University of Southern California Dissertations and Theses

Validation of a neuroimaging task to investigate decisions involving visceral immediate rewards

(USC Thesis Other)

Validation of a neuroimaging task to investigate decisions involving visceral immediate rewards

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

VALIDATION OF A NEUROIMAGING TASK TO INVESTIGATE DECISIONS
INVOLVING VISCERAL IMMEDIATE REWARDS

by
Xiaobei Zhang

A Thesis Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Requirements for the Degree
MASTER OF ARTS
(PSYCHOLOGY)

August 2017
Copyright 2017 Xiaobei Zhang

ii
Table of Contents
List of Tables iii
List of Figures iv
Abstract 5
Introduction 6
Experimental design
11
Overview of Experimental Protocol 11
Materials and methods 11
Data analysis and results 15
Discussion 20
References 24

iii

List of Tables
Table 1: Model comparison for hypothesis 2 16
Table 2: Model comparison for hypothesis 2 18

iv

List of Figures
Figure 1: Scanner task 12
Figure 2: Brain activation predicting food choices 19

5
Abstract
Neuroeconomic investigation of now-later trade-offs have relied on intertemporal monetary
choice tasks. While monetary rewards allow precise modeling, they do not generate the strong
visceral states associated with immediately available primary rewards such as eating. The study
verified the feasibility of the hybrid task including both visceral content (visual depiction of
immediately available food) and elements of a conventional delay discounting task (monetary
bonuses), so that the effect of visceral factors could be precisely characterized and quantified.
Twenty-four volunteers with no history of eating disorders, participated in the study in which
repeated choices were made between immediately available food items and different monetary
bonuses available in one month (Food vs. Future Money Task – FFMT). Participants completed
the FFMT in multiple sessions in different metabolic states. We found that 1) Food choices were
selected more when the delayed money alternative was smaller; 2) The reaction time of decision
making increased when the rewards were similarly valued; 3) preference for immediate food
over delayed money was greater when the task was completed in greater metabolic deficit; and 4)
greater brain activation in orbital frontal cortex, ventral striatum, caudate and lateral occipital
cortex during initial presentation of food items predicted choice of immediate food over delayed
money. Taken together, these results support use of the FFMT for the investigation of brain
mechanisms relevant in now-later decisions involving visceral reward.
Keywords: Decision making, Neuroeconomic, Intertemporal Choice, Visceral factors

6
Introduction
How would you make the decision if you got a chance to choose between 5 dollars
now or 20 dollars one month later (call this scenario A)? What if you were a chocolate lover, and
now the options are chocolate bar in front of you and 20 dollars a month later when your mouth
is already watering (call this scenario B).
In scenario B, a more emotional or visceral situation where you are in a huger drive
state may make the chocolate choice more appealing to you. A broad range of drive states (e.g.,
hunger, thirst, sexual desire), negative emotions (e.g., disgust, fear), and feeling states (e.g., pain)
can be regarded as visceral factors. Those visceral factors attract people’s attention and
motivation, which can lead to specific behaviors (Loewenstein, 2000). The disproportional effect
of visceral factors experience resulted in the main focus on mitigating the visceral factor,
regardless of all other goals. Visceral factors play an important role in intertemporal
choice(Loewenstein, 1996). People seem to become shortsighted after visceral stimulation,
instead of behaving in accordance with long-term goals, they have extreme discounting of the
future (Kyu & Zauberman, 2013)(Field, Santarcangelo, Sumnall, Goudie, & Cole, 2006). In fact,
self-control problems often involve visceral factors, and likewise, almost all visceral factors are
related to self-control problems: hunger and dieting (Konttinen, Haukkala, Sarlio-Lähteenkorva,
Silventoinen, & Jousilahti, 2009), sadness and impulsive suicide (Apter, Plutchik, & van Praag,
1993), and so on. Visceral factors can cause dramatic changes in preference because they are
affected by internal deviations from homeostasis and by external stimuli that signal specific
threat or reward opportunities (Loewenstein, 2000).
Delay discounting, also called intertemporal choice, refers to the situation when
people discount the value of future rewards as a function of the delay to receiving them. People

