ORIGINAL RESEARCH ARTICLE

published: 22 May 2014

BEHAVIORAL NEUROSCIENCE doi: 10.3389/fnbeh.2014.00184

Sweet lies: neural, visual, and behavioral measures reveal

a lack of self-control conflict during food choice in
weight-concerned women
Laura N. van der Laan 1*, Denise T. D. de Ridder 2 , Lisette Charbonnier 1 , Max A. Viergever 1 and
Paul A. M. Smeets 1,3
1
Image Sciences Institute, University Medical Center Utrecht, Utrecht, Netherlands
2
Department of Clinical and Health Psychology, Utrecht University, Utrecht, Netherlands
3
Division of Human Nutrition, Wageningen University and Research Centre, Wageningen, Netherlands

Edited by: Despite their intentions, weight-concerned individuals generally fail to control their eating
Jean-Claude Dreher, Centre National behavior. However, it is unknown whether this failure is due to a lack of effortful
de la Recherche Scientifique, France
self-control, or to not experiencing an internal conflict between weight goals and food
Reviewed by:
temptations. The present study used fMRI, eye tracking and reaction times to assess
Frauke Nees, Central Institute of
Mental Health, Germany the degree of conflict experienced by weight-concerned women during food choices
Cline Cruciani-Guglielmacci, that posed either a self-control dilemma (i.e., requiring a choice between healthy and
University Paris Diderot, France palatable foods), or not. Contrary to the common assumption in self-control theory that
*Correspondence: food choices posing a self-control dilemma evoke internal conflict, we found that choices
Laura N. van der Laan, Image
requiring self-control induced no conflict, as demonstrated by lower reaction times,
Sciences Institute, University
Medical Center Utrecht, fixation durations, number of gaze switches between snacks, and lower activation of the
Heidelberglaan 100, 3584 CX anterior cingulate cortex. Our results suggest that self-control failure might be due to a lack
Utrecht, Netherlands of experienced conflict, rather than to failing to act upon the perception of such conflict.
e-mail: Nynke@isi.uu.nl
This implies that effectiveness of weight maintenance interventions might be improved if
they also focus on increasing the ability to detect a self-control dilemma, in addition to the
current focus on increasing self-regulatory capacity.

Keywords: decision-making, food choice, visual attention, conflict monitoring, fMRI

INTRODUCTION At the heart of self-regulation theory lies the assumption

Approximately 5060% of the Western female population guard that if someone with the long-term goal to limit intake is
themselves against the risks of the obesogenic environment by faced with a food choice that threatens the accomplishment of
attempting to limit their energy intake (Rideout and Barr, 2009; this goal, this results in the experience of an internal conflict
Fayet et al., 2012; de Ridder et al., in press). However, despite their (Fishbach et al., 2003). Consequently, self-regulatory failure is
intentions, weight-concerned individuals generally fail to control usually attributed to the subsequent inability to resolve the con-
their food intake: they do not eat less than non-weight-concerned flict in favor of the long term goal. This has led to a large
counterparts (Stice et al., 2004, 2007, 2010; de Witt Huberts et al., number of studies aimed at improving individuals ability to
2013) and their ratings of weight concerns even predict future overcome the self-control dilemma [e.g., by cognitive strategies,
weight gain (French et al., 1994; Mann et al., 2007). (Siep et al., 2012; Giuliani et al., 2013)]. However, the assump-
The absence of a relation between self-reports of weight- tion that a self-control dilemma evokes internal conflict has so
consciousness and food intake indicates that high scores on far been untested and therefore it remains unknown whether
restraint and weight-concerns are a measure of intention rather self-regulatory failure is due to a lack of cognitive control, or
than a predictor of behavior (e.g., Stice et al., 2010). For to the absence of experienced conflict. This topic is of major
weight-concerned individuals, choosing between healthy and relevance for interventions aimed at weight loss or mainte-
(more tasty) unhealthy foods is regarded as a classic self- nance since it will elucidate whether interventions should aim
control dilemma involving the trade-off between immediate eat- only at strengthening self-regulatory capacity and cognitive con-
ing enjoyment and the future benefits of being slim and healthy trol or also at strengthening the ability to detect a self-control
(Fishbach et al., 2003). In order to resolve this dilemma and conflict.
behave in line with their intentions, self-control should be exer- Conflict monitoring is the process by which the brain deter-
cised (Baumeister and Heatherton, 1996; Fishbach et al., 2003). mines when control is required (Botvinick et al., 2001) and thus
Examples of such self-control strategies are bolstering the value precedes the actual act of self-control (exerting cognitive con-
of the long-term goal or using other cognitive strategies aimed at trol). Although there is a large body of research on self-control
keeping the long-term goal in the focus of attention (e.g., Metcalfe and the neural correlates of self-control in food choice (Hare
and Mischel, 1999; Peake et al., 2002). et al., 2009, 2011; Heatherton and Wagner, 2011; de Ridder et al.,

