Received: 1 August 2016

| Revised: 25 January 2017

| Accepted: 30 January 2017
DOI 10.1111/psyp.12860

ORIGINAL ARTICLE

Testing food-related inhibitory control to high- and low-calorie

food stimuli: Electrophysiological responses to high-calorie food
stimuli predict calorie and carbohydrate intake

Kaylie A. Carbine1 | Edward Christensen2 | James D. LeCheminant2 |

Bruce W. Bailey2 | Larry A. Tucker2 | Michael J. Larson1,3

1
Department of Psychology, Brigham
Young University, Provo, Utah
2 Maintaining a healthy diet has important implications for physical and mental health.
Department of Exercise Sciences,
Brigham Young University, Provo, Utah
3
Neuroscience Center, Brigham Young
University, Provo, Utah
Correspondence
Michael J. Larson, Ph.D., Department of
Psychology and Neuroscience Center,
Brigham Young University, 244 TLRB,
Provo, UT 84602, USA.
Email: michael_larson@byu.edu
Funding information
Brigham Young University Mentored
Environment Grant, Brigham Young
University College of Family, Home, and
Social Sciences
Abstract
One factor that may influence diet and food consumption is inhibitory control—the
ability to withhold a dominant response in order to correctly respond to environ-
mental demands. We examined how N2 amplitude, an ERP that reflects inhibitory
control processes, differed toward high- and low-calorie food stimuli and related to
food intake. A total of 159 participants (81 female; M age 5 23.5 years; SD 5 7.6)
completed two food-based go/no-go tasks (one with high-calorie and one with low-
calorie food pictures as no-go stimuli) while N2 amplitude was recorded. Participants
recorded food intake using the Automated Self-Administered 24-hour Dietary Recall
system. Inhibiting responses toward high-calorie stimuli elicited a larger (i.e., more
negative) no-go N2 amplitude; inhibiting responses toward low-calorie stimuli eli-
cited a smaller no-go N2 amplitude. Participants were more accurate during the high-
calorie than low-calorie task, but took longer to respond on go trials toward high-
calorie rather than low-calorie stimuli. When controlling for age, gender, and BMI,
larger high-calorie N2 difference amplitude predicted lower caloric intake (b 5 0.17);
low-calorie N2 difference amplitude was not related to caloric intake (b 5 20.03).
Exploratory analyses revealed larger high-calorie N2 difference amplitude predicted
carbohydrate intake (b 5 0.22), but not protein (b 5 0.08) or fat (b 5 0.11) intake.
Results suggest that withholding responses from high-calorie foods requires increased
recruitment of inhibitory control processes, which may be necessary to regulate food
consumption, particularly for foods high in calories and carbohydrates.

KEYWORDS
ASA24, ERPs, food intake, food-related inhibitory control, N2

1 | INTRODUCTION global deaths; World Health Organization [WHO], 2013),

Maintaining a healthy diet is essential for sustaining physical
and mental health. For example, a healthy diet reduces the
risk for Type 2 diabetes, obesity, and depression (Kastorini
& Panagiotakos, 2009; Kumanyika, Jeffery, Morabia, Riten-
baugh, & Antipatis, 2002; Lai et al., 2014; Rahati, Shahraki,
Arjomand, & Shahraki, 2014). Cardiovascular diseases,
which are the number one cause of death worldwide (31% of
are also significantly influenced by diet (Bazzano et al.,
2002, 2014; Keogh, 2013; Walker, 2013). Furthermore, diet-
related health risk factors, such as high blood pressure, high
cholesterol, and obesity, account for 19% of global deaths
(WHO, 2009). The importance of maintaining a healthy diet
has become enough of a public health concern that the WHO
has issued a global strategy on how to improve diet (World
Health Assembly, 2004).

Psychophysiology. 2017;1–16 wileyonlinelibrary.com/journal/psyp V

C 2017 Society for Psychophysiological Research | 1
2
| CARBINE
Multiple factors influence diet and food consumption,
such as portion control and serving sizes (Smith & Ditschun,
2009), visibility and convenience (Painter, Wansink, & Hieg-
gelke, 2002), accessibility and availability in food retail
(White, 2007), activities (e.g., watching TV, listening to
music; Stroebele & de Castro, 2006), hormones (Begg &
Woods, 2013), and social influences (Herman, Roth, &
Polivy, 2003; Vartanian, Spanos, Herman, & Polivy, 2015).
One aspect that may relate to food intake, but is less under-
stood, is inhibitory control toward food and food cues. Inhib-
itory control is defined as one’s ability to withhold a
dominant response in order to correctly respond to environ-
mental demands or task-relevant information (Ko & Miller,
2013). Inhibitory control may relate to food intake in multi-
ple ways, such as inhibiting the generally automatic response
to eat palatable foods, not acting on spontaneous food crav-
ings, or refraining from overeating in response to emotional
states (Blundell & Gillett, 2001; Davis et al., 2007; Guerrieri,
Nederkoorp, Stankiewicz et al., 2007; Jasinska et al., 2012).
Behavioral measures of inhibitory control (e.g., reaction
times, accuracy) are negatively correlated with weight, sug-
gesting that individuals with reduced inhibitory control may
have difficulty withholding from palatable foods, particularly
those high in fat and sugar (Appelhans et al., 2011; Hall,
2012; Vainik, Dagher, Dube, & Fellows, 2013). Self-report
and behavioral measures of impulsivity also reveal that indi-
viduals with higher impulsivity tend to eat more when pre-
sented with food (Guerrieri, Nederkoorn, & Jansen, 2007;
Guerrieri Nederkoorn, Stankiewicz et al., 2007; Jansen et al.,
2009).
In addition to behavioral and self-report measures, under-
standing the neural mechanisms associated with food-related
inhibitory control can help elucidate the relationship between
inhibitory control and food intake. One way to observe neu-
ral indicators of inhibitory control is through ERPs. ERPs
are scalp-recorded changes of the brain’s electrical activity
that reflect neural processing of stimuli. The amplitude of an
ERP waveform fluctuates due to the strength of a presented
stimulus or depending on how an individual responds to the
stimulus (Luck, 2005). One ERP that reflects inhibitory con-
trol is the N2, which is a negative-going deflection in the
ERP waveform that peaks 200–350 ms after the onset of a
stimulus and is associated with competing response options
or the inhibition of a response (Folstein & Van Petten,
2008). The amplitude of the N2 is more negative when an
individual suppresses a dominant response toward a stimu-
lus, supporting the idea that the N2 is an indicator of inhibi-
tory control processes (Folstein & Van Petten, 2008).
In regard to food-related inhibitory control and ERPs,
Watson and Garvey (2013) found that N2 amplitude was
larger (i.e., more negative) when females had to inhibit
responses toward food stimuli relative to nonfood stimuli,
ET AL.
suggesting participants had to recruit more cognitive resour-
ces to inhibit their response toward food cues. Their study
shows promise that ERPs can be used as neurological meas-
ures of food-related inhibitory control. As noted previously,
studies utilizing surveys and behavioral measures suggest
that individuals may require more inhibitory control when
inhibiting toward rewarding/palatable foods that are high in
fat and sugar compared to unappetizing foods (Appelhans
et al., 2011; Hall, 2012), but these possibilities have not been
examined from a neural perspective. Therefore, it is neces-
sary to understand how neural manifestations of inhibitory
control may differ depending on the type of food presented.
One aim of our study is to examine how neural indices of
food-related inhibitory control, as measured by N2 ERP
amplitude, may differ toward pictures of high- and low-
calorie foods.
In addition to understanding how food-related inhibition
may vary toward high- and low-calorie foods, it is also
important to establish how neural indices of inhibition relate
to food intake. The relationship between neural indices of
inhibitory control and food intake has rarely been examined.
In the fMRI literature, increased activation in the dorsal lat-
eral prefrontal cortex (DLPFC) when viewing hedonic food
images (e.g., cake, cookies, pastries) was related to decreased
caloric intake when meals were presented ad libitum style
(Cornier, Salzberg, Endly, Bessessen, & Tregellas, 2010).
Direct activation of the DLPFC has also been related to
reduced self-reported cravings and fewer calories consumed
from foods presented ad libitum (Lapenta, Sierve, de Mac-
edo, Fregni, & Boggio, 2014).
We are not aware of any studies that directly relate ERP
indices of inhibitory control to food intake either in con-
trolled (i.e., laboratory) or naturalistic settings. Our second
aim, therefore, was to examine the relationship between N2
amplitude toward high- and low-calorie food cues and food
intake. The relationship between ERP indices of inhibition
and food intake is important to examine in order to determine
if experimental paradigms and technologies used in research
labs can provide researchers with valid information about
real-world eating behavior. Once validity is established with
eating behaviors, indices such as the N2 ERP that reflect
neural processing of food cues may be used to strengthen
health and diet interventions by seeing what interventions
affect food-related inhibition and potentially food intake, as
well as to determine if inhibitory control can be directly
trained as a potential diet-related intervention.
As foods high in calories (e.g., foods high in fat and
sugar) may require more inhibitory control (Appelhans et al.,
2011; Hall, 2012), we first hypothesized that inhibiting
responses toward high-calorie food pictures would elicit
larger (i.e., more negative) N2 amplitudes compared to low-
calorie foods. Second, given the absence of previous studies
CARBINE ET AL.
| 3

TAB LE 1 Demographic information and questionnaires Baldwin, & Larson, 2013; Schimmel, 1967), and 2 for mis-
understanding task instructions (i.e., withheld from go stim-
Mean SD Range
uli and responded to no-go stimuli). The final sample
Age (years) 23.51 7.62 18–55 included 159 participants (81 [51%] female) aged 18–55
BMI (kg/m2) 24.24 6.54 15.90–51.44 years (M age 5 23.51; SD 5 7.62; see Table 1 for full demo-
graphic information).
Education (years) 14.46 2.13 11–22