7
discount future rewards differently. While the word has different meanings (Evenden, 1999)
“impulsive” individuals in this sense of the word are defined as those individuals with a
particularly strong tendency to underweight foreseeable consequences that are delayed. In
contrast, “patient” tend to choose later/larger rewards (Lempert & Phelps, 2016).
Several studies examined the effect of visceral stimuli on delay discounting, and
found erotic stimuli(Van den Bergh, Dewitte, & Warlop, 2008); attractive faces(Wilson & Daly,
2004); or desserts stimuli(Li, 2008) before intertemporal choice may lead to more impulsive
choices. A hot-cold system theory is widely used to explain this tendency (McClure, Laibson,
Loewenstein, & Cohen, 2004)(Metcalfe & Mischel, 1999). Emotion, as the byproduct of visceral
stimuli, can evoke the impulsive, emotional, ‘hot’ system, though leads to a shortsighted
decision. Also the arousal effect of visceral factors may be directly linked to intertemporal
choice(Lempert, Glimcher, & Phelps, 2015).
It is important to incorporate visceral factors into the models of human decision
making behavior to have a better understanding and strategy about decision making. While most
of those studies mentioned above emphasized the priming effect of the visceral stimuli or
“incidental” visceral states on intertemporal choice where either a smaller sooner money reward
or a larger later money reward needed to be chosen. In the current study, we are interested in
how “reward-relevant visceral states” will influence people’s decision. A hybrid task of visceral
content contained choice itself (food cue) and conventional delay discounting task (monetary
choices) should be established to more precisely characterize and quantify the effect of visceral
factors. In the new paradigm(Luo, Monterosso, Sarpelleh, & Page, 2015), the visceral states,
evoked by the food choice it self, are integral to the alternatives. The innovation of using a
different type of reward (food) can not only extend the money-alternatives only temporal choice

8
paradigm, but also the visceral component food-cue related response can be specified. Instead of
using two money choices, participants are presented with visual food cues (selected based on
individual preference) in a context where they must choose between the depicted food and a
delayed monetary bonus. This task is intended to serve as a model for situations in which an
individual chooses between two alternatives, only one of which is associated with satisfaction of
a current visceral state (here hunger as a result of experimentally required 12-h deprivation). A
critical feature of the design is that the monetary alternative to each food option is titrated to
remain near each participant’s “indifference point” for the item (the amount at which they are
about equally likely to choose either alternative, which is referred as willingness to pay later).
This provides maximum sensitivity in identifying the association between within-participant
variation (e.g., trials in which they exhibit more or less willingness to wait for the money than
usual) and trial-by-trial variance in brain activity. Unlike the visceral stimuli priming designs,
this task will allow identification of, and differentiation between brain activity associated with
choosing an immediate food (we use “Food Approach” throughout) and brain activity associated
with instead choosing the one month delayed monetary bonus (we use “Restraint” throughout).
This new task is the extension of the existing delay discounting task, and mimics a real life
situation where the fulfillment of the hunger drive and future benefits are weighed under the
visceral states.
The goal of the task is to use “willingness to pay” (delayed money) to precisely
quantify the valuation of visually presented foods in a way that is compatible with neuroimaging.
The work builds on many past temporal discounting studies that used money as the only reward.
Unlike money, food is subject to homeostatic regulation, and so naturally undergoes dramatic

9
changes in valuation as a function of metabolic state. To verify the the feasibility of the new
paradigm, several hypotheses are made based on conventional monetary delay discounting task.
Hypothesis 1: Food choices are selected more when the delayed money alternative is
smaller.
This assumption is to confirm that a directional selection between food and money
choices is made, rather than random pick or else a mismatch between rewards such that the
valuation of the food was either always greater or less than the delayed money alternative.
Hypothesis 2: The reaction time will increase when the delayed reward is close to the
willingness to pay (“indifference point” for the food item).
This “near-indifference slowing” has been observed in other choice domains (e.g.,
intertemporal choice with two monetary rewards(Krajbich, Bartling, Hare, & Fehr, 2015). This
pattern a critical input for drift-diffusion models of value comparison (Mormann, Malmaud,
Huth, Koch, & Rangel, 2010). It also may allow a window into brain correlates of temptation
and or ambivalence. Of particular interest in this literature is the possibility that near-
indifference slowing might allow identification of competing valuation systems, as hypothesized
by dual process theory. Based on the dual-process theory, two kinds of processing lead to human
behavior. One is a slow and deliberative process in which people are carefully weigh existing
options, the other is a faster and more automatic process while prone to certain biases. These two
processes may compete with each other in the context of value-based choice to determine the
final decision. We can label these two types of strategies as ‘intuitive/automatic’ (Type I) in
which choice details are not that important, or ‘deliberative’ (Type II) (Kahneman,
2011)(Stanovich, 1999) in which the features of the choices are considered.