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 1

van der Laan et al. Neural correlates of food choice

2012), studies specifically assessing conflict monitoring during endocrine disease, or any gastro-intestinal disorder. Smokers and
food choice in weight-concerned women are lacking. individuals having a current alcohol consumption of >28 units
In response to this, the study aim was to investigate the lev- per week were excluded because these factors have been shown
els of conflict experienced by weight-concerned women during to affect the neural response to rewarding stimuli. We excluded
food choices that posed either a self-control dilemma or not. women that followed a medically prescribed diet in the past 6
As the perception of conflict is not necessarily a conscious phe- months or that had weight fluctuations of more than five kg in the
nomenon, we used measures that assess implicit cognitive (reac- past six months so as to exclude participants who may show biases
tion times during food choice, reaction times in a lexical decision in their food choices for medical reasons and as extra caution to
task), attentional (eye tracking) and neural processes [functional exclude individuals with eating disorders or problematic eating
Magnetic Resonance Imaging (fMRI)]. Previous studies have behavior, who might have aberrant brain responses to food and
demonstrated that response conflict in a task (i.e., task difficulty) thereby introduce heterogeneity in the sample. Participants were
is accompanied by higher reaction times (e.g., Panayiotou and recruited with posters at the University Medical Center Utrecht
Vrana, 2004), more gaze switches between items (e.g., Causse and the adjacent university campus.
et al., 2011), and stronger activation of the anterior cingulate
cortex (Botvinick et al., 2001; van Veen et al., 2001). Implicit mea- PROCEDURES
sures are less susceptible to demand characteristics and socially The study consisted of two sessions (on two separate days with
desirable responding (Fazio and Olson, 2003). This is especially 18 days in between). During the first session, participants com-
relevant for weight-concerned women who might be inclined to pleted several questionnaires and computer tasks, including a
respond in line with their intention to limit their food intake primed lexical decision task (see Lexical decision task). Moreover,
rather than their actual behavior. participants evaluated the expected tastiness and perceived energy
content of the food stimuli on 9-point scales ranging from 1 =
METHODS very untasty/very few calories to 9 = very tasty/very many calo-
ETHICS STATEMENT ries. No imaging data was collected during this first session.
The study was approved by the Medical Ethical Committee of In addition to this, the complete test-battery of question-
the University Medical Center Utrecht and participants provided naires and tasks included: the Behavioral Avoidance/Inhibition
written informed consent. (BIS/BAS) Scales (Carver and White, 1994), the Health and Taste
Attitudes Scales (Roininen et al., 2001), the Barrats Impulsiveness
PARTICIPANTS Scale (Patton et al., 1995), the Brief Self-Control Scale (Tangney
The study comprised of 20 women as participants (age in et al., 2004), the N-Back task (Kirchner, 1958), the traditional
years: M = 21.2, SD = 2.8; BMI in kg/m2 : M = 21.3, SD = 1.7). STROOP task (Stroop, 1935), and the Kirby Delay Discounting
Participant selection was limited to women because they generally task (Kirby et al., 1999). Results from these measures are not
score higher on weight concerns and because of known gender reported here.
differences as well in reasons for dieting as in brain anatomy and To make sure participants were craving for a snack, they were
function (Pingitore et al., 1997; Neumark-Sztainer et al., 1999; instructed to refrain from eating and drinking (except water)
Cahill, 2006; Luders et al., 2009). Earlier studies have shown for at least 2 h prior to the second session but to have prefer-
that there are gender differences in brain responses to food cues ably eaten a meal within 23 h before the session (mean time
(Smeets et al., 2006; Cornier et al., 2010; Frank et al., 2010; Haase since last food intake in minutes SD: 140 22). Upon arrival,
et al., 2011). As self-control conflict is only relevant for indi- participants received instructions and rated hunger on a VAS
viduals who are weight-concerned, inclusion criteria consisted scale ranging from 0 (not hungry) to 100 (very hungry) (mean
of a restraint-score above average or high (Dutch Eating behav- hunger SD: 59 12). Subsequently, to ensure the relevance
ior questionnaire Van Strien et al., 1986) and a rating of 6 or of their weight-concerns participants filled out a questionnaire
higher on each of two questions: To what extent are you weight- about a pro-claimed new type of snack-biscuit (questions about
concerned? and To what extent are you occupied with being expected tastiness, expected energy content and to what extent
slim? (ranging from 1 = not at all to 9 = very much; adapted eating the biscuit is appropriate for individuals who are watching
from Fishbach et al., 2003). Additional inclusion criteria were their weight). Thereafter, participants were scanned with a 3 Tesla
having an age between 18 and 30 years and having a normal Philips Achieva scanner (Philips Healthcare, Best, Netherlands),
weight (BMI between 18.5 and 25 kg/m2 ). while performing a food choice fMRI task. During this task, eye
Being left-handed was an exclusion criterion because the movements were recorded using an MR-compatible eye-tracker
brains of left and right-handed individuals are generally thought (Nordic Neurolabs, 60 Hz). Besides the food choice task reported
to differ in anatomical and functional characteristics (Guadalupe here, participants also performed another, unrelated, food choice
et al., 2014; Ooki, 2014). By only including right-handed indi- task. Finally, participants received one of the snacks they chose in
viduals (right-handed because the majority of the population the food choice task, were thanked and reimbursed.
is right-handed), we aimed to reduce the possible variation in
functional responses in our study population. Further exclusion LEXICAL DECISION TASK
criteria consisted of having a food allergy, having an eating dis- Since temptation-goal associations have been shown to mediate
order, and having a history of medical or surgical events that successful self-control, a primed lexical decision task (Figure 1)
might significantly affect the study outcome, such as metabolic or with temptation/neutral primes and diet/neutral target words

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 2

van der Laan et al. Neural correlates of food choice

FIGURE 1 | Task structure of the primed lexical decision task measuring temptation-goal associations. Smullen is a temptation-prime (Dutch for
feasting). Slank is a diet-target (Dutch for slim).

was used to measure the strength of temptation-goal associations

(adapted from Kroese et al., 2011).
The lexical decision task consisted of 144 trials comprising a
fixation cross (1000 ms), a forward mask (XXXXXX; 500 ms),
a prime word (50 ms), a backward mask (XXXXXX; 500 ms),
and a target letter string. The target letter string stayed on the
screen until participants pressed the z or m button of the key-
board to indicate that the target was a word or a non-word, FIGURE 2 | fMRI food choice task structure.
respectively. Half of the 144 targets were non-word targets (12
non-words, each repeated 6 times), 54 were neutral targets (9 neu-
tral words, e.g., bell, purple, finding, each repeated six times), and required (SC) trials], posing a self-control dilemma concerning
18 were diet targets (3 diet related words, i.e., dieting, slim, thin, the trade-off between immediate eating enjoyment (choosing the
each repeated six times). Three temptation primes were used [i.e., appealing HE snack) and weight watching intentions (choosing
chocolate, feasting (smullen) and eating candies (snoepen), the less appealing LE snack). In the other half of the trials, the
each repeated 6 times] and 21 neutral primes (e.g., letter, contact, LE and HE snacks were matched on tastiness [equal or 1 point
normal, each repeated six times). The temptation and neutral in tastiness rating; No Self-Control required (NSC) trials], such
primes were matched on word length, as were the diet-related and that no trade-off between eating enjoyment and weight watching
neutral targets. intentions was needed to choose the long-term superior LE snack.
The primes and targets were combined such that four trial cat- Participants were not told explicitly that the choices were always
egories of interest were created: (1) a neutral prime with a diet between pairs of high and low energy snacks.
target (nine trials), (2) a temptation prime with a neutral tar- To avoid that participants had to choose between products
get (nine trials), (3) a temptation prime with a diet target (nine that they highly disliked, only stimuli with a tastiness rating of
trials), and (4) a neutral prime with a neutral target (45 trials). four or higher were used. For two participants, the tastiness rat-
For responses in the lexical decision task, (log) reaction times ings allowed the construction of only 40 SC trials (instead of 50),
were aggregated over participants, prime (neutral or tempta- without repeating stimuli more than 10 times. Successful SC tri-
tion) and target type (diet target or neutral target). A two-level als were defined as those in which the LE snack was chosen and
(trial types nested within participants) regression model was per- unsuccessful SC trials as those in which the HE snack was chosen.
formed with prime type, target type and the effect of interest: the
interaction between prime and target. EYE TRACKING DATA ANALYSIS
Fixation detection was established by marking fixations with an
fMRI FOOD CHOICE TASK adaptive velocity threshold method (Arrington Research) with
In the food choice task (Figure 2), participants made a total of the default lower cut-off for fixation duration of 75 ms. In order to
100 choices. In every trial, a high energy (HE) (energy content analyze fixation behavior, the screen was divided in three regions,
in kcal/100 gram: M = 419, SD = 103) and a low energy (LE) namely: left product, the left 42% of the screen; right product, the
(M = 56, SD = 37) snack were shown side by side. Participants right 42% of the screen and middle, the middle 16% of the screen.
had 3000 ms to indicate which of the two products they would A fixation was considered last fixation if it was the final fixation
most like to eat a portion of by pushing the left or right but- on a product preceding the button press indicating the partici-
ton of a button box. After indicating their choice, a yellow box pants choice. Total fixation duration was defined as the sum of
appeared around the chosen product for 500 ms. The trials were the durations of all fixations on a product in a trial.
interspersed with a random interval between 2000 and 5000 ms. For four participants no stable eye tracking data could be
To investigate response conflict during food choice partici- acquired. These participants were removed from the analyses
pants were required to choose between pairs of HE and LE snacks concerning eye tracking data.
matched such (on the basis of their own tastiness ratings given
in the first session) that either a self-control dilemma was posed BEHAVIORAL DATA ANALYSIS
or not. In half of the trials, LE snacks were combined with HE In the food choice task stimuli were nested within trials, and trials
snacks rated two or three points higher on tastiness [Self-Control were nested within participants. Therefore, a series of multi-level