DEBQ-Emotional scale 2.45 0.68 1.00–4.17

2.2 | Experimental protocol
DEBQ-External scale 3.32 0.58 1.10–4.70
The study protocol is outlined in Figure 1. All participants
G-FCQ-S-Total score 49.00 8.44 19–71
were instructed to get at least 7 hr of sleep the night before
VAS-Current hunger 5.56 2.12 0.00–9.85 participating and refrain from vigorous physical activity 24 hr
VAS-Fullness (reverse scored) 7.50 1.93 1.60–10.00
before in order to control for potential confounding effects
due to differing amounts of sleep (St-Onge, Wolfe, Sy,
VAS-Desire to eat 5.51 2.21 0.30–9.60 Shechter, & Hirsch, 2014) and exercise (Hanlon, Larson,
VAS-How much could eat 6.06 1.67 0.50–10.00 Bailey, & LeCheminant, 2012). Participants were also asked
to refrain from consuming caffeine 24 hr before participation.
VAS-Urge to eat 5.31 2.13 0.00–9.50
To control for hunger levels, participants were asked to
VAS-Preoccupation 3.99 2.20 0.00–9.00 stop eating and drinking (except water) by 9 pm the night
before if individuals reported to the lab during morning
VAS-Total 5.66 1.65 2.00–8.80
hours (i.e., 7–10 am) and 4 hr before participation if individ-
Note. For education, 12 completed high school. DEBQ 5 Dutch Behavior Eat- uals reported during evening hours (i.e., 7–10 pm). Such
ing Questionnaire; G-FCQ-S 5 General Food-Cravings Questionnaire, State
scale; VAS 5 visual analog scale.
fasting requirements allowed us to observe neural responses
to food cues when participants were about to consume their
next meal, as we aimed to model regular eating patterns. In
relating N2 amplitude to calorie consumption, we tested addition, having all participants fast controlled for caloric
competing hypotheses that either (a) individuals with larger
N2 amplitudes are less efficient at resisting environmental
Report to lab between 7-10am or 7-10pm
food cues and will eat more, or (b) individuals with larger
N2 amplitudes are more successful at resisting environmental
food cues and will eat less. Written consent and confirm sleep, exercise, and fasting requirements.

2 | METHOD
Complete demographics questionnaire, DEBQ, and G-FCQ-S.