10
Reaction time in a choice task depends critically on how different the decision maker
finds the options are(Henmon, 1906). If we plot the expected RT as a function of the difference
in subjective value (preference) between reward now and delayed reward, the curve will peak at
the “indifference point”, and fall off as the strength of the preference increases in either
direction(Krajbich et al., 2015). Similarly, if the participants are weighing between the food and
money choice, then they will slow down when the delayed reward is close to the willingness to
pay.
Hypothesis 3: The willingness to pay is higher when there is greater metabolic deficit.
Different level of the visceral factors can influence impulsivity differently. At higher
intensities, people may experience a feeling of “out of control”, for the visceral factors
progressively seize control over behavior (Loewenstein, 1996). With the current framework of
the research design, a lower intensity of hunger level is manipulated by having participants
ingest glucose (75 g), which will be compared with a controlled water intake condition where a
greater metabolic deficit is formed. So we hypothesize that compared with glucose intake group,
the “indifference point” for the item (willingness to pay) is higher for the controlled water group.
It should be noted that this manipulation would not be expected to lead to maximal separation of
hunger, since 75g of glucose does not fully satiate participants after 12 h of fasting. However, it
may be sufficient to drive a divergence in subjective hunger and associated decision making.
As the second part of the project, a preliminary brain data analysis was conducted to
see the brain correlates of the food priming cues and how brain signal can predict the choice.

11
Experimental design
Overview of Experimental Protocol
All participants underwent a screening visit during which height and weight were
measured and 24-h dietary and physical activity recalls were administered. A 75-g fructose
tolerance test was also performed. Only individuals who reported no gastrointestinal discomfort
(e.g., bloating, nausea, diarrhea) on a questionnaire administered 1 h after the ingestion of 75 g
fructose were included in the study to limit confounding effects of fructose malabsorption. Two
participants were excluded on this basis.
Materials and Methods
Participants
Twenty-four volunteers (14 female; 10 male; mean age 21.6 ± 2, range 16–25 y; mean
BMI 29.0 ± 7.4, range 19.6–45.4 kg/m
2
) with no history of eating disorders, fructose intolerance,
diabetes, or other medical illnesses participated in the study. Participants were all right-handed
with normal or corrected-to-normal vision, nonsmokers, and not on weight-loss diets or taking
medications (with the exception of oral contraceptives). Participants were asked to maintain their
typical diet and physical activity levels throughout this study, and female participants were
studied during the follicular phase of their menstrual cycle. Participants gave written informed
consent to all experimental procedures approved by the Institutional Review Board of the
University of Southern California. For some participants, (n = 16) an additional water session
was included.
Decision-Making Task. Participants made choices between (i) a visually presented high-
calorie food reward and (ii) a visually presented monetary reward, always delayed by 1 mo. The
delay was used to model real-life situations in which the benefits of turning down high-calorie

12
foods come later in time. Participants were also asked to express whether each indicated
preference was “strong” or “weak.” Ten individualized high-calorie food items were used during
the task. These included only food items rated as very attractive by the individual participant
during pretesting. Thirty-seven food items were included in the pretest and top ten attractive food
items for each subject were used in the decision making task. The food item presentation was
pseudorandom, with each session including six presentations of each of the 10 food items. Each
trial began with a 2-sec visual presentation of one of the food items. Next the screen changed to
include a smaller depiction of the same food item on one half of the screen and the 1-month
monetary bonus alternative on the other (side randomized). Participants were given up to 6
seconds to respond with their selection.
Figure 1 Scanner task

On a food item’s first presentation within a session, the monetary alternative was set to
the market price for the item (either searched online or went to store get the price), “discounted”
for the 1-mo delay using participant-specific estimated discounting. Market prices ranged from
$0.50 to $6.80. This discounting estimate was obtained based on a monetary intertemporal
choice procedure completed at the baseline session (Kirby, Petry, & Bickel, 1999). On

13
subsequent presentations of the food item, the amount of money offered as its alternative was
adjusted according to the following rules: (i) it increased after the food alternative was selected
and decreased after the money was selected; (ii) if the item had been presented in two or more
prior trials and if the same alternative (whether food or money) was selected in the previous two
or more presentations, then the magnitude of the adjustment of the money alternative was either
25% or 50%, based on whether the preference was indicated to be weak or strong; and (iii) if the
item had been presented in only one prior trial, or if the choice in the two most recent
presentations of the item included one selection of food and one of money, then the magnitude of
the adjustment was either 10% or 20%, based on whether the preference was indicated to be
weak or strong. At the end of each fMRI session, bonus earnings were determined by randomly
drawing a trial from the food-decision task. If the food reward was selected, participants were
provided the selected food item to eat immediately after the scan as a bonus reward.
Alternatively, if the delayed monetary reward was selected, participants received a Visa gift card
in that amount 1 mo after the study session. To control for the extra time involved in eating food,
the reimbursement session had a fixed duration of 30 min for all participants. During this time,
participants either consumed the selected food item or were required to sit in and wait (in the
case where a money reward was drawn) until the end of the session. Participants were instructed
that they were not allowed to take the bonus food reward home.
MRI Parameters. MRI data were collected using a 3T Siemens MAGNETOM Tim/Trio
scanner with a standard birdcage head coil. Participants laid supine on a scanner bed, viewing
stimuli through a mirror mounted on the head coil. For each session, 278 functional T2*-
weighted echo planar imaging (EPI) volumes of data were acquired with following parameters:
repetition time (TR), 2 s; echo time (TE), 30 ms; flip angle, 90°; field of view, 192; in-plane

