hungry and fed states), overallimplicit wanting decreased after the 12-week exercise inter-vention, whereas no changes in explicit liking were found[54]. This is corroborated by another study that found a de-crease in implicit wanting for high-fat relative to low-fat foodin women who underwent a 3-month exercise intervention(300 kcal, 5 days/week) [62]. When food reward was mea-sured in response to an exercise bout post-intervention in thatstudy, there was also a decrease in liking for savory foodswhereas liking for savory foods increase after acute exerciseat baseline. Thus, it appears that chronic exercise training may

modulate the food reward response to acute exercise in inac-tive individuals, also shown in an fMRI study [56].

Interestingly, changes in food reward in the study by Beaulieuet al. [54] were not associated with changes in body weight orcomposition, suggesting a potential independent effect of exer-cise. Indeed, an inverse association was found between changesin leptin (adjusted for percentage body fat) and changes in liking,but not wanting, for high-fat food [69]. Thus, leptin may have arole inmediating changes in food reward during exercise trainingin individuals with overweight/obesity. In contrast, a 6-monthexercise intervention led to a reduction in the fasted neuronalresponse to food compared with non-food, and some of thesechanges were associated with changes in fat mass, body weightand leptin [56]. These findings are interesting in light of theseminal study by Rosenbaum [70], who reported that leptin re-placement after > 10% weight loss using a liquid formula diet,modulated food cue-elicited neuronal activation in reward-related regions (consistent with wanting), but did not affect likingfor the diet. Beyond leptin, it is known that weight loss alsoimpacts fasting levels of ghrelin. For example, the RESOLVEstudy (NCT00917917) showed that long-term physical activitymay reverse the early enhancing effect of body weight loss onplasma ghrelin [71]. Future studies should examine whether fa-vorable effects of exercise-induced weight loss on food rewardcan be partly explained by modulation of ghrelin as well asleptin.

Differences in food preferences have also been found be-tween individuals who compensated (less than expectedweight loss based on median split) compared with non-compensators during a 6-month intervention expending either~ 700 kcal/week or ~ 1760 kcal/week [58]. Compensators hadreduced preference for low-fat and high-carbohydrate foodrelative to non-compensators. Indeed, the hedonic responseto acute exercise also appears to impact weight loss outcomesduring chronic exercise training [57]. We have shown thatafter acute exercise, liking for all foods and wanting forhigh-fat sweet food increased only in compensators to a 12-week exercise intervention (those who achieved less than ex-pected weight loss), compared with no change in non-compensators (those achieving at or above expected weightloss). These differences were independent of the exercise in-tervention and weight loss [57] and add to other evidence ofimproved appetite control in this group [72]. In a further un-published study from our laboratory where sub-groups of 12-week exercise intervention non-compensators and compensa-tors were compared with a non-exercising control group(Fig. 1), we observed that prior to the exercise intervention,compensators showed a strong liking and wanting for high-fatfood whereas non-compensators showed no difference be-tween high-fat and low-fat food. Secondly, we found that thegreater reward for high-fat food in compensators reduced fol-lowing the exercise intervention, compared with no change incontrols. Lastly, there appeared to be a unique pattern of

Curr Obes Rep (2020) 9:63–8076

change in liking and wanting in the non-compensators whoshowed a small increase in liking for high-fat food after exer-cise training, but a simultaneous decrease in wanting for high-fat food.

In adolescents with obesity, eccentric cycling exercise as partof a 12-week inpatient multidisciplinary weight loss interventionincreased the relative preference for high-fat food and increasedboth the relative preference and implicit wanting for savory food,whereas no changes were observed in response to concentricexercise training [63]. Another study in adolescents with obesityshowed that during a 10-month inpatient multidisciplinaryweight loss intervention including physical activity, liking forfood in the hungry state increased from baseline to 5 months,then returned to baseline values at 10 months, whereas liking forfood in the fed state decreased (Miguet et al., under review).There were no changes in wanting observed.