2.1 | Participants
Complete 6 VAS hunger questions
Participants were recruited via flyers posted in the commu-
nity and from undergraduate classes recruited for course
credit. All participants were native English speakers, psy-
EEG Protocol- Complete following tasks in counterbalanced fashion:
chiatrically healthy, and had no chronic or metabolic dis-
-High-calorie task (no-go stimuli = high-calorie foods; go stimuli = low-calorie foods)
eases. Participants were excluded if they reported a head
-Low-calorie task (no-go stimuli = low-calorie foods; go stimuli = high-calorie foods)
injury resulting in loss of consciousness, use of recreational
drugs, or were pregnant or lactating. Exclusion criteria were
assessed via phone calls from trained research assistants prior Complete first ASA24 dietary recall in the lab
to study participation and confirmed during study participa-
tion using a demographic questionnaire. Data were initially
collected from 181 participants that met inclusion criteria. Complete a minimum of three more ASA24 recalls after lab session:
Five participants were subsequently excluded due to com- -One for the day they reported to the lab
puter malfunction during data acquisition, 13 for fewer than
-At least one randomly assigned weekday
20 useable ERP trials for each condition (Leue, Klein,
-One randomly assigned weekend
Lange, & Beauducel, 2013; Rietdijk, Franken, & Thurik,
2014), 2 for noisy ERP data (i.e., noise levels > 2; Clayson, FIGURE 1 Experimental protocol
4
| CARBINE
TAB LE 2 Means and standard deviations (SD) for food data
Mean SD Range
Calories 2,191.61 710.19 969.21–4,347.29
Carbohydrates (grams) 271.75 88.41 104.50–511.44
Protein (grams) 83.58 39.72 24.23–370.57
Fat (grams) 87.98 34.64 35.01–233.23
intake, which may influence neural responses to food cues.
Hunger levels were quantified by averaging across six visual
analog scales (VAS; see below). An independent samples t
test between morning and evening participants indicated
greater levels of hunger in evening participants, M 5
6.20 cm, SD 5 1.69, than individuals in the morning,
M 5 5.46 cm, SD 5 1.60, t(154) 5 2.52, p 5 .01, d 5 0.45.
However, overall hunger levels did not correlate with go or
no-go N2 amplitudes on the high- or low-calorie tasks (all
rs < |.09|, all ps > .27), suggesting differences in hunger do
not relate to N2 amplitude. As time of day was not a variable
of interest and hunger differences between time of day were
not related to N2 amplitude, we collapsed across time of day
for all subsequent data analyses.
Upon arrival at the lab, participants provided written con-
sent and verbally confirmed completion of the sleep, exer-
cise, caffeine, and fasting requirements described above. If
participants did not meet the study requirements, they were
rescheduled to participate at another time when they fol-
lowed study instructions. Participants then completed a
demographic survey, the Dutch Eating Behavior Question-
naire (DEBQ), and the General Food-Cravings Question-
naire, State scale (G-FCQ-S, see Table 1 for scores). Hunger
was then assessed using six VAS items, which asked about
current levels of hunger, fullness (reverse scored), desire to
eat, how much you could eat, urge to eat, and preoccupation
with thoughts of food (see Table 1). Each VAS consisted of
a 100-mm (10 cm) line, anchored from not at all to
extremely on each end (Blundell et al., 2010; Stratton et al.,
1998). VAS measurements reliably and consistently predict
meal initiation, amount eaten, and are sensitive to experimen-
tal manipulations (Stubbs et al., 2000).
Participants then completed two food go/no-go tasks, in
a counterbalanced fashion, while wearing a 128-channel
EEG sensor net (see below). Finally, participants reported
the food they had eaten the previous day using the Auto-
mated Self-Administered 24-hour Dietary Recall (ASA24;
National Cancer Institute [NCI], 2014). Additionally, partici-
pants completed a minimum of three more dietary recalls
after their lab session, one for the day they reported to the
lab, at least one randomly assigned weekday, and one ran-
domly assigned weekend (see Table 2). Participants were
ET AL.
unaware of what days they would be recording until they
were asked via text message the following day to record the
previous day’s food intake.
2.3 | Dutch Eating Behavior Questionnaire
The DEBQ is a 33-item survey that assesses restrained, exter-
nal, and emotional eating styles (van Strien, Frijters, Bergers,
& Defares, 1986). Original development of the DEBQ
resulted in a four-factor (i.e., restrained, external, diffuse emo-
tions, clearly labeled emotions) structure, but follow-up analy-
ses have also supported a reliable three-factor (i.e., restrained,
external, emotional) structure (Wardle, 1987). Factors on the
DEBQ demonstrated strong internal consistency when exam-
ined in men, women, obese, and lean individuals (Cronbach’s
alphas > .79; Wardle, 1987). The DEBQ has accurately iden-
tified characteristic eating styles in women dieting or with
anorexia nervosa or bulimia nervosa (Wardle, 1987) and pre-
dicts weight gain over a 3-year time period (Tucker & Bates,
2009), supporting its external validity. The DEBQ was scored
following the three-factor structure by averaging across items
in each of the three factors for factor scores.
2.4 | General Food-Cravings Questionnaire,
State scale
The G-FCQ-S is a 15-item survey that measures current
cravings for food via factors that may perpetuate those crav-
ings, such as preoccupation and lack of control toward food
or reward value of food (Nijs, Franken, & Muris, 2007).
Exploratory and confirmatory factor analyses show the five
factors of the G-FCQ-S explain 80% of the total variance
with a Cronbach’s alpha of .92–.93 (Nijs, Franken, & Muris,
2007). Scores on the G-FCQ-S were significantly higher
when participants were deprived of food compared to sati-
ation, confirming the validity of the G-FCQ-S as an indicator
of state food cravings (Nijs et al., 2007). The G-FCQ-S was
scored by summing items in each of the five factors for fac-
tor scores and summing across all items for an overall score.
The G-FCQ-S was chosen as opposed to the trait scale in
order to better examine current food cravings in relation to
current neural responses to food.
2.5 | Go/no-go tasks
During EEG data collection, participants completed two food
go/no-go tasks in order to elicit the N2 ERP component and
assess inhibitory control toward high- and low-calorie foods.
In the high-calorie task, participants were instructed to
respond with a button press when they saw pictures of low-
calorie foods (go stimuli) and inhibit responses when they
saw pictures of high-calorie foods (no-go stimuli). In the
CARBINE ET AL.
low-calorie task, participants were instructed to do the oppo-
site and respond when they saw pictures of high-calorie
foods (go stimuli) and inhibit when they saw pictures of
low-calorie foods (no-go stimuli). Task order was counter-
balanced across participants. For each task, there were two
blocks of 100 trials, 70 of which were go trials and 30 were
no-go trials to establish a prepotency to responding to go tri-
als. Pictures were presented for 500 ms, with a random inter-
stimulus interval fixation cross that varied between 1,200
and 1,400 ms.
Sixty high-calorie and 60 low-calorie food pictures were
provided from Killgore and colleagues (2003), who have
used the pictures in multiple papers (e.g., Killgore &
Yurgelun-Todd, 2005, 2007; Killgore et al, 2013). Prior to
this study, 26 separate participants from the undergraduate
research participant pool rated the 120 pictures as either
high- or low-calorie foods. Only pictures that were accu-
rately categorized as high- or low-calorie foods 95% of the
time or better were used in the final tasks, resulting in 38 pic-
tures for each category (Christensen, 2014). The low-calorie
food images consisted of 13 vegetables (e.g., carrots, broc-
coli) and 25 fruits (e.g., apples, oranges). The high-calorie
food images consisted of 16 desserts (e.g., cake, ice cream),
15 high-calorie dinner meals (e.g., hamburgers, hot dogs),
and 7 high-calorie breakfast meals (e.g., waffles, pancakes).
2.6 | EEG data acquisition and analyses
EEG data were recorded using a high-density 128-channel
EEG sensor net and Electrical Geodesics, Inc. amplifier sys-
tem (20K nominal gain, band-pass 5 0.10–100 Hz). During
collection, EEG data were referenced to the vertex electrode,
digitized continuously at 250 Hz, and electrode impedance
was maintained at 50 kX or less per manufacturer recom-
mendations. Data were subsequently average-referenced and
digitally low-pass filtered at 30 Hz. Eyeblinks were removed
using independent components analysis (ICA) in the ERP
PCA Toolkit (Dien, 2010). If ICA components correlated at
.9 or higher with two blink templates (one generated by cur-
rent data and one provided by ERP PCA Toolkit), they were
removed from the data (Dien, Michelson, & Franklin, 2010).
Channels were defined as bad if the fast average amplitude
exceeded 100 microvolts (lV) or if the differential average
amplitude exceeded 50 lV, as recommended by the author
and creator of the ERP Toolkit (Dien, 2010).
ERP data were segmented into epochs from 200 ms prior
to stimulus onset to 400 ms after stimulus onset. Epochs
were baseline adjusted using the 200-ms window prior to
stimulus onset. As the N2 consistently peaks around 200–
350 ms after the onset of a stimulus (Clayson & Larson,
2013; Folstein & Van Petten, 2008), N2 amplitudes were
extracted from an a priori determined window as the mean
| 5
amplitude between 200 and 300 ms (Clayson et al., 2013).
ERP data were averaged across four frontocentral electrode
sites (6, 7, 107, Cz) to improve reliability of the signal rela-
tive to using only a single electrode (Clayson & Larson,
2013; see Larson, Farrer, & Clayson, 2011, for electrode
montage). Electrode locations were chosen to be consistent
with multiple previous studies from our lab using the N2
(e.g., Clawson, Clayson, & Larson, 2013; Clayson, Clawson,
& Larson, 2011; Clayson & Larson, 2012, 2013) and were
chosen a priori. For the final sample of 159 participants, the
number of useable no-go trials ranged from 25–68 trials
(M 5 49.28, SD 5 7.08) and useable go trials ranged from
35–150 trials (M 5 128.70, SD 5 13.02).
2.7 | ASA24 dietary recalls
The ASA24 (NCI, 2014) is an online automated multiple-pass
dietary recall system where participants record their food
intake for the previous day. Automated multiple-pass dietary
recalls have been shown to be fairly accurate and deal well
with underreporting of food intake (Moshfegh et al.,
2008). The ASA24 has been validated against interviewer-
administered multiple-pass recalls in terms of accuracy of the
food recall (Kirkpatrick et al., 2014). As previously men-
tioned, participants were instructed to record their food intake
for a minimum of four times over the course of the study: one
during the lab session for the previous day, one for the day
they reported to the lab, at least one randomly assigned week-
day (i.e., Monday–Thursday), and one randomly assigned
weekend (i.e., Friday–Sunday). If participants recorded less
than 130% of their resting metabolic rate (calculated using the
Harris & Benedict, 1918, equation), they were contacted by
telephone and their logs reviewed with them to ensure accu-
racy of the recording. If inaccurate, participants were asked to
record another randomized day. Resting metabolic rate is
defined as the necessary energy required to perform vital
body functions at rest and has been identified as the main con-
tributor to total energy expenditure (Sabounchi, Rahmandad,
& Ammerman, 2013). Therefore, if participants indicated that
they were not consuming enough calories for basic body func-
tions and energy expenditure, it was important to contact
them to ensure their food recordings were correct. Average
caloric intake was calculated for participants who successfully
recorded at least 2 days of food intake (see Table 2), as some
data were lost due to ASA24 malfunctions (e.g., total calories
were not calculated) or participants not completing their
make-up recall. If only one day was successfully recorded,
that day’s caloric intake was used in place of the average
(n 5 5). Excluding the five participants who recorded one day
did not change any patterns of significance. Therefore, all par-
ticipants remained in the analyses for completeness of data
(Gelman & Loken, 2014).
6
| CARBINE
2.8 | Statistical analyses
To test the first hypothesis that indices of inhibitory control
differ by high- and low-calorie food stimuli, 2 Task (high-
calorie, low-calorie) 3 2 Trial (go, no-go) repeated measures
analyses of variance (ANOVAs) were used to assess differ-
ences in accuracy and N2 amplitude toward high- and low-
calorie foods. The Greenhouse-Geisser correction was used
for violations of sphericity, and partial eta squared (hp2) is
reported as a measure of effect size. Correlations among
residuals are reported for future power analyses (see Larson
& Carbine, 2017). Differences in correct go-trial mean reac-
tion times (RTs) during the high- and low-calorie tasks were
analyzed using within-subject paired samples t test. Within-
subject Cohen’s d is reported as a measure of effect size.
To test the second hypothesis if N2 amplitude predicts
caloric intake, we conducted two multiple linear regressions
with age, sex (male 5 0, female 5 1), body mass index
(BMI), and either high- or low-calorie N2 difference ampli-
tude predicting caloric intake. Standardized betas from the
regression analyses are reported. High-calorie N2 difference
amplitude was calculated as no-go minus go trial amplitude
on the high-calorie task (i.e., when inhibiting toward high-
calorie images). Low-calorie N2 difference amplitude was
calculated as no-go minus go trial amplitude on the low-
calorie task (i.e., when inhibiting toward low-calorie images).
For the N2 difference amplitudes, more negative numbers
reflect a larger inhibitory response. As high- and low-calorie
N2 difference amplitudes are significantly correlated (r 5
2.28, p < .001), we entered them in separate models to
ensure results were not affected by multicollinearity. It is
important to note that we did not include hunger levels
(measured by VAS) in the regression equations, as VAS
scores reflected current hunger levels but our outcome of
interest was average caloric intake across multiple days.
However, as previously noted, hunger levels were not related
to N2 amplitudes during the high- or low-calorie task. Var-
iance inflation factor (VIF) scores are reported as measures
of multicollinearity and adjusted R2, DR2, and Cohen’s f 2 are
reported as measures of effect sizes.
Finally, as food-related inhibitory control may be influ-
enced by cravings, emotional eating, and palatable food cues
(Blundell & Gillett, 2001; Davis et al., 2007; Guerrieri,
Nederkoorn, Stankiewicz et al., 2007; Jasinska et al., 2012),
we conducted exploratory analyses (at a p 5 .01 significance
level to adjust for number of comparisons) to see if N2 dif-
ference amplitudes correlated with self-reports of cravings
(G-FCQ-S total score), emotional eating (DEBQ emotional
eating subscale), or external eating (DEBQ external eating
subscale). The DEBQ restrained eating subscale was not
examined because, although highly restrained eaters tend to
overeat when they have lower levels of inhibitory control
(Jansen et al., 2009; Price, Lee, & Higgs, 2016), restrained
ET AL.
eating has not been suggested as a determinant of food-
related inhibitory control. All analyses were performed using
SPSS 23 software.
2.9 | Power analyses
Power analyses are not commonly reported in ERP literature,
but are important to help improve the rigor of scientific
research and ensure sample sizes are adequate to detect
effects of interest (Larson & Carbine, 2017). To calculate the
sample size needed to test our first hypothesis (i.e., differen-
ces in inhibitory control toward high- and low-calorie foods),
we conducted a one group, two repeated measures, within-
factors power analyses in G*Power (v3.1) based on Watson
and Garvey’s (2013) observed effect between food and non-
food N2 no-go amplitude (hp2 5 .05, which was converted
into an F effect size of .23). Correlations among repeated
measures was set to a conservative .4, as the correlation was
not reported in Watson and Garvey. To achieve 95% power,
at least 77 participants were needed to detect the effects of
interest (alpha 5 .05). As we are unaware of any multiple lin-
ear regression analyses relating N2 amplitude to caloric intake,
we followed the recommendations of Green (1991; 104 plus
the number of predictors for testing individual predictors) to
calculate the sample size needed for our second hypothesis.
Assuming a medium-sized relationship, at least 108 participants
are needed to detect the effects of interest based on this rule of
thumb. Thus, we are more than adequately powered to detect
the effects of interest with our current sample size.
3 | RESULTS
3.1 | Hypothesis 1: Inhibitory control toward
high- and low-calorie stimuli
3.1.1 | ERP data
Grand-averaged ERP waveforms for go trials, no-go trials, and
difference waves (no-go minus go trials) for the high- and low-
calorie tasks are presented in Figure 2a–c; scalp distributions
for high- and low-calorie task no-go trials and the high- and
low-calorie task difference waves are presented in Figure 3.
For Hypothesis 1, as expected, there was a main effect for trial
type, with no-go trials eliciting a larger (i.e., more negative) N2
amplitude than go trials, collapsed across the high- and low-
calorie tasks, F(1, 158) 5 14.01, p < .001, hp2 5 .08 (see Table
3 for mean and standard deviations for all ERP amplitudes; see
Table 4 for correlations between residuals). There was no main
effect for high- versus low-calorie task collapsed across go and
no-go trial types, F(1, 158) 5 1.00, p 5 .32, hp2 5 .01.
The Task 3 Trial interaction was significant, F(1, 158) 5
55.03, p < .001, hp2 5 .26. Post hoc paired samples t tests
revealed that no-go N2 amplitude was more negative during the
high-calorie task (i.e., when inhibiting responses toward high-
CARBINE ET AL.
| 7

