![Mean scores at the VAS-P in the sample by Exposure Condition. Error bars represent standard error (SE). Pairwise comparisons (with Bonferroni adjustment) indicated that the RE food stimuli were significantly ( p < 0.05) more palatable than the PH stimuli (VAS-P RE food stimuli: 46.7 [SD = 3.64]; VAS-P PH food stimuli: 36.7 [SD = 3.91]). No difference was found in the food's palatability between RE and AR (42.7 [SD = 4.14]) or between PH and AR. *p < 0.01.](https://www.researchgate.net/profile/Giuseppe-Riva-2/publication/292577447/figure/fig2/AS:330245375840257@1455748041617/Mean-scores-at-the-VAS-P-in-the-sample-by-Exposure-Condition-Error-bars-represent_Q320.jpg)
Content uploaded by Giuseppe Riva
Author content
All content in this area was uploaded by Giuseppe Riva on Feb 17, 2016
Content may be subject to copyright.
Testing Augmented Reality
for Cue Exposure in Obese Patients:
An Exploratory Study
Federica Pallavicini, PhD,
1
Silvia Serino, PhD,
1
Pietro Cipresso, PhD,
1
Elisa Pedroli, PSYD,
1
Irene Alice Chicchi Giglioli, MA,
1
Alice Chirico, MA,
1
Gian Mauro Manzoni, PhD,
2,3,
*
Gianluca Castelnuovo, PhD,
2,3
Enrico Molinari, PhD,
2,3
and Giuseppe Riva, PhD
1,3
Abstract
Binge eating is one of the key behaviors in relation to the etiology and severity of obesity. Cue exposure with
response prevention consists of exposing patients to binge foods while actual eating is not allowed. Augmented
reality (AR) has the potential to change the way cue exposure is administered, but very few prior studies have
been conducted so far. Starting from these premises, this study was aimed to (a) investigate whether AR foods
elicit emotional responses comparable to those produced by the real stimuli, (b) study differences between
obese and control participants in terms of emotional responses to food, and (c) compare emotional responses to
different categories of foods. To reach these goals, we assess in 15 obese (age, 44.6 – 13 years; body mass index
[BMI], 44.2 – 8.1) and 15 control participants (age, 43.7 – 12.8 years; BMI, 21.2 – 1.4) the emotional responses
to high-calorie (savory and sweet) and low-calorie food stimuli, presented through different exposure conditions
(real, photographic, and AR). The State-Trait Anxiety Inventory was used for the assessment of state anxiety,
and it was administered at the beginning and after the exposure to foods, along with the Visual Analog Scale
(VAS) for Hunger and Happiness. To assess the perceived pleasantness, the VAS for Palatability was ad-
ministered after the exposure to food stimuli. Heart rate, skin conductance response, and facial corrugator
supercilii muscle activation were recorded. Although preliminary, the results showed that (a) AR food stimuli
were perceived to be as palatable as real stimuli, and they also triggered a similar arousal response; (b) obese
individuals showed lower happiness after the exposure to food compared to control participants, with regard to
both psychological and physiological responses; and (c) high-calorie savory (vs. low-calorie) food stimuli were
perceived by all the participants to be more palatable, and they triggered a greater arousal response.
Introduction
O
besity is a major public health problem worldwide, and
its prevalence in the world population is dramatically
increasing.
1
Numerous studies have pointed out that obesity
is a multifactorial disorder, including psychological factors,
such as anxiety and distress, that involve individuals in a
variety of behaviors that serve to regulate their emotions.
2
A
key behavior in relation to the etiology and severity of obesity
is binge eating (BE),
3,4
a dysfunctional behavior that can be
triggered by food craving,
5
which is an appetitive motiva-
tional–emotional state that triggers the search for food and
the consequent intake behaviors even in a state of satiety.
The conditioning model of BE postulates that, in addition
to internal states (e.g., psychological emotional states), ex-
posure to certain stimuli associated with BE (e.g., the pres-
ence of high-calorie meals) provokes physiological
responses and subjective craving. Cue exposure with re-
sponse prevention of binge eating (ERP-B) is a type of
treatment derived from this model, which consists of ex-
posing patients to binge-triggering foods and preventing BE
while the food is present and can be smelled.
6–9
The main objective of ERP-B is to extinguish food craving
by breaking the link between the conditioned and the un-
conditioned stimulus (for a recent review, see Ref.
10
). While
the cue exposure approach has been proven to be an effective
1
Applied Technology for Neuro-Psychology Lab, IRCCS Istituto Auxologico Italiano, Milan, Italy.
2
Psychology Research Laboratory, Ospedale San Giuseppe, Istituto Auxologico Italiano, Verbania, Italy.
3
Department of Psychology, Catholic University of the Sacred Heart, Milan, Italy.
*Current affiliation: Faculty of Psychology, eCampus University, Novedrate, Como, Italy.
CYBERPSYCHOLOGY,BEHAVIOR, AND SOCIAL NETWORKING
Volume 19, Number 2, 2016
ª Mary Ann Liebert, Inc.