14
resolution, 64 × 64; voxel dimensions, 2 × 2 × 2 mm. A total of 32 axial slices was used to cover
the whole brain with no gap. The slices were tilted 30° along the anterior commissure–posterior
commissure plane to gain better signal in the orbital frontal cortex. Additionally, during the same
session, a high-resolution anatomical image (matrix size: 256 × 256 × 176) with 1 × 1 × 1 mm3
resolution was obtained using a T1-weighted 3D magnetization prepared rapid gradient echo
(MP-RAGE) sequence (inversion time, 900 ms; TR, 1,950 ms; TE, 2.26 ms; flip angle, 90°).
MRI Analysis. All fMRI data were processed using fMRI Expert Analysis Tool
version 6.00, part of the Oxford University Centre for Functional MRI of the Brain Software
Library (www.fmrib.ox.ac.uk/fsl). A total of four functional volumes (four TRs) was discarded
to account for magnetic saturation effects. Translational movement parameters never exceeded
one voxel in any direction for any participant. The fMRI data were motion-corrected, high pass-
filtered (100 s), and spatially smoothed with a Gaussian kernel of full-width at half-maximum of
5 mm. The functional volumes were realigned to each participant’s respective T1-weighted
anatomical image and then normalized into standard space (Montreal Neurological Institute;
MNI) using affine transformation with FLIRT (52) to the avg152 T1 MNI template.
For the present report, a preliminary analysis was carried out using brain activity during
the presentation of the food item alone at the beginning of each trial. We set out to answer the
question, “Can brain activity during the presentation of the food item predict how often it will be
selected by a participant?” While prediction of WTP might be a more intuitive objective, we
chose number of selections of the food item instead since WTP could not be computed on items
in which the same response was made for all 6 presentations of an item. For the imaging
contrast, the primary regressor was the food frequency. Decision was coded as either 1 or 2 (1
for food, 2 for money). All the 1s and 2s were averaged and used for analysis.

15
Data Analysis and Results
Hypothesis 1: Food choices are selected more when the delayed money alternative is smaller.
We performed a simple logistic regression (base model) with choice (Food or Money) as
the dependent variable, delayed money alternative as the predictor. Alternatively, we analyzed
data with logistic hierarchical regression models using the R lmer function of the lme4 library to
take into account trial-by-trial variances in individuals’ choices. For the second model, choice
was the dependent variable, delayed reward was included as a fixed-effects predictor, also as
random slope and random intercept effects with the grouping variable “ID” (look it by each
subject). The random slope and intercept effects of delayed reward with the grouping variable
“Food” was added to second model, so the third model can take care of the difference in both the
individual and the Food item level. The fourth model included delayed rewards as both a fixed-
effects predictor and random slope and intercept effects with the grouping variable as “Food
nested within subject ID”, in this way each subject and a specific Food will have a specific slope
and intercept. Data were fitted using logit and probit model separately. The logit model assumes
the underlying distribution is logistic, whereas probit assumes a normal distribution. Results
from these two models were compared based on Akaike Information Criterion (AIC). AIC has
been widely used in model comparison and selection (Anderson & Burnham, 2002). The smaller
the AIC, the better the model. Using an information-theoretic approach (Pine et al., 2009), the
AIC was summed over all subjects for each model separately, and the absolute difference
between these two models was calculated (referred to as ΔAIC). As a rule of thumb, it has been
suggested that if the absolute difference is greater than 2, it favors the better fitting model.
We found that for all the models and methods we used, Food choices are significantly

16
selected more when the delayed money alternative is smaller (p<0.05). After model comparison
(see table1), the fourth model using “logit” method has the smallest AIC is the best fitting model.
Table 1: Model comparison for hypothesis 1

Hypothesis 2: The reaction time will increase when the delayed reward is close to the
willingness to pay (“indifference point” for the food item).
WTP could not be obtained for some food item within each subject. In general, 78% of
the food items within each subject were included in the analysis for WTP can be calculated.
The difference between the delayed reward (will call this as “distance”) and the
willingness to pay is calculated as the absolute value of (WTP - Delayed reward) / (WTP +
Delayed reward). Willingness to pay for each food item and each session was based on
observed “cross-over” points for each item during titration. Cross-over points were cases where
either (i) an increase in the monetary alternative to a particular food resulted in a switch to
preference for the money or (ii) a decrease in the monetary alternative to a particular food
resulted in a switch to preference for the food item. The average of the amount offered in the two
trials comprising the cross-over was computed as a WTP estimate. When multiple cross-over