(a) High-Calorie Task: No-Go Trials (b) Low-Calorie Task: No-Go Trials
1 High-Calorie Task: Go Trials 1 Low-Calorie Task: Go Trials

Amplitude (µV)
Amplitude (µV)
0 0

-1 -1

-2 -2

N2 N2
-3 -3
-200 -100 0 100 200 300 400 -200 -100 0 100 200 300 400
Time (ms) Time (ms)

(c)
1

0
Amplitude (µV)

-1

-2 High- Calorie Task: GNG Difference

Low-Calorie Task: GNG Difference

-3
-200 -100 0 100 200 300 400
Time (ms)
F I G U R E 2 (a) N2 amplitudes during the high-calorie task by trial type. (b) N2 amplitudes during the low-calorie task by trial type. (c) N2 difference
wave (no-go minus go trials) for the high- and low-calorie tasks

calorie stimuli) than during the low-calorie task (i.e., when
inhibiting responses toward low-calorie stimuli, t(158) 5 2
2.94, p 5 .004, dz 5 .23). Go N2 amplitude was more negative
during low-calorie task (i.e., when responding to high-calorie
foods) than the high-calorie task (i.e., when responding to low-
calorie foods, t(158) 5 5.10, p < .001, dz 5 .41). Additionally,
no-go N2 amplitude was more negative than go N2 amplitude
during the high-calorie task, t(158) 5 8.07, p < .001, dz 5 .68,
whereas go N2 amplitude was more negative than no-go N2
amplitude during the low-calorie task, t(158) 5 3.39, p 5 .001,
dz 5 .28 (see Figure 2a,b). In sum, inhibiting responses toward
high-calorie foods on the high-calorie task elicited a larger
inhibitory response, as indicated by a larger no-go N2 ampli-
tude. However, inhibiting responses toward low-calorie foods
on the low-calorie task did not elicit an inhibitory response, as
indicated by a smaller no-go N2 amplitude.
3.1.2 | Behavioral data
For accuracy, there was a main effect for trial type as
expected—individuals were more accurate on go trials than
no-go trials, collapsed across the high- and low-calorie tasks,
F(1, 144) 5 214.13, p < .001, hp2 5 .60 (see Table 3; see
Table 5 for correlations between residuals). There was also a
main effect of task, with individuals being more accurate on
the high-calorie task than the low-calorie task, collapsed
across trial type, F(1, 144) 5 6.56, p 5 .01, hp2 5 .04. The
Task 3 Trial interaction was not significant, F(1, 144) 5
3.03, p 5 .08, hp2 5 .02. Correct go trial mean RTs (note that
there are only RTs for go trials as participants were asked to
inhibit responses for no-go trials) were significantly faster
during the high-calorie task (i.e., when responding to low-
calorie stimuli) than the low-calorie task (i.e., when respond-
ing to high-calorie stimuli, t(144) 5 8.12, p < .001, dz 5 .67;
see Table 2). Overall, individuals were more accurate when
having to inhibit responses toward high-calorie food cues,
but were slower when having to respond toward high-calorie
food cues.
As results indicated a larger inhibitory response toward
high-calorie food stimuli and slower RTs to high-calorie
stimuli, we conducted follow-up zero-order correlations at a
significance level of p 5 .01 (to reduce Type I error) to see if
8
| CARBINE
High-Calorie Task: No-Go Trials
High-Calorie Task: No-Go minus Go Difference
F I G U R E 3 Scalp distributions for high-calorie task no-go trials, low-calorie task no-go trials, and the corresponding no-go minus go trial difference
on the high- and low-calorie tasks
neural responses to food cues related to the differences seen
in RTs. Mean go trial RTs toward high-calorie stimuli were
positively correlated with high-calorie N2 difference ampli-
tude (r 5 .25, p 5 .002). In other words, the larger inhibitory
response there was toward high-calorie food stimuli (i.e.,
more negative N2 difference amplitude), the faster individu-
als correctly responded toward high-calorie food stimuli.
Mean go trial RTs toward low-calorie stimuli were not corre-
lated with low-calorie N2 difference amplitude (r 5 2.15,
p 5 .08; see Figure 4a,b).1
1
After examining scatter plots (see Figure 4a,b), three RTs were identi-
fied as potential outliers for both the high- and low-calorie tasks. After
removing the outliers, RTs toward high-calorie stimuli were still posi-
tively correlated (at p 5 .01 level) with high-calorie N2 difference ampli-
tude (r 5 .24, p 5 .004), and RTs toward low-calorie stimuli were still
ET AL.
Low-Calorie Task: No-Go Trials
Low-Calorie Task: No-Go minus Go Difference
3.2 | Hypothesis 2: Predicting caloric intake
To test Hypothesis 2, age, sex, BMI, and high-calorie N2 dif-
ference amplitude (i.e., no-go minus go amplitude on the
high-calorie task) were entered into a multiple linear regres-
sion predicting caloric intake (see Table 6). The model met
basic assumptions for multicollinearity, and was acceptable
for homoscedasticity and normality of residuals. Age
(b 5 0.04, p 5 .56) and BMI (b 5 0.09, p 5 .22) did not sig-
nificantly predict caloric intake. Sex predicted caloric intake
(b 5 20.47, p < .001), with females consuming fewer calo-
ries than males. High-calorie N2 difference amplitude also
predicted caloric intake (b 5 0.18, p 5 .01), where individu-
als with more negative N2 difference amplitudes (i.e., a
not significantly correlated with low-calorie N2 difference amplitude
(r 5 2.17, p 5 .04).
CARBINE ET AL.
| 9

TAB LE 3 Means and standard deviations (SD) for ERP and p 5 .01 significance level to see if the increase in caloric
behavioral data intake was driven by an increase in carbohydrate, protein, or
fat consumption (all measured in grams broken down by
Mean SD Range
ASA24 reporting). One participant was excluded from analy-
High-calorie task ses, as macronutrient data could not be calculated due to
N2 no-go amplitude (lV) 22.27 2.60 28.50–3.42 ASA24 malfunction. Age, sex, BMI, and high-calorie N2
difference amplitude were entered into three separate multi-
N2 go amplitude (lV) 21.69 2.35 27.23–5.03
ple linear regressions, one predicting carbohydrate intake,
N2 difference amplitude (lV) 20.46 0.74 23.06–1.43 one predicting protein intake, and one predicting fat intake
(see Table 7). Models were acceptable for assumptions of
No-go trial accuracy (%) 87.62 9.19 46.67–100
multicollinearity and homoscedasticity and normality of
Go trial accuracy (%) 97.70 4.88 47.86–100 residuals.
Correct go RTs (ms) 421.77 67.31 287.22–736.50 For carbohydrates, neither age (b 5 20.03, p 5 .74) nor
BMI (b 5 0.03, p 5 .73) significantly predicted intake. Sex
Low-calorie task
predicted carbohydrate intake (b 5 20.39, p < .001), with
N2 no-go amplitude (lV) 21.95 2.41 27.87–4.52 females consuming fewer carbohydrates than males. High-
calorie N2 difference amplitude also predicted carbohydrate
N2 go amplitude (lV) 22.19 2.51 28.48–3.12
intake (b 5 0.22, p 5 .003), where individuals with more
N2 difference amplitude (lV) 0.22 0.77 21.79–3.97 negative high-calorie N2 difference amplitudes (i.e., larger
No-go trial accuracy (%) 85.63 10.36 53.33–100 inhibitory response toward high-calorie foods) consumed
fewer carbohydrates (see Figure 4d).3 For protein, neither
Go trial accuracy (%) 97.31 3.74 71.43–100
age (b 5 0.06, p 5 .45), BMI (b 5 0.09, p 5 .43), nor high-
Correct go RTs (ms) 448.62 62.67 324.09–749.79 calorie N2 difference amplitude (b 5 0.09, p 5 .24) signifi-
cantly predicted intake. Sex predicted protein intake
Note. lV 5 microvolts; N2 difference amplitude 5 no-go amplitude minus go
amplitude. (b 5 20.42, p < .001), with females consuming less protein
than males. Finally, for fat intake, neither age (b 5 0.13,
larger inhibitory response toward high-calorie foods) con- p 5 .10), BMI (b 5 0.06, p 5 .41), nor high-calorie N2 dif-
sumed fewer calories (see Figure 4c).2 ference amplitude (b 5 0.12, p 5 .11) significantly predicted
Age, sex, BMI, and low-calorie N2 difference amplitude intake. Sex predicted fat intake (b 5 20.41, p < .001), with
(i.e., no-go minus go amplitude on the low-calorie task) were females consuming less fat than males. In short, a larger
entered into a separate multiple linear regression predicting inhibitory response toward high-calorie foods was related
caloric intake (see Table 6). The model met basic assump- only to decreased carbohydrate consumption.
tions for multicollinearity and was acceptable for homo-
scedasticity and normality of residuals. Age (b 5 0.06,
p 5 .42), BMI (b 5 0.10, p 5 .21), and low-calorie N2 differ-
ence amplitude (b 5 20.03, p 5 .67) did not predict caloric
3.3 | Exploratory questionnaire analyses
intake. Only sex predicted caloric intake in this model High-calorie N2 difference amplitude was not significantly
(b 5 20.46, p < .001), with females consuming fewer calo- correlated with the G-FCQ-S total score (r 5 .01, p 5 .87),
ries than males. DEBQ emotional eating subscale (r 5 .14, p 5 .08), or
DEBQ external eating subscale (r 5 .09, p 5 .22). Similarly,
low-calorie N2 difference amplitude was not significantly
3.2.1 | Macronutrient regressions
correlated with G-FCQ-S total score (r 5 2.01, p 5 .87),
As high-calorie N2 difference amplitude significantly pre- DEBQ emotional eating subscale (r 5 2.04, p 5 .65), or
dicted caloric intake, we conducted exploratory analyses at DEBQ external eating subscale (r 5 2.05, p 5 .53).