DOI: 10.1089/cyber.2015.0235
107
treatment of BE behaviors,
8,9
it has important constraints that
make its implementation difficult, such as logistical diffi-
culties, the time required to carry out the treatment, and the
need for natural environments adapted to the requirements of
each patient.
10
Among the technologies that have emerged in more
recent years, augmented reality (AR) represents an inno-
vative tool that has the potential to dramatically change the
way psychological treatments are administered,
11
includ-
ing cue exposure treatments.
12
Essentially, the concept of
AR is that synthetic objects can be added to the real world
in real time, enriching reality with helpful and relevant
information.
13–17
Despite the opportunities that AR could offer in the ad-
ministration of cue exposure, including the prevention of BE
behavior in obesity, very few prior studies have compared
emotional reactions to AR and real stimuli. Nevertheless, a
necessary condition for the successful use of AR in cue ex-
posure is that it must elicit roughly the same emotional re-
action as a real-life stimulus. In particular, the studies
conducted so far have mainly investigated the level of
presence, defined as ‘‘the feeling of being in a world that
exists outside of the self,’’
18,19
as experienced in AR sys-
tems, compared to that experienced in a real environment
20
and virtual reality (VR) environments.
21
To our knowledge, only one previous study has investigated
the differences in terms of emotional responses to stimuli
presented through AR compared to other exposure meth-
ods—particularly VR
21
—while no study has yet compared
the responses with respect to those shown in response to real-
world stimuli. Moreover, previous studies have focused on
AR only in terms of exposure to phobic stimuli, showing the
usefulness of AR in the treatment of a specific phobia, such
as cockroach and spider phobias,
15,22–25
and acrophobia
20,21
(for a recent review, see Ref.
11
). As yet, nothing is known
about its possible application of AR for cue exposure in the
treatment of other conditions, including obesity.
Starting from these premises, the first goal of this study was
to investigate whether AR food stimuli elicit emotional re-
sponses comparable to those produced by real exposure
stimuli. The second aim of this study was to investigate dif-
ferences between obese and control group participants in
terms of emotional responses to food stimuli. Finally, the third
aim was to compare emotional responses to the different ca-
tegories of food stimuli adopted to extend to which the effects
of cue exposure are related to a specific category of food.
Materials and Methods
Subjects
The experimental sample included 15 obese (OB) patients
(7 males and 8 females), as well as a control group (CTR) of
15 individuals (7 males and 8 females). OB patients were
recruited from the inpatient units of a public hospital in
Verbania (Italy), while CTR subjects were recruited through
local advertisements among college students and adminis-
trative and workers’ staff at the hospitals. The mean body
mass index (BMI)
26
was 44.2 – 8.1 in the OB group (fre-
quency of OB Class II = 33.3%; Class III = 66.7%) and
21.2 – 1.4 in the CTR group. The mean age and years of
education were 44.6 – 13 and 11 – 3.83 in the OB group and
43.7 – 12.8 and 15.7 – 2.43 in the CTR group, respectively.
To be included in the study, individuals were required to
meet the following criteria: (a) no Axis I disorders as defined
in the Diagnostic and Statistical Manual of Mental Dis-
orders, Fourth Edition (DSM-IV-TR)
27
; (b) age between 18
and 60 years; (c) not to be presently using pharmacotherapy;
(d) no history of neurological diseases, psychosis, alcohol,
or drug dependence; (e) no migraine, headache, or vestibular
abnormalities; (f) no food allergies or intolerances; (g) not
currently dieting for weight loss or in any dietary restric-
tions; and (h) no BMI less than 18.5.
A semistructured clinical interview and the Mini Inter-
national Neuropsychiatric Interview Plus (MINI)
28,29
were
used to exclude the presence of any psychiatric diseases,
including actual or past eating disorders, according to the
DSM-IV-TR.
27
Before participating in the study, each par-
ticipant was provided with written information about the
study and required to give written consent for inclusion in the
study. The study received ethical approval by the Ethical
Committee of the Istituto Auxologico Italiano.
Psychological assessment
The following questionnaires were administered to each
participant:
State-Trait Anxiety Inventory, Form Y-1 (STAI-
Y1):
30,31
The STAI-Y1 is a validated and widely used
measure of state anxiety.
Visual Analog Scale (VAS):
32
The VAS is a hori-
zontal line, 100 mm in length, anchored by word de-
scriptors at each end. The participants mark on the
line the point that they feel represents their perception
of their current state. Participants’ level of hunger
was measured using Visual Analog Scale for Hunger
(VAS-H), a rating scale headed ‘‘How strong is your
hunger right now?’’ A similar scale, the Visual
Analog Scale for Happiness (VAS-HP), was also
used. All the scales were anchored by the phrases
‘‘Not at all’’ and ‘‘Extremely.’’ Moreover, since the
hedonic component of sensation (or the ‘‘pleasant-
ness’’) of food stimuli is an integral part of their
sensory profile and it can influence appetite and eat-
ing rate,
33–36
the Visual Analog Scale for Palatability
(VAS-P) was administered.
Physiological assessment
At the beginning of the experiment and during the labo-
ratory session, skin conductance response (SCR), heart rate
(HR; in particular, NN50 values), and facial corrugator su-
percilii muscle electromyography (fEMG) were recorded to
obtain participants’ physiological responses to food stimuli.
While SCR
37
and NN50
38
are considered indexes of arousal
responses, fEMG is considered a good measure of negative
emotional state.
39,40
The physiological signals were acquired using a ProComp
Infiniti device from Thought Technology, including Biograph
Infiniti 5.0.2 software to record and export all raw signals.
Every signal was synchronously acquired at 2,048 Hz and
exported at a minimum of a 256 Hz sampling rate (every
3.9 ms). Amplitude is measured in microvolts.
108 PALLAVICINI ET AL.
Food stimuli and exposure conditions
We selected nine food items—three in each of the three
categories—based on nutritional estimates (i.e., total fat and
saturated fat):
high-calorie savory foods (HC-SAVs) included three
salty foods (salami, potato chips, and crackers) rela-
tively high in total fat (>20 g/100 g) and saturated fat
(>6.25 g/100 g),
high-calorie sweet foods (HC-SW) consisted of three
sweets (chocolates, cookies, and muffins) containing at
least moderate levels of total fats (>10 g/100 g) and
saturated fats (>5 g/100 g), and
low-calorie (LC) foods were three food stimuli (pears,
tomatoes, and carrots) containing low total fats (<4g/
100 g) and saturated fats (<1.25 g/100 g).
Food stimuli were presented in three exposure conditions
(Fig. 1): (a) real (RE), (b) photographic (PH), and (c) AR.
Experimental procedure
The order of presentation of each exposure condition (RE,
PH, and AR), as well as the order of appearance of each food
category (HC-SAV, HC-SW, and LC) within the three dif-
ferent conditions, was counterbalanced for each participant
following a previously established randomization schema.
Participants were told to freely examine the food stimuli,
without any specific request. All participants were tested not
more than 4 hours after a meal to avoid effects related to
excessive hunger or overeating.
Before the start of the experiment, participants were
administered the MINI interview. Once this first phase was
completed, there was a 3-minute baseline during which
subjects were asked to stay completely relaxed while their
physiological parameters were recorded. Once the physi-
ological baselines were recorded, participants were also
asked to complete the STAI-Y1, the VAS-H, and the VAS-
HP. Thereafter, the experimental session started, and SCR
and fEMG were continuously recorded until the end of the
experiment.
To measure the variations that occurred during the three
different exposure conditions, subjects completed the STAI-
Y1 and the VAS-H, and the VAS-HP again immediately
after each session. In addition, at the end of each exposure
condition, the VAS-P was administered. There was a pause
of 5 minutes between exposure condition sessions.
FIG. 1. Experimental exposure conditions. Food stimuli were presented in the following exposure conditions. (A) Real (RE)
food stimuli: Each participant was exposed to real food, presented on a plate on a table in front of the subject. During the
pause, all food items were covered with red plastic lids, so that the subjects could not see them. (B) Photographic (PH) food
stimuli: Subjects were asked to view food stimuli presented in picture format. Photographs were adjusted to balance the images
for contrast and luminance using Photoshop software. Particular care was taken to ensure identical lighting and arrangement of
the plate across food items and portion sizes. Images (180 · 260 mm) were printed on A4 paper (210 · 297 mm), centered
vertically. (C) Augmented reality (AR) food stimuli: In the AR condition, subjects were asked to hold a plate that was marker
based. A camera positioned on the PC recognized the marker on the plate projecting each food onto the plate. Subjects were
able to move and turn the plate to explore and observe the food. In particular, the AR setting included the following hardware
units: (a) Microsoft’s HD LifeCam camera, which offers true HD capture in 720p resolution, able to capture 30 frames per
second (Microsoft, Redmond, WA, USA); (b) the marker to decode the AR stimulus; and (c) a portable computer (Acer Aspire
with Intel
Corei5, graphics processor NVIDIA GeForce GT 540M, and Bluetooth support).
AUGMENTED REALITY AND CUE EXPOSURE IN OBESE PATIENTS 109
Statistical analyses and experimental design
The analyses were performed using SPSS version 22.0
(Statistical Package for the Social Sciences for Windows,
Chicago, IL) for PC. All collected biosignals were analyzed
using MATLAB 7.0 (The Mathworks, Natick, MA) for the
signal processing and computation of psychophysiological
measures.
Independent t tests were conducted to verify the base-
line homogeneity of the sample in age. Since the sample
was characterized by statistically significant differences at
baseline, both psychological and physiological, differences
were calculated in STAI-Y1, VAS-H, VAS-HP, NN50,
SCR, and fEMG data measured after the exposure com-
pared to the corresponding baseline (delta scores).
Next, mixed ANOVAs were conducted to test whether the
psychological and/or physiological responses changed de-
pending on the group (OB or CTR), the exposure condition (RE,
PH, or VR), or the type of food (HC-SAV, HC-SW, or LC).
One participant in the OB group was excluded from the
SCR and fEMG analyses due to problems with the recording.
Regarding analyses of NN50, eight OB patients and five
CTR participants were excluded due to problems with the
ECG recording.
Finally, a 2 · 3 · 3(Group[OBandCTR]· Exposure
Condition [RE, PH, and AR] · Food Stimulus Category [HC-
SAV, HC-SW, and LC]) mixed ANOVA was conducted on
VAS-P scores assessed after the exposure to the food stimuli.
The level of significance was set at a = 0.05. Post hoc tests (with
Bonferroni adjustment) were carried out to compare significant
differences. In graphs, error bars represented standard errors.
Results
Psychological responses to food stimuli
First, analyses of the VAS-P showed a statistically sig-
nificant main effect of Exposure Condition, F(2, 56) = 5.1,
p < 0.01, g
2
p
¼ 0:154, and Food Stimulus Category, F(2, 56) =
7.5, p < 0.01, g
2
p
¼ 0:212 (Figs. 2 and 3 for detailed results).
Second, analyses of the VAS-HP revealed a statistically
significant main effect of Group, F(1, 28) = 5.9, p < 0.05,
g
2
p
¼ 0:176 (Fig. 4). No other statistically significant main
effects and/or interactions were observed.
Regarding the VAS-H and the STAI-Y1, we did not ob-
serve any statistically significant main effects or interactions
(Table 1).
Physiological responses to food stimuli
First, findings of the analyses of SCR showed a statisti-
cally significant main effect of Food Stimulus Category, F(2,
54) = 3.2, p < 0.05, g
2
p
¼ 0:108. Pairwise comparisons (with
Bonferroni adjustment) were not significant.
For SCR, we did not observe any other statistically sig-
nificant main effects and/or interactions of Exposure
Condition or Group. Second, findings on NN50 showed a
statistically significant main effect of Exposure Condition,
F(2, 30) = 3.79, p < 0.05, g
2
p
¼ 0:202. Pairwise comparisons
(with Bonferroni adjustment) indicated that RE food stimuli
were characterized by higher NN50 delta values, indicating
higher arousal responses, compared to PH food stimuli (RE:
20.6 [SD = 5.47]; PH: 5.33 [SD = 3.02]). No difference was
found between RE and AR (AR: 28 [SD = 10.1]). No other
statistically significant main effects or interactions were
found.
Finally, analyses of fEMG revealed a statistically sig-
nificant main effect of Group, F(1, 27) = 6.7, p < 0.01,
g
2
p
¼ 0:199. In particular, data showed that OB patients were
characterized by significantly higher fEMG values, indicating
higher negative emotional responses to food stimulus exposure,
compared to CTR participants (OB patients: 0.237 [SD = 0.489]
and CTR participants: -1.59 [SD = 0.987]). We did not observe
statistically significant main effects and/or interactions of Ex-
posure Condition or Food Stimulus Category.
Discussion
Regarding the first main goal of this study (i.e., to inves-
tigate whether AR food stimuli elicit emotional responses
comparable to those produced by real exposure), analyses of
psychological responses showed that stimuli presented in AR
were perceived by individuals to be as palatable as real
foods. In fact, the results revealed a significant difference in
the entire sample in palatability scores between the real and
photographic stimuli but not between the real and AR
stimuli.
FIG. 2. Mean scores at the VAS-P in the
sample by Exposure Condition. Error bars
represent standard error (SE). Pairwise
comparisons (with Bonferroni adjustment)
indicated that the RE food stimuli were
significantly ( p < 0.05) more palatable than
the PH stimuli (VAS-P RE food stimuli:
46.7 [SD = 3.64]; VAS-P PH food stimuli:
36.7 [SD = 3.91]). No difference was found
in the food’s palatability between RE and
AR (42.7 [SD = 4.14]) or between PH and
AR. *p < 0.01.
110 PALLAVICINI ET AL.
This result appears relevant since AR stimuli must be
perceived to be as palatable as real ones to be effective in cue
exposure. Palatability (the hedonic component of sensation
or the ‘‘pleasantness’’ of food) is, in fact, an integral part of
the sensory profile for a given food.
35,41
As reported in the
literature, palatable foods can stimulate appetite and eating
rate, leading to excessive food consumption.
33,34
With regard to physiological responses, analyses showed a
significantly higher arousal response (particularly in the
NN50) after real food stimulus exposure than after exposure
to photographic stimuli; values after exposure to real and AR
foods were very similar. Such a difference was not observed
in the other physiological measure related to arousal re-
sponse (SCR) and in the hunger experienced by individuals.
However, these results may be linked to a longer response
latency of SCR compared to the HR index
42
and the partic-
ipants’ difficulty in assessing their subjective hunger, re-
spectively. Finally, analyses showed no differences in the
other emotional responses: No differences were found be-
tween the exposure conditions regarding the VAS-HP and
the fMEG.
Taken together, these results offer preliminary evidence
that AR food stimuli were perceived to be as palatable as real
ones, and they triggered an arousal response (with regard to
NN50) similar to that shown in response to real stimuli. One
possible explanation for this result could be related to the
specific features of AR.
As stated by Botella et al.,
15
in particular, perceiving a
virtual object in a real environment can enhance the sense of
presence and reality judgment. In their study, she and her
colleagues used an AR system developed for cockroach
phobia,
15
and they analyzed the emotional responses (i.e.,
subjective level of anxiety) shown by individuals in response
to AR stimuli (i.e., virtual cockroaches). The results fit with
what would be expected if an in vivo exposure procedure had
been applied.
The first main finding of our study could have important
implications in relation to cue exposure for obese patients.
AR could potentially offer a more realistic and interactive
experience compared to fully immersive VR
12
by enhancing
the ecological validity of the ‘‘mixed reality’’
43,44
through
the direct interaction of the individuals and patients with
FIG. 3. Mean scores at the VAS-P in the
sample by Food Stimulus Category. Error
bars represent standard error (SE). Pairwise
comparisons (with Bonferroni adjustment)
indicated that HC-SAV stimuli (48.7
[SD = 4.07]) were significantly ( p < 0.01)
more palatable than LC stimuli (32.9
[SD = 3.99]). No difference was found in
the food’s palatability between HC-SW
(44.3 [SD = 4.55]) and HC-SAV stimuli or
between HC-SW and LC food stimuli.
HC-SW, high-calorie sweet food; LC, low
calorie. *p < 0.01.
FIG. 4. Mean delta scores at the VAS-HP
in the sample by Group. Error bars represent
standard error (SE). In particular, pairwise
comparisons (with Bonferroni adjustment)
indicated that OB patients showed signifi-
cantly ( p < 0.05) lower happiness delta
scores compared to CTR participants, indi-
cating more intense negative emotional re-
actions to food stimuli (OB patients: -6.84
[SD = 3.84]; CTR participants: 6.41
[SD = 3.81]). *p < 0.05. OB, obese.
AUGMENTED REALITY AND CUE EXPOSURE IN OBESE PATIENTS 111
virtual stimuli. Moreover, AR could become a part of obesity
rehabilitation programs. Such technology could make it
possible to administer cue exposure at home, enabling psy-
chologists and patients to save time and money and poten-
tially offering more engaging and interactive experiences
compared to traditional programs.
Regarding the second main aim of this study (i.e., to in-
vestigate differences between obese patients and control
group participants in terms of emotional responses to food
stimuli), we found that obese patients were characterized by
more intense negative emotional reactions to food stimuli
compared to control group participants.
Results on psychological responses showed that obese pa-
tients were characterized by lower happiness scores compared
to control group participants, regardless of the exposure con-
dition or the food stimulus category. Conversely, and in line
with what has been observed in a previous study,
36
control
group participants showed an increase in happiness after ex-
posure to food stimuli.
Analyses of the fEMG seem to confirm this result, showing
stronger activations in the obese patients than in the control
group participants and indicating higher negative emotional
responses to food stimuli.
39,40
A similar result has been also
reported in a previous study, in which obese participants
showed strong negative emotional responses to foods high in
fat.
45
One possible explanation of this observed result could be
related to the specific sample of obese patients included in this
study: The patients were hospitalized and undergoing diets for
weight management. Therefore, certain foods may have been
perceived by the obese participants as ‘‘prohibited,’’ thus
potentially triggering inhibitory emotional strategies.
Regarding the self-reported hunger and the physiological
values related to arousal responses (SCR and NN50), we did
not find any significant differences after exposure to food
stimuli between obese and control group participants. One
possible explanation could be that obese participants may
have been very hungry at the sight of food, but they lied
because of social stigma, creating bias in the VAS-H col-
lected data.
46,47
However, results regarding physiological arousal re-
sponses seem to disagree with this hypothesis: No differ-
ences were observed between obese and control group
participants in SCR and NN50 values. This result must be
interpreted with caution since in the literature there are still
mixed results regarding differences in terms of arousal in
response to food stimuli between obese patients and healthy
individuals. In particular, while one previous study reported
no differences
48
(similar to what we observed in this re-
search), another study found that obese patients were char-
acterized by higher arousal responses.
49
Finally, results showed that all the participants perceived
food stimuli to be more palatable and showed a greater
arousal response after exposure to high-calorie foods (par-
ticularly, the savory ones) versus low-calorie food stimuli.
Analyses of psychological responses showed that HC-SAV
stimuli were assessed as significantly more palatable than
low-calorie stimuli. Along these lines, analyses of physio-
logical responses showed greater arousal activation in the
whole sample (particularly with regard to SCR) after expo-
sure to HC-SAVs than after exposure to low-calorie foods,
regardless of exposure conditions.
Taken together, these results seem to confirm previous
studies reporting that the most palatable and arousing foods
are those that are both energy dense and high in fat content.
50–
53
High-calorie foods appear to be the most effective in in-
ducing an emotional response, and thus, they seem to be the
best type of food stimuli to use in cue exposure treatments.
Although our results are interesting for their possible ap-
plications to cue exposure treatments for obesity, our study
has some limitations that could affect the generalizability of
the results or that may have influenced the findings. The main
issues are related to the small sample size and the specific
sample of obese patients included in this study (who were
hospitalized and undergoing diets).
Table 1. Mean (and Standard Deviation) of Values
for STAI, VAS-H, VAS-HP, and VAS-P of the Sample by Group
Obese group (n = 15) Control group (n = 15)
Exposure condition Real Picture Augmented reality Real Picture Augmented reality
STAI-Y1
High-calorie sweet 27.2 (6) 27.1 (6.1) 26.2 (5.9) 28.8 (9) 29.3 (9.3) 28.3 (8.6)
High-calorie salt 27.1 (5.7) 26.5 (5.7) 26.1 (6.3) 28.6 (8.6) 28.8 (9.7) 28.3 (8.6)
Low calorie 26.1 (4.1) 26.5 (6.7) 25 (9.2) 28.1 (9.3) 29.4 (12.2) 28.1 (8.8)
VAS-Hunger
High-calorie sweet 24.4 (25.9) 20.20 (23.5) 20.8 (27.5) 42.1 (31.6) 36.1 (26.8) 38.6 (25.9)
High-calorie salt 22.1 (21.8) 16.1 (19.5) 21.1 (27.6) 40.8 (25.3) 41.6 (25.3) 45.4 (23.1)
Low calorie 17.4 (20.7) 15.1 (20.3) 19.4 (26.4) 37.0 (29.2) 35.8 (27.2) 40.8 (24.3)
VAS-Happiness
High-calorie sweet 48.8 (36.8) 52 (35.2) 54.4 (36.6) 63.1 (21.3) 55.2 (21.5) 63.8 (16.8)
High-calorie salt 48.5 (36.7) 54 (36.7) 56 (36.7) 61.0 (19.6) 58.4 (22.7) 61.0 (17.8)
Low calorie 45 (35.3) 54 (35.2) 56.47 (37) 64.7 (21) 55.6 (25.9) 63.1 (18)
VAS-Palatability
High-calorie sweet 43.8 (28) 38.2 (23.6) 41.4 (29.9) 56.5 (30.2) 37.5 (24.1) 48.4 (31.5)
High-calorie salt 47.9 (26.2) 42.5 (26.7) 49.1 (29.8) 60.9 (19.7) 42.3 (23.4) 49.7 (21.4)
Low calorie 24.9 (21) 26.6 (19.8) 26.6 (24.3) 45.8 (2.7) 32.8 (30.1) 41 (26.2)
STAI-Y1, State-Trait Anxiety Inventory, Form Y-1; VAS-H, Visual Analog Scale for Hunger; VAS-HP, Visual Analog Scale for
Happiness; VAS-P, Visual Analog Scale for Palatability.
112 PALLAVICINI ET AL.
Despite these limitations, the present study found that (a)
AR food stimuli were perceived to be as palatable as real
stimuli, and they also triggered a similar arousal response;
(b) obese individuals showed lower happiness after the ex-
posure to food compared to control participants, with regard
to both psychological and physiological responses; and (c)
high-calorie savory (vs. low-calorie) food stimuli were per-
ceived by all the participants to be more palatable, and they
triggered a greater arousal response.
Future studies are needed to investigate the differences in
emotional responses between food stimuli presented through
AR compared to other exposure conditions. In particular, the
relationship between levels of stimulus realism, interactivity,
and the intensity of emotional response should be deeply
investigated (e.g., assessing the level of presence experi-
enced by users and comparing AR food exposure with the
same stimuli presented through a VR system).
Moreover, in the future, additional studies could expand
knowledge about the emotional responses to food stimuli of
obese individuals compared to healthy individuals (e.g.,
measuring other informative physiological responses, such
as cortical and visual attention responses).
Finally, it would be interesting to replicate the assessment
of emotional responses to high-calorie versus low-calorie
foods to define a set of effective food stimuli to be adopted in
cue exposure treatments (e.g., conducting studies focused on
a larger sample size of obese individuals).
Acknowledgments
The authors wish to thank the company Regola S.r.l. of
Torino (Italy) and Fabio Ferrara.
Author Disclosure Statement
No competing financial interests exist.
References
1. Ng M, Fleming T, Robinson M, et al. Global, regional, and
national prevalence of overweight and obesity in children
and adults during 1980–2013: a systematic analysis for the
Global Burden of Disease Study 2013. The Lancet 2014;
384:766–781.
2. Kloiber S, Ising M, Reppermund S, et al. Overweight and
obesity affect treatment response in major depression.
Biological Psychiatry 2007; 62:321–326.
3. Hudson JI, Hiripi E, Pope HG, et al. The prevalence and
correlates of eating disorders in the National Comorbidity
Survey Replication. Biological Psychiatry 2007; 61:348–358.
4. Fairburn CG. (2008) Cognitive behavior therapy and eating
disorders. New York: Guilford Press.
5. Booth DA, Lewis VJ, Blair AJ. Dietary restraint and binge
eating: pseudo-quantitative anthropology for a medicalised
problem habit? Appetite 1990; 14:116–119.
6. Jansen A, Broekmate J, Heymans M. Cue-exposure vs self-
control in the treatment of binge eating: a pilot study. Be-
haviour Research and Therapy 1992; 30:235–241.
7. Jansen A, Van den Hout MA, De Loof C, et al. A case of
bulimia successfully treated by cue exposure. Journal of
Behavior Therapy and Experimental Psychiatry 1989; 20:
327–332.
8. Martinez-Mallen E, Castro-Fornieles J, Lazaro L, et al. Cue
exposure in the treatment of resistant adolescent bulimia
nervosa. International Journal of Eating Disorders 2007;
40:596–601.
9. Toro J, Cervera M, Feliu MH, et al. Cue exposure in the
treatment of resistant bulimia nervosa. International Journal
of Eating Disorders 2003; 34:227–234.
10. Koskina A, Campbell IC, Schmidt U. Exposure therapy in
eating disorders revisited. Neuroscience and Biobehavioral
Reviews 2013; 37:193–208.
11. Chicchi Giglioli IA, Pallavicini F, Pedroli E, et al. Aug-
mented reality: a brand new challenge for the assessment
and treatment of psychological disorders. Computational and
Mathematical Methods in Medicine 2015; 2015:862942.
12. Baus O, Bouchard S. Moving from virtual reality exposure-
based therapy to augmented reality exposure-based therapy:
a review. Frontiers in Human Neuroscience 2014; 8:112.
13. Azuma R, Baillot Y, Behringer R, et al. Recent advances in
augmented reality. IEEE Computer Graphics 2001; 21:34–47.
14. Azuma RT. A survey of augmented reality. Presence—
Teleoperators and Virtual Environments 1997; 6:355–385.
15. Botella C, Banos R, Guerrero B, et al. Mixing realities?: an
augmented reality system for the treatment of cockroach
phobia. CyberPsychology & Behavior 2005; 8:305–306.
16. Berryman DR. Augmented reality: a review. Medical Re-
ference Services Quarterly 2012; 31:212–218.
17. Du
¨
nser A, Billinghurst M. (2011) Evaluating augmented
reality systems. In: Fuhrt B, ed. Handbook of augmented
reality. New York: Springer, pp. 289–307.
18. Riva G, Waterworth JA, Waterworth EL, et al. From in-
tention to action: the role of presence. New Ideas in Psy-
chology 2011; 29:24–37.
19. Waterworth JA, Waterworth EL, Mantovani F, et al. On
feeling (the) present: an evolutionary account of the sense of
presence in physical and electronically-mediated environ-
ments. Journal of Consciousness Studies 2010; 17:167–178.
20. Juan MC, Botella B, Alcaniz M, et al. (2004) An aug-
mented reality system for treating psychological disorders:
application to phobia to cockroaches. In: Titsworth FM, ed.
Proceedings of the 3rd IEEE/ACM International Sympo-
sium on Mixed and Augmented Reality, pp. 256–257. IEEE
Computer Society, Los Alamitos, CA.
21. Juan MC, Pe
´
rez D. Using augmented and virtual reality for
the development of acrophobic scenarios. Comparison of
the levels of presence and anxiety. Computers & Graphics
2010; 34:756–766.
22. Botella C, Breto
´
n-Lo
´
pez J, Quero S, et al. Treating cock-
roach phobia using a serious game on a mobile phone and
augmented reality exposure: a single case study. Computer
in Human Behavior 2011; 27:217–227.
23. Botella C, Breto
´
n-Lo
´
pez J, Quero S, et al. Treating cock-
roach phobia with augmented reality. Behavior Therapy
2010; 41:401–413.
24. Wrzesien M, Alcan
˜
iz M, Botella C, et al. The therapeutic
lamp: treating small-animal phobias. IEEE Computer
Graphics and Application Magazine 2013; 33:80–86.
25. Wrzesien M, Burkhardt J-M, Alcan
˜
iz M, et al. (2011) How
technology influences the therapeutic process: a compara-
tive field evaluation of augmented reality and in vivo
exposure therapy for phobia of small animals. In: Pro-
ceedings of the Human-Computer Interaction–INTERACT
2011, pp. 523–540. Lisbon: Portugal Springer.
26. Anandacoomarasamy A, Caterson ID, Leibman S, et al.
Influence of BMI on health-related quality of life: com-
parison between an obese adult cohort and age-matched
population norms. Obesity 2009; 17:2114–2118.
AUGMENTED REALITY AND CUE EXPOSURE IN OBESE PATIENTS 113
27. Association AP. (2000) Diagnostic and statistical man-
ual of mental disorders. 4th edition text review ed.
Washington, DC: American Psychiatric Association.
28. Sheehan DV, Lecrubier Y, Sheehan KH, et al. The validity
of the Mini International Neuropsychiatric Interview
(MINI) according to the SCID-P and its reliability. Euro-
pean Psychiatry 1997; 12:232–241.
29. Rossi A, Alberio R, Porta A, et al. The reliability of the MINI-
International Neuropsychiatric Interview—Italian version.
Journal of Clinical Psychopharmacology 2004; 24:561–563.
30. Spielberger CD, Gorsuch RL, Lushene RE. (1970) Manual
for the state-trait anxiety inventory. Palo Alto: Consulting
Psychologists Press.
31. Macor A, Pedrabissi L, Santinello M. Ansia di stato e di
tratto: ulteriore contributo alla verifica della validita psi-
cometrica e teorica dello S.T.A.I. forma Y di Spielberger.