AIC ΔAIC
logistic regression Logit Probit
First: Base Model 6644.1 6643.9 0.2
Second: Random slope and intercept, look it by ID 5527.1 5526.1 1
Third: Random slope and intercept, look it by ID and Food 5514.2 5512.7 1.5
Fourth: Random slope and intercept, each food is nested within each subject 5322.7 5327.8 -5.1

17
points were present for the same food item during the same session, these points were averaged
to obtain the item’s overall WTP for that session. Items in which no cross-over point was
obtained during a session were excluded. It is worth noting that the presence of these cross-over
points do not imply price sensitivity since they would occur from a participant that responded
randomly. Rather, the use of switching as implying a WTP is justified by evidence that
participants are responding in a way that is sensitive to price (Hypothesis 1).
Using a subset of participants who had at least one WTP estimates, we conducted linear
regression and hierarchical regression. A base model included Reaction time as dependent
variable, the distance as the predictor. Alternatively, we analyzed data with linear hierarchical
regression models using the R lmer function of the lme4 library to take into account trial-by-trial
variances. For the second model, RT was the dependent variable, distance was included as a
fixed-effects predictor, also as random slope and random intercept effects with the grouping
variable “ID” (look it by each subject). The random slope and intercept effects of distance with
the grouping variable “Food” was added to second model, so the third model can take care of the
difference both the individual and the Food item level. The fourth model included distance as
both a fixed-effects predictor and random slope and intercept effects with the grouping variable
as “Food nested within subject ID”, in this way each subject and a specific Food will have a
specific slope and intercept. We found that for all the models we used, the reaction time will
increase significantly when the delayed reward is close to the willingness to pay (p<0.05). After
model comparison (see table 2), the fourth model has the smallest AIC is the best fitting model.

18
Table 2: Model comparison for hypothesis 2
Linear regression AIC
First: Base Model 9807.2
Second: Random slope and intercept, look it by ID 8941
Third: Random slope and intercept, look it by ID and Food 8945.9
Fourth: Random slope and intercept, each food is nested within each subject 8925.7

Hypothesis 3: The willingness to pay is higher when there is greater metabolic deficit.
Using a subset of participants who additionally completed a water session, we observed
that, relative to water, glucose resulted in significantly decreased WTP-delayed (Z = −2.245, P =
0.025 for glucose vs. water).

Order effect analysis:
Because of the titration procedure used to determine willingness to pay, the money offers
presented to participants were not independent from trial order. This confound, if not addressed,
could lead to misleading results about the affect of the size of the money alternative. We
therefore included trial order as a nuisance covariate, and report on the affect of money offer
after removing variance associated with presentation number of the food (1-6).
For Hypothesis 1, order effect was tested by adding trial sequence number (1 to 6) within
each food item as another predictor in the existing model. We found delayed money alternative
(p=0.005 with probit model, p=0.006 with logit model) can significantly predict the later choice
after controlling for sequence. For Hypothesis 2, after adding the new predictor trial sequence
number, the “distance” is still a significant predictor (p<0.001) for RT.

19
Brain data analysis:
For the preliminary analysis of brain data, we want to answer the question that whether
brain activity during the presentation of the food item can predict how often the food choice is
selected. Brain activity during the presentation of the food item alone at the beginning of each
trial is used to predict the frequency of the food choices. The primary regressor was the food
frequency. These contrast maps were based on whole-brain analysis during the food-cue period
in every voxel (Z > 2.3, p < 0.05 corrected for multiple comparison problems). Greater activation
in orbital frontal cortex, ventral striatum, caudate, lateral occipital cortex (see fig.1) is more
likely when the food option is chosen more frequently.