2 3
After examining scatter plots (see Figure 4c), one high-calorie N2 dif- After examining scatter plots (see Figure 4d), one high-calorie N2 dif-
ference value was identified as a potential outlier. Once removed, age ference value was identified as a potential outlier. Once removed, age
(b 5 0.04, p 5 .56) and BMI (b 5 0.09, p 5 .22) still did not signifi- (b 5 20.03, p 5 .73) and BMI (b 5 0.03, p 5 .72) still did not signifi-
cantly predict caloric intake. Sex still predicted caloric intake cantly predict carbohydrate intake. Sex still predicted carbohydrate
(b 5 20.47, p < .001), with females consuming fewer calories. High- intake (b 5 20.39, p < .001), with females consuming fewer carbohy-
calorie N2 difference amplitude still predicted caloric intake (b 5 0.16, drates. High-calorie N2 difference amplitude still predicted carbohydrate
p 5 .03), where individuals with more negative N2 difference amplitudes intake (b 5 0.20, p 5 .008), where individuals with more negative N2
consumed fewer calories. difference amplitudes consumed fewer carbohydrates.
10
| CARBINE ET AL.

TAB LE 4 Correlations between repeated measures residuals for N2 amplitude

High-calorie High-calorie Low-calorie Low-calorie

task: No-go trials task: Go trials task: No-go trials task: Go trials

High-calorie task: No-go trials 1.00 – – –

High-calorie task: Go trials .94 1.00 – –

Low-calorie task: No-go trials .85 .86 1.00 –

Low-calorie task: Go trials .89 .88 .94 1.00

Note. Dashes indicate no data available.

TAB LE 5 Correlations between repeated measures residuals for percent accuracy

High-calorie High-calorie Low-calorie Low-calorie

task: No-go trials task: Go trials task: No-go trials task: Go trials

High-calorie task: No-go trials 1.00 – – –

High-calorie task: Go trials .02 1.00 – –

Low-calorie task: No-go trials .47 2.07 1.00 –

Low-calorie task: Go trials .10 .04 2.13 1.00

Note. Dashes indicate no data available.

(a) (b)
5 5
High-Calorie N2 Difference

Low-Calorie N2 Difference

4 4
3
Amplitude (µm)

3
Amplitude (µm)

2 2
1
* 1
*
0 0
-1 -1
-2 -2
-3 -3
0 200 400 600 800 0 200 400 600 800
Reaction Time (ms) Reaction Time (ms)

(c) (d)
2 2
High-Calorie N2 Difference

High-Calorie N2 Difference

1 1
Amplitude (µm)

Amplitude (µm)

0 0

-1 -1

-2 -2

-3
* -3
*
-4 -4
0 1000 2000 3000 4000 5000 0 100 200 300 400 500 600
Calories (kcal) Carbohydrates (grams)
F I G U R E 4 Scatter plots showing the relationships between (a) high-calorie N2 difference amplitude and RTs, (b) low-calorie N2 difference amplitude
and RTs, (c) high-calorie N2 difference amplitude and caloric intake, (d) high-calorie N2 difference amplitude and carbohydrate intake. *Potential outlier
CARBINE ET AL.
| 11

TAB LE 6 Multiple linear regressions predicting average caloric intake

b t DR2 VIF F df Adj. R2 Cohen’s f2

High-calorie difference N2 amplitude model 14.69*** 4, 154 .28 .39

Age 0.04 0.59 .002 1.17
Sex 20.47 26.76*** .21 1.03
BMI 0.09 1.23 .01 1.18
High-calorie N2 difference amplitude 0.18 2.54** .03 1.02

Low-calorie difference N2 amplitude model 12.61*** 4, 154 .23 .30

Age 0.06 0.82 .003 1.16
Gender 20.46 26.44*** .20 1.03
BMI 0.10 1.27 .01 1.19
Low-calorie N2 difference amplitude 20.03 20.43 .001 1.02

Note. Dependent variable for both models was average caloric intake. VIF 5 variance inflation factor.
a
*p < .05. **p < .01. ***p < .001.

4 | DISCUSSION amplitude and increased accuracy during the high-calorie

We aimed to understand how neural indices of food-related
inhibitory control differed toward high- and low-calorie
food cues and if N2 ERP amplitude was related to daily
food intake. Our first hypothesis that there would be a
larger inhibitory response to high-calorie foods was sup-
ported, as we found multiple behavioral and ERP differen-
ces when inhibiting responses toward high- compared to
low-calorie foods. Compared to the low-calorie task, indi-
viduals had larger no-go N2 amplitudes and increased
accuracy on the high-calorie task (i.e., when inhibiting
responses toward high-calorie foods). The larger no-go N2
task supports the idea that accurately responding to high-
calorie foods necessitates increased recruitment of inhibi-
tory control processes, possibly due to the more palatable
and appetizing nature of high-calorie foods (Appelhans
et al., 2011; Hall, 2012).
Individuals also had longer RTs when responding to
high-calorie foods compared to low-calorie foods, and larger
high-calorie N2 difference amplitudes were related to shorter
RTs toward high-calorie food stimuli. Research has sug-
gested that longer RTs toward high- relative to low-calorie
foods indicates increased attention toward high-calorie foods
(Meule & Kubler, 2014). Our results suggest that the

TAB LE 7 Multiple linear regressions predicting average protein, fat, and carbohydrate intake

b t DR2 VIF F df Adj. R2 Cohen’s f2

Protein intake (g) 10.31*** 4, 153 .19 .23

Age 0.06 0.77 .003 1.18
Sex 20.42 25.76*** .17 1.03
BMI 0.09 1.17 .01 1.19
High-calorie N2 difference amplitude 0.09 1.18 .01 1.03

Fat intake (g) 11.28*** 4, 153 .21 .27

Age 0.13 1.68 .01 1.18
Sex 20.41 25.76*** .17 1.03
BMI 0.06 .83 .003 1.19
High-calorie N2 difference amplitude 0.12 1.60 .01 1.03

Carbohydrate intake (g) 9.16*** 4, 153 .17 .20

Age 20.03 20.34 .001 1.18
Sex 20.39 25.32*** .15 1.03
BMI 0.03 .35 .001 1.19
High-calorie N2 difference amplitude 0.22 3.00** .05 1.03

Note. VIF 5 variance inflation factor.