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 3

van der Laan et al. Neural correlates of food choice

regression analyses was performed to investigate how visual mea- in the WFU-pickatlas toolbox (Maldjian et al., 2003). To deter-
sures relate to choice. For outcomes on the stimulus level (the mine whether the anterior cingulate shows differential activation
stimulus being chosen or not, total fixation duration on HE/LE for SC and NSC trials, the contrast images were entered into a
snacks and last fixation) three-level regression analyses were per- one-sample t-test. For the anterior cingulate ROIs a statistical
formed, and for outcomes on the trial level (reaction times and threshold of p < 0.05 Family Wise Error (FWE)-corrected over
total fixation duration for HE and LE snack summed) two-level the ROI volume (i.e., small volume correction) was used.
regression analyses were performed. For continuous outcome For completeness and to enable future meta-analysis, we also
variables (fixation duration and reaction times) linear regres- report whole brain results at a statistical threshold of p < 0.005
sion analyses were performed. For binary outcome variables (the uncorrected and a cluster extent k 10 voxels, in Supplementary
stimulus being chosen or not) logistic regression analyses were Table S1.
performed.
To correct for a non-normal distribution, natural log- RESULTS
transformed reaction times were used in all analyses. Extreme CHOICE BEHAVIOR
reaction times [>3 SD from the (log- transformed) mean] were In the NSC condition, 45.9% HE choices were made and 54.1%
set to missing. were LE choices. In the SC condition, 78.9% were HE choices and
The statistical program R (packages lme4 and languageR) was 21.1% were LE choices.
used to perform multi-level regression analyses.
REACTION TIME DURING FOOD CHOICE
fMRI DATA Mean reaction time during the food choice task was 1412 ms
Image acquisition and preprocessing (SD = 452 ms). Regression analysis showed that reaction time
MRI scanning was performed on a 3 Tesla scanner (Philips was significantly higher for NSC (M = 1462 ms, SD = 460 ms)
Achieva, Philips Health- care, Best, Netherlands), equipped with compared to SC trials (M = 1363 ms, SD = 439 ms, p < 0.01,
a SENSE head coil. A T1 -weighted structural image was acquired Figure 3A, Table S2). The regression analysis revealed that in the
at a resolution of 1 1 1 mm (TR = 8.4 ms, TE = 3.8 ms, total successful SC trials reaction times (M = 1430 ms, SD = 468 ms)
scan duration = 284 s). Functional scans were acquired with a were significantly higher compared to unsuccessful SC trials
2D-EPI sequence (TE = 23 ms, flip angle = 70 , nr slices = 30, (M = 1345 ms, SD = 429 ms, p < 0.01, Figure 4A).
voxel size = 4 4 4 mm, acquisition time of one 3D volume =
TOTAL FIXATION DURATION AND NUMBER OF FIXATIONS
1400 ms). The total number of volumes acquired differed between
The regression analysis showed that total fixation duration (sum
participants because of the random inter-trial interval. Data
of fixation durations on left and right product summed) was
were preprocessed and analyzed using the SPM8 software pack-
higher in the NSC (M = 973 ms, SD = 450 ms) compared to the
age (Wellcome Department of Imaging Neuroscience, London,
SC trials (M = 873 ms, SD = 437 ms, p < 0.01, Figure 3B, Table
United Kingdom) ran with MATLAB R2012A (The Mathworks
S3). The same pattern was found for the number of fixations
Inc, Natick, MA). Functional images were realigned to the first
(Figure 3C).
image of the time series. Functional and structural images were
A regression analysis explaining fixation duration in SC tri-
co-registered and normalized (retaining 4 4 4 mm voxels) to
als with factors energy content of product fixated on (HE or LE),
MNI space (Evans et al., 1993) by using linear and non-linear
and success (i.e., successful if LE is chosen and unsuccessful if HE
transformations.
is chosen), and an interaction term of energy content success
revealed a significant interaction (p < 0.01) between energy con-
Participant level analysis
tent and success. That is, in successful SC trials LE stimuli were
Statistical maps were generated for each participant by fitting a fixated on longer than in unsuccessful SC trials. This interaction
boxcar function to the time series, convolved with the canoni- effect is plotted in Figure 4C. The same pattern was found for the
cal hemodynamic response function. Data were high-pass filtered number of fixations (Figure 4D).
with a cutoff of 128 s. Two conditions were modeled: the choice
periods of the SC trials and the choice periods of the NSC tri- NUMBER OF SWITCHES BETWEEN ITEMS
als. For each participant, two contrast images were calculated: (1) The regression analysis showed that the number of gaze switches
to establish the brain regions that responded more strongly dur- between snacks was higher in the NSC (M = 1.4, SD = 0.9) than
ing SC food choices, we performed a mean subtraction analysis the SC condition (M = 1.3, SD = 0.9, p < 0.05, Figure 3D). In
between SC and NSC trials, (2) to establish the brain regions that successful SC trials (M = 1.5, SD = 1.0) participants switched
respond stronger during NSC food choices, we performed a mean their gaze significantly more often between snacks than in unsuc-
subtraction analysis between NSC and SC trials. cessful SC trials (M = 1.3, SD = 0.9, p < 0.01, Figure 4B).