These studies are suggestive that chronic exercise improvesfood reward (reduced response to high-energy foods and in-creased response to low-energy foods). However, the effectsizes were relatively small and inter-individual variabilitytended to be large. Two studies found a reduction in implicitwanting for high-fat relative to low-fat foods after exercisetraining [54, 62]. This may be a result of a direct effect ofexercise on brain regions related to food reward, as shownby the fMRI studies included in the current review, and others[73, 74]. Furthermore, as exercise affects cognition and exec-utive function, it has been proposed that processes such asinhibitory control could have a moderating effect on wantingand modulate eating behavior [66].

Another two studies found an increase in liking after exer-cise training, which might be explained by concomitant im-provements in homeostatic appetite control in these studies (asmall increase in hunger or a reduction in fasting leptin con-centrations). Individual differences in food reward appear toact as pre-existing moderators of the impact of exercise train-ing on weight loss and suggest that those with healthier pref-erences or better satiety signaling at baseline appear to losemore weight with exercise. No clear evidence exists regardingthe optimal mode, frequency, intensity, duration, and time ofday for exercise to have the most impact on food reward.Further systematic research into these factors is warranted.

Conclusions

One of the perceived barriers for engaging in exercise is itspotential to promote hedonic eating. Food reward plays animportant role in weight management through its interveningstatus between the nutrient requirements of the body and he-donic inputs from the food environment that promote foodintake. This review brings together current evidence from ob-servational, acute, and chronic exercise training studies to in-form public debate on the impact of physical activity on food

reward. A conceptual model, building on previous theory[19••] is shown in Fig. 2. Observational studies show thatperformance of moderate-to-vigorous physical activity is as-sociated with lower liking and wanting for high-fat or high-energy food, and higher liking for low-fat/low-energy food.These findings may reflect improved appetite control and aresupported by evidence from chronic exercise training inter-ventions. Where exercise training leads to successful weightloss, it appears to be accompanied by a dissociation betweenliking and wanting evidenced by a reduction in wanting forhigh-energy food but increase in liking for low-energy food.Acute bouts of exercise tend to only impact behavioral indicesof food reward in less active individuals or those with poorappetite control, where it tends to result in reduced food re-ward. These findings are corroborated by observational stud-ies that demonstrate greater liking and especially wanting forhigh-energy foods (and greater susceptibility to food cravings)in inactive individuals. Food reward does not counteract thebenefit of physical activity for obesity management. Rather,exercise appears to accompany positive changes in food pref-erences in line with improvements in appetite control.

Compliance with Ethical Standards

Conflict of Interest The authors declare no conflict of interest.

Human and Animal Rights and Informed Consent This article does notcontain any studies with human or animal subjects performed by any ofthe authors.

Open Access This article is licensed under a Creative CommonsAttribution 4.0 International License, which permits use, sharing, adap-tation, distribution and reproduction in any medium or format, as long asyou give appropriate credit to the original author(s) and the source, pro-vide a link to the Creative Commons licence, and indicate if changes weremade. The images or other third party material in this article are includedin the article's Creative Commons licence, unless indicated otherwise in acredit line to the material. If material is not included in the article'sCreative Commons licence and your intended use is not permitted bystatutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of thislicence, visit http://creativecommons.org/licenses/by/4.0/.

References

Papers of particular interest, published recently, have beenhighlighted as:• Of importance•• Of major importance

1 Guess N. A qualitative investigation of attitudes towards aerobic andresistance exercise amongst overweight and obese individuals. BMCRes Notes. 2012;5:191–12. https://doi.org/10.1186/1756-0500-5-191.

2 Leone LA, Ward DS. A mixed methods comparison of perceivedbenefits and barriers to exercise between obese and nonobese wom-en. J Phys Act Health. 2013;10(4):461–9.

Curr Obes Rep (2020) 9:63–80 77

3 Schreiber K.Why your workout leaves you so hungry you could eata horse. Greatist. 2014. https://greatist.com/grow/when-exercise-makes-you-overeat#1.

4 Hill A. Exercise helps you diet - if it’s the right kind. The Guardian.2009. https://www.theguardian.com/lifeandstyle/2009/feb/22/diet-exercise-appetite.