a
*p < .05. **p < .01. ***p < .001.
12
| CARBINE
increased recruitment of cognitive resources when process-
ing high-calorie foods not only aids individuals in being
more accurate, but may also help them manage their
increased attention toward high-calorie foods. The longer
RTs when responding to high-calorie foods but increased
accuracy when inhibiting toward high-calorie foods sug-
gest a speed-accuracy tradeoff (Heitz, 2014; van Maanen,
2016), where inhibiting toward high- compared to low-
calorie foods is more demanding, resulting in individuals
slowing down in order to be more accurate. However,
future research comparing high- and low-calorie go/no-go
tasks to a control go/no-go task is needed to further support
this conclusion.
There was also a Task 3 Trial interaction when examin-
ing N2 amplitude. Inhibiting responses toward high-calorie
stimuli on the high-calorie task elicited a larger (i.e., more
negative) no-go N2 amplitude compared to go trials, whereas
inhibiting responses toward low-calorie stimuli on the low-
calorie task resulted in a smaller no-go N2 amplitude com-
pared to go trials. Our study adds incremental validity to the
literature by comparing inhibitory control toward high- and
low-caloric stimuli as opposed to just examining high-calorie
foods (e.g., Batterink, Yokum, & Stice, 2010; Kakoschke,
Kemps, & Tiggemann, 2015; Price et al., 2016; Watson &
Garvey, 2013). The interaction suggests that inhibiting domi-
nant responses to high-calorie foods necessitates increased
recruitment of inhibitory control processes, but inhibiting
dominant responses to low-calorie foods does not elicit
increased recruitment of inhibitory control processes. In
other words, the more appetizing nature of high-calorie foods
may influence why the brain recruits more cognitive resour-
ces to withhold automatic responses from high-calorie foods
(Appelhans et al., 2011; Hall, 2012).
Our findings demonstrate that the brain may process
high- and low-calorie foods differently in terms of inhibi-
tory control, as just by switching high- and low-calorie
foods as no-go stimuli, neural responses during the two
tasks (which contained the same timings, same pictures,
and same ratio of go/no-go trials) changed. Future analyses
examining if specific palatability ratings or macronutrient
content of the pictures influence differences in inhibitory
control could also help clarify how the brain processes var-
ious food cues differently. Another possible interpretation
for our differences between high- and low-calorie N2
amplitudes is that the high-calorie task may be a more
valid measure of food-related inhibition, as the high- but
not low-calorie N2 difference amplitude predicted caloric
intake.
We are aware of only one additional study, an fMRI
study that examined inhibitory control differences between
high- and low-calorie food stimuli, but in an adolescent/
young adult sample. Similar to our results, He et al. (2014)
ET AL.
found a main effect of trial type, with greater inhibitory
control activity during no-go relative to go trials collapsed
across tasks. Our results indicated the main effect was
mainly driven by inhibiting toward high-calorie foods,
whereas He and colleagues found no differences between
inhibiting toward high- and low-calorie food stimuli. It is
possible that age and developmental differences between
the two samples may account for the discrepancies, as
frontal brain regions involved in inhibitory control proc-
esses are not fully developed in adolescents (Casey et al.,
1997; Zelazo & Carlson, 2012). Future research should
probably examine how developmental factors may influ-
ence neural processing and distinction between high- and
low-calorie food cues.
For our second hypothesis, we found that, when control-
ling for BMI, age, and sex, individuals with larger high-
calorie N2 difference amplitudes (i.e., no-go minus go trials
on the high-calorie task) consumed fewer calories. Results
suggest that high-calorie-specific inhibitory control, as
opposed to inhibitory control toward food in general, may
help regulate food intake (particularly for more appetizing
and high-calorie foods) and eating habits. Complementing
the behavioral findings and previous research (Cornier et al.,
2010; Lapenta et al., 2014), individuals who recruit the nec-
essary cognitive resources to inhibit responses to high-
calorie foods are not only at a task level more accurate, but
also more successful in day-to-day life at regulating con-
sumption of foods higher in calories. Our results also show
that experimental paradigms and technologies used in
research labs may be beneficial in examining food-related
inhibitory control, as neural measures can provide valid
information about real-world eating behavior and habits.
Exploratory analyses revealed that larger high-calorie N2
difference amplitude predicted lower carbohydrate intake,
but not protein or fat intake. Health interventions aimed not
only at managing urges to eat foods higher in calories, but
specifically foods higher in carbohydrates, may aid individu-
als in regulating their food intake and improving eating hab-
its. As Watson and Garvey (2013) found that inhibiting
responses to food cues relative to neutral pictures elicited a
larger N2 amplitude only in female participants, controlling
for sex in our model was essential as it showed that, while
sex does influence eating behaviors, targeting inhibitory con-
trol processes may still help regulate food intake and eating
habits, regardless of sex. It is also important to note that
measuring caloric and carbohydrate intake was how we
quantified eating habits and behaviors. We believe that inhib-
itory processes may be more involved in regulating eating
habits and what types of food people eat, not necessarily reg-
ulating the exact number of calories an individual consumes.
Given that this is the first study to show such a relationship
between food consumption and neural indices such as the
CARBINE ET AL.
N2, the results are considered tentative, and we recommend
future research to replicate the N2 amplitude and food
intake/eating behavior regressions.
Health interventions may benefit from utilizing food-
based inhibition training, in addition to diet and exercise,
to help facilitate weight loss and manage food consump-
tion. For example, individuals who complete tasks where
they continually inhibit responses toward specific food
stimuli will consume less of that food when it is presented
to them following the task (Allom, Mullan, & Hagger,
2016; Houben, 2011; Houben & Jansen, 2015; Jones et al.,
2016). We are unaware of any studies assessing the long-
term effects of such food-related inhibition training, or
how such inhibition training may alter neural processing of
food cues and subsequently food intake or eating habits. In
light of the relationship between neural indices of inhibi-
tion and food intake and the success of altering food intake
through the use of food-related inhibition training, we
believe that there is promise that such inhibition training
could affect food-related inhibitory control at the neural
level.
Exploratory analyses showed that neural indices of food-
related inhibitory control were not related to self-reports of
cravings, emotional eating, or external eating. As cravings
and emotional and external eating are believed to drive food-
related inhibition (Blundell & Gillett, 2001; Davis et al.,
2007; Guerrieri, Nederkoorn, Stankiewicz et al., 2007; Jasin-
ska et al., 2012), our results suggest self-reports and self-
perspectives (i.e., surveys) of one’s inhibitory control toward
food may not be accurate compared to more objective meas-
ures (i.e., N2 amplitude) of inhibitory control. Previous
research examining dietary intake (e.g., Kretsch, Fong, &
Green, 1999; Macdiarmid & Blundell, 1997; Nydahl, Gus-
tafsson, Mohsen, & Becker, 2009), and attention toward
food cues (Carbine et al., 2017) has also demonstrated simi-
lar discrepancies between objective and subjective measures.
As food-related inhibition, particularly for high-calorie foods,
may be an important factor in regulating eating habits, an
objective measure (e.g., ERPs) could help individuals
become accurately aware of how efficient they are at inhibi-
ting toward food.
Our study had several strengths, including using both
high- and low-calorie foods as no-go stimuli in order to
better understand food-related inhibitory control and con-
trolling for variables like sex, BMI, and chronic or meta-
bolic diseases that could influence the relationship between
N2 amplitude and food intake. However, the study also
had limitations. As we had a limited BMI range, we did
not examine how neural indices of inhibitory control
directly relate to BMI status, something future research
should examine. Also, future research utilizing more objec-
tive measures of food intake, such as weighed food buffets
| 13
or taste tests, could help replicate and confirm our findings.
Finally, individuals participating in the evening showed
different hunger levels than the individuals participating in
the morning; however, hunger levels did not affect N2
amplitude (refer to Method). We also did not manipulate
hunger, but had individuals come in after fasting in order
to model a time when participants were about to consume
their next meal. We hoped that this would provide greater
insight on daily food consumption and regular eating hab-
its. Furthermore, there was no potential confound across
participants of caloric intake before task participation. Spe-
cifically manipulating hunger levels or having participants
report to the lab during greater periods of hunger (i.e., after
longer fasting periods) could clarify how larger differences
in hunger may influence inhibitory control toward food.
Further examination of how inhibitory control relates to
unhealthy eating habits (e.g., eating in the absence of hun-
ger) would also be beneficial.