Group level analysis LOCATION OF LAST FIXATION

The anterior cingulate is regarded as a primary region in con- The last fixation was on a HE stimulus in 50.3% of the trials.
flict monitoring (Botvinick et al., 2001; van Veen et al., 2001). To investigate whether visual patterns differed between success-
Therefore the left and right anterior cingulate were our a priori ful and unsuccessful SC trials, we conducted a regression analysis
Regions Of Interest (ROIs). The ROI masks were generated using explaining last fixation location in SC trials with factors energy
the AAL-atlas (Tzourio-Mazoyer et al., 2002) as implemented content of product fixated on last (HE or LE), and success (i.e.,

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 4

van der Laan et al. Neural correlates of food choice

FIGURE 3 | Mean reaction times (A), total fixation duration on HE and LE snack summed (B), number of fixations on HE and LE snack summed (C)
and number of switches between HE and LE snacks (D), for SC and NSC trials. Barplots show mean s.e.m. p < 0.05.

successful if LE is chosen and unsuccessful if HE is chosen) and ROI volume, Z = 3.51, peak coordinate MNI (10, 28, 26)] was
an interaction term of energy content success. This analysis stronger activated during NSC compared to SC trials (Figure 5).
revealed a significant interaction (p < 0.01) between energy con- Results from whole-brain analysis are reported in Table S1.
tent of last product fixated on and success. That is, in successful
SC trials, the last fixation was in 70.2% of the trials on the LE DISCUSSION
snack (29.8% on HE snack); in unsuccessful SC trials, the last Our aim was to investigate levels of experienced conflict during
fixation was in 40.5% of the trials on the LE snack (59.5% on HE). food choices made by weight concerned women. In accordance
with previous studies (Stice et al., 2004, 2007, 2010), we found
TEMPTATION-GOAL ASSOCIATIONS that participants were generally unsuccessful in choosing in line
Reaction times in the lexical decision task were used to establish with their long term weight-watching goal: the LE snack was cho-
temptation-goal associations. If the prime activates the weight- sen in only 21.2% of the trials posing a self-control dilemma
watching goal, the reaction to diet targets preceded by a temp- (choosing between a LE and a more tasty HE snack, SC trials) and
tation prime would be quicker than the reaction to diet targets in 54.1% of the trials that did not pose a self-control dilemma
preceded by a neutral prime. Therefore, we conducted a linear (choosing between equally liked HE and LE snacks, NSC tri-
regression (trial type nested within participants) on reaction time als). Our results showed that during the choices that posed no
with target (diet or neutral), prime (temptation or neutral) and self-control dilemma the reaction times, fixation duration and
the interaction term of target prime. This analysis showed that number of gaze switches were higher compared to trials that did
neither the effect of interest (the interaction between target and pose a self-control dilemma. It is well established that reaction
prime; p = 0.83), nor the main effects of target (p = 0.64) and times, fixation durations and number of gaze switches are higher
prime (p = 0.63) were significant (Table S4). in difficult tasks requiring the recruitment of cognitive resources
(Panayiotou and Vrana, 2004; Causse et al., 2011). Furthermore, a
fMRI RESULTS SC vs. NSC TRIALS cluster in the anterior cingulate cortex, which has previously been
The subtraction analysis of NSC vs. SC trials revealed that a shown to activate during the perception of conflict (Botvinick
cluster in the left anterior cingulate [p = 0.017 FWE-corrected for et al., 2001; van Veen et al., 2001), was more strongly activated

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 5

van der Laan et al. Neural correlates of food choice

FIGURE 4 | Mean reaction times (A) and number of gaze switches between snacks (B) in successful and unsuccessful SC trials. Barplots show mean
s.e.m. Mean total fixation duration (C) and number of fixations (D) on HE and LE snacks in successful and unsuccessful SC trials. p < 0.05.

FIGURE 5 | fMRI results reveal stronger anterior cingulate activation in for contrast NSC vs. SC trials. Circle indicates anterior cingulate cluster. For
NSC vs. SC trials. Left panel: Mean s.e.m. parameter estimates in anterior visualization purposes, fMRI-results are thresholded at T > 2.87, p < 0.005
cingulate cluster for both conditions vs. baseline. Right panel: fMRI results uncorrected. ACG, Anterior cingulate cluster; INS, Insula. p < 0.05.

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 6

van der Laan et al. Neural correlates of food choice

during trials that did not pose a self-control dilemma (NSC weight-watching goal in response to different choice sets varying
trials). in aspects like the need for self-control, is an important direction
Hence, in sharp contrast with psychological theories on self- for future research.
control, the behavioral, eye tracking and neural findings suggest From theory it follows that having a long-term goal is a
that a stronger response conflict was experienced during the tri- prerequisite for perceiving an internal conflict in response to a
als in which no self-control dilemma was posed (NSC trials). For self-control dilemma (e.g., Fishbach et al., 2003). For this rea-
weight-concerned individuals the NSC choices should have con- son, we only included participants that were weight-concerned,
stituted an easy choice that requires no self-control because the LE according to self-reports. However, the finding that our partici-
snack was as tasty as the HE snack while being in line with their pants did not experience conflict questions whether they are truly
weight-watching goal. In contrast, the SC trials should have con- dedicated to restrict their energy intake. Interesting to note is that
stituted a difficult choice because the preferred HE snack is not we found exactly what would be expected for individuals who are
conducive to their weight-watching goal and thus self-control was not weight-concerned (i.e., people without the goal to limit their
required to choose the LE snack over the tastier HE snack. Our energy intake). For non-weight-concerned individuals, tastiness
findings suggest that weight-concerned women do not experience is the main predictor of choice, while the energy content of the
difficulty or internal conflict during choices posing a self-control food does not play a role (Arvola et al., 1999; Ayres et al., 2012).
dilemma and that this leads to food choices that are primarily Therefore, the SC trials would be easy since there is a large dif-
guided by tastiness and not by energy content. This supports ference in tastiness between the options, and energy content is
the notion that self-regulatory failure might be due to a lack of relatively irrelevant. The NSC trials on the other hand would pose
experienced internal conflict rather than a failure to act upon the a hedonic dilemma for non-weight-concerned individuals since
perception of such conflict. they were required to choose between two equally liked snacks.
Having established that self-regulatory failure might be due to The finding, that our population of weight-concerned women
a lack of internal conflict, the question arises why self-proclaimed perceived more conflict in the NSC trials in which the HE and LE
weight-concerned women do not appear to experience internal foods were equal in tastiness and that in approximately half (54%)
conflict when confronted with a self-control dilemma. A possible of these trials the LE snack was chosen suggests that choices were
explanation might be that exposure to the appealing HE snack mainly based on tastiness considerations, and not influenced by
decreased the accessibility of their long term goal to watch energy content. The higher conflict in the NSC trials might there-
their weight. This mechanism, concerning the facilitating role fore be interpreted as a hedonic dilemma of choosing between
of temptations on indulgence, has been provided by the goal- two equally liked snacks. Thus, intriguingly, both the choice pat-
conflict model (Stroebe et al., 2008), which posits that exposure terns and the responses on the implicit measures in our popula-
to temptation inhibits the accessibility of the opposing long- tion of self-proclaimed weight-concerned women resemble those
term goal. However, there is a contrasting line of research based expected for non-weight-concerned individuals. Because we used
on counteractive-control theory (Trope and Fishbach, 2000) highly stringent inclusion criteria, it is likely that our study popu-
that posits that temptations may assist, rather than undermine, lation truly had an authentic long-term goal to watch their weight.
long-term goal attainment (Kroese et al., 2011; Smeets et al., Therefore our results suggest that weight-concerned womens
2013). That is, temptation cues activate rather than inhibit long- choices are made primarily on basis of taste considerations and
term goal accessibility. To explore how exposure to tempting not on basis of energy content. Thus, our results confirm other
HE foods might have influenced the accessibility of the weight- findings that suggest that self-reports of weight-concerns and
watching goal, we used a primed lexical decision task to measure restraint are reflective of intentions and wishes to restrict intake,
temptation-goal associations. This task revealed that exposure to rather than of actual eating behavior (Stice et al., 2004, 2007,
temptation cues did not appear to activate the long term weight- 2010; de Witt Huberts et al., 2013; de Ridder et al., in press). It is of
watching goal. This is in line with an earlier study that showed high interest to repeat this paradigm in a non-weight-concerned
that temptation cues activate long term goals only in successful, population. By comparing our results with a group of non-
and not in unsuccessful self-controllers (Papies et al., 2008). Thus, weight-concerned women, we could rule out whether the effects
our results suggest that the lack of experienced conflict might be seen in the present study are general effects that occur also in non-
explained by the absence of temptation-goal associations, imply- weight-concerned women or whether they are specific for weight-
ing that exposure to appealing HE snacks within binary choice concerned women in which the self-control dilemma is relevant.
sets requiring self-control does not increase the accessibility of Importantly, the present study is distinct from earlier neu-
the weight-watching goal. This could be further tested in future roimaging studies on self-control because we focused specifically
studies that measure the accessibility of the weight-watching goal on conflict-monitoring in response to choice sets that either
directly in response to different choice pairs either requiring self- threaten the accomplishment of the long-term goal or not.
control or not, such as the choices that we used in our study. Conflict monitoring is the process by which the brain deter-
For example, reaction times to dieting and neutral words, after mines when control is required (Botvinick et al., 2001) and
a prime choice-set that either requires self-control or not, could thus precedes the actual act of self-control. Other neuroimag-
be measured. Most current studies on temptation-goal associ- ing studies on self-control have established how factors such as
ations assess goal accessibility by measuring reaction times to healthiness and tastiness are incorporated in the brain during
diet and neutral words after primes with single words or pic- food choices, how explicitly cueing people to consider healthiness
tures of temptation stimuli. Investigating the accessibility of the of foods influences the neural response during food choice, and

