VU Research Portal

Brain Activation by Visual Food-Related Stimuli and Correlations with Metabolic and
Hormonal Parameters: A fMRI Study
Jakobsdottir, S.; de Ruiter, M.B.; Deijen, J.B.; Veltman, D.J.; Drent, M.L.

published in
The Open Neuroendocrinology Journal
2012

DOI (link to publisher)

10.2174/1876528901205010005

document version
Publisher's PDF, also known as Version of record

Link to publication in VU Research Portal

citation for published version (APA)

Jakobsdottir, S., de Ruiter, M. B., Deijen, J. B., Veltman, D. J., & Drent, M. L. (2012). Brain Activation by Visual
Food-Related Stimuli and Correlations with Metabolic and Hormonal Parameters: A fMRI Study. The Open
Neuroendocrinology Journal, 2012(5), 5-12. https://doi.org/10.2174/1876528901205010005

General rights
Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners
and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights.

• Users may download and print one copy of any publication from the public portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain
• You may freely distribute the URL identifying the publication in the public portal

Take down policy

If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.

E-mail address:
vuresearchportal.ub@vu.nl

Download date: 21. Sep. 2025

 The Open Neuroendocrinology Journal, 2012, 5, 5-12 5

Open Access
Brain Activation by Visual Food-Related Stimuli and Correlations with
Metabolic and Hormonal Parameters: A fMRI Study
Sigridur Jakobsdottir1, Michiel de Ruiter2, Jan Berend Deijen*,3, Dick J. Veltman4 and
Madeleine L. Drent1

Department of Internal Medicine, Endocrine Section, VU University Medical Center, Amsterdam, The Netherlands
Department of Psychiatry, VU University Medical Center, Amsterdam, The Netherlands
Department of Clinical Neuropsychology, VU University, Amsterdam, The Netherlands
4
Department of Psychiatry, VU University Medical Center & University of Amsterdam Academic Medical Center,
Amsterdam, The Netherlands

Abstract: Regional brain activity in 15 healthy, normal weight males during processing of visual food stimuli in a satiated
and a hungry state was examined and correlated with neuroendocrine factors known to be involved in hunger and satiated
states. Two functional Magnetic Resonance Imaging (fMRI) sessions were performed with a one week interval, after
overnight fasting or 1 hour after a standardized meal. Blood samples and appetite assessment were obtained after each
fMRI session. Main effects of processing food versus non-food stimuli were observed in the ventral visual stream,
including the fusiform gyrus and hippocampal areas bilaterally, significantly more in the fasting state. Leptin
concentration correlated negatively with activity in the left hippocampal area and right insula during the satiation
condition. A positive correlation between ghrelin and “thought of food” hunger scores were found. The positive
correlation between ghrelin and food related activation in the insula areas and the right hippocampus during fasting did
not reach significance.
Conclusion: The increased activation of food vs non-food pictures in the ventral visual stream reflects increased salience
of food pictures when subjects are hungry. Leptin was associated with activations in areas involved in processing of new
information and emotion.
Keywords: Amygdala, fMRI, ghrelin, hippocampus, leptin, visual stream.

INTRODUCTION cognitive factors and changes in regional cerebral blood flow

in the amygdala and orbitofrontal cortex using positron
The neurophysiological processing of food-related emission tomography (PET) has been demonstrated,
stimuli is increasingly considered relevant for the
supporting the theory that these areas constitute an integrated
understanding of appetite regulation and the pathogenesis of
neural system critically involved in making adaptive
obesity. Single-cell recordings in primates have shown
responses and guiding decision making [5,7,8]. In a 15O-PET
neural activity related to food stimuli in the amygdala and
study, it was shown that other limbic structures such as the
the orbitofrontal cortex (OFC) [1,2]. In the amygdala,
nucleus accumbens and the insula seem to participate
information regarding biologically relevant stimuli such as similarly in mediating physiological and motivational states
food is generally perceived, and forwarded to other brain
[9]. LaBar et al., used fMRI to assess the influence of hunger
regions such as the ventromedial cortex for further
on the response of the amygdala and its related cortical
processing of its rewarding value or motivational salience [3-
structures. Their results showed that food related visual
5]. Functional neuroimaging modalities such as functional
stimuli were associated with a greater response in the
Magnetic Resonance Imaging (fMRI) have been used to
amygdala, parahippocampal gyrus, and anterior fusiform
locate specific brain areas related to the perception and gyrus when participants were hungry. These findings further
processing of food related stimuli in humans. Regional brain
support the hypothesis that these regions are involved not
activity associated with perception of food stimuli is likely to
only in visual processing but also in the integration of
depend on the state of hunger and satiety, presumably
subjective interoceptive states [10]. However, in both
reflecting processing of motivational significance of food
studies, subjects were scanned using a fixed-order design,
stimuli in addition to sensory effects [4-6 for an overview].
which may have biased their results [8,9].
An interaction between perceptional, motivational and
Several hormones, for example ghrelin, leptin and
insulin, have been shown to be involved in the central
*Address correspondence to this author at the Department of Clinical regulation of appetite, hunger and satiation [11-13].
Neuropsychology, VU University, Van der Boechorststraat 1, 1081 BT Receptors for these hormones have been found in the
Amsterdam, The Netherlands; Tel: +31-20-5988756; Fax: +31-20-5988971;
E-mail: jb.deijen@psy.vu.nl
hypothalamus, and the involvement of these hormones in the
energy homeostasis and food intake has widely been