4.1 | Conclusion
Inhibiting responses to high-calorie foods elicited a larger
inhibitory response than when inhibiting toward low-calorie
foods, as evidenced by a larger no-go N2 amplitude.
Behavioral results suggest that the increased recruitment of
inhibitory control resources may help individuals be more
accurate and efficient when inhibiting toward high-calorie
foods. Individuals who recruited more cognitive resources
to inhibit toward high-calorie foods consumed fewer calo-
ries, and specifically carbohydrates. Future research should
examine how food-related inhibitory control training may
affect the relationship between neural indices of inhibitory
control, food intake, and eating habits. Finally, self-reports
of food-related inhibitory control may not directly relate to
N2 amplitude, further supporting the use of objective meas-
ures in future research when examining food-related inhibi-
tory control.
ACKNOWLEDGMENT
A Brigham Young University Mentored Environment
Grant and the Brigham Young University College of Fam-
ily, Home, and Social Sciences funded this research.
REF ER ENC ES
Allom, V., Mullan, B., & Hagger, M. (2016). Does inhibitory control
training improve health behaviour? A meta-analysis. Health Psychol-
ogy Review, 10(2), 1682186. doi:10.1080/17437199.2015.1051078
Appelhans, B. M., Woolf, K., Pagoto, S. L., Schneider, K. L.,
Whited, M. C., & Liebman, R. (2011). Inhibiting food reward:
Delay discounting, food reward sensitivity, and palatable food
intake sensitivity intake in overweight and obese women. Obesity,
19(11), 2175–2182. doi:10.1038/oby.2011.57
14
| CARBINE
Batterink, L., Yokum, S., & Stice, E. (2010). Body mass correlates
inversely with inhibitory control in response to food among ado-
lescent girls: An fMRI study. NeuroImage, 52(4), 1696–1703.
doi:10.1016/j.neuroimage.2010.05.059
Bazzano, L. A., He, J., Ogden, L. G., Loria, C. M., Vupputuri, S.,
Myers, L., & Whelton, P. K. (2002). Fruit and vegetable intale
and risk of cardiovascular disease in U.S. adults: The first
national health and nutrition examination survey epidemiologic
follow-up study. American Journal of Clinical Nutrition, 76(1),
93–99.
Bazzano, L. A., Hu, T., Reynolds, K., Yao, L., Bunol, C., Liu, Y., . . .
He, J. (2014). Effects of low-carbohydrate and low-fat diets. Annals
of Internal Medicine, 161, 309–318. doi:10.7326/M14-0180
Begg, D., & Woods, S. (2013). Endocrinology of food intake. Nature
Reviews Endocinology, 9(10), 584–597. doi:10.1038/nrendo.2013.136
Blundell, J., de Graaf, C., Hulshof, T., Jebb, S., Livingstone, B.,
Lluch, A., . . . Westerterp, M. (2010). Appetite control: Methodo-
logical aspects of the evaluation of foods. Obesity Review, 11(3),
251–70. doi:10.1111/j.1467-789X.2010.00714.x
Blundell, J. E., & Gillett, A. (2001). Control of food intake in the
obese. Obesity, 9(S11), 263S–270S. doi:10.1038/oby.2001.129
Carbine, K. A., Larson, M. J., Romney, L., Bailey, B. W., Tucker, L.
A., Christensen, W. F., & LeCheminant, J. D. (2017). Disparity
in neural and subjective responses to food images in women with
obesity and normal-weight women. Obesity, 25, 384–390. doi:
10.1002/oby.21710
Casey, B. J., Trainor, R. J., Orendi, J. L., Schubert, A. B., Nystrom,
L. E., Geidd, J. N., . . . Rapoport, J. L. (1997). A developmental
functional MRI study of prefrontal activation during performance
of a go-no-go task. Journal of Cognitive Neuroscience, 9(6), 835–
846. doi:10.1162/jocn.1997.9.6.835
Christensen, E. (2014). Neural responses to food pictures and their
association with dietary intake (Unpublished master thesis). Brig-
ham Young University, Provo, UT.
Clawson, A., Clayson, P. E., & Larson, M. J. (2013). Cognitive con-
trol adjustments and conflict adaptation in major depressive disor-
der. Psychophysiology, 50, 711–721. doi:10.1111/psyp.12066
Clayson, P. E., Baldwin, S. A., & Larson, M. J. (2013). How does
noise affect amplitude and latency measurement of event-
related potentials (ERPs)? A methodological critique and simu-
lation study. Psychophysiology, 50, 174–186. doi:10.1111/
psyp.12001
Clayson, P. E., Clawson, A., & Larson, M. J. (2011). Sex differen-
ces in electrophysiological indices of conflict monitoring. Bio-
logical Psychology, 87(2), 282–289. doi:10.1016/j.biopsycho.
2011.03.011
Clayson, P. E., & Larson, M. J. (2012). Cognitive performance and
electrophysiological indices of cognitive control: A validation
study of conflict adaptation. Psychophysiology, 49, 627–637. doi:
10.1111/j.1469-8986.2011.01345.x
Clayson, P. E., & Larson, M. J. (2013). Psychometric properties of
conflict monitoring and conflict adaptation indices: Response time
and conflict N2 event-realted potentials. Psychophysiology, 50(2),
1209–1219. doi:10.1111/psyp.12138
Cornier, M. C., Salzberg, A. K., Endly, D. C., Bessesen, D. H., &
Tregellas, J. R. (2010). Sex-based differences in the behavioral
ET AL.
and neuronal responses to food. Physiology & Behavior, 99(4),
538–543. doi:10.1016/j.physbeh.2010.01.008
Davis, C., Patte, K., Levitan, R., Reid, C., Tweed, S., & Curtis, C.
(2007). From motivation to behavior: A model of reward sensitiv-
ity, overeating, and food preferences in the risk profile for obesity.
Appetite, 48(1), 12–19. doi:10.1016/j.appet.2006.05.016
Dien, J. (2010). The ERP PCA toolkit: An open source program for
advanced statistical analysis of event-related potential data. Jour-
nal of Neuroscience Methods, 187(1), 138–145. doi:10.1016/j.
jneumeth.2009.12.009
Dien, J., Michelson, C. A., & Franklin, M. S. (2010). Separating the
visual sentence N400 effect from the P400 sequential expectancy
effect: Cognitive and neuroanatomical implications. Brain
Research, 1355, 126–140. doi:10.1016/j.brainres.2010.07.099
Folstein, J. R., & Van Petten, C. (2008). Influence of cognitive con-
trol and mismatch on the N2 component of the ERP: A review.
Psychophysiology, 45, 152–170. doi: 10.1111/j.1469-8986.2007.
00602.x
Gelman, A., & Loken, E. (2014). The statistical crisis in science.
American Scientist, 102(6), 460–465. doi:10.1511/2014.111.460
Green, S. B. (1991). How many subjects does it take to do a regres-
sion analysis? Multivariate Behavioral Research, 26, 499–510.
doi:10.1207/s15327906mbr2603_7
Guerrieri, R., Nederkoorn, C., & Jansen, A. (2007). How impul-
siveness and variety influence food intake in a sample of
healthy women. Appetite, 48, 119–122. doi:10.1016/j.appet.
2006.06.004
Guerrieri, R., Nederkoorn, C., Stankiewicz, K., Alberts, H., Gesch-
wind, N., Martijn, C., & Jansen, A. (2007). The influence of trait
and induced state impulsivity on food intake in normal-weight
healthy women. Appetite, 49(1), 66–73. doi:10.1016/j.appet.2006.
11.008
Hall, P. A. (2012). Executive control resources and frequency of fatty
food consumption: Findings from an age-stratified community sam-
ple. Healthy Psychology, 31(2), 235–241. doi:10.1037/a0025407
Hanlon, B., Larson, M. J., Bailey, B. W., & LeCheminant, J. D. (2012).
Neural response to pictures of food after exercise in normal-weight
and obese women. Medicine and Science in Sports and Exercise, 44
(10), 1864–1870. doi:10.1249/MSS.0b013e31825cade5
Harris, J. A., & Benedict, F. G. (1918). A biometric study of human
basal metabolism. Proceedings of the National Academy of Scien-
ces, 4(12), 370–373. doi:10.1073/pnas.4.12.370
He, Q., Xiao, L., Xue, G., Wong, A., Ames, S. L. Schembre, S. M.,
& Bechara, A. (2014). Poor ability to resist tempting caloric rich
food is linked to altered balance between neural systems involved
in urge and self-control. Nutrition Journal, 13(92). doi:10.1186/
1475-2891-13-92
Heitz, R. P. (2014). The speed-accuracy tradeoff: History, physiology,
methodology, and behavior. Frontiers in Neuroscience, 8, 150.
doi:10.3389/fnins.2014.00150
Herman, C. P., Roth, D. A., & Polivy, J. (2003). Effects of the presence
of others on food intake: A normative interpretation. Psychological
Bulletin, 129(6), 873–886. doi:10.1037/0033-2909.129.6.873
Houben, K. (2011). Overcoming the urge to splurge: Influencing eat-
ing behavior by manipulating inhibitory control. Journal of
CARBINE ET AL.
Behavior Therapy and Experimental Psychiatry, 42, 384–388.
doi:10.1016/j.jbtep.2011.02.008
Houben, K., & Jansen, A. (2015). Chocolate equals stop. Chocolate-
specific inhibition training reduces chocolate intake and go associ-
ations with chocolate. Appetite, 87, 3182323. doi:10.1016/j.
appet.2015.01.005
Jansen, A., Nederkoorn, C., van Baak, L., Keirse, C., Guerrieri, R., &
Havermans, R. (2009). High-restrained eaters only overeat when
they are also impulsive. Behavior Research and Therapy, 47(2),
105–110. doi:10.1016/j.brat.2008.10.016
Jasinska, A. J., Yasuda, M., Burant, C. F., Gregor, N., Khatri, S.,
Sweet, M. & Falk, E. B. (2012). Impulsivity and inhibitory con-
trol deficits are associated with unhealthy eating in young adults.
Appetite, 59(3), 738–747. doi:10.1016/j.appet.2012.08.001
Jones, A., Di Lemma, L. C. G., Robinson, E., Christiansen, P.,
Nolan, S., Tudur-Smith, C., & Field, M. (2016). Inhibitory con-
trol training for appetitive behaviour change: A meta-analytic
investigation of mechanisms of action and moderators of effec-
tiveness. Appetite, 97, 16–18. doi:10.1016/j.appet.2015.11.013
Kakoschke, N., Kemps, E., & Tiggemann, M. (2015). Combined
effects of cognitive bias for food cues and poor inhibitory control
on unhealthy food inake. Appetite, 97, 358–364. doi:10.1016/j.
appet.2015.01.004
Kastorini, C. M., & Panagiotakos, D. B. (2009). Dietary patterns and
prevention of Type 2 diabetes: From research to clinical practice;
A systematic review. Current Diabetes Reviews, 5(4), 221–227.
doi:10.2174/157339909789804341
Keogh, J. (2013). Nutrition and vascular health. Nutrition and Die-
tetics, 70, 3–4. doi:10.1111/1747-0080.12029
Killgore, W. D. S., Weber, M., Schwab, Z. J., Kipman, S. R., Del-
Donno, S. R., Webb, C. A., & Rauch, S. L. (2013). Cortico-lim-
bic responsiveness to high-calorie food images predicts weight