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 7

van der Laan et al. Neural correlates of food choice

how depletion affects the neural response to rewarding food cues to less-preferred LE snacks could break this loop and help peo-
(Hare et al., 2009, 2011; Wagner et al., 2013). ple choose LE snacks. This idea is in line with the suggestion that
It is important to note that we chose to use pairs of HE and distracting attention away from attractive stimuli might facilitate
LE snacks matched on tastiness as no self-control choices. From a self-control for which Van Dillen et al. (2013) provided initial
theoretical perspective, choosing between equally liked HE and evidence. This topic deserves further investigation.
LE snacks should have constituted an easy choice for weight- Our study population consisted of women with a healthy
concerned women: choosing LE snacks requires no self-control weight and therefore their motivation for weight-concern might
when they are equally tasty as the HE snack and also in line with not arise from medical or health reasons. Rather, since earlier
their weight-watching goal. An alternative approach would have studies showed a clear link between worries about appearance and
been to let participants choose between two LE foods differing weight-concerns/restraint (Putterman and Linden, 2004, 2006;
in tastiness: this choice set also requires no self-control. However, OBrien et al., 2007; de Ridder et al., in press) we think that the
contrasting choices between pairs of HE/LE snacks and pairs of high level of self-reported weight-concerns in our study popu-
LE/LE snacks has other disadvantages. For example, many ear- lation might indicate an intention to lose weight for cosmetic
lier studies have shown that the neural response to HE and LE reasons or general concerns about healthy eating. A large repre-
foods differs (van der Laan et al., 2011). Therefore, using pairs of sentative community study has also confirmed that high levels of
LE snacks instead of our current NSC trials (pairs of HE and LE restraint eating are mainly associated with concerns about weight
snacks), would have confounded our factor of interestthe need and appearance and to a lesser extent associated with concerns
for self-controlwith the energy content of the stimuli. For this about the perceived health consequences of dietary habits (de
reason, we chose to use pairs of equally liked HE and LE snacks as Ridder et al., in press). We would like to note that, although
NSC trials. our population was of normal weight, research has shown that
Another factor which might have played a role is that tastiness normal-weight individuals who report high weight-concerns are
ratings to match the choice pairs were collected prior to actual at increased risk for gaining weight (French et al., 1994; Mann
choice. It could be that LE snacks were rated as being tastier et al., 2007). This is even worse for individuals that diet for
before the choice as compared to during choice, which is sup- appearance (as opposed to health) reasons: they show more
ported by an earlier study (Myrseth et al., 2009) that showed lapses in restraint and disinhibited eating (Putterman and Linden,
that activities in line with the long-term goal are rated higher in 2004). Therefore, normal weight females reporting high levels
perceived appeal before making a decision than at the moment of weight-concerns are a very important population to focus on
of (or after) deciding, when a more tempting alternative is con- in research and weight-maintenance interventions, especially if
currently available. However, it is unlikely that this would have you take into consideration that approximately 5060% of the
affected the results for our contrast between NSC and SC trials Western female population reports to be high in weight-concerns
because (1) the bias would be small (Myrseth et al., 2009), (2) (Rideout and Barr, 2009; Fayet et al., 2012; de Ridder et al.,
the difference in tastiness between HE and LE snacks is still larger in press).
in the SC trials, posing a stronger conflict, than in the NSC tri- We did not include self-reported measures of conflict in our
als, (3) using tastiness ratings collected prior to actual choices to study, that is, we did not ask them explicitly how difficult they
match choice pairs on basis of pre-existing preference is an often found the choices or whether they consciously perceived con-
used and well-acknowledged approach to study self-control (e.g., flict. Measuring subjective or self-reported measures of conflict
Hare et al., 2009), and (4) the finding that the proportion of HE in addition to implicit measures might have provided further
and LE snacks in the NSC trials was approximately 50/50 suggests understanding about consciously perceived conflict. However,
that choices were based on tastiness and that this tastiness was we used implicit instead of subjective self-reported measures of
approximately equal. conflict for several reasons. Firstly, the perception of conflict
Although the number of successful SC trials (SC trials in which is not necessary a conscious phenomenon. Therefore, subjec-
the LE snack was chosen) was low, the eye tracking data revealed tive measures might not validly capture the process. Secondly,
interesting patterns that hint toward a possibly effective strategy implicit measures are less susceptible to demand characteristics
for successful self-control. The higher reaction times and num- and socially desirable responding (Fazio and Olson, 2003) than
ber of gaze switches during successful compared to unsuccessful self-reported measures. Thirdly, ACC activity, reaction times, and
SC trials confirm the general assumption that it is difficult to fixation durations are well acknowledged measures to investigate
choose a LE snack over a tastier HE alternative. The findings response conflict (Kawashima et al., 1996; Macdonald et al., 2000;
that participants fixated longer on LE than HE items in the suc- Botvinick et al., 2001; van Veen et al., 2001; Kerns et al., 2004;
cessful SC trials and that the last fixation was significantly more Panayiotou and Vrana, 2004; Causse et al., 2011). Finally, explic-
often on the LE snack might have increased preference for the itly asking for response conflict during the food choices focusses
LE snack. Research has shown that fixation duration both reflects attention on conflict which may also affect their choice.
and influences preference. That is, people look longer at preferred Our findings have implications for weight management
or chosen items (Chandon et al., 2009; Atalay et al., 2012) and interventions. Since recognizing the self-control dilemma is a pre-
manipulating gaze duration for an option increases preference requisite for acting upon it, this study implies that weight man-
for it (Shimojo et al., 2003), i.e., there is a positive feedback loop agement interventions might benefit from focusing on increasing
between looking and liking. It could be speculated that directing the ability to detect a self-control dilemma, in addition to the
attention away from preferred HE snacks and allocating attention current focus on increasing self-regulatory capacity and cognitive

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 8

van der Laan et al. Neural correlates of food choice

control. Evidently, if no conflict is experienced in the first place, REFERENCES

improving the ability to perform effortful self-control is likely Arvola, A., Lahteenmaki, L., and Tuorila, H. (1999). Predicting the intent to pur-
to be ineffective. As outlined above, a possible explanation for chase unfamiliar and familiar cheeses: the effects of attitudes, expected liking
and food neophobia. Appetite 32, 113126. doi: 10.1006/appe.1998.0181
the lack of experienced conflict during food choice could be that Atalay, A. S., Bodur, H. I., and Rasolofoarison, D. (2012). Central gaze cascade on
the relevance of the long-term goal is over-ruled by the expo- product choice. J. Consum. Res. 39, 848866. doi: 10.1086/665984
sure to tempting foods. Therefore, a useful strategy might be Ayres, K., Conner, M. T., Prestwich, A., and Smith, P. (2012). Do implicit measures
to strengthen temptation-goal associations to change the auto- of attitudes incrementally predict snacking behaviour over explicit affect-related
measures? Appetite 58, 835841. doi: 10.1016/j.appet.2012.01.019
matic response to tempting foods to activating the long-term
Baumeister, R., and Heatherton, T. F. (1996). Self-regulation failure: an overview.
goal, rather than inhibiting it. Forming implementation inten- Psychol. Inq. 7, 115. doi: 10.1207/s15327965pli0701_1
tions as an intervention has proven successful in strengthening Botvinick, M. M., Braver, T. S., Barch, D. M., Carter, C. S., and Cohen, J. D. (2001).
temptation-goal associations (e.g., Kroese et al., 2011). Conflict monitoring and cognitive control. Psychol. Rev. 108, 624652. doi:
The lack of internal conflict in response to a food choice set 10.1037/0033-295X.108.3.624
Cahill, L. (2006). Why sex matters for neuroscience. Nat. Rev. Neurosci. 7, 477484.
posing a self-control dilemma may extrapolate to other behav-
doi: 10.1038/nrn1909
ioral domains involving a trade-off between direct impulses and Carver, C. S., and White, T. L. (1994). Behavioral inhibition, behavioral activa-
future goals, such as smoking cessation, financial management, tion, and affective responses to impending reward and punishment: the BIS/BAS
interpersonal relations and aggression. Future studies should elu- scales. J. Pers. Soc. Psychol. 67, 319333. doi: 10.1037/0022-3514.67.2.319
cidate whether a lack of experienced conflict also accounts for Causse, M., Baracat, B., Pastor, J., and Dehais, F. (2011). Reward and uncer-
tainty favor risky decision-making in pilots: evidence from cardiovascular and
self-regulatory failure in these domains. oculometric measurements. Appl. Psychophysiol. Biofeedback 36, 231242. doi:
To conclude, in contrast with the common assumption that 10.1007/s10484-011-9163-0
food choices posing a self-control dilemma evoke an internal con- Chandon, P., Hutchinson, J. W., Bradlow, E. T., and Young, S. H. (2009). Does in-
flict in weight-concerned women, we found that these choices store marketing work? Effects of the number and position of shelf facings on
appeared to induce no conflict, as indicated by lower reaction brand attention and evaluation at the point of purchase. J. Mark. 73, 117. doi:
10.1509/jmkg.73.6.1
times, lower fixation durations, lower number of gaze switches Cornier, M. A., Salzberg, A. K., Endly, D. C., Bessesen, D. H., and Tregellas, J. R.
between snacks, and decreased activation of the anterior cingulate (2010). Sex-based differences in the behavioral and neuronal responses to food.
cortex. These findings provide support for the notion that self- Physiol. Behav. 99, 538543. doi: 10.1016/j.physbeh.2010.01.008
regulatory failure might be due to a lack of experienced conflict, de Ridder, D., Adriaanse, M., Evers, C., and Verkes, R. J. (in press). Who diets? Most
people and especially when they worry about food. Appetite.
rather than failing to act upon the experience of such conflict.
de Ridder, D. T., Lensvelt-Mulders, G., Finkenauer, C., Stok, F. M., and Baumeister,
Since recognizing the self-control conflict is a prerequisite for act- R. F. (2012). Taking stock of self-control: a meta-analysis of how trait self-
ing upon it, this study suggests that the effectiveness of weight control relates to a wide range of behaviors. Pers. Soc. Psychol. Rev. 16, 7699.
maintenance interventions might be increased if they also focus doi: 10.1177/1088868311418749
on increasing the ability to detect a self-control dilemma, in addi- de Witt Huberts, J. C., Evers, C., and de Ridder, D. T. (2013). Double trouble:
restrained eaters do not eat less and feel worse. Psychol. Health 28, 686700.
tion to the current focus on increasing self-regulatory capacity
doi: 10.1080/08870446.2012.751106
and cognitive control. Evans, A. C., Collins, D. L., Mills, D. R., Brown, E. D., Kelly, R. L., and Peters, T.
M. (1993). 3D statistical neuroanatomical models from 305 MRI volumes. Proc.
AUTHOR CONTRIBUTIONS IEEE Nucl. Sci. Symp. Med. Imaging 3, 18131817.
Fayet, F., Petocz, P., and Samman, S. (2012). Prevalence and correlates of dieting in
Laura N. Van der Laan, Paul A. M. Smeets, and Denise T. D. de
college women: a cross sectional study. Int. J. Womens Health 4, 405411. doi:
Ridder conceived and designed the experiment. Laura N. Van der 10.2147/IJWH.S33920
Laan collected and analyzed the data. Laura N. Van der Laan, Fazio, R. H., and Olson, M. A. (2003). Implicit measures in social cogni-
Paul A. M. Smeets, and Denise T. D. de Ridder interpreted the tion. research: their meaning and use. Annu. Rev. Psychol. 54, 297327. doi:
data. Laura N. Van der Laan wrote the manuscript and Paul A. 10.1146/annurev.psych.54.101601.145225
Fishbach, A., Friedman, R. S., and Kruglanski, A. W. (2003). Leading us not unto
M. Smeets and Denise T. D. de Ridder provided critical revisions.
temptation: momentary allurements elicit overriding goal activation. J. Pers.
Lisette Charbonnier and Max A. Viergever contributed to the final Soc. Psychol. 84, 296309. doi: 10.1037/0022-3514.84.2.296
version of the manuscript by interpreting results, reviewing and Frank, S., Laharnar, N., Kullmann, S., Veit, R., Canova, C., Hegner, Y. L., et al.
critically revising text. Lisette Charbonnier provided the food pic- (2010). Processing of food pictures: influence of hunger, gender and calorie
tures. All authors approved the final version for submission and content. Brain Res. 1350, 159166. doi: 10.1016/j.brainres.2010.04.030
French, S. A., Jeffery, R. W., and Wing, R. R. (1994). Food intake and physical activ-
agreed to be accountable to for all aspects of the work. ity: a comparison of three measures of dieting. Addict. Behav. 19, 401409. doi:
10.1016/0306-4603(94)90063-9
ACKNOWLEDGMENTS Giuliani, N. R., Calcott, R. D., and Berkman, E. T. (2013). Piece of cake. Cognitive
Support by governmental research grant Agentschap NL reappraisal of food craving. Appetite 64, 5661. doi: 10.1016/j.appet.2012.
12.020
(FND08009). Stimuli were created in the Full4Health project
Guadalupe, T., Willems, R. M., Zwiers, M. P., Arias, V. A., Hoogman, M., Hagoort,
European Union Seventh Framework Programme (FP7/2007 P., et al. (2014). Differences in cerebral cortical anatomy of left- and right-
2013), grant agreement nr. 266408. handers. Front. Psychol. 5:261. doi: 10.3389/fpsyg.2014.00261
Haase, L., Green, E., and Murphy, C. (2011). Males and females show differential
SUPPLEMENTARY MATERIAL brain activation to taste when hungry and sated in gustatory and reward areas.
Appetite 57, 421434. doi: 10.1016/j.appet.2011.06.009
The Supplementary Material for this article can be found online Hare, T. A., Camerer, C. F., and Rangel, A. (2009). Self-control in decision-making
at: http://www.frontiersin.org/journal/10.3389/fnbeh.2014. involves modulation of the vmPFC valuation system. Science 324, 646648. doi:
00184/abstract 10.1126/science.1168450