1876-5289/12 2012 Bentham Open

6 The Open Neuroendocrinology Journal, 2012, Volume 5 Jakobsdottir et al.

hypothesized [14-16]. These receptors are also expressed in METHODS

other areas of the brain such as the hippocampus and
Subjects
pituitary. One of the most important adipokines, leptin, has
been implicated in a variety of functions of the central Fifteen male subjects were included. Data from one
nervous system such as learning and memory processes, subject had to be discarded due to scanner failure, leaving 14
neuroendocrine regulation, and possibly neuroprotection [17- subjects for subsequent analyses. Inclusion criteria: male,
20]. Leptin replacement in genetically leptin-deficient adults age: 20-40 years, BMI 20-25, right handed, healthy. The
modulates the sensitivity to visual food stimuli, with reduced subjects’ health was determined by medical history, physical
activation in hunger related areas and enhanced response in examination and laboratory screening tests. Exclusion
regions involved in satiation [21,22]. Likewise, metabolic criteria: left handed, a history of neurological or psychiatric
factors such as glucose and free fatty acids (FFA) are likely disorder, use of medication.
to serve as regional modulators of postprandial neuronal
Mean age was 23.4 ± 3.5 years, range 19-27 years, mean
events [23]. It is therefore relevant to include measurements
BMI was 22.4 ± 2.0 kg/m
. Written informed consent was
of these factors when studying brain responses to food
obtained, the study was approved by the Medical Ethical
related stimuli under various conditions, but to date studies
Committee of the VU University Medical Center and was
on this issue have been scarce.
conducted according to the principles of the Helsinki
Another relevant aspect of the reaction to food stimuli is Declaration.
the role of craving, which can be defined as an intense desire
to eat specific food types and is likely to influence the Design
development of obesity, although the process of craving is With a one-week interval two fMRI sessions were
still insufficiently understood. The neural correlates of food performed at the same time in the morning, either after 12-
craving have been investigated with fMRI, showing hour overnight fasting or 1 hour after consuming a 1600 kcal
activation in areas such as the hippocampus, insula and standard meal in randomised order. The meal consisted of
caudate nucleus, which have also been implicated in drug 44.4 energy % carbohydrates, 15.8 energy % protein and
craving [24]. In another study, viewing chocolate pictures 39.8 energy % fat to provide a clear contrast between the
was associated with increased activation of the ventral fasting and satiated setting.
striatum in cravers compared to non-cravers, highlighting the
importance of these regions in making salience judgments After each fMRI session, a 10-point Likert scale Appetite
regarding food stimuli [25]. assessment score was obtained, consisting of items on
hunger, fullness, desire to eat, prospective consumption, and
The present study was designed to further investigate total appetite [26]. Also, blood samples were collected for
processing of visual food stimuli both in a satiated (one hour measurement of plasma glucose, free fatty acids,
after food ingestion with a standardized meal) and hungry triglycerides, insulin, leptin, and ghrelin.
(after 12 hour fasting) state, in randomised order on two
separate occasions to avoid order effects [8,9]. A second aim Scanning Procedure
was to evaluate correlations between hormonal A 1.5 Tesla Sonata MR scanner (Siemens, Erlangen,
measurements and regional brain activity. To this end, we Germany) was used to obtain echo-planar images (EPI) with
employed fMRI during presentation of food and non-food blood oxygenation level dependent (BOLD) contrast. The
pictures in a crossover design, during which memory subject’s head was fixed with foam pads to minimize head
encoding and retrieval performance was registered, since it movements. During the scanning procedure, pictures were
has been shown that memory for food items is associated presented by projection on a screen at the back end of the
with amygdala activity [9]. Also to evaluate some important scanner, which could be seen through a mirror mounted
modulatory factors, ghrelin, leptin, insulin, glucose and free above the subject’s head. For functional MRI, an echo planar
fatty acids were measured both during fasting and satiated imaging sequence (interpulse interval or TR = 3.306 s, time
conditions. We hypothesized that responses to food vs non- to echo or TE = 45ms, flip angle = 90º) was used creating
food pictures would be greater during the fasting state transversal whole brain acquisitions (38 slices, 3 x 3 mm in-
relative to the satiated state, in areas known to be involved in plane resolution, slice thickness 2.5 mm with a 0.5 mm inter-
processing visual food stimuli. Furthermore, a positive slice gap). Two series of pictures were presented (software:
correlation between activity in these areas and ghrelin was E-Prime) in a block design, depicting food or non-food items
expected, as ghrelin is an orexigenic hormone and its such as landscapes, people, and houses. These pictures were
receptors have been located in the hippocampus and visually matched for visual complexity, both for the objects
ventromedial hypothalamus. In contrast, during the satiated shown and their background. They were not systematically
state we expected to find correlations between the satiation matched for color. The pictures were not selected from the
signals leptin and insulin and activity within the IAPS set so that independent ratings for valence/arousal
hypothalamus. were not available. For the encoding phase 48 pictures and
In conclusion, to further investigate the processing of for the retrieval phase 96 pictures (half new, half already
visual food stimuli during a satiated and fasting condition, seen during the encoding phase) were presented randomly
food and non-food pictures were presented during an fMRI selected for food (half) and nonfood (half). Each picture was
procedure. Memory encoding, retrieval performance and displayed for 4 seconds, followed by a pause of 2 seconds
several important modulatory factors were evaluated. before the next picture was shown. A button box was used to
Food Perception and Brain Activity The Open Neuroendocrinology Journal, 2012, Volume 5 7