status among women. International Journal Obesity, 37, 1435–
1442. doi:10.1038/ijo.2013.26
Killgore, W. D., Young, A. D., Femia, L. A., Bogorodzki, P., Rogow-
ska, J., & Yurgelun-Todd, D. A. (2003). Cortical and limbic activa-
tion during viewing of high- verses low-calorie foods. NeuroImage,
19(4), 1381–1394. doi:10.1016/S1053-8119(03)00191-5
Killgore, W. D. S. & Yurgelun-Todd, D. A. (2005). Developmental
changes in the functional brain responses of adolescents to images
of high and low-caloire foods. Developmental Psychobiology, 47
(4), 377–397. doi:10.1002/dev.20099
Killgore, W. D. S. & Yurgelun-Todd, D. A. (2007). Positive affect
modulates activity in the visual cortex to images of high calorie
foods. International Journal of Neuroscience, 117, 643–653. doi:
10.1080/00207450600773848
Kirkpatrick, S. I., Subar, A. F., Douglass, D., Zimmerman, T. P.,
Thompson, F. E., Kahle, L. L., . . . Potischman, N. (2014). Per-
formance of the automated self-administered 24-hour recall rela-
tive to a measure of true intakes and to an interviewer-
administered 24-h recall. American Journal of Clinical Nutrition,
100(1), 233–240. doi:10.3945/ajcn.114.083238
Ko, Y., & Miller, J. (2013). Signal-related contributions to
stopping-interference effects in selective response inhibition.
Experimental Brain Research, 228, 205–212. doi:10.1007/
s00221-013-3552-y
| 15
Kretsch, M. J., Fong, A. K., & Green, M. W. (1999). Behavioral and
body size correlates of energy intake underreporting by obese and
normal-weight women. Journal of the American Dietetic Associa-
tion, 99(3), 300–306. doi:10.1016/S0002-8223(99)00078-4
Kumanyika, S., Jeffery R. W., Morabia, A., Ritenbaugh, C., & Antipa-
tis, V. J. (2002). Obesity prevention: The case for action. Interna-
tional Journal of Obesity, 26(3), 425–436. doi:10.1038/sj.
ijo.0801938
Lai, J. S., Hiles, S., Bisquera, A., Hure, A. J., McEvoy, M., & Atttia,
J. (2014). A systematic review and meta-analysis of dietary pat-
terns and depression in community-dwelling adults. American
Journal of Clinical Nutrition, 99(1), 181–197. doi:10.3945/
ajcn.113.069880
Lapenta, O. M., Sierve, K. D., de Macedo, E. C., Fregni, F., & Bog-
gio, P. S. (2014). Transcranial direct current stimulation modulated
ERP-indexed inhibitory control and reduces food consumption.
Appetite, 83, 42–48. doi:10.1016/j.appet.2014.08.005
Larson, M. J., & Carbine, K. A. (2017). Sample size calculations in
human electrophysiology (EEG and ERP) studies: A systematic
review and recommendations for increased rigor. International
Journal of Psychophysiology, 111, 33–41. doi:10.1016/j.
ijpsycho.2016.06.015
Larson, M. J., Farrer, T. J., & Clayson, P. E. (2011). Cognitive con-
trol in mild traumatic brain injury: Conflict monitoring and con-
flict adaptation. International Journal of Psychophysiology, 82,
69–78. doi:10.1016/j.ijpsycho.2011.02.018
Leue, A., Klein, C., Lange, S., & Beauducel, A. (2013). Inter-individ-
ual and intra-individual variability of the N2 component: On reli-
ability and signal-to-noise ratio. Brain and Cognition, 83(1), 61–
71. doi:10.1016/j.bandc.2013.06.009
Luck, S. J. (2005). An introduction to the event-related potential tech-
nique. Cambridge, MA: Massachusetts Institute of Technology.
Macdiarmid, J. I., & Blundell, J. E. (1997). Dietary under-reporting: What
people say about recording their food intake. European Journal of
Clinical Nutrition, 51(3), 199–200. doi:10.1038/sj.ejcn.1600380
Meule, A., & Kubler, A. (2014). Double trouble. Trait food craving and
impulsivity interactively predict food-cue affected behavioral inhi-
bition. Appetite, 79, 174–182. doi:10.1016/j.appet.2014.04.014
Moshfegh, A. J., Rhodes, D. G., Baer, D. J., Murayi, T., Clemens, J.
C., Rumpler, W. V., . . . Cleveland, L. E. (2008). The US Depart-
ment of Agriculture automated multiple-pass method reduces bias
in the collection of energy intakes. American Journal of Clinical
Nutrition, 88(2), 324–332.
National Cancer Institute. (2014). Automated Self-Administered 24-Hour
Recall (ASA24)22014. Bethesda, MD: National Cancer Institute.
Nijs, I. M., Franken, I. H., & Muris, P. (2007). The modified Trait
and State Food-Cravings Questionnaires: Development and vali-
dation of a general index of food craving. Appetite, 49(1),
38246. doi:10.1016/j.appet.2006.11.001
Nydahl, M., Gustafsson, I. B., Mohsen, R., & Becker, W. (2009).
Comparison between optical readable and open-ended weighed
food records. Food & Nutrition Research, 53. doi:10.3402/fnr.
v53i0.1889
Painter, J. E., Wansink, B., & Hieggelk, J. B. (2002). How visibility
and convenience influence candy consumption. Appetite, 38, 237–
238. doi:10.1006/appe.2002.0485
16
| CARBINE
Price, M., Lee, M., & Higgs, S. (2016). Food-specific response inhibi-
tion, dietary restraint and snack intake in lean and overweight/obese
adults: A moderated-mediation model. International Journal of
Obesity, 40(5), 877–882. doi:10.1038/ijo.2015.235
Rahati, S., Shahraki, M., Arjomand, G., & Shahraki, T. (2014). Food
pattern, lifestyle, and diabetes mellitus. International Journal of
High Risk Behaviors and Addiction, 3(1), e8725. doi:10.5812/
ijhrba.8725
Rietdijk, W. J. R., Franken, I. H. A., & Thurik, A. R. (2014). Internal
consistence of event-related potentials associated with cognitive
control: N2/P3 and ERN/Pe. PLOS ONE, 9(7), e102672. doi:
10.1371/journal.pone.0102672
Sabounchi, N. S., Rahmandad, H., & Ammerman, A. (2013). Best-fit-
ting prediction equations for basal metabolic rate: Informing obesity
interventions in diverse populations. International Journal of Obe-
sity, 37(10), 1364–1370. doi:10.1038/ijo.2012.218.
Schimmel, H. (1967). The (1/-) reference: Accuracy of estimated
mean components in average response studies. Science, 157, 92–
94. doi:10.1126/science.157.3784.92
Smith, J. A., & Ditschun, T. L. (2009). Controlling satiety: How envi-
ronmental factors influence food intake. Trends in Food Science &
Technology, 20(6–7), 271–277. doi:10.1016/j.tifs.2009.03.009
St-Onge, M. P., Wolfe, S., Sy, M., Shechter, A., & Hirsch, J. (2014).
Sleep restriction increases the neuronal response to unhealthy
food in normal-weight individuals. International Journal of Obe-
sity, 38(3), 411–416. doi:10.1038/ijo.2013.114.
Stratton, R. J., Stubbs, R. J., Hughes, D., King, N., Blundell, J. E., &
Elia, M. (1998). Comparision of the traditional paper visual ana-
log scale questionnaire with an Apple Newton electronic appetite
rating systeam (EARS) in free living subjects feeding ab libitum.
European Journal of Clinical Nutition, 52, 737–741. doi:10.1038/
sj.ejcn.1600636
Stroebele, N., & de Castro, J. M. (2006). Listening to music while eat-
ing is related to increases in people’s food intake and meal duration.
Appetite, 47(3), 285–289. doi:10.1016/j.appet.2006.04.001
Stubbs, R. J., Hughes, D. A., Johnstone, A. M., Rowley, E., Reid, C.,
Elia, M., . . . Blundell, J. E. (2000). The use of visual analogue scales
to assess motivation to eat in human subjects: A review of their reli-
ability and validity with an evaluation of new hand-held computer-
ized systems for temporal tracking of appetite ratings. British Journal
of Nutrition, 84(4), 405–415. doi:10.1017/S0007114500001719
Tucker, L. A., & Bates, L. (2009). Restrained eating and risk of gain-
ing weight and body fat in middle-aged women: A 3-year pro-
spective study. American Journal of Health Promotion. 23(3),
187–194. doi:10.4278/ajhp.07061456
Vainik, U., Dagher A., Dube, L., & Fellows, L. K. (2013). Neurobeha-
vioural correlates of body mass index and eating behaviours in
ET AL.
adults: A systematic review. Neuroscience and Biobehavioral
Reviews, 37, 279–299. doi:10.1016/j.neubiorev.2012.11.008
van Maanen, L. (2016). Is there evidence for a mixture of processes
in speed-accuracy trade-off behavior? Topics in Cognitive Sci-
ence, 8, 279–290. doi:10.1111/tops.12182
van Strien, T., Frijters, J. E. R., Bergers, G. P. A., & Defares, P. B.
(1986). The Dutch eating behavior questionnaire (DEBQ) for assess-
ment of restrained, emotional, and external eating behavior. Interna-
tional Journal of Eating Disorders, 5(2), 295–315. doi:0.1002/1098-
108X(198602)5:2 < 295::AID-EAT2260050209 > 3.0.CO;2-T
Vartanian, L. R., Spanos, S., Herman, C. P., & Polivy, J. (2015). Mod-
eling of food intake: A meta-analytic review. Social Influence, 10
(3), 119–136. doi:10.1080/15534510.2015.1008037
Walker, J. (2013). Reducing cardiovascular disease risk: Cholesterol and
diet. Nursing Standard, 28(2), 48–55. doi:10.7748/ns2013.09.28.2.48.
e7747
Wardle, J. (1987). Eating style: A validation study of the Dutch eat-
ing behavior questionnaire in normal subjects and women with
eating disorders. Journal of Psychosomatic Research, 31(2), 161–
169. doi:10.1016/0022-3999(87)90072-9
Watson, T. D., & Garvey, K. T. (2013). Neurocognitive correlates of
processing food-related stimuli in a go/no-go paradigm. Appetite,
71, 40–47. doi:10.1016/j.appet.2013.07.007
White, M. (2007). Food access e-059 and obesity. Obesity Reviews,
8(1), 99–107.
World Health Assembly. (2004). Global strategy on diet and physical activ-
ity (WHA57.17). Geneva, Switzerland: World Health Organization.
World Heath Organization. (2009). Global health risks: Mortality and
burden of disease attribute to selected major risks. Geneva, Swit-
zerland: World Health Organization.
World Health Organization (2013). Cardiovascular diseases (CVDs) (Fact
Sheet No. 317). Geneva, Switzerland: World Health Organization.
Zelazo, P. D., & Carlson, S. M. (2012). Hot and cool executive func-
tion in childhood and adolescence: Development and plasticity.
Child Development Perspectives, 6(4), 354–360.
How to cite this article: Carbine KA, Christensen E,
LeCheminant JD, Bailey BW, Tucker LA, Larson MJ.
Testing food-related inhibitory control to high- and low-
calorie food stimuli: Electrophysiological responses to
high-calorie food stimuli predict calorie and carbohy-
drate intake. Psychophysiology. 2017;00:1–16. https://
doi.org/10.1111/psyp.12860