5 Thivel D, Finlayson G, Blundell JE. Homeostatic andneurocognitive control of energy intake in response to exercise inpediatric obesity: a psychobiological framework. Obes Rev.2019;20(2):316–24. https://doi.org/10.1111/obr.12782.

6•• Oustric P, Gibbons C, Beaulieu K, Blundell J, Finlayson G. Changesin food reward duringweight management interventions - a systematicreview. Obes Rev. 2018;19(12):1642–58. https://doi.org/10.1111/obr.12754. Systematic review on the effect of weight managementinterventions on food reward, highlighting the very limitedevidence pertaining to the role of exercise-induced weight lossinterventions on food reward.

7 Blundell JE, Gibbons C, Caudwell P, Finlayson G, Hopkins M.Appetite control and energy balance: impact of exercise. ObesRev. 2015;16:67–76. https://doi.org/10.1111/obr.12257.

8 Horner KM, Schubert MM, Desbrow B, Byrne NM, King NA.Acute exercise and gastric emptying: a meta-analysis and implica-tions for appetite control. Sports Med. 2015;45(5):659–78. https://doi.org/10.1007/s40279-014-0285-4.

9 Schubert MM, Sabapathy S, Leveritt M, Desbrow B. Acute exerciseand hormones related to appetite regulation: a meta-analysis. SportsMed. 2014;44(3):387–403. https://doi.org/10.1007/s40279-013-0120-3.

10 Donnelly JE, Blair SN, Jakicic JM, Manore MM, Rankin JW, SmithBK, et al. Appropriate physical activity intervention strategies forweight loss and prevention of weight regain for adults. Med SciSports Exerc. 2009;41(2):459–71. https://doi.org/10.1249/MSS.0b013e3181949333.

11 ShawK, Gennat H, O'Rourke P, DelMar C. Exercise for overweightor obesity. Cochrane Database Syst Rev. 2006;4:CD003817. https://doi.org/10.1002/14651858.CD003817.pub3.

12 Stiegler P, Cunliffe A. The role of diet and exercise for the mainte-nance of fat-free mass and resting metabolic rate during weight loss.Sports Med. 2006;36(3):239–62.

13 Goodyear LJ, Kahn BB. Exercise, glucose transport, and insulinsensitivity. Annu Rev Med. 1998;49:235–61. https://doi.org/10.1146/annurev.med.49.1.235.

14 Dyck DJ. Leptin sensitivity in skeletal muscle is modulated by dietand exercise. Exerc Sport Sci Rev. 2005;33. https://doi.org/10.1097/00003677-200510000-00007.

15 Steinberg GR, Smith AC, Wormald S, Malenfant P, Collier C, DyckDJ. Endurance training partially reverses dietary-induced leptin re-sistance in rodent skeletal muscle. Am J Physiol Endocrinol Metab.2004;286(1):E57–63. https://doi.org/10.1152/ajpendo.00302.2003.

16 Hopkins M, Duarte C, Beaulieu K, Finlayson G, Gibbons C,Johnstone AM, et al. Activity energy expenditure is an independentpredictor of energy intake in humans. Int J Obes. 2019;43:1466–74.https://doi.org/10.1038/s41366-018-0308-6.

17 King NA, Caudwell PP, Hopkins M, Stubbs RJ, Naslund E,Blundell JE. Dual-process action of exercise on appetite control:increase in orexigenic drive but improvement in meal-induced sati-ety. Am J Clin Nutr. 2009;90(4):921–7. https://doi.org/10.3945/ajcn.2009.27706.

18 Beaulieu K, Hopkins M, Blundell JE, Finlayson G. Does habitualphysical activity increase the sensitivity of the appetite control sys-tem? A Systematic Review. Sports Med. 2016;46(12):1897–919.https://doi.org/10.1007/s40279-016-0518-9.