register the subject’s response and reaction times. The Statistical Analysis
subjects were not asked to memorize the pictures, but were
Laboratory measurements were analyzed using a
requested to press the button (yes or no) whether the pictures
standard statistical package (SPSS version 13; SPSS,
were taken indoor or outdoor to control for attention
Chicago, IL, USA). Paired T-tests were used to evaluate
differences. During retrieval, subjects performed a two-
choice recognition task (seen before/new). 96 fMRI volumes changes between the fasting and the satiated state. In
addition, Pearson’s correlations (one-tailed or two-tailed,
were collected in the encoding phase, whereas during
depending on specific hypotheses) were calculated between
retrieval 176 volumes were acquired. Between the encoding
the subscales of the Hunger questionnaire and blood sample
and retrieval subsessions, a T1-weighted structural MRI scan
parameters as well as the correlations between imaging and
(MP-RAGE, magnetization prepared rapid acquisition
laboratory data. Correlations were computed separately for
gradient echo, resolution 1mm in-plane, 160 slices) was
obtained. the hungry and satiated conditions.
RESULTS
Image Processing and Analysis
Psychometric Measurements
Statistical Parametric Mapping (SPM5) software
(Wellcome Department of Imaging Neuroscience, Institute To determine the effects of conditions satiated-food,
of Neurology, London UK) was used for imaging analysis. satiated-nonfood and fasting-food, fasting-nonfood on
The images were realigned to compensate for subject reaction time (RT) and number of correct responses two
movement and were corrected for differences in slice separate repeated-measures analyses of variance (ANOVA)
acquisition timing. The T1-weighted anatomical images were used with RT and number of correct responses under
obtained from each subject were thereafter coregistered to each condition as repeated measurements factor. Only RTs
the mean echo planar images (EPI). Spatial normalization for correct trials were evaluated. With respect to RT, within-
was performed to match a standard template, and images subjects contrasts indicated that the RT was significantly
were spatially smoothed using a 6 mm Full Width at Half longer for fasting-food than for fasting-nonfood items (F
Maximum (FWHM) filter. Next, imaging data were analysed (1,14) = 25.22, p < 0.0005,
2 = 0.64). In addition, RTs were
within the context of the General Linear Model, using boxcar significantly longer for satiated-food than for satiated-
regressors convolved with a synthetic haemodynamic nonfood items (F (1,14) = 46.90, p < 0.0005,
2 = 0.77).
response function. For each subject, we performed food vs There were neither significant differences between the RTs
non-food comparisons and entered the resulting contrast under the fasting and satiated conditions separately analysed
images into second-level (random effects analyses). For for the food and non-food conditions, nor for those averaged
main effects (food vs non-food) results were thresholded at across food and non-food conditions.
p<0.05 corrected for multiple comparisons using the False
Furthermore, within-subjects contrasts indicated that the
Discovery Rate (FDR) method [27], unless indicated
number of correct responses was significantly lower for
otherwise. Stimulus x condition interaction effects are
fasting-food than for fasting-nonfood items (F (1,15) = 4.56,
reported at p<0.001 uncorrected masked with the relevant
p = 0.05,
2 = 0.23). In addition, there was neither any
main effect. Additionally, analyses of covariance were
performed using hormone measurements as regressors. significant difference between the number of correct
responses under the fasting and satiated conditions
Anatomical regions as identified by Montreal Neurological
separately analyzed for the food and non-food condition nor
Institute (MNI) coordinates for peak effects were verified
for those averaged across food and non-food conditions.
using a standard brain atlas.
Laboratory Measurements
Biochemical Measurements
Mean glucose levels did not differ between the fasting
Serum leptin and ghrelin were measured with a radio-
immunoassay (Linco Research Inc, St. Charles, Missouri, and satiated state in this study population. As expected,
during fasting mean levels of ghrelin and FFA were
USA). The intra-assay coefficient of variation (CV) for
significantly higher compared to the satiated state. During
leptin was 3% at 25 ng/ml and 8% at 5 ng/ml and the inter-
the satiated state, insulin and triglyceride levels were higher
assay CV was 4% at 25 ng/ml and 8% at 5 ng/ml. For ghrelin
than during the fasting state. Leptin levels were also slightly
the intra-assay CV was 5% at 3000 ng/l, 8% at 2000 ng/l and
higher during satiation (Table 1).
10% at 1000 ng/l and the inter-assay CV was 5%. Insulin
was measured using the Immunoradiometric assay Table 1. Laboratory Measurements in Healthy Men During
Biosource/Medgenix Diagnostics, Fleurus, Belgium. The Fasting and Satiated Conditions presented as Mean
intra-assay CV is 2% at 318 pmol/l and 5% at 40 pmol/l and Levels with Standard Deviation
inter-assay CV 6%. Serum concentrations of glucose were
measured with Hexokinase method (Roche diagnostics, Glucose (mmol/l) 4.79 ± 0.30 4.57 ± 0.33
Mannheim, Germany) with the inter-assay CV of <2% at the
Triglycerides (mmol/l) 0.77 ± 0.33 1.35 ± 0.57**
mean values of 4.7 and 18.3 mmol/l. Triglycerides were
measured with the Enzymatic colorimetric assay (Roche Free fatty acids (mmol/l) 0.31 ± 0.11 0.15 ± 0.11**
diagnostics, Mannheim, Germany). The inter-assay CV was
Insulin (pmol/l) 46.8 ± 21.5 245.7 ± 127.3**
<4% at 1 mmol/l and <3% at 2 mmol/l. FFA were measured
with an enzymatic colorimetric test (ELAN, Merck, Leptin (ng/ml) 3.0 ± 1.5 3.5 ± 2.1*
Darmstadt, Germany), the inter-assay CV was <4.5% at 0.73
Ghrelin (ng/l) 1826.7 ± 402 1386.9 ± 223.8**
mmol/l and <4% at 1.02 mmol/l.
* p < 0.05, ** p < 0.01.
8 The Open Neuroendocrinology Journal, 2012, Volume 5 Jakobsdottir et al.