Figure 2: Brain activation predicting food choices

20
Discussion
In the current study, we tried to establish a new paradigm incorporating visceral factor
into the existing intertemporal monetary choice task. The visceral state was evoked by the visual
depiction and immediate availability of desirable foods, after 12-hour food deprivation of the
participants. This hybrid task includes both visceral content (food cue) and conventional delay
discounting task (monetary choices), so that we can more precisely characterize and quantify the
effect of visceral factors. This paradigm also allows us to separate the brain correlates of “Food
approach” (brain regions more active when participants choose immediate food) and “Restraint”
(brain regions more active when participants choose delayed money). The paradigm was a
simple laboratory model of the common real-world situation in which the individual conceives a
long-term potential reward (e.g., health, mobility, appearance) for avoiding attractive visceral
foods options. Here that long-term reward was made explicit as a 1-mo delayed monetary bonus.
We used a titration procedure that allowed us to quantify the long-term reward the individual
was willing to give up for each food.
We verified the feasibility of this new paradigm by three hypotheses based on the
conventional monetary delay discounting task. It’s worth mentioning that the delay in the new
paradigm was used to model real-life situations in which the benefits of turning down foods
come later in time. It is the extension of the monetary delay discounting task, rather than a
strictly delay discounting task. Since the two time points (immediate and 1 month delay) are
associated with distinct rewards (food and money respectively) the task does not allow
separation of differences in reward valuation from temporal discounting. For example, an
extreme preference for immediate food could either reflect high valuation of food over money,
high valuation of now over later rewards, or some combination of the two.

21
As predicted, food choices were selected more when the delayed money alternative was
smaller. This means the decisions were not randomly picked, and the amounts of delayed money
available were at least sometimes within the right range for assessing decision making (neither so
large as to be always preferred, nor so small as to be consistently dispreferred). Model
comparison was conducted and the best fit model is the hierarchical regression model included
delayed rewards as both a fixed-effects predictor and random slope and intercept effects with the
grouping variable as “Food nested within subject ID”. Both the individual difference and the
data structure were taken care of with this best fitting model.
Also as expected, participants slowed down when the delayed reward was close to the
willingness to pay. This suggests participants were weighing the choices and that brain activity
during this epoch may be investigated for learning about substrates of self-control success and
failure. It’s worth noting that individual difference of slowing down allowed us to separate brain
correlates of obvious temptation and ambivalence of the decision for the future study. People
with a significant slowing down may use different strategies compared with those without a
significant slowing down. As the continuing of the current analysis, brain activation of the
participants with different “strategies” will be compared, then the underlying mechanism can be
understood. This near-indifference slowing might allow identification of competing valuation
systems, as hypothesized by dual process theory. Hierarchical model comparison was used to
find the best fit model and the best fit model turned out to be the one included distance as both a
fixed-effects predictor and random slope and intercept effects with the grouping variable as
“Food nested within subject ID”. This model can take care of both the individual differences and
the data structure.

22
The best fitting models for both hypotheses take care of the individual differences and the
data structure. Future study using the same model with different visceral-factor alternatives
should also consider the individual differences and the data structure.
In this study, we measured the differential effects of water vs. glucose ingestion on
decisions between immediate food rewards and delayed monetary rewards in healthy volunteers.
We found that the willingness to pay was higher when acutely food-deprived participants
ingested only water compared with ingestion of water in which 75g glucose had been dissolved.
This result seemed to be intuitive, while it confirmed that the effectiveness of the manipulation.
Even though it did not fully satiate participants after 12 h of fasting, 75g of glucose was
sufficient to drive a divergence in subjective hunger and associated decision making.
The results of the preliminary analysis of the brain data showed that, the activation in
orbital frontal cortex, ventral striatum, caudate and lateral occipital cortex during the initial
presentation of the available food item predicted the subsequent choice of that food. In a meta-
analysis of food-cue activation of the the brain (Tang, Fellows, Small, & Dagher, 2012), peak
areas of activation was found in the amygdala, insula, lateral orbital frontal cortex (OFC), ventral
striatum, fusiform gyrus and thalamus. The OFC and ventral striatum play an important role in
reward processing, research found a positive relationship between BOLD signal change and
reward magnitude in those region(Bartra, McGuire, & Kable, 2013)(Clithero & Rangel,
2014)(Diekhof, Kaps, Falkai, & Gruber, 2012). For food-cue rewards, prior research has shown
an increase in OFC activity in response to high-calorie food cues compared with nonfood cues
(Tang et al., 2012)(van der Laan, de Ridder, Viergever, & Smeets, 2011). Greater food-cue
reactivity in the OFC and ventral striatum was positively related to the frequency of food

23
choices, this suggests greater reward and motivation signaling for food cues can predict the later
preference for the food choice.
In summary, we successfully verified the feasibility of the new paradigm incorporating
visceral-factor alternative into the conventional monetary delay discounting task. This model can
be used for the future study to separate the brain correlates of the “Food approach” and
“Restraint”. We also found that the brain activation in orbital frontal cortex, ventral striatum,
caudate and lateral occipital cortex can predict the later food choice.