19•• Beaulieu K, Hopkins M, Blundell J, Finlayson G. Homeostatic andnon-homeostatic appetite control along the spectrum of physical ac-tivity levels: An updated perspective. Physiol Behav. 2018;192:23–9.https://doi.org/10.1016/j.physbeh.2017.12.032.Narrative review on

the impact of habitual physical activity on the mechanisms ofappetite control that proposes an updated model of the J-shaperelationship between physical activity level and energy intake,with body composition, satiety signaling and non-homeostaticfactors playing a role.

20 Berthoud HR, Munzberg H, Morrison CD. Blaming the brain forobesity: integration of hedonic and homeostatic mechanisms.Gastroenterology. 2017;152(7):1728–38. https://doi.org/10.1053/j.gastro.2016.12.050.

21 Pool E, Sennwald V, Delplanque S, Brosch T, Sander D. Measuringwanting and liking from animals to humans: a systematic review.Neurosci Biobehav Rev. 2016;63:124–42. https://doi.org/10.1016/j.neubiorev.2016.01.006.

22 Berridge KC, Robinson TE, Aldridge JW. Dissecting components ofreward: ‘liking’, ‘wanting’, and learning. Curr Opin Pharmacol.2009;9(1):65–73. https://doi.org/10.1016/j.coph.2008.12.014.

23 Dalton M, Finlayson G. Psychobiological examination of liking andwanting for fat and sweet taste in trait binge eating females. PhysiolBehav. 2014;136:128–34. https://doi.org/10.1016/j.physbeh.2014.03.019.

24 French SA, Mitchell NR, Wolfson J, Finlayson G, Blundell JE,Jeffery RW. Questionnaire and laboratory measures of eating behav-ior. Associations with energy intake and BMI in a community sam-ple of working adults. Appetite. 2014;72:50–8. https://doi.org/10.1016/j.appet.2013.09.020.

25 de Castro JM, Bellisle F, Dalix AM, Pearcey SM. Palatability andintake relationships in free-living humans. Characterization and in-dependence of influence in North Americans. Physiol Behav.2000;70(3–4):343–50. https://doi.org/10.1016/s0031-9384(00)00264-x.

26 Cox DN, Perry L, Moore PB, Vallis L, Mela DJ. Sensory and he-donic associations with macronutrient and energy intakes of leanand obese consumers. Int J Obes Relat Metab Disord. 1999;23(4):403–10. https://doi.org/10.1038/sj.ijo.0800836.

27 de Araujo IE, Schatzker M, Small DM. Rethinking food reward.Annu Rev Psychol. 2019. https://doi.org/10.1146/annurev-psych-122216-011643.

28 Mela DJ. Eating for pleasure or just wanting to eat? Reconsideringsensory hedonic responses as a driver of obesity. Appetite.2006;47(1):10–7. https://doi.org/10.1016/j.appet.2006.02.006.

29 Finlayson G, King N, Blundell J. The role of implicit wanting inrelation to explicit liking and wanting for food: implications forappetite control. Appetite. 2008;50(1):120–7. https://doi.org/10.1016/j.appet.2007.06.007.

30 Dalton M, Finlayson G, Hill A, Blundell J. Preliminary validation andprincipal components analysis of the Control of Eating Questionnaire(CoEQ) for the experience of food craving. Eur J Clin Nutr.2015;69(12):1313–7. https://doi.org/10.1038/ejcn.2015.57.

31 Hill AJ, Weaver CF, Blundell JE. Food craving, dietary restraint andmood. Appetite. 1991;17(3):187–97.

32 White MA, Whisenhunt BL, Williamson DA, Greenway FL,Netemeyer RG. Development and validation of the food-cravinginventory. Obes Res. 2002;10(2):107–14. https://doi.org/10.1038/oby.2002.17.

33 Stoeckel LE, Kim J, Weller RE, Cox JE, Cook EW 3rd, Horwitz B.Effective connectivity of a reward network in obese women. BrainRes Bull. 2009;79(6):388–95. https://doi.org/10.1016/j.brainresbull.2009.05.016.

34 Beaulieu K, Hopkins M, Blundell JE, Finlayson G. Impact of phys-ical activity level and dietary fat content on passive overconsump-tion of energy in non-obese adults. Int J Behav Nutr Phys Act.2017;14(1):14. https://doi.org/10.1186/s12966-017-0473-3.

35 Beaulieu K, Hopkins M, Long C, Blundell JE, Finlayson G. Highhabitual physical activity improves acute energy compensation innonobese adults. Med Sci Sports Exerc. 2017;49(11):2268–75.https://doi.org/10.1249/mss.0000000000001368.

Curr Obes Rep (2020) 9:63–8078

36• Drummen M, Dorenbos E, Vreugdenhil ACE, Raben A,Westerterp-Plantenga MS, Adam TC. Insulin resistance, weight,and behavioral variables as determinants of brain reactivity to foodcues: a prevention of diabetes through lifestyle intervention andpopulation studies in Europe and around the World - aPREVIEW study. Am J Clin Nutr. 2019;109(2):315–21. https://doi.org/10.1093/ajcn/nqy252. Using data from the Horizon2020 PREVIEW study the authors measured fasting glucoseand glucose tolerance as potential mechanisms to explain foodcompared to non-food brain activation and physical activitylevels.

37 Baecke JA, Burema J, Frijters JE. A short questionnaire for themeasurement of habitual physical activity in epidemiological stud-ies. Am J Clin Nutr. 1982;36(5):936–42. https://doi.org/10.1093/ajcn/36.5.936.

38 Horner KM, Finlayson G, Byrne NM, King NA. Food reward inactive compared to inactive men: roles for gastric emptying andbody fat. Physiol Behav. 2016;160:43–9. https://doi.org/10.1016/j.physbeh.2016.04.009.

39 KillgoreWD, KipmanM, Schwab ZJ, Tkachenko O, Preer L, GogelH, et al. Physical exercise and brain responses to images of high-calorie food. Neuroreport. 2013;24(17):962–7. https://doi.org/10.1097/wnr.0000000000000029.

40• Luo S, O'Connor SG, Belcher BR, Page KA. Effects of physicalactivity and sedentary behavior on brain response to high-caloriefood cues in young adults. Obesity (Silver Spring). 2018;26(3):540–6. https://doi.org/10.1002/oby.22107. Tested the interactionbetween BMI status and physical activity on neuronalactivation to high calorie foods and showed that inverseassociations were stonger in young adults with obesity.

41 Oustric P, Myers A, Gibbons C, Buckland N, Dalton M, Long C,et al. Are objectively measured free-living physical activity andsedentary behaviour associated with control over eating and foodpreferences in women? Appetite. 2018;123:465. https://doi.org/10.1016/j.appet.2017.11.067.

42 Alkahtani SA, Byrne NM, Hills AP, King NA. Acute interval exer-cise intensity does not affect appetite and nutrient preferences inoverweight and obese males. Asia Pac J Clin Nutr. 2014;23(2):232–8. https://doi.org/10.6133/apjcn.2014.23.2.07.

43 Alkahtani SA, Aldayel A, Hopkins M. Effects of acute eccentricexercise on appetite-related hormones and food preferences inmen. Am J Mens Health. 2019;13(4):1557988319861587. https://doi.org/10.1177/1557988319861587.

44 Crabtree DR, Chambers ES, Hardwick RM, Blannin AK. The ef-fects of high-intensity exercise on neural responses to images offood. Am J Clin Nutr. 2014;99(2):258–67.

45 Evero N, Hackett LC, Clark RD, Phelan S, Hagobian TA. Aerobicexercise reduces neuronal responses in food reward brain regions. JAppl Physiol. 2012;112(9):1612–9.

46 Farah NM, Brunstrom JM, Gill JM. Using a novel computer-basedapproach to assess the acute effects of exercise on appetite-relatedmeasures. Appetite. 2012;58(1):196–204.

47 Finlayson G, Bryant E, Blundell JE, King NA. Acute compensatoryeating following exercise is associated with implicit hedonic want-ing for food. Physiol Behav. 2009;97(1):62–7.