Imaging Data During the fasting condition there was a positive

correlation between ghrelin levels and activation of the right
Results of imaging data are summarized in Table 2. During
and left insula, posterior OFC, left thalamus and right
encoding main effects for food related stimuli vs non-food
hippocampus but these correlations did not reach
related stimuli were observed in the left fusiform gyrus and the
significance.
right anterior and posterior hippocampus during the fasting
state, and the occipital gyrus bilaterally. Amygdala activity only Correlations Between Appetite Assessment Score and
approached significance. During retrieval there was significant Laboratory Data
activation in the fusiform gyrus and occipital gyrus bilaterally
Under the fasting condition, a significant correlation was
and the right medial frontal cortex. The right anterior
found between ghrelin levels and ‘thought of food’ scale
hippocampus approached significance during the satiated state.
scores (r = 0.48, p = 0.04, one-tailed). In addition, higher
Interaction effects (increased activation of food vs non-food
pictures during the fasting compared to the satiated condition) ghrelin concentrations were related to lower insulin
concentrations (r = - 0.55, p = 0.035, two-tailed). Finally,
were observed during encoding in right occipital cortex and
glucose correlated positively with insulin (r = 0.54, p =
bilateral fusiform gyrus and amygdala (Table 3). During
0.047, two-tailed) and with leptin (r = 0.60, p = 0.02, two-
retrieval there was more activation during the satiated condition
tailed).
compared to the fasting state in the left fusiform gyrus and
occipital cortex. With respect to the satiated condition, ghrelin concentrat-
ions correlated negatively with scores on the ‘eagerness to
Correlations Between Imaging and Laboratory Data eat’ scale (r = - 0.61, p = 0.008). Finally, ghrelin correlated
negatively with insulin (r = - 0.58, p = 0.02, two-tailed) and
During the satiated condition a strong negative
glucose correlated positively with triglycerides (r = 0.66, p <
correlation was found between leptin levels and activation of
0.007, two-tailed).
the left hippocampus (x=-21, y=-21, z=-9 peak Z score:5.09)
and activation of the right insula (x=33, y=18, z=-3 peak Z DISCUSSION
score:4.74) (Fig. 1). There was a negative correlation be-
In the present study we investigated the effects of
tween leptin levels and the temporal lobe bilaterally (x=-57,
processing visual food vs non-food stimuli on brain
y=-60, z=-3 peak Z score:4.7 and x=66, y=-33, z=9 peak Z
activation during standardized fasting and satiated
score:3.93 respectively) as well as the frontal gyrus (x=27,
conditions. In order to mimic a physiological fasting state in
y=48, z=9 peak Z score:3.90).
the brain, we adopted a 12-hour fasting period, rather than an