24
References
Anderson, D. R., & Burnham, K. P. (2002). Avoiding pitfalls when using information-theoretic
methods. The Journal of Wildlife Management, 66(3), 912–918.
https://doi.org/10.2307/3803155
Apter, A., Plutchik, R., & van Praag, H. M. (1993). Anxiety, impulsivity and depressed mood in
relation to suicidal and violent behavior. Acta Psychiatrica Scandinavica, 87(1), 1–5.
https://doi.org/10.1111/j.1600-0447.1993.tb03321.x
Bartra, O., McGuire, J. T., & Kable, J. W. (2013). The valuation system: A coordinate-based
meta-analysis of BOLD fMRI experiments examining neural correlates of subjective
value. NeuroImage, 76, 412–427. https://doi.org/10.1016/j.neuroimage.2013.02.063
Clithero, J. A., & Rangel, A. (2014). Informatic parcellation of the network involved in the
computation of subjective value. Social Cognitive and Affective Neuroscience, 9(9),
1289–1302. https://doi.org/10.1093/scan/nst106
Diekhof, E. K., Kaps, L., Falkai, P., & Gruber, O. (2012). The role of the human ventral striatum
and the medial orbitofrontal cortex in the representation of reward magnitude – An
activation likelihood estimation meta-analysis of neuroimaging studies of passive reward
expectancy and outcome processing. Neuropsychologia, 50(7), 1252–1266.
https://doi.org/10.1016/j.neuropsychologia.2012.02.007
Evenden, J. L. (1999). Varieties of impulsivity. Psychopharmacology, 146(4), 348–361.
https://doi.org/10.1007/PL00005481
Field, M., Santarcangelo, M., Sumnall, H., Goudie, A., & Cole, J. (2006). Delay discounting and
the behavioural economics of cigarette purchases in smokers: the effects of nicotine

25
deprivation. Psychopharmacology, 186(2), 255–263. https://doi.org/10.1007/s00213-006-
0385-4
Henmon, V. A. C. (1906). The time of perception as a measure of differences in sensations.
University of Michigan Library.
Kahneman, D. (2011). Thinking, Fast and Slow. Macmillan.
Kirby, K. N., Petry, N. M., & Bickel, W. K. (1999). Heroin addicts have higher discount rates for
delayed rewards than non-drug-using controls. Journal of Experimental Psychology:
General, 128(1), 78–87. https://doi.org/10.1037/0096-3445.128.1.78
Konttinen, H., Haukkala, A., Sarlio-Lähteenkorva, S., Silventoinen, K., & Jousilahti, P. (2009).
Eating styles, self-control and obesity indicators. The moderating role of obesity status
and dieting history on restrained eating. Appetite, 53(1), 131–134.
https://doi.org/10.1016/j.appet.2009.05.001
Krajbich, I., Bartling, B., Hare, T., & Fehr, E. (2015). Rethinking fast and slow based on a
critique of reaction-time reverse inference. Nature Communications, 6, 7455.
https://doi.org/10.1038/ncomms8455
Kyu, B., & Zauberman, G. (2013). Can Victoria’s Secret change the future? A subjective time
perception account of sexual-cue effects on impatience. Journal of Experimental
Psychology: General, 142(2), 328–335. https://doi.org/10.1037/a0028954
Lempert, K. M., Glimcher, P. W., & Phelps, E. A. (2015). Emotional arousal and discount rate in
intertemporal choice are reference-dependent. Journal of Experimental Psychology.
General, 144(2), 366–373. https://doi.org/10.1037/xge0000047
Lempert, K. M., & Phelps, E. A. (2016). The malleability of intertemporal choice. Trends in
Cognitive Sciences, 20(1), 64–74. https://doi.org/10.1016/j.tics.2015.09.005

26
Li, X. (2008). The effects of appetitive stimuli on out-of-domain consumption impatience.
Journal of Consumer Research, 34(5), 649–656. https://doi.org/10.1086/521900
Loewenstein, G. (1996). Out of Control: Visceral influences on behavior. Organizational
Behavior and Human Decision Processes, 65(3), 272–292.
https://doi.org/10.1006/obhd.1996.0028
Loewenstein, G. (2000). Emotions in economic theory and economic behavior. The American
Economic Review, 90(2), 426–432.
Luo, S., Monterosso, J. R., Sarpelleh, K., & Page, K. A. (2015). Differential effects of fructose
versus glucose on brain and appetitive responses to food cues and decisions for food
rewards. Proceedings of the National Academy of Sciences, 112(20), 6509–6514.
https://doi.org/10.1073/pnas.1503358112
McClure, S. M., Laibson, D. I., Loewenstein, G., & Cohen, J. D. (2004). Separate neural systems
value immediate and delayed monetary rewards. Science (New York, N.Y.), 306(5695),
503–507. https://doi.org/10.1126/science.1100907
Metcalfe, J., & Mischel, W. (1999). A hot/cool-system analysis of delay of gratification:
Dynamics of willpower. Psychological Review, 106(1), 3–19.
https://doi.org/10.1037/0033-295X.106.1.3
Mormann, M. M., Malmaud, J., Huth, A., Koch, C., & Rangel, A. (2010). The drift diffusion
model can account for the accuracy and reaction time of value-based choices under high
and low time ressure (SSRN Scholarly Paper No. ID 1901533). Rochester, NY: Social
Science Research Network. Retrieved from https://papers.ssrn.com/abstract=1901533