48 Martins C, Stensvold D, Finlayson G, Holst J, Wisloff U, KulsengB, et al. Effect of moderate- and high-intensity acute exercise onappetite in obese individuals. Med Sci Sports Exerc. 2015;47(1):40–8. https://doi.org/10.1249/mss.0000000000000372.

49 McNeil J, Cadieux S, Finlayson G, Blundell JE, Doucet E. Theeffects of a single bout of aerobic or resistance exercise on foodreward. Appetite. 2015;84:264–70. https://doi.org/10.1016/j.appet.2014.10.018.

50 Miguet M, Fillon A, Khammassi M, Masurier J, Julian V, Pereira B,et al. Appetite, energy intake and food reward responses to an acutehigh intensity interval exercise in adolescents with obesity. Physiol

Behav. 2018;195:90–7. https://doi.org/10.1016/j.physbeh.2018.07.018.

51 Thivel D, Fillon A, Genin PM,MiguetM, KhammassiM, Pereira B,et al. Satiety responsiveness but not food reward is modified inresponse to an acute bout of low versus high intensity exercise inhealthy adults. Appetite. 2019;145:104500. https://doi.org/10.1016/j.appet.2019.104500.

52 Saanijoki T, Nummenmaa L, Tuulari JJ, Tuominen L, Arponen E,Kalliokoski KK, et al. Aerobic exercise modulates anticipatory re-ward processing via the mu-opioid receptor system. Hum BrainMapp. 2018;39(10):3972–83. https://doi.org/10.1002/hbm.24224.

53 Alkahtani SA, Byrne NM, Hills AP, King NA. Interval trainingintensity affects energy intake compensation in obese men. Int JSport Nutr Exerc Metab. 2014;24(6):595–604.

54 Beaulieu K, HopkinsM, Gibbons C, Oustric P, Caudwell P, BlundellJ, et al. Exercise training reduces reward for high-fat food in adultswith overweight/obesity. Med Sci Sports Exerc. 2019.

55 Beaulieu K. The influence of physical activity level on the sensitiv-ity of the appetite control system. Leeds: University of Leeds; 2017.

56 Cornier MA,Melanson EL, Salzberg AK, Bechtell JL, Tregellas JR.The effects of exercise on the neuronal response to food cues.Physiol Behav. 2012;105(4):1028–34.

57 Finlayson G, Caudwell P, Gibbons C, Hopkins M, King N, BlundellJE. Low fat loss response after medium-term supervised exercise inobese is associated with exercise-induced increase in food reward. JObes. 2011;2011:615624. https://doi.org/10.1155/2011/615624.

58 Martin CK, Johnson WD, Myers CA, Apolzan JW, Earnest CP,Thomas DM, et al. Effect of different doses of supervised exerciseon food intake, metabolism, and non-exercise physical activity: theE-MECHANIC randomized controlled trial. Am J Clin Nutr. 2019.https://doi.org/10.1093/ajcn/nqz054.

59 Geiselman PJ, Anderson AM, DowdyML,West DB, Redmann SM,Smith SR. Reliability and validity of a macronutrient self-selectionparadigm and a food preference questionnaire. Physiol Behav.1998;63(5):919–28. https://doi.org/10.1016/s0031-9384(97)00542-8.

60 Martins C, Aschehoug I, Ludviksen M, Holst J, Finlayson G,Wisloff U, et al. High-intensity interval training, appetite, and re-ward value of food in the obese. Med Sci Sports Exerc. 2017;49(9):1851–8. https://doi.org/10.1249/mss.0000000000001296.

61 Martins C, Kazakova I, Ludviksen M, Mehus I, Wisloff U, KulsengB, et al. High-intensity interval training and isocaloric moderate-intensity continuous training result in similar improvements in bodycomposition and fitness in obese individuals. Int J Sport Nutr ExercMetab. 2016;26(3):197–204. https://doi.org/10.1123/ijsnem.2015-0078.

62 RiouME, Jomphe-Tremblay S, Lamothe G, Finlayson GS, BlundellJE, Decarie-Spain L, et al. Energy compensation following a super-vised exercise intervention in women living with overweight/obesity is accompanied by an early and sustained decrease in non-structured physical activity. Front Physiol. 2019;10:1048. https://doi.org/10.3389/fphys.2019.01048.

63 Thivel D, Julian V, Miguet M, Pereira B, Beaulieu K, Finlayson G,et al. Introducing eccentric cycling during a multidisciplinary weightloss intervention might prevent adolescents with obesity from in-creasing their food intake: the TEXTOO study. Physiol Behav.2019;112744. https://doi.org/10.1016/j.physbeh.2019.112744.

64 Pate RR, Heath GW, Dowda M, Trost SG. Associations betweenphysical activity and other health behaviors in a representative sam-ple of US adolescents. Am J Public Health. 1996;86(11):1577–81.https://doi.org/10.2105/ajph.86.11.1577.

65 Hart PD, Benavidez G, Erickson J. Meeting recommended levels ofphysical activity in relation to preventive health behavior and healthstatus among adults. J Prev Med Public Health. 2017;50(1):10–7.https://doi.org/10.3961/jpmph.16.080.

Curr Obes Rep (2020) 9:63–80 79

66 Joseph RJ, Alonso-Alonso M, Bond DS, Pascual-Leone A,Blackburn GL. The neurocognitive connection between physicalactivity and eating behaviour. Obes Rev. 2011;12(10):800–12.https://doi.org/10.1111/j.1467-789X.2011.00893.x.

67 Annesi JJ, Porter KJ. Behavioural support of a proposedneurocognitive connection between physical activity and improvedeating behaviour in obese women. Obes Res Clin Pract. 2014;8(4):e325–30. https://doi.org/10.1016/j.orcp.2013.08.001.

68 Oustric P, Thivel D, Dalton M, Beaulieu K, Gibbons C, Hopkins M,et al. Measuring food preference and reward: application and cross-cultural adaptation of the Leeds Food Preference Questionnaire inhuman experimental research. Food Qual Prefer. 2020;80:103824.https://doi.org/10.1016/j.foodqual.2019.103824.

69 Hopkins M, Gibbons C, Caudwell P, Webb DL, Hellstrom PM,Naslund E, et al. Fasting leptin is a metabolic determinant of foodreward in overweight and obese individuals during chronic aerobicexercise training. Int J Endocrinol. 2014;2014:323728. https://doi.org/10.1155/2014/323728.

70 Rosenbaum M, Kissileff HR, Mayer LES, Hirsch J, Leibel RL.Energy intake in weight-reduced humans. Brain Res. 2010;1350:95–102. https://doi.org/10.1016/j.brainres.2010.05.062.

71 Tremblay A, Dutheil F, Drapeau V, Metz L, Lesour B, Chapier R,et al. Long-term effects of high-intensity resistance and enduranceexercise on plasma leptin and ghrelin in overweight individuals: theRESOLVE Study. Appl Physiol Nutr Metab. 2019;44(11):1172–9.https://doi.org/10.1139/apnm-2019-0019.

72 Gibbons C, Blundell JE, Caudwell P, Webb DL, Hellstrom PM,Naslund E, et al. The role of episodic postprandial peptides inexercise-induced compensatory eating. J Clin Endocrinol Metab.2017;102(11):4051–9. https://doi.org/10.1210/jc.2017-00817.

73 Legget KT, Wylie KP, Cornier MA, Melanson EL, Paschall CJ,Tregellas JR. Exercise-related changes in between-network connec-tivity in overweight/obese adults. Physiol Behav. 2016;158:60–7.https://doi.org/10.1016/j.physbeh.2016.02.031.

74 McFadden KL, Cornier MA, Melanson EL, Bechtell JL, TregellasJR. Effects of exercise on resting-state default mode and saliencenetwork activity in overweight/obese adults. Neuroreport.2013 ;24 (15 ) : 866–71 . h t t p s : / / do i . o rg / 10 . 1097 /wn r.0000000000000013.

Publisher’s Note Springer Nature remains neutral with regard to jurisdic-tional claims in published maps and institutional affiliations.

Curr Obes Rep (2020) 9:63–8080