Table 2. Regions Showing an Increase in Brain Activity (Blood Oxygen Level Dependent Contrasts) in Response to Visual Food
Versus Non Food Stimuli

Fasting+Satiated Fasting Satiated

Left/Right
x y z Peak Z x y z Peak Z x y z Peak Z

Encoding
Amygdala L -18 -9 -21 2.89b) -18 0 -12 3.07a)

R 33 0 -24 2.92b)

Anterior hippocampus R 30 -12 -24 2.61b)

Posterior hippocampus R 30 -24 -6 2.58b)

Fusiform gyrus L -36 -84 -12 4.20

R 36 -51 -18 4.50a) 30 -84 3 4.52

Occipital cortex L -36 -84 -12 4.06a) -36 -84 -12 4.20

R 39 -75 -9 3.90a) 42 -54 -18 4.42

Retrieval

Anterior hippocampus R 33 -9 -27 2.71b) 36 -12 -24 3.02b)

Fusiform gyrus L -45 -57 -21 4.28 -45 -57 -21 3.13a) -30 -75 -12 4.94

R 36 -63 -15 4.00 45 -45 -18 3.56a) 30 -72 -6 4.24

Occipital cortex L -33 -81 -9 4.76 -33 -81 -9 3.75a) -27 -81 -12 4.45

R 33 -75 -15 3.84 33 -75 15 3.22a) 18 -90 -9 3.45

Medial frontal cortex R 3 51 9 3.32 3 51 9 3.52a)

x, y, z = coordinates of peak voxel from the Montreal Neurological Institute (MNI) brain peak Z = standardized significant value.
Main effects significant corrected for multiple comparisons (False Discovery Rate method) unless indicated otherwise
a) P uncorrected <0.001
b) P uncorrected <0.005
Food Perception and Brain Activity The Open Neuroendocrinology Journal, 2012, Volume 5 9

Table 3. Regions Showing Group Interactions in Response to Visual Food Versus Non Food Stimuli

Fasting>Satiated Satiated>Fasting

Left/Right x y z Peak Z x y z Peak Z

Encoding

Amygdala L -18 -6 -24 2.66

R 33 0 -24 2.74
Fusiform gyrus L -33 -84 -6 3.25

-45 -60 -6 3.38

R 42 -54 -18 3.55

Occipital cortex R 30 -84 3 4.73

Retrieval

Fusiform gyrus L -27 -75 -12 3.42

Occipital cortex L -12 -90 -6 3.58

x, y, z = coordinates of peak voxel from the Montreal Neurological Institute (MNI) brain peak Z = standardized significant value.
Main effects significant corrected for multiple comparisons (False Discovery Rate method) unless indicated otherwise.
P <0.001.

extremely long fasting period as used by others [28]. Also, others [9], although we suggest that this could be due to
we performed a second measurement one hour after greater semantic cohesion of the food stimuli [34] relative to
ingestion of a standardized meal, expecting the acute nonfood stimuli.
metabolic and hormonal changes following a meal to be
Previous research has demonstrated that the gut-brain
effective in producing a satiated state. Functional MRI
axis and various hypothalamic factors are of importance in
sessions were performed in balanced order, with a one-week
regulating the energy balance and the perception of hunger
interval to separate the two conditions (fasting and satiated)
and satiety [15,35-37]. In the present study, we found a
to control for order effects. Main effects of food vs non-food positive correlation between ghrelin and food-related
stimuli were observed during encoding in the ventral visual
activation in the insula areas bilaterally, as well as in the
stream, including fusiform gyrus and parahippocampal areas
right hippocampus, during the fasting state although these
bilaterally, although hippocampus and amygdala were only
associations did not reach significance. As ghrelin is the only
found at a lower threshold. Given the fact that food and non-
orexigenic hormone [38,39] this association would be
food pictures were matched for visual complexity, these
expected. Also, it is known to stimulate meal initiation, and
effects are likely to be due to increased salience of food receptors for ghrelin have been located in the hippocampus,
pictures, in particular when subjects were hungry. These
arcuate nucleus and ventromedial hypothalamus. When
results are in accordance with previous reports investigating
correlating the hunger scores with ghrelin we found a
visual processing of food stimuli using fMRI [10,29]. In
positive correlation between ghrelin and “thought of food”
addition to the medial temporal regions observed in the
hunger scores, consistent with the findings of others [13].
present study, the orbitofrontal cortex has been demonstrated
to be involved in feeding related behaviour both in animals In the satiated condition, there was a strong negative
and humans [23,30,31]. In our study, amygdala activity was correlation between leptin and left hippocampus and right
observed for food vs non-food pictures and interaction insula activity. Thus, high leptin levels correlated with a
analyses showed that amygdala activity was increased in the lowered response to food related pictures. The insula has
fasting state as expected. Due to susceptibility artefacts been implicated in multiple processes, including
(signal loss due to the presence of bone-air transitions, in interoceptive awareness of body states, food craving, and
particular nasal sinuses) we were not able to adequately basic emotions. Hippocampal activation is typically
measure BOLD activation in the medial OFC; also, ventral increased during memory processes (registration of new
striatum activity was not found even though the ventral information and retrieval), and our data indicate that high
striatum is likely to be involved in signalling rewarding levels of leptin are associated with a blunted response.
properties of food [25]. For this reason, some researchers Stimulation of the hypothalamus by leptin results in the
have used 15O-PET rather than fMRI when investigating suppression of food intake, stimulation of satiated behaviour,
OFC function [31-33]. In addition, detecting amygdala and energy expenditure [40-43]. Also, in obese subjects,
activity may be problematic due to rapid habituation [10]. correlations have recently been reported between leptin and
regional grey matter volumes [44].
During the two conditions of fasting and satiation, mean
reaction time for food pictures was longer than for non-food In the present study, we did not find significant
pictures. Also, memory performance for food-pictures was correlations between insulin and activation of limbic areas
lower than for non-food pictures. The latter finding was during the fasting state as might be expected. Nevertheless,
somewhat unexpected given the association between salience interpretations of the correlations between insulin and
and memory performance for food stimuli as reported by regional brain activation areas are not straightforward as
10 The Open Neuroendocrinology Journal, 2012, Volume 5 Jakobsdottir et al.