Psychological Science 1990; 15:67–74.
32. Flint A, Raben A, Blundell JE, et al. Reproducibility, power
and validity of visual analogue scales in assessment of
appetite sensations in single test meal studies. International
Journal of Obesity and Related Metabolic Disorders 2000;
24:38–48.
33. Yeomans MR, Blundell JE, Leshem M. Palatability: re-
sponse to nutritional need or need-free stimulation of ap-
petite? British Journal of Nutrition 2004; 92:S3–S14.
34. Yeomans MR, Lee MD, Gray RW, et al. Effects of test-
meal palatability on compensatory eating following dis-
guised fat and carbohydrate preloads. International Journal
of Obesity 2001; 25:1215–1224.
35. Young PT. (1967) Palatability: the hedonic response to
foodstuffs. In: Code CF, ed. Handbook of physiology,
section 6: alimentary canal, Vol 1 control of food and water
intake. Washington, DC: American Physiology Society,
pp. 353–366.
36. Moore DJ, Konrath S. ‘‘I can almost taste it’’: why people
with strong positive emotions experience higher levels of
food craving, salivation and eating intentions. Journal of
Consumer Psychology 2015; 25:42–59.
37. Boucsein W. (1992) Electrodermal activity. New York:
Plenum Press.
38. Electrophysiology TFotESoCatNASoPa. Heart rate variabil-
ity: standards of measurement, physiological interpretation,
and clinical use. European Heart Journal 1996; 17:354–381.
39. Larsen JT, Norris CJ, Cacioppo JT. Effects of positive and
negative affect on electromyographic activity over zygo-
maticus major and corrugator supercilii. Psychophysiology
2003; 40:776–785.
40. Dimberg U, Thunberg M. Rapid facial reactions to emo-
tional facial expressions. Scandinavian Journal of Psychol-
ogy 1998; 39:39–45.
41. Yeomans MR. Taste, palatability and the control of appetite.
Proceedings of the Nutrition Society 1998; 57:609–616.
42. Figner B, Murphy RO. (2011) Using skin conductance in
judgment and decision making research. In Schulte
Mecklenbeck M, Kuehberger A, Ranyard R, eds. A hand-
book of process tracing methods for decision research.
New York, NY: Psychology Press, pp. 163–184.
43. Milgram P, Kishino F. A taxonomy of mixed reality visual
displays. EICE Transactions on Information and Systems D
1994; 77:1321–1329.
44. Milgram P, Haruo T, Akira U, et al. Augmented reality: a
class of displays on the reality-virtuality continuum. Tele-
manipulator and Telepresence Technologies 1994; 2351:
282–292.
45. Roefs A, Jansen A. Implicit and explicit attitudes toward
high-fat foods in obesity. Journal of Abnormal Psychology
2002; 111:517–521.
46. Brewis A, Wutich A. Explicit versus implicit fat-stigma.
American Journal of Human Biology: The Official Journal
of the Human Biology Council 2012; 24:332–338.
47. Major B, Eliezer D, Rieck H. The psychological weight of
weight stigma. Social Psychological and Personality Sci-
ence 2012; 3:651–658.
48. Craeynest M, Crombez G, Koster EH, et al. Cognitive-
motivational determinants of fat food consumption in
overweight and obese youngsters: the implicit association
between fat food and arousal. Journal of Behavior Therapy
and Experimental Psychiatry 2008; 39:354–368.
49. Ferriday D, Brunstrom JM. ‘I just can’t help myself’: ef-
fects of food-cue exposure in overweight and lean indi-
viduals. International Journal of Obesity 2011, 35:142–149.
50. Drewnowski A. Taste preferences and food intake. Annual
Review of Nutrition 1997; 17:237–253.
51. Drewnowski A. Why do we like fat? Journal of the
American Dietetic Association 1997; 97:1081S–1085S.
52. Folkenberg DM, Martens M. Sensory properties of low
fat yoghurts. Part B: Hedonic evaluations of plain yo-
ghurts by consumers correlated to fat content, sensory
profile and consumer attitudes. Milchwissenschaft 2003;
58:154–157.
53. Privitera GJ, Antonelli DE, Creary HE. The effect of food
images on mood and arousal depends on dietary histories
and the fat and sugar content of foods depicted. Journal of
Behavioral and Brain Science 2013; 3:1–6.
Address correspondence to:
Dr. Federica Pallavicini
Applied Technology for Neuro-Psychology Lab
IRCCS Istituto Auxologico Italiano
Via Magnasco, 2
Milan 20149
Italy
E-mail: f.pallavicini@auxologico.it
114 PALLAVICINI ET AL.
This article has been cited by:
1. Riva Giuseppe, Gutiérrez-Maldonado José, Wiederhold Brenda K.. 2016. Virtual Worlds versus Real Body: Virtual Reality Meets
Eating and Weight Disorders. Cyberpsychology, Behavior, and Social Networking 19:2, 63-66. [Citation] [Full Text HTML] [Full
Text PDF] [Full Text PDF with Links]