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 9

van der Laan et al. Neural correlates of food choice

Hare, T. A., Malmaud, J., and Rangel, A. (2011). Focusing attention on the Pingitore, R., Spring, B., and Garfield, D. (1997). Gender differences in body
health aspects of foods changes value signals in vmPFC and improves satisfaction. Obes. Res. 5, 402409. doi: 10.1002/j.1550-8528.1997.tb00662.x
dietary choice. J. Neurosci. 31, 1107711087. doi: 10.1523/JNEUROSCI.6383- Putterman, E., and Linden, W. (2004). Appearance versus health: does the rea-
10.2011 son for dieting affect dieting behavior? J. Behav. Med. 27, 185204. doi:
Heatherton, T. F., and Wagner, D. D. (2011). Cognitive neuroscience of self- 10.1023/B:JOBM.0000019851.37389.a7
regulation failure. Trends Cogn. Sci. 15, 132139. doi: 10.1016/j.tics.2010. Putterman, E., and Linden, W. (2006). Cognitive dietary restraint and cortisol:
12.005 importance of pervasive concerns with appearance. Appetite 47, 6476. doi:
Kawashima, R., Satoh, K., Itoh, H., Ono, S., Furumoto, S., Gotoh, R., et al. (1996). 10.1016/j.appet.2006.02.003
Functional anatomy of GO/NO-GO discrimination and response selectiona Rideout, C. A., and Barr, S. I. (2009). Restrained eating vs trying to lose
PET study in man. Brain Res. 728, 7989. weight: how are they associated with body weight and tendency to overeat
Kerns, J. G., Cohen, J. D., Macdonald, A. W. III., Cho, R. Y., Stenger, V. among postmenopausal women? J. Am. Diet. Assoc. 109, 890893. doi:
A., and Carter, C. S. (2004). Anterior cingulate conflict monitoring and 10.1016/j.jada.2009.02.009
adjustments in control. Science 303, 10231026. doi: 10.1126/science.10 Roininen, K., Tuorila, H., Zandstra, E. H., de, G. C., Vehkalahti, K., Stubenitsky, K.,
89910 et al. (2001). Differences in health and taste attitudes and reported behaviour
Kirby, K. N., Petry, N. M., and Bickel, W. K. (1999). Heroin addicts among Finnish, Dutch and British consumers: a cross-national validation
have higher discount rates for delayed rewards than non-drug-using of the Health and Taste Attitude Scales (HTAS). Appetite 37, 3345. doi:
controls. J. Exp. Psychol. Gen. 128, 7887. doi: 10.1037/0096-3445. 10.1006/appe.2001.0414
128.1.78 Shimojo, S., Simion, C., Shimojo, E., and Scheier, C. (2003). Gaze bias both
Kirchner, W. K. (1958). Age differences in short-term retention of rapidly reflects and influences preference. Nat. Neurosci. 6, 13171322. doi: 10.1038/
changing information. J. Exp. Psychol. 55, 352358. doi: 10.1037/h00 nn1150
43688 Siep, N., Roefs, A., Roebroeck, A., Havermans, R., Bonte, M., and Jansen,
Kroese, F. M., Adriaanse, M. A., Evers, C., and de Ridder, D. T. (2011). A. (2012). Fighting food temptations: the modulating effects of short-term
Instant success: turning temptations into cues for goal-directed behav- cognitive reappraisal, suppression and up-regulation on mesocorticolim-
ior. Pers. Soc. Psychol. Bull. 37, 13891397. doi: 10.1177/01461672114 bic activity related to appetitive motivation. Neuroimage 60, 213220. doi:
10889 10.1016/j.neuroimage.2011.12.067
Luders, E., Gaser, C., Narr, K. L., and Toga, A. W. (2009). Why sex matters: Smeets, P. A., de, G. C., Stafleu, A., van Osch, M. J., Nievelstein, R. A., and van der,
brain size independent differences in gray matter distributions between men G. J. (2006). Effect of satiety on brain activation during chocolate tasting in men
and women. J. Neurosci. 29, 1426514270. doi: 10.1523/JNEUROSCI.2261- and women. Am. J. Clin. Nutr. 83, 12971305. Available online at: http://ajcn.
09.2009 nutrition.org/content/83/6/1297.short
Macdonald, A. W. III., Cohen, J. D., Stenger, V. A., and Carter, C. S. (2000). Smeets, P. A., Kroese, F. M., Evers, C., and de Ridder, D. T. (2013).
Dissociating the role of the dorsolateral prefrontal and anterior cingu- Allured or alarmed: counteractive control responses to food tempta-
late cortex in cognitive control. Science 288, 18351838. doi: 10.1126/sci- tions in the brain. Behav. Brain Res. 248, 4145. doi: 10.1016/j.bbr.2013.
ence.288.5472.1835 03.041
Maldjian, J. A., Laurienti, P. J., Kraft, R. A., and Burdette, J. H. (2003). An Stice, E., Cooper, J. A., Schoeller, D. A., Tappe, K., and Lowe, M. R. (2007). Are
automated method for neuroanatomic and cytoarchitectonic atlas-based inter- dietary restraint scales valid measures of moderate- to long-term dietary restric-
rogation of fMRI data sets. Neuroimage 19, 12331239. doi: 10.1016/S1053- tion? Objective biological and behavioral data suggest not. Psychol. Assess. 19,
8119(03)00169-1 449458. doi: 10.1037/1040-3590.19.4.449
Mann, T., Tomiyama, A. J., Westling, E., Lew, A. M., Samuels, B., and Stice, E., Fisher, M., and Lowe, M. R. (2004). Are dietary restraint scales valid mea-
Chatman, J. (2007). Medicares search for effective obesity treatments: diets sures of acute dietary restriction? Unobtrusive observational data suggest not.
are not the answer. Am. Psychol. 62, 220233. doi: 10.1037/0003-066X. Psychol. Assess. 16, 5159. doi: 10.1037/1040-3590.16.1.51
62.3.220 Stice, E., Sysko, R., Roberto, C. A., and Allison, S. (2010). Are dietary restraint scales
Metcalfe, J., and Mischel, W. (1999). A hot/cool-system analysis of delay of valid measures of dietary restriction? Additional objective behavioral and bio-
gratification: dynamics of willpower. Psychol. Rev. 106, 319. doi: 10.1037/0033- logical data suggest not. Appetite 54, 331339. doi: 10.1016/j.appet.2009.12.009
295X.106.1.3 Stroebe, W., Mensink, W., Aarts, H., Schut, H., and Kruglanski, A. W. (2008). Why
Myrseth, K. O., Fishbach, A., and Trope, Y. (2009). Counteractive self- dieters fail: testing the goal conflict model of eating. J. Exp. Soc. Psychol. 44,
control. Psychol. Sci. 20, 159163. doi: 10.1111/j.1467-9280.2009. 2636. doi: 10.1016/j.jesp.2007.01.005
02268.x Stroop, J. R. (1935). Studies of interference in serial verbal reactions. J. Exp. Psychol.
Neumark-Sztainer, D., Sherwood, N. E., French, S. A., and Jeffery, R. Gen. 18, 643662. doi: 10.1037/h0054651
W. (1999). Weight control behaviors among adult men and women: Tangney, J. P., Baumeister, R. F., and Boone, A. L. (2004). High self-control pre-
cause for concern? Obes. Res. 7, 179188. doi: 10.1002/j.1550-8528.1999. dicts good adjustment, less pathology, better grades, and interpersonal success.
tb00700.x J. Pers. 72, 271324. doi: 10.1111/j.0022-3506.2004.00263.x
OBrien, K., Venn, B. J., Perry, T., Green, T. J., Aitken, W., Bradshaw, A., et al. (2007). Trope, Y., and Fishbach, A. (2000). Counteractive self-control in overcom-
Reasons for wanting to lose weight: different strokes for different folks. Eat. ing temptation. J. Pers. Soc. Psychol. 79, 493506. doi: 10.1037/0022-3514.
Behav. 8, 132135. doi: 10.1016/j.eatbeh.2006.01.004 79.4.493
Ooki, S. (2014). An overview of human handedness in twins. Front. Psychol. 5:10. Tzourio-Mazoyer, N., Landeau, B., Papathanassiou, D., Crivello, F., Etard, O.,
doi: 10.3389/fpsyg.2014.00010 Delcroix, N., et al. (2002). Automated anatomical labeling of activations in SPM
Panayiotou, G., and Vrana, S. (2004). The role of self-focus, task difficulty, task self- using a macroscopic anatomical parcellation of the MNI MRI single-subject
relevance, and evaluation anxiety on reaction time performance. Motiv. Emot. brain. Neuroimage 15, 273289. doi: 10.1006/nimg.2001.0978
28, 171196. doi: 10.1023/B:MOEM.0000032313.69675.0d van der Laan, L. N., de Ridder, D. T., Viergever, M. A., and Smeets, P. A.
Papies, E. K., Stroebe, W., and Aarts, H. (2008). Healthy cognition: processes of self- (2011). The first taste is always with the eyes: a meta-analysis on the neu-
regulatory success in restrained eating. Pers. Soc. Psychol. Bull. 34, 12901300. ral correlates of processing visual food cues. Neuroimage 55, 296303. doi:
doi: 10.1177/0146167208320063 10.1016/j.neuroimage.2010.11.055
Patton, J. H., Stanford, M. S., and Barratt, E. S. (1995). Factor structure of the Van Dillen, L. F., Papies, E. K., and Hofmann, W. (2013). Turning a blind eye to
Barratt impulsiveness scale. J. Clin. Psychol. 51, 768774. temptation: how cognitive load can facilitate self-regulation. J. Pers. Soc. Psychol.
Peake, P. K., Mischel, W., and Hebl, M. (2002). Strategic atten- 104, 427443. doi: 10.1037/a0031262
tion deployment for delay of gratification in working and waiting Van Strien, T., Frijters, J. E. R., Bergers, G. P. A., and Defares, P. B. (1986). The Dutch
situations. Dev. Psychol. 38, 313326. doi: 10.1037/0012-1649.38. eating behavior questionnaire (DEBQ) for assessment of restrained, emotional,
2.313 and external eating behavior. Int. J. Eat. Disord. 5, 195215.

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 10

van der Laan et al. Neural correlates of food choice

van Veen, V., Cohen, J. D., Botvinick, M. M., Stenger, V. A., and Carter, Received: 20 March 2014; accepted: 05 May 2014; published online: 22 May 2014.
C. S. (2001). Anterior cingulate cortex, conflict monitoring, and Citation: van der Laan LN, de Ridder DTD, Charbonnier L, Viergever MA and
levels of processing. Neuroimage 14, 13021308. doi: 10.1006/nimg.20 Smeets PAM (2014) Sweet lies: neural, visual, and behavioral measures reveal a lack
01.0923 of self-control conflict during food choice in weight-concerned women. Front. Behav.
Wagner, D. D., Altman, M., Boswell, R. G., Kelley, W. M., and Heatherton, Neurosci. 8:184. doi: 10.3389/fnbeh.2014.00184
T. F. (2013). Self-regulatory depletion enhances neural responses to This article was submitted to the journal Frontiers in Behavioral Neuroscience.
rewards and impairs top-down control. Psychol. Sci. 24, 22622271. doi: Copyright 2014 van der Laan, de Ridder, Charbonnier, Viergever and Smeets.
10.1177/0956797613492985 This is an open-access article distributed under the terms of the Creative Commons
Attribution License (CC BY). The use, distribution or reproduction in other forums is
Conflict of Interest Statement: The authors declare that the research was con- permitted, provided the original author(s) or licensor are credited and that the original
ducted in the absence of any commercial or financial relationships that could be publication in this journal is cited, in accordance with accepted academic practice. No
construed as a potential conflict of interest. use, distribution or reproduction is permitted which does not comply with these terms.

Frontiers in Behavioral Neuroscience www.frontiersin.org May 2014 | Volume 8 | Article 184 | 11