Fig. (1). The negative correlation between leptin levels and activation of the left hippocampus for the food vs non-food contrast. Colour bar
shows statistical T-values. See text for details.

insulin signalling in the brain is highly complex, sharing In summary, in the present study, main effects of food
common pathways with leptin and serotonin [45,46]. versus non-food visual stimuli during encoding were
observed in the ventral visual stream, including fusiform
In contrast to the findings of Gautier et al, we did not
gyrus and hippocampal areas and occipital cortex bilaterally,
find correlations between postprandial FFA and activation in
and at lower threshold the amygdala areas. There was
hippocampal and parahippocampal regions [28]. Also, there
significantly more activation in the fusiform and
were no significant correlations between glucose and the
activation in the limbic system during the fasting and hippocampus gyrus bilaterally during the fasting condition
when compared to the satiation state, most likely due to
satiated condition when viewing food vs nonfood food
increased salience of food pictures. Furthermore, there was a
pictures.
strong negative correlation between leptin and left
Our study has several potential limitations. The sample hippocampus and right insula activity when the subjects
size was only moderate, although sample sizes of 12-15 are were satiated.
customary in fMRI studies. Lately it has been shown to be of
Further insights in the neural correlates of processing
importance to categorize the visual food stimuli in low
food stimuli in obesity, binge eating and anorexia nervosa
versus high calorie stimuli as this could influence the results
will be of importance in the search for individualized
especially during the fasting state [47]. Also, recently
treatment such as behavioural counselling and in the search
published data have implicated adiponectin in regulating
for effective drug treatments.
food intake and energy expenditure [48,49]. Moreover,
glucagon-like peptide-1 (GLP-1), which is synthesized in the ACKNOWLEDGEMENTS
brain as well [50,51], may also be involved. Therefore,
future research should attempt to investigate the correlations Declared none.
between these hormones/adipokines and regional brain CONFLICT OF INTEREST
activation using a similar fMRI paradigm.
Declared none.
Finally, various local regulatory circuits and locally
produced/derived adipokines are presumably involved in the REFERENCES
processing of food intake regulation and appetite, but such [1] Rolls ET, Yaxley S, Sienkiewicz ZJ. Gustatory responses of single
factors are as yet difficult to assess. This regulation is likely neurons in the caudolateral orbitofrontal cortex of the macaque
to be highly complex, involving the dopaminergic system, monkey. J Neurophysiol 1990; 64(4): 1055-66.
and psychosocial factors including stress [52-55].
Food Perception and Brain Activity The Open Neuroendocrinology Journal, 2012, Volume 5 11

[2] Nishijo H, Ono T, Nishino H. Single neuron responses in amygdala [27] Genovese CR, Lazar NA, Nichols T. Thresholding of statistical
of alert monkey during complex sensory stimulation with affective maps in functional neuroimaging using the false discovery rate.
significance. J Neurosci 1988; 8(10): 3570-83. Neuroimage 2002; 15: 870-8.
[3] Killgore WD, Young AD, Femia LA, Bogorodzki P, Rogowska J, [28] Gautier JF, Chen K, Salbe AD, et al. Differential brain responses to
Yurgelun-Todd DA. Cortical and limbic activation during viewing satiation in obese and lean men. Diabetes 2000; 49: 838-46.
of high- versus low-calorie foods. Neuroimage 2003; 19: 1381-94. [29] Killgore WD, Yurgelun-Todd DA. Body mass predicts
[4] LeDoux JE. Emotion circuits in the brain. Annu Rev Neurosci orbitofrontal activity during visual presentations of high-calorie
2000; 23: 155-84. foods. Neuroreport 2005; 16: 859-63.
[5] Rolls ET. The orbitofrontal cortex and reward. Cereb Cortex 2000; [30] Rolls ET. The orbitofrontal cortex. Philos Trans R Soc Lond B Biol
10: 284-94. Sci 1996; 351: 1433-43.
[6] Tataranni PA, DelParigi A. Functional neuroimaging: a new [31] Tataranni PA, Gautier JF, Chen K, et al. Neuroanatomical
generation of human brain studies in obesity research. Obes Rev correlates of hunger and satiation in humans using positron
2003; 4: 229-38. emission tomography. Proc Natl Acad Sci U S A 1999; 96: 4569-
[7] Bechara A, Damasio H, Damasio AR, Lee GP. Different 74.
contributions of the human amygdala and ventromedial prefrontal [32] Frey S, Petrides M. Orbitofrontal cortex: A key prefrontal region
cortex to decision-making. J Neurosci 1999; 19: 5473-81. for encoding information. Proc Natl Acad Sci U S A 2000; 97:
[8] Baxter MG, Parker A, Lindner CC, Izquierdo AD, Murray EA. 8723-7.
Control of response selection by reinforcer value requires [33] Morris JS, Frith CD, Perrett DI, et al. A differential neural response
interaction of amygdala and orbital prefrontal cortex. J Neurosci in the human amygdala to fearful and happy facial expressions.
2000; 20: 4311-9. Nature 1996; 383: 812-5.
[9] Morris JS, Dolan RJ. Involvement of human amygdala and [34] Maratos EJ, Allan K, Rugg MD. Recognition memory for
orbitofrontal cortex in hunger-enhanced memory for food stimuli. J emotionally negative and neutral words: an ERP study.
Neurosci 2001; 21: 5304-10. Neuropsychologia 2000; 38: 1452-65.
[10] LaBar KS, Gitelman DR, Parrish TB, Kim YH, Nobre AC, [35] de Fatima Haueisen Sander Diniz M, de Azeredo Passos VM, Diniz
Mesulam MM. Hunger selectively modulates corticolimbic MT. Gut-brain communication: how does it stand after bariatric
activation to food stimuli in humans. Behav Neurosci 2001; 115: surgery? Curr Opin Clin Nutr Metab Care 2006; 9: 629-36.
493-500. [36] Dhillo WS. Appetite regulation: an overview. Thyroid 2007; 17:
[11] Cummings DE, Overduin J. Gastrointestinal regulation of food 433-45.
intake. J Clin Invest 2007; 117: 13-23. [37] Romijn JA, Corssmit EP, Havekes LM, Pijl H. Gut-brain axis. Curr
[12] de Graaf C, Blom WA, Smeets PA, Stafleu A, Hendriks HF. Opin Clin Nutr Metab Care 2008; 11: 518-21.
Biomarkers of satiation and satiety. Am J Clin Nutr 2004; 79: 946- [38] Cummings DE. Ghrelin and the short- and long-term regulation of
61. appetite and body weight. Physiol Behav 2006; 89: 71-84.
[13] Cummings DE, Frayo RS, Marmonier C, Aubert R, Chapelot D. [39] Malik S, McGlone F, Bedrossian D, Dagher A. Ghrelin modulates
Plasma ghrelin levels and hunger scores in humans initiating meals brain activity in areas that control appetitive behavior. Cell Metab
voluntarily without time- and food-related cues. Am J Physiol 2008; 7: 400-9.
Endocrinol Metab 2004; 287: E297-304. [40] Ahima RS, Prabakaran D, Mantzoros C, et al. Role of leptin in the
[14] Figlewicz DP. Insulin, food intake, and reward. Semin Clin neuroendocrine response to fasting. Nature 1996; 382: 250-2.
Neuropsychiatry 2003; 8: 82-93. [41] Bloom S. Hormonal regulation of appetite. Obes Rev 2007; 8
[15] Ahima RS, Lazar MA. Adipokines and the peripheral and neural (Suppl 1): 63-5.
control of energy balance. Mol Endocrinol 2008; 22: 1023-31. [42] Karhunen LJ, Lappalainen RI, Vanninen EJ, Kuikka JT, Uusitupa
[16] Viveros MP, de Fonseca FR, Bermudez-Silva FJ, McPartland JM. MI. Serum leptin and regional cerebral blood flow during exposure
Critical role of the endocannabinoid system in the regulation of to food in obese and normal-weight women. Neuroendocrinology
food intake and energy metabolism, with phylogenetic, 1999; 69: 154-9.
developmental, and pathophysiological implications. Endocr Metab [43] Meier U, Gressner AM. Endocrine regulation of energy
Immune Disord Drug Targets 2008; 8: 220-30. metabolism: review of pathobiochemical and clinical chemical
[17] Farr SA, Banks WA, Morley JE. Effects of leptin on memory aspects of leptin, ghrelin, adiponectin, and resistin. Clin Chem
processing. Peptides 2006; 27: 1420-5. 2004; 50: 1511-25.
[18] Tang BL. Leptin as a neuroprotective agent. Biochem Biophys Res [44] Pannacciulli N, Le DS, Chen K, Reiman EM, Krakoff J.
Commun 2008; 368: 181-5. Relationships between plasma leptin concentrations and human
[19] Harvey J. Leptin regulation of neuronal excitability and cognitive brain structure: a voxel-based morphometric study. Neurosci Lett
function. Curr Opin Pharmacol 2007; 7: 643-7. 2007; 412: 248-53.
[20] Funahashi H, Yada T, Suzuki R, Shioda S. Distribution, function, [45] Gerozissis K. Brain insulin, energy and glucose homeostasis;
and properties of leptin receptors in the brain. Int Rev Cytol 2003; genes, environment and metabolic pathologies. Eur J Pharmacol
224: 1-27. 2008; 585: 38-49.
[21] Farooqi IS, Bullmore E, Keogh J, Gillard J, O'Rahilly S, Fletcher [46] Sharma MD, Garber AJ, Farmer JA. Role of insulin signaling in
PC. Leptin regulates striatal regions and human eating behavior. maintaining energy homeostasis. Endocr Pract 2008; 14: 373-80.
Science 2007; 317: 1355. [47] Goldstone AP, de Hernandez CG, Beaver JD, et al. Fasting biases
[22] Baicy K, London ED, Monterosso J, et al. Leptin replacement brain reward systems towards high-calorie foods. Eur J Neurosci
alters brain response to food cues in genetically leptin-deficient 2009; 30: 1625-35.
adults. Proc Natl Acad Sci U S A 2007; 104: 18276-9. [48] Psilopanagioti A, Papadaki H, Kranioti EF, Alexandrides TK,
[23] Del Parigi A, Gautier JF, Chen K, et al. Neuroimaging and obesity: Varakis JN. Expression of adiponectin and adiponectin receptors in
mapping the brain responses to hunger and satiation in humans human pituitary gland and brain. Neuroendocrinology 2009; 89:
using positron emission tomography. Ann N Y Acad Sci 2002; 38-47.
967: 389-97. [49] Rodriguez-Pacheco F, Martinez-Fuentes AJ, Tovar S, et al.
[24] Pelchat ML, Johnson A, Chan R, Valdez J, Ragland JD. Images of Regulation of pituitary cell function by adiponectin. Endocrinology
desire: food-craving activation during fMRI. Neuroimage 2004; 23: 2007; 148: 401-10.
1486-93. [50] Lerche S, Brock B, Rungby J, et al. Glucagon-like peptide-1
[25] Rolls ET, McCabe C. Enhanced affective brain representations of inhibits blood-brain glucose transfer in humans. Diabetes 2008; 57:
chocolate in cravers vs non-cravers. Eur J Neurosci 2007; 26: 1067- 325-31.
76. [51] Sandoval DA, Bagnol D, Woods SC, D'Alessio DA, Seeley RJ.
[26] Hill AJ, Rogers PJ, Blundell JE. Techniques for the experimental Arcuate glucagon-like peptide 1 receptors regulate glucose
measurement of human eating behaviour and food intake: a homeostasis but not food intake. Diabetes 2008; 57: 2046-54.
practical guide. Int J Obes Relat Metab Disord 1995; 19: 361-75.
12 The Open Neuroendocrinology Journal, 2012, Volume 5 Jakobsdottir et al.

[52] Pliquett RU, Fuhrer D, Falk S, Zysset S, von Cramon DY, [54] Wilkinson M, Brown R, Imran SA, Ur E. Adipokine gene
Stumvoll M. The effects of insulin on the central nervous system-- expression in brain and pituitary gland. Neuroendocrinology 2007;
focus on appetite regulation. Horm Metab Res 2006; 38: 442-6. 86: 191-209.
[53] Rothemund Y, Preuschhof C, Bohner G, et al. Differential [55] Woods SC. Signals that influence food intake and body weight.
activation of the dorsal striatum by high-calorie visual food stimuli Physiol Behav 2005; 86: 709-16.
in obese individuals. Neuroimage 2007; 37: 410-21.

Received: November 24, 2011 Revised: February 14, 2012 Accepted: February 17, 2012

© Jakobsdottir et al.; Licensee Bentham Open.

This is an open access article licensed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-
nc/3.0/) which permits unrestricted, non-commercial use, distribution and reproduction in any medium, provided the work is properly cited.