27
Pine, A., Seymour, B., Roiser, J. P., Bossaerts, P., Friston, K. J., Curran, H. V., & Dolan, R. J.
(2009). Encoding of marginal utility across time in the human brain. Journal of
Neuroscience, 29(30), 9575–9581. https://doi.org/10.1523/JNEUROSCI.1126-09.2009
Stanovich, K. E. (1999). Who Is rational?: Studies of individual differences in reasoning.
Mahwah, N.J: Psychology Press.
Tang, D. W., Fellows, L. K., Small, D. M., & Dagher, A. (2012). Food and drug cues activate
similar brain regions: A meta-analysis of functional MRI studies. Physiology & Behavior,
106(3), 317–324. https://doi.org/10.1016/j.physbeh.2012.03.009
Van den Bergh, B., Dewitte, S., & Warlop, L. (2008). Bikinis instigate generalized impatience in
intertemporal choice. Journal of Consumer Research, 35(1), 85–97.
https://doi.org/10.1086/525505
van der Laan, L. N., de Ridder, D. T. D., Viergever, M. A., & Smeets, P. A. M. (2011). The first
taste is always with the eyes: A meta-analysis on the neural correlates of processing
visual food cues. NeuroImage, 55(1), 296–303.
https://doi.org/10.1016/j.neuroimage.2010.11.055
Wilson, M., & Daly, M. (2004). Do pretty women inspire men to discount the future?
Proceedings of the Royal Society of London B: Biological Sciences, 271(Suppl 4), S177–
S179. https://doi.org/10.1098/rsbl.2003.0134

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Homeostatic imbalance and monetary delay discounting: effects of feeding on RT, choice, and brain response

PDF

Sequential decisions on time preference: evidence for non-independence

PDF

Behabioral and neural evidence of state-like variance in intertemporal decisions

PDF

Behavioral and neural evidence of incentive bias for immediate rewards relative to preference-matched delayed rewards

PDF

The acute impact of glucose and sucralose on food decisions and brain responses to visual food cues

PDF

Reward immediacy and subjective stress modulate anticipation of primary and secondary rewards in temporarily-abstinent cigarette smokers

PDF

Choice biases in making decisions for oneself vs. others

PDF

Reward substitution: how consumers can be incentivized to choose smaller food portions

PDF

Effect of mindfulness training on attention, emotion-regulation and risk-taking in adolescence

PDF

Taking the temperature of the Columbia Card Task

PDF

Toward a more realistic understanding of decision-making

PDF

Food choice dimensions and the relationship with BMI

PDF

Value-based decision-making in complex choice: brain regions involved and implications of age

PDF

Insula activity during safe-sex decision-making in sexually risky men suggests negative urgency and fear of rejection drives risky sexual behavior

PDF

Using Pavlovian cues to boost self-control

PDF

Cumulative risk as a moderator of multisystemic therapy effects for juvenile offenders

PDF

The evolution of decision-making quality over the life cycle: evidence from behavioral and neuroeconomic experiments with different age groups

PDF

Regularized structural equation modeling

PDF

Indexing cerebrovascular resistance in cognitive decline and Alzheimer's disease

PDF

Selflessness takes time: altruistic (but not cooperative) prosocial behavior increases with decision time

Asset Metadata

Creator Zhang, Xiaobei (author)

Core Title Validation of a neuroimaging task to investigate decisions involving visceral immediate rewards

School College of Letters, Arts and Sciences

Degree Master of Arts

Degree Program Psychology

Publication Date 07/14/2017

Defense Date 06/20/2017

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Decision making,intertemporal choice,neuroeconomic,OAI-PMH Harvest,visceral factors

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Monterosso, John (committee chair), Bechara, Antoine (committee member), John, Richard (committee member)

Creator Email xiaobeiz@usc.edu,zxbwlyzgw@sina.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c40-397998

Unique identifier UC11266376

Identifier etd-ZhangXiaob-5504.pdf (filename),usctheses-c40-397998 (legacy record id)

Legacy Identifier etd-ZhangXiaob-5504.pdf

Dmrecord 397998

Document Type Thesis

Rights Zhang, Xiaobei

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the a...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA