University of Southern California Dissertations and Theses

The brain's virtuous cycle: an investigation of gratitude and good human conduct

(USC Thesis Other)

The brain's virtuous cycle: an investigation of gratitude and good human conduct

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content T H E B R A I N ’ S V I RT U O U S C Y C L E : A N I N V E S T I G AT I O N O F
G R AT I T U D E A N D G O O D H U M A N C O N D U C T
glenn r. fox
A Dissertation Presented to
the FACULTY OF THE GRADUATE SCHOOL of
the UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the Requirements for the Degree
DOCTOR OF PHILOSOPHY
December, 2014

To: Melissa, Lisa, Dad, Dave, Kathy
For: Mom

A B S T R A C T
When we are the beneficiaries of good human conduct, we experience a concert
of positive emotions ranging from relief to elation. These emotions can in turn
motivate us to expend great sums of energy to reward those near the source
of the good conduct, creating, quite literally, a virtuous cycle. If this cycle has
a fuel or catalyst, it goes by the name of gratitude. I approach gratitude us-
ing an interdisciplinary method. I use modern brain imaging combined with
a powerful resource: testimony from survivors of the Holocaust, housed in the
Shoah Foundation Institute’s Visual History Archive at the University of South-
ern California. By using this resource, my research can shape our understanding
of gratitude, offering an investigation of the cognitive and physiological routes
gratitude can take.
To reach this end, I compiled a set of stimuli that probe the experience of
gratitude in the context of gift-giving. The stimuli set is comprised of situations
derived from the Shoah Foundation Institute’s Visual History Archive, housed
at U.S.C. The archive is comprised of over 50,000 videotaped testimonials from
survivors of the Holocaust. I spent hundreds of hours watching and listening
to the archives, out of which I selected a collection of stories, or scenarios, in
which the survivor receives a gift. In my experiments, participants read these
scenarios and they were asked to imagine the perspective of the survivor and
feel as close to how they would feel if they were in the same situation. While they
imagined themselves in the midst of these scenarios, I measured their cognitive,
physiological, and brain responses.
This dissertation is comprised of three separate experiments. The first, a be-
havioral experiment was used to identify the necessary and sufficient criteria
for the generation of gratitude. I found that the felt need for a gift better pre-
dicted gratitude that the perceived effort for a gift. The second measured heart
rate changes during the experience gratitude. Higher levels of gratitude were
associated with overall lower patterns of heart rate response, indicating an in-
fluence on gratitude from the parasympathetic nervous system. The third mea-
sured brain activity during the experience of gratitude. Gratitude ratings cor-
related with regions of the brain associated with social cognition, reward and
self-processing.
If there is an emotion capable of promoting good human conduct and respect
between people, that emotion is gratitude. Gratitude arms us against despair
and discrimination, fortifies us against trauma, and impels us to help our neigh-
bors. The account of good human conduct created in my dissertation unfolds
a story told by the body, revealing volumes about the necessary and sufficient
conditions for the flowering of human goodness at the level of the person and
at the level of society.
v

Gratitude is a desire or eagerness of love, by which we try
to benefit one who has benefitted us from a like affect of love
— Spinoza
A C K N O W L E D G M E N T S
“There is no vice greater than ingratitude” says Seneca. Well, this is after all a
dissertation on gratitude, so I would be remiss to not write a thorough acknowl-
edgment section in which I publicly acknowledge a small portion of the people
who helped me get to where I am today.
Before I begin to thank some of the individuals responsible for helping me suc-
ceed in graduate school, I would like to recognize the institutions that, through
financial or institutional support, have made my doctoral education possible.
The individuals in the Brain and Creativity Institute, especially Cinthya, Pamela
and Susan, Denise and Faith have been a tremendous support to me in their help-
fulness and kind demeanor. In the Neuroscience Graduate Program, Vanessa
Clark and Gloria Wong stood out for their great friendship and helpful advice.
At the Shoah Foundation Institute where I was supported as a summer fellow,
Dan Leshem, Syuzanna Petrosyan and Crispin Brooks have been an absolute
treasure to work with, providing amazing insight from their disciplines and
many spirited conversations. I also need to express a very strong level of grati-
tude to the Irving and Jeanne Glovin Award, given by the Oskar Schindler Hu-
manities Foundation. This award set off a fantastic string of events, enabling me
to achieve much greater things than if I had missed out on in. Finally, the Greater
Good Science Center and the Templeton Foundation provided a generous grant
to pay for my brain scans. Dacher Keltner and Emiliana Simon-Thomas have
been wonderful people to work with and deserve my deep thanks for their sup-
port.
Now, on to the various people who supported me in other ways.
My mom, dad and sister deserve the greatest thanks. My dad never hesitated
to ask me to make connections between abstract thoughts, he constantly chal-
lenged me to be able to explain why something happened, or how something
would work in the future. Whereas many parents lament their children asking
them “why” all the time, it was my dad that always asked me why things hap-
pened. He raised me to see the world in terms of puzzles and curiosity and to
not only ask questions, but to see how the best questions led to more questions.
My sister, Lisa, who as a quiet and calm child had to endure a household with
a perpetually impetuous and hyperactive sibling deserves much credit. My big
sister always inspired me and challenged me to think harder. In and throughout
graduate school, Lisa has been a source of endless support and guidance. I quite
literally could not have done this without her.
My mom, of course, was the one who taught me to imagine the biggest goals
I could fathom, and then chase them down. Growing up, this meant that I was
vii
able to do all those things that I did that I am proud of: being in Back to the
Future III, earning a black belt, and playing soccer in Europe, all before I was
14. What was so amazing about my mom is that for all those accomplishments,
I always felt that I was the special one, but really all it proved was that my mom
was willing to carry me the distance if I was willing to put in some effort. She
and my dad drove five or six hours on every weekend to drive me to soccer
tournaments, and to push me to be the best I could be. If I have a single phrase
to remember my mom by, it’s her saying, “Oh come on! Just go for it!!” Now I
have another fantastic goal achieved, earning my Doctorate, I can’t wait to do
justice to her memory by going as big as possible from here on out.
In college at U.C. Santa Cruz, I was blessed to work closely with Meg Wilson.
Meg was a tremendous mentor, spending hours working side by side with me,
teaching me the craft of science and experimental design. Meg is an extraordi-
nary scientist, detail obsessed and constructs near bulletproof experiments. She
instilled in me the need to chase down details in conducting an experiment and
to select the best of the compromises when making scientific decisions.
I also need to recognize Anthony Pratkanis, a tremendous influence on me in
college. I literally went to Anthony’s office hours every week, and every week
I pitched to him ideas for research on social influence. Anthony instilled in
me an eye for good research as he learned from Eliot Aronson. Anthony is
a master at experimental design and those hours spent in his office learning
about how to construct experiments have proven invaluable as I conduct my
own investigations.
When I graduated college, I knew I wanted to do research in cognitive neuro-
science, but I did not have any experience with brain imaging. Meg connected
me then to Karen Emmorey. Karen and I travelled to Gallaudet University where
we ran a number of experiments on American Sign Language. I learned from
her there the power of being organized and calculated, and that week taught
me that with enough organization, one could take on many different studies at
once.
My next stop was Stanford, where I was determined to find a job as a pro-
fessional research assistant. I subscribed to Stanford’s job postings, applying for
every laboratory job available. I grew tired of not hearing replies of any sort, so
one day when I saw a job posting, I decided to take matters into my own hands
and walk into the laboratory and ask about the job. By chance, I stumbled in
Bob Dougherty’s office and asked about the job. Bob was kind enough to ex-
plain that I was a little late (and I gathered, under-qualified) for the job. So I
ended up working there as a volunteer, studying frontal lobotomy and reading
development. Four months later though, they were kind enough to hire me as a
paid research assistant, a job I kept until I left for USC.
When I joined USC’s Neuroscience Graduate Program, two of the first people
I met were Bosco Tjan and Lisa Aziz-Zadeh. Spending a semester in Bosco’s
lab was one of the smartest things I did in all of graduate school. Bosco, apart
from being a truly funny and wonderful person, taught me everything I know
about brain imaging. As near as I can tell, Bosco literally knows everything.
viii
On top of that, he is an excellent teacher and mentor and I would absolutely
be nowhere near as prepared for my career as I am if it was not for Bosco’s
influence. Likewise, I made what I consider to be a perfect decision when I
spent the rest of my first year rotating with Lisa. Lisa deserves an award for
patience alone, as we spent the next four years trying to complete one project.
Lisa has been a wonderful mentor and role model, showing that progress in
science does not have to come at the expense of good will or kindness.
I applied to work with Antonio and Hanna as they were first building the
Brain and Creativity Institute. The notion that I could work with someone I had
read so much of, and come to admire deeply was something of a mind-blowing
thought to me. In truth, after I applied, I did not expect to hear back. I figured
that a man as busy as Antonio Damasio would not be one to field cold calls.
One day however, when my phone rang and it was Antonio Damasio calling
me; I thought I was going to fall out of my chair. And when he offered me a
job, I just about keeled over. It is hard to believe that that was almost eight years
ago. I owe so much to Hanna and Antonio, not the least for employing me for
the better part of a decade. Obviously, their accomplishments as scientists are
nearly unfathomable. But I am a firm believer in the notion that we learn more
from how people behave than from what they say or accomplish. In this, I see in
Hanna and Antonio people that dearly care for those around them. I see them
as excellent leaders and motivators. They have assembled a family at the BCI,
one in which the intensity is low, expectations are high, interpersonal conflict is
rare or non-existent, and people are happy to come to work every day. This as an
accomplishment cannot be emphasized enough. Hanna and Antonio represent
much of the best of all my other mentors, and if I am given the chance to lead
my own laboratory, their leadership culture is one I will attempt to emulate.
In graduate school, I have made much closer friends than I ever thought possi-
ble. Dave Herman, an insane man and one of my favorite all time people. Fei, for
her constantly cheery demeanor and vast statistical knowledge is a true buddy
and friend, and definitely the only person I could share a desk with for all this
time. Kingson is a dear friend, someone in whom I admire his tenacity and will-
ingness to tackle new adventures, both in life and in career. Helder has become
a true friend, someone I consider to be a deeply good and caring person. Helder
has been an unflinching friend and a true compatriot, his support in the past
year is a perfect example of how to be a true friend. Without these friends, I
would have graduated many years ago, and been much worse off for it.
I also need to pay special tribute to the many wonderful undergraduates
that helped me gather all this data. In graduate school, with their help, I col-
lected data from roughly 700 experimental participants. My first students, Ash-
noo Nanavati and Kalee Shah were tremendous students, and provided years
of great feedback and hard work. Chris Shafer joined shortly thereafter and
worked for me as the most loyal student I could imagine. Mike Metke was a gift
to find, and he is off to an amazing career.
Finally I need to acknowledge my amazing, one of a kind, beautiful, kind,
generous, amazing wife, Melissa. Melissa is probably the best person I have ever
ix
met. Through all of graduate school, and the hurdles and penury it involves,
Melissa has unflinchingly supported me. As a human being, Melissa dedicates
her life to helping others, both professionally and personally. Melissa is a person
of the highest degree of integrity and an unbreakable heart and work ethic. I
admire her and love her and I take so much pride in being associated with
Melissa. I can only try to imitate and honor the amazing person she is.
There is a Buddhist meditation that involves generating gratitude for one’s
shoes. The notion is that our shoes carry us, expecting nothing, while we may
often forget the need they fulfill in separating us from the dirty world around
us. In being grateful for them, we find a way to love something that we easily
take for granted. Next, we are instructed to, as we reflect on our shoes, consider
what it took to create these shoes. Someone had to assemble the shoes, design
the shoes, harvest the material for the shoes. Soon we realize that if we are truly
grateful to our shoes, we will become grateful to the world. I’d like to close this
lengthy acknowledgment section then, by offering my own deepest gratitude, to
my shoes.
x
C O N T E N T S
1 introduction to gratitude 1
1.1 Seneca and gratitude 3
1.2 Previous research on gift-exchange and gratitude 5
1.3 Goals of this dissertation 6
i the gift-space model of gratitude 7
2 introduction to gift space 9
2.1 Method 12
2.1.1 Participants 12
2.1.2 Procedure 13
2.2 Results 16
2.3 Discussion 24
2.4 Conclusion 28
ii the heart’s memory: autonomic activity during the ex-
perience of gratitude 31
3 the heart’s memory 33
3.1 Method 35
3.1.1 Participants 35
3.1.2 Stimuli 36
3.1.3 Procedure 37
3.1.4 Analysis 39
3.2 Results 40
3.3 Discussion 43
3.4 Conclusion 47
4 gratitude ratings correlate with vagal tone 49
4.1 Method 52
4.1.1 Participants 52
4.1.2 Stimuli 53
4.1.3 Procedure 54
4.1.4 ECG Analysis 56
4.2 Results 56
4.3 Discussion 58
4.4 Conclusion 59
iii the brain’s virtuous cycle: mapping the neural corre-
lates of gratitude 61
5 the brain’s virtuous cycle: mapping the neural corre-
lates of gratitude 63
5.1 Method 65
5.1.1 Participants 65
xi
xii contents
5.1.2 Procedure 65
5.1.3 Image Acquisition 67
5.1.4 Analysis 68
5.2 Results 73
5.3 Discussion 80
5.4 Conclusion 85
6 conclusion: the gratitude prescription 87
6.1 Summary of findings 87
6.2 Seneca’s Oversight 89
6.3 How to be grateful. 92
iv appendix 95
bibliography 99
L I S T O F F I G U R E S
Figure 1 Gift Space Diagram 11
Figure 2 Behavioral Experiment Protocol 16
Figure 3 Three-Dimensional Regression of need, effort, and grati-
tude 17
Figure 4 Principle Component Analysis of need, effort, and grati-
tude 18
Figure 5 Correlation of IRI Perspective-Taking with participants’
mean and standard deviation of effort ratings 19
Figure 6 Histogram of responses from participants scoring high
and low on trait perspective-taking 20
Figure 7 Participants’ mean need, effort, and gratitude ratings in
gift-space 22
Figure 8 Mean need, effort, and gratitude ratings for commonly
reported feelings 23
Figure 9 Heart Rate Stimulus Presentation Paradigm 41
Figure 10 Heart rate patterns associated with high and low levels of
gratitude. 42
Figure 11 Heart rate patterns associated with varying levels of need 43
Figure 12 Heart rate patterns associated with varying levels of grat-
itude 44
Figure 13 Heart rate patterns associated with varying levels of ef-
fort 45
Figure 14 Vagal Tone Correlation 57
Figure 15 Vagal Tone Rating Histogram 57
Figure 16 fMRI Stimulus Presentation Paradigm 68
Figure 17 Neural Correlates of Gratitude 73
Figure 18 Gifts analyzed by condition 75
Figure 19 Task-state correlation for Differing Levels of Gratitude 76
Figure 20 Task-state correlation Matrices of High and Low Trait-
Gratitude Participants 79
L I S T O F TA B L E S
Table 1 Correlations Between Ratings and Personality 20
Table 2 Behavioral Personality Correlation Matrix 21
xiii
Table 3 Statistical analysis of heart rate time series features by
condition and rating level. 48
Table 4 Summary of brain regions from separate analyses 74
Table 5 Summary of functional connectivity comparisons between
trial ratings sorted by condition 77
Table 6 Summary of task-state correlation comparisons between
different personality types. 78
Table 7 Guidelines for the placement of ROIs in each individual’s
anatomical brain image. 98
A C R O N Y M S
ECG electrocardiogram
RSA respiratory sinus arrhythmia
CVT cardiac vagal tone
ROI region of interest
fMRI functional magnetic resonance imaging
MRI magnetic resonance imaging
ACC anterior cingulate cortex
VMPFC ventral medial prefrontal cortex
MPFC medial prefrontal cortex
Temp Pole temporal pole
MTG middle temporal gyrus
Nuc Acc nucleus accumbens
AI anterior insula
PI posterior insula
PMC posteromedial cortex
Inf PMC inferior posteromedial cortex
Sup PMC superior posteromedial cortex
S2 secondary somatosensory cortex
xiv
acronyms xv
STS superior temporal sulcus
KS Kilmogorov-Smirnov test
BOLD blood oxygenation level dependent

1
I N T R O D U C T I O N T O G R AT I T U D E
Within the wide range of mistakes made by those who live recklessly and without
reflection... there is almost nothing, I would claim, more harmful than our ignorance of
how to give and receive benefits.
But what is more honorable than gratitude? The scope for practicing this virtue is as
broad as life itself.
— Seneca (On Benefits, 1.1.1 and 4.19.4 respectively)
When we are the beneficiaries of good human conduct, we experience a con-
cert of positive emotions ranging from relief to elation. These emotions can in
turn motivate us to expend great sums of energy to reward those near the source
of the good conduct, creating, quite literally, a virtuous cycle. If this cycle has a
fuel or a catalyst, it goes by the name of gratitude. Gratitude, then, resides at the
center of ethical human conduct, serving as a fulcrum with which people seek to
do right by others. In so doing, gratitude seeds the fields of human flourishing.
Gratitude is thought to be the product of receiving a gift that fulfills an im-
portant need, and that comes as a result of great effort from the giver. Gratitude
is thought to be something to be enhanced and cultivated, as it is the ‘glue that
binds souls together’ according to the Maori tribespeople. In fact, nearly every
major philosopher and religious tradition stresses the importance of gratitude,
yet in spite of this, empirical investigations of gratitude are notably rare (Em-
mons and McCullough, 2004).
Although it may seem like the lack of investigation into gratitude is an over-
sight, there is good reason for this lack. Gratitude is complicated, high-dimensional
and difficult to predict and elicit. People often do not feel gratitude when they
1
2 introduction to gratitude
should, and sometimes gratitude is felt when none should be expected. Grati-
tude can emerge in large amounts, or in small amounts. Gratitude can be di-
rected toward another person, but also toward the world, or to the universe at
large. A study of gratitude needs to account for this high-dimensional space, re-
stricting the approach enough to gather useful data, but also expansive enough
to gather data that are meaningful. My approach in this dissertation is to strike
a balance between the restrictions necessary to conduct rigorous science, while
still paying homage to the rich concert of emotional feelings involved in grati-
tude.
In this introductory chapter, I will draw briefly from Seneca’s philosophy of
benefits, then I will describe previous research that examines gratitude. These
will form a brief framework of gratitude leading to the first chapter of my thesis.
There I will introduce the concept of ‘gift-space’ which is a term I coined to
describe the space in which reactions a gift can inhabit. I will show that the felt
need for a gift, combined with the perceived effort in providing a gift are the
primary determinants of gratitude, but both need not be present for gratitude
to occur. In chapter 2, I will describe another experiment in which I measured
participants’ heart rate while they reflected on gifts, and I correlated their heart
rate responses to the level of gratitude that they reported. This chapter revealed
that higher ratings of gratitude were associated with overall lower heart rate.
In chapter 3, I will explore individual differences in autonomic nervous system
function, measured using cardiac vagal tone (CVT), correlated with participants’
gratitude ratings. This analysis revealed a negative correlation between CVT and
gratitude ratings, but a positive correlation between CVT and the standard devi-
ation of the ratings. This result indicates that more gratitude may not always be
ideal, and points to questions about what is the appropriate level of gratitude
for a given gift. In chapter 4, I will describe my experiment in which I mea-
sured brain activity in the participants while they reflected on the gifts. This
study found that the ratings of gratitude for a gift correlated with brain regions
1.1 seneca and gratitude 3
in the ventral medial prefrontal cortex (VMPFC), a region associated with social
cognition, reward, and self-processing.
In sum, my dissertation represents one of the few studies of gratitude that fo-
cus on the experience of gratitude itself, rooted in a validated framework, with
special attention paid to the known neurobiological underpinnings of feeling
and emotion. My dissertation also represents the only study of gratitude to link
separate methodologies, leading to the opportunity to draw broad and mean-
ingful conclusions about gratitude and its enhancement in our daily lives.
1.1 seneca and gratitude
Seneca’s On Benefits is a treatise on the nature of virtue, gift-giving, and grati-
tude. In this piece, Seneca details what are the critical facets of giving a benefit to
another person. Seneca is perhaps best known for his statement that there “is no
greater vice than ingratitude.” In Seneca’s thinking, gratitude is a fundamental
tenet of human nature; to deny an opportunity for gratitude is to deny our true
nature. In this, Seneca holds the belief that the nature of virtue and gratitude is
intrinsic to ourselves. Seneca summarizes this belief: “[T]he convictions that we
should behave virtuously towards our fellow-men were not devised by man to
cope with his weakness, but are innate”(Griffin, 2013). This sentiment informs
whether or not we think man came to live in communities as part of a natural
predilection, or as part of an innate drive to socialize. Insofar as gratitude, if
living in a community is part of a natural drive, and gratitude is our way of un-
derstanding and keeping track of social relationships, then we have support for
a notion of gratitude as an innate response to understanding and maintaining
relationships.
On Benefits describes the multi-dimensional aspect of gratitude, “Nothing is
fitting for anyone in the abstract; it makes a difference who the giver is, who the
4 introduction to gratitude
recipient is, when, why, where, and so forth–all the factors necessary to think
through an action properly (2.16.1).” Seneca notes that some gifts that should
be treated with joy, and some gifts should be rejected depending on the context
perceived of the donor and the gift itself. He correctly identified gratitude as
being a set of related emotional feelings, falling on a continuum from gifts that
are basically useless, to gifts that generate strong feelings of gratitude.
Seneca proposes that the intent of a gift is the primary determinant of grat-
itude. He says, “It is the intention that exalts what is petty and brings light to
what is shabby; intention humbles those things that are grand and generally
regarded as valuable (1.6.2).” He also says, “If benefits consisted in the things
and not precisely in the intention of the benefactor, then the benefits would be
greater to the extent that the things received are greater. But that is not the case
(1.6.3).” As researchers have conceived of gratitude as being generated based
upon the need for a gift combined with the benefactors’ effort for a gift, then
Seneca would say that it is the effort that drives gratitude. As we will see, this
is one of Seneca’s oversights. The data presented later in my dissertation will
demonstrate that while a benefactor’s effort in producing a benefit is a deter-
minant of gratitude, it is in fact the felt need for a benefit that better predicts
gratitude.
Seneca also commits an oversight in his view that gratitude can and should
inhabit any space. Most notably, Seneca is know for proposing that gratitude can
even emerge from the relationships between a slave and a master. Seneca says
on the matter: “when the good will [a slave] feels for his master goes beyond
the boundaries set by his position in life and when he aims at something higher,
... and so surpasses the expectations of his master, then we have found a case
of benefit given within the household (2.22.1).” As we will see, this also is not
an accurate prescription of gratitude. The data in my dissertation will indicate
that there are appropriate levels of gratitude that are related to the context and
1.2 previous research on gift-exchange and gratitude 5
relationship of the exchange. In other words, more gratitude is not inherently
better gratitude.
From Seneca we learn that gratitude is a social emotion, built into relation-
ships and cultures. Gratitude is decomposable into two primary dimensions:
‘need’ and ‘effort.’ Altering the need and effort of a gift then will have an effect
on the gratitude one feels for a gift, or on one’s ability to recognize receiving a
benefit. Given this philosophical context, we can examine modern research into
gratitude.
1.2 previous research on gift-exchange and gratitude
The research that has examined the link between emotion and gift-exchange
has shown that how we perceive the gifts we receive depends on two factors:
how we perceive the effort put into a gift, and our need for the gift (Lane and
Anderson,1976; Tesser et al.,1968). In terms of connecting feelings and emotions
to receiving gifts, it has been shown that gifts that are perceived as coming at
great effort from the giver and that fulfill a personal need for the recipient will
elicit the emotion of gratitude (Emmons,2008). It has also been shown that gifts
for which we feel a strong sense of gratitude will lead us to perceive a closer
relationship to the giver (Algoe et al., 2008, 2013; Ruth et al., 2004) and benefits
that we are grateful for will inspire us to take great efforts to repay the original
gift-donor (Bartlett and DeSteno, 2006).
The benefits of gratitude in the treatment of mental illness are well-documented
(Wood et al., 2010). Gratitude has been show to be a protective factor in the de-
velopment of post-traumatic stress disorder in war veterans (Kashdan, Uswatte
and Julian,2006; Kashdan, Uswatte, Steger and Julian,2006; Vernon et al.,2009).
As a personality trait, gratitude is associated with higher levels of grit and per-
severance (Kleiman et al.,2013) and lower levels of depression (Froh et al.,2011).
6 introduction to gratitude
Finally, gratitude is associated with higher levels of well-being and life satisfac-
tion (Wood et al., 2008, 2009).
1.3 goals of this dissertation
The previous research has provided a solid foundation on which to build, but
it has left three questions crucial for understanding gratitude. First, from this
research it is obvious that gratitude has many benefits, but what are the nec-
essary and sufficient conditions for gratitude to be generated? Second, given
gratitude’s connection to health and well-being, what changes occur in the body
during the experience of gratitude? Third, how can we use these findings to
enhance gratitude in our daily lives?
The data that comprise my thesis will address the first two of these three
points. For the third point involving gratitude in our daily lives, I will conclude
by drawing together the conclusions of my studies to build a prescription for
how to enhance and cultivate gratitude. The prescription includes focusing on
how the things we have and the things we receive fulfill deeper needs than we
may originally perceive. In addition, the data indicate that instead of trying to
force ourselves to be more grateful for every thing, we should focus on differ-
entiating situations from one another, finding the appropriate level of gratitude
for a given circumstance.
Part I
T H E G I F T- S PA C E M O D E L O F G R AT I T U D E

2
I N T R O D U C T I O N T O G I F T S PA C E
The experience of receiving a gift can inspire a concert of positive emotions
ranging from relief to elation. In less ideal situations, receiving a gift can also
lead to feelings of resentment, anger and even bitterness. In the best of cases
however, a properly aimed and received gift can inspire a feeling of gratitude in
the recipient. To the extent that this gift is well-intended and needed, the level of
gratitude can range from a general uplifting feeling for a small gift, to feelings
of overwhelming gratitude for gifts that may be life-saving.
We know, from previous investigations, that gratitude generally arises when
we receive or recognize a needed and well-intentioned benefit (Emmons and
McCullough, 2004). We also know that gratitude can command great sums of
energy in the name of reciprocation (Bartlett and DeSteno, 2006; DeSteno et al.,
2010) and that it strengthens social relationships (Algoe et al., 2008). The ef-
fects of “counting blessings” in daily life have also been associated with robust
changes in personal well-being and pro-social behavior (Emmons and McCul-
lough, 2003; Mccullough et al., 2002).
Together, the above findings provide a foundation to the understanding of gift-
exchange and gratitude, but many fascinating questions remain as to what are
the basic mechanisms of the generation of gratitude. Namely, we often receive
gifts that come at great ‘effort,’ but do not fulfill a tangible need or utility, and yet
these gifts can inspire strong feelings of gratitude. Conversely, we often receive
gifts that we ‘need’ very much, but come at little or even accidental effort, and
still again we may have strong feelings of gratitude–or none at all. In other
9
10 introduction to gift space
words, how does the context surrounding receiving a gift color our feeling of
gratitude?
Here, I seek to clarify the relationship among need, effort, and gratitude in
the context of gift-giving, focusing on the perspective of the recipient. I am
using a new model for exploring how we respond to the gifts we receive. The
model examines gifts as a class of emotionally competent stimuli capable of
generating a wide array of feelings (Damasio, 1994, 1996, 2011; Feinstein et al.,
2011). It has been suggested that the link between emotion and gift-exchange,
i.e. a recipient’s appreciation of a gift, tends to depend on two factors: the need
for the gift, and the effort put into the gift-giving (Lane and Anderson, 1976;
Tesser et al.,1968). It has been shown that gratitude is elicited by gifts perceived
as coming at great effort from the giver when they fulfill a personal need of the
recipient (Ames et al., 2004; McCullough et al., 2001; Tsang, 2006).
The model uses the terms effort and need as labels for a set of dimensions
or axes that initially define a two-dimensional space, with need on the x-axes,
and effort on the y-axes; I also add gratitude on the z-axis. I manipulate effort
and need of a gift, together or independently, and map their relationship to
gratitude. The model allows, for any gift, to independently vary effort and need.
These variations will alter how a gift is perceived, for instance the highest ratings
of gratitude are expected to take place for gifts that are rated as both high-need
and high-effort (See Figure 1). Other gifts, say those that are rated as low-need
and high-effort may also elicit some gratitude, but these gifts also may also elicit
other feelings, such as awkwardness, or anxiousness.
Gratitude is high-dimensional, meaning that there are many factors that can
influence one’s gratitude for a gift. The approach here, however, is to reduce
the dimensions of gift-exchange to those that are most currently accepted as the
primary factors of gratitude; that is: the combination of the recipient’s needs for
the gift, and the perception of the donor’s efforts to provide a benefit (Emmons
and McCullough, 2004; Lane and Anderson, 1976; Tesser et al., 1968). I opera-
introduction to gift space 11
(y) Effort
(x) Need
High-Need/High Effort
Low-Need/High Effort
High-Need/Low Effort
Figure 1: Two-dimensional depiction of gift-space, showing need on the x-axis and effort
on the y-axis
tionalize ‘need’ and ‘effort’ broadly to refer to the state of the recipient’s need,
and the perceived state of the donor’s effort. Specifically, need refers broadly to
the degree to which the recipient has a need for the gift; it can be a basic need,
a physiological need or even a social or cultural need. Effort refers broadly to
the amount of perceived intention, cost, inconvenience, or risk that is involved
in providing the gift.
The model evaluates gratitude by examining it as a product of effort and need.
To visualize this relationship, I add a third-dimensional axis for gratitude to the
original 2-D plot of effort and need. For each participant’s rating of each gift, I
plot their responses as X, Y, and Z coordinates–X for need, Y for effort and Z
for gratitude. This provides a 3-D representation where each gift resides in the
gift-space and shows the relationship among the three parameters (need, effort,
and gratitude). In so doing, the model reveals a set of necessary and sufficient
criteria for gratitude to emerge.
Given that the goal of this study is to examine the varieties of grateful expe-
rience, I need a set of stimuli capable of creating strong, but varied emotional
reactions to the gifts. To this end, I have compiled a set of stimuli from scenarios
from the Shoah Foundation Institute’s Visual History Archive, housed at U.S.C.
The archive comprises over50,000 videotaped testimonials from survivors of the
Holocaust. Out of these testimonies I selected a collection of situations in which
the survivor received a gift. In the experiment, participants read these scenarios
12 introduction to gift space
and I ask them to take the perspective of the survivor and try to feel as close to
how they may have felt if they had been in the same situation.
My aim is to determine the validity of the model of gift-space as a tool to
(1) investigate the necessary and sufficient criteria for gratitude, (2) examine
qualitatively other emotions reported in response to the gifts, and (3) examine
individual personality differences in how gifts are rated. I hypothesize that: 1.
Reactions to gifts will vary along dimensions related to the perceived effort and
the personal need for the gift, and that effort and need will correlate with grat-
itude using multiple regression analysis. In addition, I predict that the variance
in the ratings of the gifts will best be explained in terms of effort and need, as
analyzed using principle component analysis. 2. Variations in ratings between
effort and need for a gift will alter or determine how a gift is perceived. 3. Indi-
vidual personality differences measured using personality scales will correlate
with how gifts are rated.
2.1 method
2.1.1 Participants
Subjects were invited to participate using U.S.C.’s Psychology Department sub-
ject pool. Forty three (21 male) participants enrolled in the study (mean age =
20.26, sd = 1.52). One participant had to be excluded from the analyses due to
failure to complete the study, reducing the total N to 42 subjects (21 male). All
participants provided full informed consent and all research was conducted in
accordance with U.S.C.’s Institutional Review Board policies on human subjects
research.
2.1 method 13
2.1.2 Procedure
With the goal of sampling a wide range of gifts, I created a set of stimuli capable
of probing the varied constellations of experiences associated with receiving a
gift. I assembled a collection of stories based upon testimony from survivors of
the Holocaust, looking for instances in which the survivor describes receiving a
gift, of any sort, from another person. The stories of gifts received were selected
to fall into one of three categories of our gift-space model. For instance, many
survivors tell stories in which they are taken under the protection of strangers,
feeling strong gratitude for such truly life-saving gift. This type of gift was coded
as high-effort (the risk taken by the family that is hiding the survivor), and high-
need (the life-saving consequences for the recipient). Other gifts were selected to
be high-effort/low-need. An example of these is for instance a survivor receiv-
ing peanuts from a stranger on the street, but being allergic to peanuts and thus
not able to eat them. Yet another category were gifts of high-need/low-effort.
Examples of these were gifts that involved situations in which a donor gave
something at no great cost, but that would be of critical help to the survivor, e.g.
a survivor, while trying to flee her country, has a clerk at the border that inten-
tionally ‘forgets’ to stamp ‘j’ for ‘jude’ on the survivor’s passport, thus granting
her the ability to leave the country and saving her life.
The scenarios described by the survivors were transcribed into stimuli rang-
ing from 30 to 40 words and were rephrased to be in the first person. I asked
the participants to read these scenarios as if they were the receiving protagonist
and tell for each of them: how much effort they felt was involved in the gift,
how much they needed the gift, how grateful they felt for the gift; they were
also asked to report how they felt after receiving the gift by selecting a feeling-
descriptor from a list of feelings. This list contained: Grateful, Bitter, Happy, Sad,
Anxious, Awkward, Content, Relieved, Elevated. These terms were chosen be-
14 introduction to gift space
cause of: their close relationship to gratitude, (elevation, relief, content); their
relation to general basic emotion (Happy, Sad); and feelings described in a pilot
study I conducted prior to the current study (anxious, awkward, bitter). The
goal of including the feeling labels in the task was to determine if certain com-
binations of effort and need led to unique and/or separable reported feeling
states. There were 16 stimuli in each condition (high-gratitude, high-effort and
high-need); a total of 48 stimuli.
The experiment was designed to be similar to a visit to the United States
Holocaust Museum in Washington D.C., where visitors are asked to step into
the perspective of a survivor as they go through the four phases of the Holo-
caust. The museum has four floors–one for each phase of the Holocaust, and
participants visit the floors in succession; on each floor they view videos, photos
and artifacts from that phase intended to induce the experience of what was oc-
curring during that phase of the tragedy. In our experiment, the stimuli are also
broken up into the four phases of the Holocaust. Each phase of the experiment
began with an in-house created documentary detailing the events of that epoch.
The documentaries were about two minutes long and were created in collabo-
ration with students from the U.S.C. school of Cinematic Arts so as to mimic
a Ken Burns-style documentary. The documentaries relied on powerful images
as well as a professional actor providing a voice-over description so as to elicit
strong emotional reactions in the participants. The purpose of the documentary
was to deeply involve the participants in the historical situation by giving infor-
mation enough for them to imagine the perspective of the survivors as well as
to provide brief breaks throughout the study to help guard against habituation
to the stimuli.
The participants received detailed instructions for providing their ratings of
the gifts, which are included verbatim in Appendix A.
After watching the sequence of stimuli, participants were asked to fill out a se-
ries of questionnaires, three commonly used psychological questionnaires: The
2.1 method 15
Maslow Need Satisfaction scale (Lester, 1990), the six item gratitude question-
naire (GQ-6) (Mccullough et al., 2002), and the Interpersonal Reactivity Index
(IRI) (Davis, 1994). Participants also filled out a homemade questionnaire that
asked them about their experience during the experiment. This questionnaire
was eight items in length and asked subjects to rate on a likert scale the fol-
lowing questions: (1) how involved they felt in the task, (2) how similar their
feelings during the situations matched what they would have felt if the experi-
ence was real, (3) how difficult it was to put themselves in the situations and (4)
how much they felt they had an increased understanding and sense of empathy
for the Holocaust by going through the experiment. After these questions were
answered, the participants were asked four additional open-ended questions
assessing if there were any confusing stimuli, if any stimuli were particularly
moving, if they had tried to figure out what the experiment was about, and if
they had any personal experience with the Holocaust. (See Figure 2).
Participants provided their ratings for the gifts by making a mark on a contin-
uous line regarding their judgment of the scenario’s need, effort, and gratitude
respectively. Thereafter they had to select what feeling they felt after the scenario.
Definitions of what was meant by need, effort, and gratitude respectively were
given to the participants before the experiment started. As a means to scale their
responses, participants were told to only use the highest ratings of effort, need,
and gratitude for gifts in which life or great harm was at risk. Participants were
also told to treat each story as an independent event in and of itself, and that
the stories were all based upon real events drawn from the Shoah Foundation
Institute’s Visual History Archive.
The questions were given in the form of a 20 page booklet. After the experi-
ment, the booklet was scanned into an electronic format and then a homemade
software program run via MATLAB was used to analyze the responses. For each
item, the distance along the line was computed as a percentage of the full length
of the line, and the ratings for need, effort, and gratitude were entered into a
16 introduction to gift space
Experimental Session Procedure
1. Consent and Instructions
2. Experimental Session
1. Phase Description Video
2. Stimulus
2a. Rate Effort
2b. Rate Need
2c. Rate Gratitude
2d. Select Emotion
3. Post Experimental Questionnaires
4. Debriefing
12x
4x
Figure 2: Depiction of how experimental sessions were conducted.
data matrix as the X, Y and Z coordinates to be plotted in three dimensions. The
final data were analyzed using MATLAB and SPSS using3-d multiple regression
to determine the correlation among need, effort and gratitude. A principle com-
ponent analysis (PCA) was used to identify the ‘shape’ of the data to determine
if in fact need and effort were the best dimensions to understand gratitude. To
answer the other questions of the study regarding individual differences, Pear-
son’s r tests were used to determine bivariate correlations.
2.2 results
A three-dimensional multiple regression analysis revealed a correlation between
need, effort, and gratitude (r = .52;F =789;p< .0001). (See Figure 3)
The principal component analysis (PCA) revealed a clear ‘shape’ to the data;
need and effort show a clear relationship to gratitude, and all three variables
appear to increase in relation to each other. However, it is clear that the plane
of the principal component is tilted more toward the need axis, indicating a
bias toward rating gifts as high-need and high-gratitude. By comparison, a ran-
dom plotting of dots in the 3-d space would, theoretically, reveal no such prin-
2.2 results 17
Gratitude
Need Effort
r(40) = .559
p < .001
Effort
Gratitude
Need
Gratitude
Figure 3: Three-Dimensional Regression of need, effort, and gratitude
cipal component. Plotting the raw data also shows that there are no data points
that show high-need or high-effort that have concurrently low-gratitude.(See
Figure 4).
I was also interested in determining the relationship between ratings of gifts
and individual differences in personality. I found a positive correlation between
participants’ mean need, and mean gratitude ratings with participants’ scores
on the GQ-6 questionnaire (Mccullough et al., 2002). There were also numerous
correlations between scores on the IRI (Davis, 1994). Participants’ scores on the
fantasy scale and the perspective-taking scale of the IRI correlated negatively
with the mean effort ratings, and positively with the standard deviation of the
effort ratings. (See Table 2 for the full results of the personality correlations).
Scores on the empathic concern sub-scale correlated positively with the standard
deviation of participants’ effort ratings. Participant scores on the perspective-
18 introduction to gift space
Need
Gratitude
Effort
Need
Gratitude
Gratitude
Effort
Figure 4: Principle Component Analysis of the raw data. In this analysis, each dot is
one data point, blue dots are above the plane, red are below. The plane at the
center of the data points is the principle component created by the analysis,
which represents the best way to explain variance in the ratings.
taking scale also correlated positively with the standard deviation of the need
scale, and there was a non-significant trend indicating a negative correlation
with the mean need ratings. (See Table 4 for the full results of the personality
correlations). There were no significant correlations between scores on the need
scale (Lester, 1990) and the ratings of the gifts. (See Figure 5 for scatter plots of
the IRI Perspective-Taking scores and the effort ratings). There were also no sex
differences in how the gifts were rated.
2.2 results 19
Mean Effort Rating
Standard Deviation of Effort Rating
IRI Perspective Taking
0.20
0.40
0.60
0.80
1.00 0.40
0.20
0.00
0.10
0.30
0 5 10 20 25 30 15 0 5 10 20 25 30 15
IRI Perspective Taking
Figure 5: Correlation of IRI Perspective-Taking with participants’ mean and standard
deviation of effort ratings
Given the correlations between perspective-taking and the ratings of perceived
effort, it was also of interest to understand more closely how participants high in
perspective-taking rated the stimuli different than participants low in perspective-
taking. To visualize this difference, the participants were ranked according to
their perspective-taking scores, and I compared the top and bottom ten re-
spectively. The comparison involved creating a histogram of their responses in
their rating of perceived effort, and then examining how the shape of the his-
togram differed between the high and low perspective-takers. This visualization
revealed a pattern that confirms the result given by the correlation analysis. Par-
ticipants in the high perspective-taking group have responses that populate the
range of scores, in a somewhat bi-modal distribution. Participants in the low
perspective-taking group showed a large preference for the highest effort scores,
with a linear trend in the number of their responses at each level. (See ??).
I was also interested in possible correlations within the various personality
measures and the homemade four-item questionnaire given after the experi-
ment. Scores on the GQ-6 correlated with how similarly participants felt their
emotions would have been if they were actually in the situations. Participant
scores on the IRI Fantasy sub-scale correlated with how difficult participants
20 introduction to gift space
Effort rating
Number of responses
Effort rating
Number of responses
Low IRI-PT subjects High IRI-PT subjects
140
100
60
20
1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10
140
100
60
20
Figure 6: Comparison of high and low perspective-takers using histograms of the two
group’s ratings of perceived effort.
Personality
Measure Statistic
Mean Effort
Mean Need
Mean Gratitude
Std. Dev. Effort
Std. Dev. Need
Std. Dev. Gratitude
How Involved
How Similar
How Difficult
Increased Empathy
GQ-6
r 0.184 .314* .329* -0.061 -0.037 -0.175 0.19 .334* -0.226 0.018
p<= 0.243 0.043 0.033 0.701 0.817 0.268 0.229 0.031 0.15 0.912
IRI: Fantasy
Scale
r -.390* -0.161 -0.008 .312* 0.124 0.03 -0.068 -0.285 .397*** -0.173
p<= 0.011 0.308 0.961 0.044 0.435 0.848 0.668 0.067 0.009 0.275
IRI: Empathic
Concern
r -0.212 -0.038 0.077 .454** 0.28 0.262 -0.269 -0.2 .444** -.311*
p<= 0.178 0.812 0.629 0.003 0.073 0.094 0.085 0.203 0.003 0.04
IRI: Perspective
Taking
r -.422** -0.283 -0.032 .487** .307* 0.167 -0.136 -0.233 .415** -0.092
p<= 0.005 0.07 0.841 0.001 0.048 0.29 0.39 0.138 0.006 0.562
IRI: Personal
Distress
r 0.282 0.244 0.02 -0.201 -0.141 -0.027 0.138 -0.001 -0.294 0.039
p<= 0.07 0.119 0.902 0.203 0.373 0.865 0.385 0.996 0.059 0.805
Table 1: Correlations between ratings of the stimuli with personality measures.
found it to put themselves in the scenarios. The same result was also present in
relation to the IRI Empathic Concern sub-scale were positively correlated with
how difficult participants found it to put themselves in the scenarios, but there
was a negative correlation with how much the experiment increased their empa-
thy for the Holocaust. On the IRI perspective-taking sub-scale, scores correlated
with how difficult participants found it to put themselves in the scenarios. See
Table 4 for details.
I also calculated the between measure correlation of each personality measure
to identify how certain traits, measured by each scale may correlate with each
2.2 results 21
Personality
Measure Statistic
GQ-6
IRI: Fantasy Scale
IRI: Empathic Concern
IRI: Perspective Taking
IRI: Personal Distress
Physiological
Needs
Safety Needs
Belonging Needs
Esteem Needs
Self-Actualization
Needs
GQ-6
r 1 0.001 -0.168 -0.213 0.195 0.276 .428** .342* .453** .373*
p<= 0.996 0.289 0.177 0.215 0.076 0.005 0.027 0.003 0.015
IRI: Fantasy Scale
r 0.001 1 .587** .644** -0.272 0.044 0.104 -0.006 0.107 0.056
p<= 0.996 0.001 0.001 0.082 0.78 0.511 0.971 0.501 0.727
IRI: Empathic Concern
r -0.168 .587** 1 .760** -.319* 0.108 0.106 -0.109 0.035 0.011
p<= 0.289 0.001 0.001 0.04 0.497 0.505 0.493 0.825 0.947
IRI: Perspective Taking
r -0.213 .644** .760** 1 -.559** 0.154 0.177 -0.023 0.06 0.035
p<= 0.177 0.001 0.001 0.001 0.331 0.261 0.885 0.708 0.826
IRI: Personal Distress
r 0.195 -0.272 -.319* -.559** 1 0.031 -0.082 0.134 0.041 -0.152
p<= 0.215 0.082 0.04 0.001 0.844 0.606 0.397 0.795 0.338
Physiological Needs
r 0.276 0.044 0.108 0.154 0.031 1 .767** .424** .547** .504**
p<= 0.076 0.78 0.497 0.331 0.844 0.001 0.005 0.001 0.001
Safety Needs
r .428** 0.104 0.106 0.177 -0.082 .767** 1 .463** .628** .685**
p<= 0.005 0.511 0.505 0.261 0.606 0.001 0.002 0.001 0.001
Belonging Needs
r .342* -0.006 -0.109 -0.023 0.134 .424** .463** 1 .568** .510**
p<= 0.027 0.971 0.493 0.885 0.397 0.005 0.002 0.001 0.001
Esteem Needs
r .453** 0.107 0.035 0.06 0.041 .547** .628** .568** 1 .741**
p<= 0.003 0.501 0.825 0.708 0.795 0.001 0.001 0.001 0.001
Self-Actualization
Needs
r .373* 0.056 0.011 0.035 -0.152 .504** .685** .510** .741** 1
p<= 0.015 0.727 0.947 0.826 0.338 0.001 0.001 0.001 0.001
Table 2: Correlation matrix showing correlations between participants’ scores on person-
ality measures.
other. Participant scores on the GQ-6 (Mccullough et al., 2002) correlated with
ratings of the fulfillment of their safety needs, belonging needs, esteem needs
and self-actualization needs on the Maslow Need Satisfaction index (Lester,
1990). For the IRI, (Davis, 1994) the fantasy scale correlated with the IRI’s em-
pathic concern and perspective-taking scales; the empathic concern scale with
the personal distress scale; the perspective-taking scale with the personal dis-
tress scale. The Maslow Need Satisfaction index revealed significant correlations
between each of its measures. For the full correlation matrix of these results, see
Table 2.
Participants’ responses to the events were also expected to vary between indi-
viduals, therefore, I performed several qualitative analyses to start to understand
the variability of the responses to events, how much participants themselves var-
ied in their ratings, and what the relationship was among the ratings of need,
22 introduction to gift space
effort, and gratitude and the feeling labels chosen by the participants after each
event.
Gratitude
Need Effort
Gratitude
Gratitude
Need Effort
Figure 7: Participants’ mean need, effort, and gratitude ratings plotted together in gift
space reveals that the participants’ ratings varied, but they all appear to fall in
a similar plane.
To identify how participants varied overall on their ratings, the participant’s
mean need, effort, and gratitude rating was calculated for each participant and
plotted (one point for each participant), in gift-space. Using mean values would
inherently bring the ratings toward the middle of gift-space, but with this ap-
proach, I was able to show how much the participants varied in how they rated
need, effort, and gratitude, and if any individual participants were likely to rate
one element higher than another. For instance, if a participant would systemat-
2.2 results 23
ically rate the high-need stimuli as the highest gratitude, this would shift the
data point toward that end of the gift-space. However, the results showed that
participant responses were largely clustered along a plane similar to that found
in the 3-d regression, indicating that while the participants may have varied
greatly on the average overall ratings, they appear to obey a similar ‘rule’ for
forming their relationship among need, effort, and gratitude. (See Figure 7).
Effort
Effort
Gratitude
Effort
Gratitude
Gratitude
Need
Need
Need
Figure 8: Mean coordinates for commonly reported feeling states across all trials and
participants. Please note that the axes run from0.3 to1 in this figure, allowing
for better visualization. Abbreviations: B = Bitter, C = Content, G = Grateful, H
= Happy, R = Relieved, S = Sad, W = Awkward, X = Anxious; Combinations of
letters reflect the participants choosing two or more feeling words to describe
their feelings after a stimulus.
To determine if certain ratings of need, effort, and gratitude were associated
consistently with words describing certain feelings, mean need, effort, and grat-
itude ratings were matched to feeling labels that appeared 25 or more times
across all the participants. This qualitative result shows that certain feeling la-
bels, and combinations of feeling labels were chosen repeatedly by the partici-
pants, and that when plotted together, the different labels are spread through the
24 introduction to gift space
gift space, showing that different feelings were consistently marked by different
ratings of need, effort and gratitude. (See Figure 8).
2.3 discussion
The aim of this study was to determine the relationship among need, effort,
and gratitude to test a model for the understanding of gift-exchange from the
perspective of the recipient. The framework of this model rests on a concept
I call ‘gift-space.’ If need, effort, and gratitude are plotted as labels on a 3-d
set of axes (x, y, and z respectively), the space they encompass is called gift-
space. This first investigation of gift-space used a variety of gifts to populate the
various regions of the gift-space. The gifts chosen were real life scenarios chosen
from stories of survivors of the Holocaust. For each gift, participants were asked
to put themselves in the situation of the survivor and imagine how they would
feel if they had been in the same situation. After each gift, participants rated
the gift in terms of how much they felt they needed the gift, how much effort
they perceived going into the gift, and how grateful they felt after receiving the
gift. After these three ratings, participants identified words that described how
they felt after receiving the gift. I hypothesized that: 1. Reactions to gifts would
vary along dimensions related to the perceived effort and the personal need for
the gift, and that effort and need would correlate with gratitude. In addition, I
predicted that the variance in the ratings of the gifts will best be explained in
terms of effort and need, as analyzed using principle component analysis. I also
predicted that variations in ratings between effort and need for a gift would
alter or determine how a gift is perceived. Finally, I predicted that individual
differences, measured using personality scales, would also correlate with how
gifts are rated.
2.3 discussion 25
The hypothesized relationship among need, effort, and gratitude was con-
firmed. This initial finding comes as no surprise, given that it replicates pre-
vious findings (Tesser et al., 1968; Lane and Anderson, 1976). It also fits well
the posited definition of gratitude as being the emotion resulting from receiv-
ing a well-intended gift that fulfills a serious need of the recipient (Emmons
and McCullough, 2004). However, the finding provides novelty in a number of
ways. First, in addition to replicating previous results, I add a third dimension
that determines how the factors related to the perceived effort of giving and
the felt-need for a gift interact in combination to form the gratitude response.
In addition, the results are able to determine how need or effort alone may in-
dependently lead to feelings of gratitude. For instance, I show that while high
need and high effort together generate gratitude, both are not necessary. High
levels of gratitude are reported when there are high enough levels of effort, and
relatively low levels of need, and the reverse is also true. However, while both
effort and need correlate with gratitude, need ratings were a better predictor
of gratitude. This relation was not predicted, but given the circumstances of
the study, in which participants were inherently in a position of dire need, it
should have been expected. In the PCA analysis, the relationship among need,
effort, and gratitude also supported our hypotheses. PCA revealed that the over-
all variance in gift-ratings was best explained by the chosen dimensions of need
and effort.
The correlation between participants’ ratings and individual differences on
personality measures had also been predicted. The scores on the GQ-6 (Mccul-
lough et al., 2002) correlated with the mean need ratings and mean gratitude
ratings. The finding that the GQ-6 would correlate with mean gratitude ratings
is to be expected, but that it correlates with the mean need ratings, and not mean
effort ratings is less obvious. Participants that are more grateful may indeed be
more likely to understand their own need for a gift and use that information to
help increase their gratitude ratings. This may mean the two measures are not
26 introduction to gift space
entirely independent. I found an unexpected correlation between the ratings of
the gifts and the participants’ scores on the IRI (Davis,1994). I had hypothesized
that higher scores on the perspective-taking sub-scale of the IRI would posi-
tively correlate with the participants’ mean effort ratings. Given that gratitude
is highly related to a capacity to understand the effort that goes into gift-giving,
it was to be expected that those who are more apt to take the others’ perspec-
tive would be more apt to recognize, on average, higher levels of giving effort.
The result however, was the exact opposite of our prediction. Perspective-taking
scores were negatively correlated with mean effort ratings, but on the other hand
they were positively correlated with the standard deviation of the effort ratings.
This finding may, in fact, be more readily explainable than our original predic-
tion. Participants with high scores on the perspective-taking scale may be more
likely to take the perspective of others, and because of this they also may be
more accurate and more differentiating in how they understand others’ perspec-
tives, as opposed to simply attributing higher effort levels overall. The positive
correlation with the standard deviation of the effort ratings tells us the second
half of the story. If people who are more likely to differentiate others’ perspec-
tive, this should lead to a greater variety in actual ratings. Given that the gifts
were designed to vary widely across the dimension of effort and need, it makes
sense that those more accurate in their judgments of others’ minds would also
be more likely to be more attuned to the varied levels of the effort depicted in
the stories. There is also a similar result for the IRI’s fantasy scale: a negative
correlation with the mean effort ratings, and a positive correlation with the stan-
dard deviation of the effort ratings. The same explanation is likely to hold. That
those prone to fantasizing would be more differentiating in their judgments of
others. These results in particular beg for a follow-up investigation, as it may be
that when trying to understand gift-giving, those most attuned to others’ minds
are not just more likely to see higher levels of effort, and attribute higher levels
2.3 discussion 27
of gratitude, but they will be more likely to perceive them more accurately and
with greater differentation.
To qualitatively analyze individual differences in gratitude for the different
gifts, I plotted the participants’ mean need, effort and gratitude ratings on the
same grid, i.e. every participant would be represented with one point depend-
ing on the mean value of their ratings. This plot revealed that while participants
did vary to some extent along their mean for each variable, every participant’s
gratitude was similarly related to their own need and effort ratings. This indi-
cates that while participants may vary in the degree to which they feel gratitude
for a gift, they all appear to follow a similar process to relate need, effort, and
gratitude. However, this analysis is limited in that it ignores some aspects of
the participants’ variability within their answers, as using the mean values will
inherently push the value towards the middle of the scale.
Previous studies have examined the relationship between need, effort, and
gratitude, but few have attempted to link these parameters to other types of feel-
ings. I predicted that in addition to gratitude, other types of emotions would be
elicited by our stimuli. Although each stimulus was capable of producing grat-
itude at some level, I expected other types of emotions to be reported as well.
To address this point I plotted the average coordinates for each commonly re-
ported emotion label. The plot revealed that, in line with the predictions, certain
reported feelings were associated with unique portions of gift-space. The graph
shows that content, bitter and awkward are clumped in the lower portion of the
gift-space. Also interestingly, participants regularly reported similar combina-
tions of different feeling ratings, especially in combination with gratitude. The
graph shows that grateful and bitter were common designations for gifts that
were rated as low need and relatively high gratitude. Another interesting cluster
consists of the labels anxious, sad, and grateful-anxious to those gifts rated as
mid-need and mid-effort. This exploratory analysis shows that indeed gift-space
is a useful framework for understanding the relationship between how a gift is
28 introduction to gift space
perceived and what types of feelings may actually be reported. However, this
particular analysis is limited in a number of ways: it was primarily exploratory,
and the words chosen to denote the feelings may not comprise each and every
feeling that a participant may actually feel, as is the obvious limitation in using
a forced choice task.
The terms “effort” and “need” are useful in breaking down the components
of a gift exchange, and as the data showed, capable of explaining much of the
variance in how gratitude may arise toward a gift. However, the model is some-
what limited by their usage, as they are likely not completely orthogonal to
each other. The gift-space framework in this study does an adequate job in the
positive quadrants where gifts are of some positive meaning, but can it explain
receiving something that is in fact harmful or meant to harm the recipient? Fu-
ture studies should focus on examining the experience of receiving “gifts” that
may land in the negative regions of the gift space. For instance, what are the
feeling correlates for a gift that was given with good intentions, but is in fact
quite harmful? In this expansion of gift-space, it may be that the terms “effort”
and “need” are not entirely sufficient, and future research should address what
are the best ways to describe the full range of gifts and what are the critical
dimensions for predicting how one might feel for these types of gifts.
2.4 conclusion
Our study helps to further elucidate the emotion of gratitude, and its constituent
emotional parts. A deeper understanding of how and when gratitude is gener-
ated will call attention to the circumstances in which gratitude plays a significant
and positive role, and underline the value of the phenomenon in healthy social-
ity. By relying on the Shoah Foundation Institute’s Visual History Archive, our
participants empathize with the survivors and victims of the Holocaust. In so
2.4 conclusion 29
doing, the participants make the same discovery I did while creating the stimuli:
in the midst of the tragic suffering, there were many who acted with compassion,
willing sacrifice, and profound human dignity.

Part II
T H E H E A RT ’ S M E M O RY: A U T O N O M I C A C T I V I T Y
D U R I N G T H E E X P E R I E N C E O F G R AT I T U D E

3
T H E H E A RT ’ S M E M O RY
A French proverb says, “gratitude is the heart’s memory,” a sentiment that un-
derscores the literal objectives of this section of this chapter. In this chapter, I
will examine heart rate during the experience of gratitude. As it pertains to my
model of gift-space, I will also examine the changes in heart rate related to the
ratings of need and effort.
Previous research on gratitude posits that it is a moral emotion (McCullough
et al., 2001), capable of tying us closer to others Algoe et al. (2008). Gratitude
is thought to emerge in situations in which a person receives or recognizes a
gift that they need that comes at significant and well-intentioned effort from a
benefactor (Emmons and McCullough, 2004). It is well known that gratitude is
associated with many positive changes in subjective well-being, improvements
in self-reports of physical health, and resistance to trauma(Emmons and McCul-
lough, 2003; Fredrickson et al., 2003; Kashdan, Uswatte and Julian, 2006; Lam-
bert et al., 2009; Mccullough et al., 2002). How does gratitude accomplish this
feat? What happens in the body when we experience gratitude, and what might
that tell us about gratitude’s powerful health benefits?
There is very little research linking gratitude to physiology. In one study, a
five minute period in which participants self-generate feelings of appreciation
found differences in heart rate compared to other emotions such as anger (Mc-
Craty et al., 1995). This study analyzed heart rates using heart rate variability
and spectral density measures, which provide an indication of the relative sym-
pathetic and parasympathetic influences on the heart rate. McCraty et al. (1995)
found that appreciation, compared to anger, had a strong middle frequency
33
34 the heart’s memory
component, thought to be mostly related to parasympathetic influence on the
heart rate, in addition to some influence from the sympathetic nervous system.
Previous research then, provides information about what happens in the heart
when appreciation is self-generated and purposely reflected on for a long pe-
riod of time, but does not provide information about the acute state of gratitude
brought on by the real time recognition of a gift, and does not provide infor-
mation about how varying levels of gratitude may or may not be marked by
varying levels of heart rate.
There is however, research that connects features in the heart rate response
to the presentation of emotion inducing stimuli. The degree to which heart
rate drops when presented with a stimulus has been correlated with increased
memory for emotionally-charged events (Abercrombie et al.,2008), orienting re-
sponses to salient stimuli (Bradley, 2009), and with measures of empathy and
compassion (Eisenberg et al., 1994; Stellar et al., 2012). Heart rate variability
has also been connected to components of empathy in children observing their
mother in painful situations (Tully et al., 2014). Heart rate has also been shown
to take longer to return to its baseline state following events of social rejection
compared with events of social acceptance (Moor et al., 2010).
Previous research has described some of the benefits of gratitude in health
and well-being as described above, but a gap remains in the literature insofar as
understanding what happens in the body when we experience gratitude itself.
Furhter, before we can link gratitude’s experience to its health benefits, we need
to create an understanding of how the body creates the experience of gratitude
itself. We need to understand how varying levels of gratitude will differentially
impact the heart’s behavior, and we need to understand how need and effort
relate to changes in heart rate.
To examine the heart’s behavior under a range of experiences associated with
gift-giving, we need to use stimuli capable of eliciting strong emotional reac-
tions across the range of gratitude experiences. In this study, participants were
3.1 method 35
given stimuli drawn from the stories of survivors of the Holocaust. The stories
were drawn from the USC Shoah Foundation Institute’s Visual History Archive,
comprised of 50,000 videotaped testimony given by survivors of the Holocaust.
Stories were selected from a subsample of the testimony in which the survivor
tells of receiving a gift, or help, from another person. In the experiment, par-
ticipants were given these stories and asked to imagine themselves in the same
situation and feel as much as possible how they would feel if they were in the
same situation.
As detailed in the previous chapter, the gifts vary by nature. The gifts them-
selves vary according to how much the recipient needed the gift, and how much
overall effort they perceive in a gift. These variations are the underpinnings of
the gift-space framework. As it applies to heart rate physiology, it was predicted
that patterns of heart rate responses will differ in relation to changes in how a
gift is rated and perceived. Gifts rated as eliciting high levels of gratitude will
show patterns that resemble gifts in the high-effort/high-need region of gift
space.
3.1 method
3.1.1 Participants
The study dataset consists of data from 108 participants (63 female); ten par-
ticipants had to be removed from the analysis due to signal loss from exces-
sive movement during the experiment, or electrodes falling off the skin, leaving
a final sample of 98 participants (57 female), with average age of 20 (min =
18;max = 32;sd = 2.14). The participants were recruited using USC’s psychol-
ogy subject pool as well as posted fliers and advertisements on USC’s University
Park Campus. All research participants were given informed consent, and all ac-
36 the heart’s memory
tivities were done in accordance with and approval from USC’s Institutional
Review Board policies on human subjects research.
3.1.2 Stimuli
The goal of the study was to examine gratitude in the context of a wide range of
gifts. To meet this goal, the stimuli needed to be capable of probing the varied
constellations of experiences associated with receiving a gift. The stimuli consist
of a collection of stories based upon testimony from survivors of the Holocaust.
The stories were selected from testimony housed in the USC Shoah Foundation
Institute’s Visual History Archive, comprised of 50,000 videotaped Holocaust
survivor testimonial. Out of the 50,000 testimonies, stories were selected from
instances in which the survivor describes receiving a gift, of any sort, from
another person. The stories of gifts received were selected to fall into one of three
categories of our gift-space model. For instance, many survivors tell stories in
which they are taken under the protection of strangers, feeling strong gratitude
for such truly life-saving gift. This type of gift was coded as high-effort (the risk
taken by the family that is hiding the survivor), and high-need (the life-saving
consequences for the recipient). Other gifts were selected to be high-effort/low-
need. An example of these is for instance a survivor receiving peanuts from a
stranger on the street, but being allergic to peanuts and thus not able to eat them.
Yet another category were gifts of high-need/low-effort. Examples of these were
gifts that involved situations in which a donor gave something at no great cost,
but that would be of critical help to the survivor, e.g. a survivor, while trying to
flee her country, has a clerk at the border that intentionally ‘forgets’ to stamp ‘j’
for ‘jude’ on the survivor’s passport, thus granting her the ability to leave the
country and saving her life.
3.1 method 37
The scenarios described by the survivors were transcribed into stimuli ranging
from 30 to 40 words and were rephrased to be in the first person. Participants
read these scenarios and took the role of the survivor as they imagined how they
would feel if they were in the same situation. After a period of reflection on the
gift, participants were asked to rate how much gratitude they felt for the gift
(see below for complete stimulus presentation detail). There were 16 stimuli in
each condition (high-gratitude, high-effort and high-need); a total of 48 stimuli.
3.1.3 Procedure
The experiment was designed to mimic visiting the United States Holocaust Mu-
seum in Washington D.C., where visitors imagine the perspective of a survivor
as they go through the four phases of the Holocaust. The museum has four
floors–one for each phase of the Holocaust, and participants visit the floors in
succession; on each floor they view videos, photos and artifacts from that phase
intended to induce the experience of what was occurring during that phase of
the tragedy. In this experiment, the stimuli are broken into the four phases of
the Holocaust. Each phase of the experiment began with an in-house created
documentary detailing the events of that time period. The documentaries were
about two minutes long and were created in collaboration with students from
the U.S.C. school of Cinematic Arts so as to mimic a Ken Burns-style docu-
mentary. The documentaries relied on powerful images as well as a professional
actor providing a voice-over description so as to elicit strong emotional reactions
in the participants. The purpose of the documentary was to deeply involve the
participants in the historical situation by giving information enough for them
to imagine the perspective of the survivors as well as to provide brief breaks
throughout the study to help guard against habituation to the stimuli.
38 the heart’s memory
Participants were seated in a comfortable chair and connected to the physiol-
ogy monitoring equipment. The experimental session began with a ten minute
resting baseline acquisition in which baseline heart rate activity was measured.
Consequent to the baseline measurement, participants worked with a laptop as
each stimulus was presented as black text on a light grey screen. For each gift,
they were given 10 seconds to read the text, understand the event and to begin
to form a response to the stimulus. In the event that the participant’s reaction
began to build before the 10 second duration was reached, they were told to
press a button to advance the trial manually. After viewing the stimulus, partic-
ipants were shown a blank light blue “reflection” screen. The participants were
told that this was the screen in which they should experience their reaction to
the event itself, to imagine themselves in the situation presented, and to form
as deep and realistic of a reaction as they could. Following the reflection period,
participant rated on a scale of 1-4 how much gratitude they felt in response
to the event. Participants were told to scale their gratitude such that a 1-rating
would be associated with a small amount of gratitude, as in receiving lunch
from a friend, to a 4-rating for events that overwhelmed them with gratitude.
After the rating screen, the participants were given18 seconds of rest, indicated
by a black fixation cross on a light grey screen. During the rest period, partici-
pants were told to simply return to their baseline state, putting everything out
of their mind from the previous event. The sequence of trials is randomized for
each participant within each phase of the experiment; participants were asked
to treat each event as an independent event in of itself (see to Figure 9 for the
stimulus presentation protocol). Each phase lasted approximately 10 minutes.
After participants completed all four phases of the experiment, they were dis-
connected from the psychophysiology equipment and then they re-visited each
stimulus in the same order as during the experiment, however, in revisiting the
stimuli they rated each event according to its effort, need and gratitude and they
did so on a scale of1-9. Participants were instructed to use the lowest ratings for
3.1 method 39
need in response to gifts in which they felt their physiological, psychological, or
basic needs were only barely met or improved, whereas the highest ratings were
to be given to things in which they perceived a need was met great enough that
their life was greatly impacted for the better or was saved altogether as a result
of the gift. Low ratings on the effort scale were to be given for gifts in which the
participant judged the donor’s sum effort to be small, or that their sacrifice was
such that their life was not impacted or put at risk as a result of the gift. The
highest ratings of effort, the participants were told, should then only be used
for instances in which the donor’s way of life was severely impacted, or in some
cases sacrificed to provide the gift. Participants were again told to rate each
event as an independent event in and of itself, and not to compare their ratings
from one gift to another. These guidelines were given to help provide differen-
tiation in the ratings, to help the participants judge the stimuli, and to prevent
a ceiling effect in which every gift was rated using only the highest values in
the range. The participants’ ratings of need and effort for each gift were used to
correlate to their physiological measurements during the main experiment and
to their ratings of gratitude. After they re-rated the stimuli, participants filled
out a set of questionnaires such that their responses could be correlated to their
behavior during the experiment.
3.1.4 Analysis
An electrocardiogram (ECG) was collected to measure heart rate using two elec-
trodes, one placed on the participant’s right wrist, and one on the participant’s
left ankle. ECG recordings were preprocessed in Acqknowdge 9.32 (Biopac) to
identify the R peaks of the QRS complex. The resulting intervals between R
peaks were plotted and visually inspected for artifacts. Misidentified R peaks
were manually corrected. RR interval series were then uniformly re-sampled
40 the heart’s memory
at 4 Hz using cubic spline interpolation (Kubios HRV; kubios.uku.fi/). Uni-
formly sampled RR interval series were transformed into heart rate series (in
beats per minute, BPM). For each trial, average heart rate during the last half of
the rest screen preceding the trial was used as a baseline. Heart rate response
time course during each trial was calculated as the moment-by-moment heart
rate change relative to the pre-trial baseline. For each participant, heart rate re-
sponses for each type of rating (the 1-4 ratings of gratitude during the initial
reflection of the stimuli, and the1-9 ratings of need, effort and gratitude follow-
ing their initial reactions) were averaged, creating 5 average heart rate response
time courses. The baseline for heart rate was calculated using recursive averag-
ing of the second half of the rest period during the experiment. The data were
plotted using MATLAB.
The primary goal of this analysis was the examination of change in heart
rate through the timecourse of the trials, but these data still require analysis by
statistical inference. To do this, the time series were analyzed by extracting the
minimum and maximum values, in terms of percentage of baseline BPM. These
values were calculated for each subject for each level of rating. These data were
extracted and imported to SPSS for analysis. A repeated measures ANOVA with
a Greenhouse-Geisser correction was used to determine differences in heart rate
maximum and minimum among the ratings.
3.2 results
The primary focus of this investigation was to examine differences in heart rate
in relation to ratings of need, effort and gratitude. The results showed a strong
differentiation between the trials rated as eliciting low levels of gratitude and
those rated as eliciting high levels of gratitude. In greater detail, we see that
the stimulus onset is shows an initial drop in the heart rate for the high-rated
3.2 results 41
Figure 9: Heart rate experiment stimulus presentation paradigm. This sequence was
repeated 12 times per run.
stimuli, compared to an initial acceleration in the low-rated stimuli. Overall, the
waveforms for events rated with high gratitude, compared with low gratitude,
have a similar shape but is separable over the time course of the trials (see
Figure 10).
The data was also plotted for need, effort and gratitude using the ratings
collected from subjects after the psychophysiology data collection portion of
the experiment. For each gift, the participants rated from 1 to 9 how much
the gift was needed, how much effort they felt it took to provide, and how
grateful they felt. The heart rate patterns for each rating were averaged across
all the subjects and plotted as a continuous function over the course of a full trial
(stimulus, reflection, probe, rest). The results revealed that the ratings of need
and gratitude related to differences in heart rate over the course of the trial.
For the need and gratitude ratings, higher ratings were associated with lower
heart rate. Need and gratitude both showed differentiable patterns between the
ratings (see Figure 11 and Figure 12), whereas effort did not (see Figure 13).
To statistically compare the differences in heart rate among the need, effort
and gratitude ratings, an analysis of the maximum and minimum values heart
rate value for each rating for each participant was also conducted. These data
were analyzed using a repeated measures ANOVA with a Greenhouse-Geisser
42 the heart’s memory
Stimulus
Reflect
Probe
Rest
% Change in BPM
Time (4hz)
High Gratitude
Low Gratitude
Figure 10: Average heart rate response during trials rated as high or low in gratitude.
Overall, high levels of gratitude were associated with lower heart rate, per-
haps related to increased parasympathetic function during the deep experi-
ence of gratitude.
correction. There were differences in in the maximum and minimum heart rate
values when comparing the high and low need and gratitude ratings. The heart
rate values were not associated with differences in the effort ratings. The values
were also analyzed across each of the portions of the trial time series window
(stimulus, reflect, probe, rest). In comparing gratitude ratings, heart rate differed
significantly between all the ratings for each part of the time series except for the
maximum value of the probe condition. In the need condition, heart rate differed
significantly between all the ratings for each part of the time series except for the
the maximum value during the stimulus presentation and the maximum value
during the reflect period. Heart rate values did not differ according to the rating
of effort in any of the time series points. (see Table 3)
3.3 discussion 43
Need Rating
Time (4hz)
Stimulus
Reflect
Probe
Rest
Percent BPM
Figure 11: Average heart rate response for each of the nine ratings of need. Overall,
classifying the trials according to the level of need shows a difference between
the lowest and highest ratings, and also a pattern similar to that found with
the gratitude ratings.
3.3 discussion
This investigation sought to examine differences in heart rate during the experi-
ence of receiving gifts. Specifically differences were measured in how heart rate
differed in relation to the ratings of need, effort and gratitude for a gift. It was
hypothesized that high ratings of gratitude would be related to differences in
heart rate during the processing of gifts, and that gifts rated as eliciting high
levels of gratitude would have a similar pattern to gifts rated as high-need or
high-effort. The data largely confirm the hypotheses. Different ratings of the
gifts were associated with different patterns of heart rate associated with the
44 the heart’s memory
Gratitude Rating
Time (4hz)
Stimulus
Reflect
Probe
Rest
Percent BPM
Figure 12: Average heart rate response for each of the nine ratings of gratitude. Classify-
ing the trials according to gratitude shows differences in heart rate response
for the levels of gratitude.
gifts across the timecourse of the trials. The analysis revealed that need ratings
resulted in heart rate patterns that resembled the gratitude ratings, but the effort
ratings were not associated with differences in heart rate.
The data show that higher ratings of gratitude are marked by lower overall
heart rate over the course of the trial. Interestingly, the patterns of activity were
largely similar, except for an initial dip in heart rate for the trials rated as elic-
iting high levels of gratitude. This may be related to the inhibition of the heart
rate coming courtesy of the parasympathetic nervous system, which helps lower
heart rate. This initial drop could be part of the parasympathetic nervous sys-
tem influencing the heart rate early in the processing and then holding the heart
3.3 discussion 45
Effort Rating
Time (4hz)
Stimulus
Reflect
Probe
Rest
Percent BPM
Figure 13: Average heart rate response for each of the nine ratings of effort. Overall,
classifying the trials according to the level of effort does not show differences
between the ratings.
rate low over the course of the trial, similar to mechanisms previously proposed
(Gunther Moor et al., 2010).
These data represent a novel approach compared to previous research on
heart rate and emotion. These data focus on the processes taking place during
the actual experience of gratitude that emerges after receiving a gift, as opposed
to measuring and decomposing heart rate during a static and long duration
reflection period (McCraty et al.,2004,2009). These data add to previous investi-
gations of emotion and heart rate by being the first to investigate the changes in
heart rate the occur while a participant reflects on gifts, and to correlate differ-
ent ratings of gifts to differences in heart rate. Such an analysis allows a window
into emotion processing as it unfolds in the theater of the body.
46 the heart’s memory
While there are previous reports of positive emotions and heart behavior (Kok
and Fredrickson, 2010; Kok et al., 2013), but these studies examine the heart’s
vagal tone by itself, and not the heart’s reaction to a stimulus (note that I will
examine vagal tone in the next chapter). The data in this current study show
more closely what heart processes underly the processing of gratitude as it takes
place after receiving a benefit.
It is also interesting to note that the heart rate patterns, as in the shape of the
waveform as it undulates during the trial, did not greatly vary between the low
and high-gratitude trials except for the initial dip noted above. This fits nicely
with the overall gift-space model of gratitude. The data support the idea that the
experience of gratitude takes place on a continuum, and that heart rate across
the continuum is related to the level of gratitude felt.
The approach taken in this experiment allowed us to examine the data from a
number of perspectives. These perspectives were gained by classifying the heart
rate behavior using the participants’ ratings of need, effort and gratitude. In this
case, many of the same trials will be rated as high-need or as high-gratitude, so
if need, effort and gratitude are all rated in unison, we would see no differences
in the heart rate patterns across the three classifications. However, if we do see
differences, we will learn which ratings most resemble each other in differentiat-
ing heart rate response across the trial. This analysis revealed that effort ratings
did not differentiate the heart rate response. Because effort did not explain as
well the variance in in heart rate across its ratings, we can conclude that classify-
ing the trials based on effort does not yield as much information as classifying
them based on need or gratitude, indicating that need and gratitude are better
at explaining the heart’s behavior in this experiment. This is interesting given
the previous emphasis in gratitude research on the notion that “it is the thought
that counts” (Ames et al., 2004; Algoe et al., 2008). These data indicate that this
adage may not be entirely true. It is more likely that when we respond to a
donor’s thoughtfulness, we are responding to the gifts ability to fulfill deeper
3.4 conclusion 47
psychological needs–and not accurately constructing in our minds a picture of
the effort required to produce a gift. It makes more sense given the context
of my experiment itself. In my experiment, participants are forced to imagine
themselves in situations that are dire and life threatening. In such circumstance,
one’s need for help and aid is amplified to the point that the intent behind help
become less emphasized, as even small utilitarian gifts given with little effort
can elicit strong feelings of gratitude. In this way, these data do not so much as
contradict previous findings, they merely highlight the need to understand the
flexibility of gratitude as it takes place in the context of the social situations in
which it is generated.
3.4 conclusion
These data investigate gratitude in a manner that has not previously been done.
While other attempts to study gratitude and the heart focus on static mainte-
nance of the emotion based on a self-generated emotion, these data are the first
to examine changes in heart rate during response to a gift, and to correlate
those responses to varying levels of heart rate as they are experienced, and as
they vary in relation to the ratings themselves. Gratitude’s health benefits are
well documented (Kok et al., 2013), but no research has examined how these
positive emotions take place at the level of the heart. The data presented in this
chapter provide a clue that perhaps gratitude’s healing power comes from its
ability to lower heart rate, which should be tested in future studies.
48 the heart’s memory
Statistical analysis of heart rate time series data
Condition Time Series Feature df F Sig
Gratitude Stimulus Max 8, 5.174 2.987 0.011
Gratitude Stimulus Min 8, 5.971 3.99 0.001
Gratitude Reflect Max 8, 4.015 2.929 0.021
Gratitude Reflect Min 8, 5.008 5.911 0.001
Gratitude Probe Max 8, 3.419 2.091 0.093
Gratitude Probe Min 8, 6.019 5.478 0.001
Gratitude Rest Max 8, 2.910 2.778 0.043
Gratitude Rest Min 8, 5.775 7.267 0.001
Need Stimulus Max 8,3.802 2.12 0.084
Need Stimulus Min 8,5,712 3.273 0.005
Need Reflect Max 8,2.748 1.822 0.151
Need Reflect Min 8,5.913 3.908 0.001
Need Probe Max 8,4.119 4.145 0.003
Need Probe Min 8,4.755 3.934 0.002
Need Rest Max 8,2.781 2.801 0.046
Need Rest Min 8, 6.115 4.426 0.001
Effort Stimulus Max 8, 4.020 1.533 0.193
Effort Stimulus Min 8, 5.785 1.541 0.167
Effort Reflect Max 8, 3.217 1.933 0.122
Effort Reflect Min 8, 5.798 1.25 0.281
Effort Probe Max 8, 4.406 1.973 0.092
Effort Probe Min 8, 6.191 1.542 0.161
Effort Rest Max 8, 2.951 0.207 0.888
Effort Rest Min 8, 6.471 1.1 0.362
Table 3: Time series data for each level of rating for need, effort and gratitude, respec-
tively, was entered into a multilevel repeated measures ANOVA to analyze the
maximum and minimum heart rate (in %BPM of baseline). Overall, need and
gratitude showed greater differences in heart rate between the levels of the rat-
ings.
4
G R AT I T U D E R AT I N G S C O R R E L AT E W I T H VA G A L T O N E
The previous chapters have examined characteristics of gratitude with an em-
phasis on the cognitive and physiological processes that accompany the expe-
rience of gratitude, and where possible, these findings have been connected
to individual differences. In chapter 1, for instance, it was shown that indi-
viduals scoring high on a personality measurement of perspective-taking used
greater variance in how they rated the gifts compared to individuals low on
the perspective-taking scale. This effect led to the examination of gratitude as a
process of accurately recognizing the context of a gift, as opposed to the over-
recognition of effort or the deliberate up-regulation of gratitude. In this chapter,
I explore this concept at the level of the heart’s vagal tone.
How deeply is the experience and processing of gratitude rooted in our indi-
vidual bodies? It is obvious that given the same gift to two different individuals,
they will likely have a different reaction. These reactions will be based on many
factors, such as their previous life experiences, their need-state at the time of
receiving a gift, and at some level their own biological predilection for emo-
tion processing. That people differ on these reactions is a given, but what is the
nature of their differences, and how is it connected to our physiological under-
pinnings?
As a trait, individuals higher in gratitude show many other positive traits. In-
dividuals high in gratitude show that gratitude can provide resilience in the
face of trauma (Fredrickson et al., 2003). A study of Vietnam war veterans
showed that veterans with post-traumatic stress disorder (PTSD) showed lower
trait gratitude than veterans without PTSD.(Kashdan, Uswatte and Julian,2006).
49
50 gratitude ratings correlate with vagal tone
Individuals high in gratitude show fewer suicidal ideations and greater sense
of life-meaning (Kleiman et al., 2013). Gratitude is also associated with higher
life-satisfaction and lower materialism (Lambert et al., 2009) and is linked with
higher measures of subjective well-being and pro-social behavior (Watkins et al.,
2003; Mccullough et al., 2002).
The benefits of having a grateful disposition are clear. But interestingly, pre-
vious research has not examined individual differences in the actual experience
of gratitude for receiving gifts from others. Perhaps even more interestingly, a
fascinating question remains as to how individual differences in the experience,
generation and judgment of gratitude are related to individual differences in
physiology.
To understand how individuals may differ in the processing of gratitude, we
first need to describe how the body creates emotion, and how we have come
to measure differences in how emotion unfolds in the the theater of the body.
Emotion is by its nature a process of continually monitoring bodily changes
(Damasio,1996), and these changes, when experienced as feelings, are in contin-
uous communication between our aware mind and our bodies. A large part of
how the body accomplishes this is done by the vagus nerve.
The vagus nerve is connected to the brainstem and has a synapse on the
sino-atrial node of the heart, where it provides inhibitory input over the heart’s
beating rate. The vagus nerve’s input functions as part of the parasympathetic
nervous system where it counteracts the sympathetic nervous system’s input on
the heart’s beating rate. As we inhale, the lungs block the function of the vagus
nerve, which allows the heart rate to speed up, and for the freshly oxygenated
blood to be pumped out of the heart more quickly. In exhalation, the block
on the vagus nerve is removed, its inhibitory input returns, and the heart’s rate
slows down. The difference in rate between when the vagus nerve is blocked and
when it is unblocked is the basis for vagal tone, as a greater increase indicates
a stronger inhibitory force. This respiratory sinus arrhythmia, also known as
gratitude ratings correlate with vagal tone 51
cardiac vagal tone (CVT) is a measure of the vagus nerve’s “braking force” on
heart rate.
It is well known from previous research that CVT is associated with our social
and emotional lives. Individual differences in CVT have been correlated with
benefits to emotion regulation, social cognition and pro-sociality (Porges, 2007;
Oveis, 2009). In addition, CVT scores are correlated with beneficial top-down
emotion regulation strategies (Geisler et al.,2010), handling stressors (Fabes and
Eisenberg, 1997), and reappraising situations that invoke anger (Vögele et al.,
2010). In summary, CVT is associated with adaptive self-regulation and higher
social engagement, in addition to greater propensity for using social engage-
ment strategies during times of distress, and lower tendency for disengagement
during regulation of negative emotions (Geisler et al., 2013).
In this study, the goal is to link individual differences in the experience of
gratitude to individual differences in CVT. If we are to examine how individuals
might differ in their experience of gratitude, we need an experimental paradigm
capable of generating gratitude in a variety of contexts, and to do so in a way
that is engaging and emotionally powerful. To meet this need, I created stimuli
based on testimony from survivors of the Holocaust, housed in the USC Shoah
Foundation Institute’s Visual History Archive. This archive holds more than
50,000 videotaped testimonies from survivors of the Holocaust. Out of these
testimonies were drawn stories in which the survivor tells of receiving a gift
or help from another person. These stories were rephrased into first person
language and presented to participants as text on a screen. Participants were
told to imagine themselves at the center of the scenario as a survivor receiving
the gift, and feel, as much as possible, how they would feel if they were in
the same situation. After each gift, participants rate it according to how much
gratitude they feel. These ratings of gratitude then form the basis for correlations
with CVT and other personality measures.
52 gratitude ratings correlate with vagal tone
Based upon previous research, there is strong support for the idea that CVT
will be correlated with gratitude. Intuitively, one might predict that CVT will pos-
itively correlate with gratitude, i.e. higher gratitude ratings will be associated
with higher vagal tone. On the other hand however, CVT is inherently a regula-
tory mechanism (Kok and Fredrickson, 2010), meaning that CVT is responsible
for making responses appropriate to context. As we saw in the first chapter of
my dissertation, trait perspective-taking was associated with variability in rating
the effort in a gift, as opposed to attributing more effort overall. Based on this,
I might have predicted that CVT would correlate with the standard deviation of
the gratitude ratings, however, the predictions for this experiment were made
before those data provided their insight. Thus, it was predicted that individual
differences in CVT would correlate positively with gratitude ratings.
4.1 method
4.1.1 Participants
Data was collected from 108 participants (63 female) with an average age of 20
(min = 18;max = 32;sd = 2.14). Eight participants were removed from the
dataset due to their electrodes falling off of their hands, or to excess movement
during the study, leaving a final sample of 100 participants (61 female). The
participants were recruited using USC’s psychology department subject pool as
well as posted fliers and advertisements on USC’s University Park Campus. All
research participants were given informed consent, and all activities were done
in accordance with and approval from USC’s Institutional Review Board policies
on human subjects research.
4.1 method 53
4.1.2 Stimuli
Participants were presented with a wide range of gifts designed to sample the
depth of the gratitude experience. To examine gratitude in such a way that ful-
fills the goals of the experiment, as well as creating a meaningful experience for
the participants, the stimuli assembled are drawn from a collection of stories
based upon testimony from survivors of the Holocaust. The stories are based
on instances in which the survivor describes receiving a gift, of any sort, from
another person. The stimuli were designed to be categorized according to their
level of perceived effort on the part of the donor, as well as for their felt need for
the participant. Together, effort and need predict gratitude, and in this study we
focus on how ratings from the experience of gratitude vary in accordance with
the participants’ CVT. The gifts vary by nature, For instance, many survivors tell
stories in which they are taken under the protection of strangers, feeling strong
gratitude for such truly life-saving gifts. Other gifts elicited lower levels of grat-
itude, such as a survivor receiving peanuts from a stranger on the street, but
being allergic to peanuts and thus not able to eat them, possibly diminishing
the feeling of gratitude.
The experiment was designed to mimic a visit to the United States Holocaust
Museum in Washington D.C., where visitors imagine the perspective of a sur-
vivor as they go through the four phases of the Holocaust. For each phase, pho-
tos and artifacts from that phase are introduced that induce the experience of
what was occurring during that phase of the tragedy. Similarly, in this experi-
ment the stimuli are also broken up into the four phases of the Holocaust. Each
phase of the experiment began with an in-house created documentary detailing
the events of that time period. The documentaries were about two minutes long
and were created in collaboration with students from the U.S.C. School of Cin-
ematic Arts so as to mimic a Ken Burns-style documentary. The documentaries
54 gratitude ratings correlate with vagal tone
relied on powerful images as well as a professional actor providing a voice-over
description of the historical events. The purpose of the documentaries was to
deeply involve the participants in the historical situation by giving information
to thoroughly imagine the perspective of the survivors; it also provided the ben-
efit of providing breaks throughout the study, keeping the participants engaged
in the experiment.
4.1.3 Procedure
To begin the session, participants were seated in a comfortable chair and con-
nected to the physiology monitoring equipment. They first underwent a base-
line ECG recording session for calculation of resting respiratory sinus arrhyth-
mia (RSA) to acquire CVT. During the recording, participants were instructed to
relax with their eyes closed and to synchronize their breathing to an audio cue
(delivered at 0.25 Hz) for five minutes. ECG was measured using two electrodes
placed on the participant’s right wrist and left ankle.
After the baseline recording session, the participants underwent the experi-
ment to measure their gratitude ratings. They worked with a laptop as each
stimulus was presented as black text on a light grey screen. For each gift, they
were given 10 seconds to read the text, understand the event and to begin to
form a response to the stimulus. In the event that the participant’s reaction be-
gan to build before the10 second duration was reached, they were told to press
a button to advance the trial manually. After viewing the stimulus, participants
were shown a blank light blue “reflection” screen. The participants were told
that this was the screen in which they should experience their reaction to the
event itself, to imagine themselves in the situation presented, and to form as
deep and realistic of a reaction as they could. Following the reflection period,
participants rated on a scale of 1-4 how much gratitude they felt in response
4.1 method 55
to the event. Participants were told to scale their gratitude such that a 1-rating
would be associated with a small amount of gratitude, such as receiving lunch
from a friend, to a 4-rating for events that overwhelmed with gratitude. After
the rating screen, the participants were given 18 seconds of rest, indicated by a
black fixation cross on a light grey screen. During the rest period, participants
were told to simply return to their baseline state, putting everything out of their
mind from the previous event. The sequence of trials was randomized for each
participant within each phase of the experiment; participants were asked to treat
each event as an independent event in and of itself (see to Figure 9 for the stim-
ulus presentation protocol). Each phase lasted approximately10 minutes. There
were12 stimuli in each phase; a total of48 stimuli. The mean and standard devi-
ation of the participants’ gratitude ratings from all of the ratings were correlated
with their CVT to examine the relationship between gratitude and vagal tone.
After watching the sequence of stimuli, participants were asked to fill out
a series of questionnaires. The included three commonly used psychological
questionnaires: The Maslow Need Satisfaction scale (Lester, 1990), the six item
gratitude questionnaire (GQ-6) (Mccullough et al., 2002), and the Interpersonal
Reactivity Index (IRI) (Davis, 1994). Participants also filled out a homemade
questionnaire that asked them about their experience during the experiment.
This questionnaire was eight items in length and asked subjects to rate on a
likert scale the following questions: (1) how involved they felt in the task, (2)
how similar their feelings during the situations matched what they would have
felt if the experience was real, (3) how difficult it was to put themselves in the
situations and (4) how much they felt they had an increased understanding
and sense of empathy for the Holocaust by going through the experiment. Af-
ter these questions were answered, the participants were asked four additional
open-ended questions assessing if there were any confusing stimuli, if any stim-
uli were particularly moving, if they had tried to figure out what the experiment
was about, and if they had any personal experience with the Holocaust.
56 gratitude ratings correlate with vagal tone
4.1.4 ECG Analysis
To calculate resting cardiac vagal tone, ECG recordings were preprocessed in Ac-
qknowdge 9.32 (Biopac) to identify the R peaks of the QRS complex. The result-
ing intervals between R peaks were plotted and visually inspected for artifacts.
Misidentified R peaks were manually corrected. Using Kubios HRV (Tarvainen
et al.,2009), RR interval series were then uniformly re-sampled at4 Hz using cu-
bic spline interpolation, detrended using smooth prior regularization to remove
slow fluctuation, and subjected to power spectrum analysis using fast Fourier
transformation (Tarvainen et al., 2002). High frequency (0.15-0.4Hz, roughly
the breathing frequency) power of the RR interval time series was used as a
measure of resting RSA. Resting RSA is expressed in normalized units, which
was calculated as the relative value (in percentage) of high frequency power in
proportion to the total power minus the very low frequency (VLF < 0.04Hz)
power (powerHF=(powertotal-powerVLF)).
4.2 results
The primary focus of this investigation was to correlate individual differences
on ratings of gratitude to individual differences in CVT. The data revealed a neg-
ative correlation between CVT and mean gratitude ratings, r(100) = -.273,p =
.006, and a positive correlation with the standard deviation of the participant’s
ratings of gratitude,r(100) = .262,p = .008. (see Figure 14).
CVT was also correlated to the scores from the questionnaires given to the
participants. Only the IRI’s fantasy scale correlated to participants vagal tone,
r(100) = .206,p = .039).
The results revealed a correlation between CVT and the participant’s gratitude
ratings, however it was different from my original prediction. CVT was negative
4.2 results 57
Std. Dev. of Gratitude Ratings
Vagal Tone Vagal Tone
Mean of Gratitude Ratings
r(100) = -.273
p < .006
r(100) = -.262
p < .008
Figure 14: Vagal tone correlates negatively with the mean ratings of gratitude, but posi-
tively with the standard deviation of the ratings of gratitude.
correlated with the mean gratitude ratings and positively correlated with the
standard deviation of the ratings. Given this result, it is helpful to visualize how
participants at the high and low ends of the range of vagal tone scores differed
in the distribution of their ratings. I measured this distribution by plotting a
histogram of their ratings. For this visualization, the participants were ranked
according to their vagal tone score. The top and bottom 25 participants were
analyzed to examine how often they used each of the ratings (one to four) dur-
ing the experiment. The histogram reveals differences in the distribution of the
participants’ ratings. Low vagal tone participants selected the highest gratitude
rating more often, and the lowest rating less often, compared to the high vagal
tone participants (see Figure 15).
Number of responses
Gratitude rating Gratitude rating
Number of responses
450
350
250
150
1 2 3 4 1 2 3 4
50
Low vagal tone subjects High vagal tone subjects
450
350
250
150
50
Figure 15: Participants with the lowest vagal tone used the highest rating more fre-
quently than the other ratings. Participants with the highest vagal tone used
the lower ratings more frequently.
58 gratitude ratings correlate with vagal tone
4.3 discussion
The focus of this investigation was to examine possible correlations between CVT
and gratitude. It was predicted that CVT would be positively correlated with
gratitude ratings. However, the data revealed a negative correlation between
CVT and mean gratitude ratings, and a positive correlation between CVT and the
standard deviation of the participant’s gratitude ratings. This result is similar
to the result seen in chapter 1, where perspective-taking correlated with the
standard deviation of the effort ratings. In many ways, the result of the current
investigation of CVT are even more interesting for what they reveal about the
link between individual differences in processing complex social processing and
innate physiology.
Trait gratitude, measured by how much or how frequently an individual nat-
urally generates gratitude, has been linked with many other positive traits. In-
dividuals who are high in trait gratitude show greater resilience to trauma
(Fredrickson et al., 2003), lower incidents of PTSD (Kashdan, Uswatte and Ju-
lian, 2006), and higher measures of subjective well-being (Kleiman et al., 2013;
Lambert et al., 2009; Mccullough et al., 2002; Watkins et al., 2003).
In terms of individual differences in physiology as measured using CVT, indi-
viduals high in CVT also have other benefits to life and well-being. High vagal
tone is associated with prosocial behavior and emotion regulation (Porges,2007;
Oveis, 2009), as well as better regulation of negative emotion (Geisler et al.,
2013).
In a often cited study, Kok and Fredrickson (2010) describe CVT as something
that we as individuals can draw upon as a resource that “moderates the degree
to which people experience positive emotions.” The data from the current study
add to these findings the notion that gratitude is something to be differentiable,
at its best regulated by our bodies based upon precise sensitivity to the social
4.4 conclusion 59
contexts in which it may arise. This revises the narrative that more gratitude
is better gratitude. The correlation between CVT and the standard deviation of
gratitude rating suggests (if we are to assume that it is preferable to have a
higher CVT value) that we should focus on generating an appropriate gratitude
response matching the context in which we recognize receiving a benefit from
another person.
Previous studies have provided valuable insight on the benefits of gratitude
based on self-report measures of trait-gratitude, e.g. (Mccullough et al.,2002). In
finding the correlation between the standard deviation of gratitude ratings and
CVT in this study, the investigation of individual differences in gratitude is taken
two steps further. First, the results of this study examine individual differences
in in how gratitude is experienced as opposed to relying on self-report measures
of trait gratitude. Second, these differences in how gratitude is experienced are
related to individual differences in physiology. By connecting CVT to the ratings
of gratitude as it is experienced by participants, we reach a new understand-
ing of how gratitude is generated, and how that may differ across individuals.
This finding identifies our social cognitive lives as being intrinsically linked to
emotion as it is experienced in the theater of the body.
4.4 conclusion
The current finding strengthens the connection between our bodies and healthy
sociality. The fact that something so deep in our physiology such as CVT is
related to gratitude is for its own sake fascinating. The findings from individuals
high in CVT indicate that gratitude may be, like many things, more about quality
than quantity. This leads to an application to daily life in which we seek to find
the appropriate amount of gratitude for a given situation as opposed to working
hard to generate as much gratitude as possible regardless of context.

Part III
T H E B R A I N ’ S V I RT U O U S C Y C L E : M A P P I N G T H E
N E U R A L C O R R E L AT E S O F G R AT I T U D E

5
T H E B R A I N ’ S V I RT U O U S C Y C L E : M A P P I N G T H E N E U R A L
C O R R E L AT E S O F G R AT I T U D E
In the previous chapters, I have described gratitude in increasing detail, moving
from the relatively coarse measurement of gratitude using paper and pencil to
the next level of measuring gratitude using changes in heart rate. In this chapter
we examine gratitude at another source: the brain. This chapter will also allow
for the further investigation of the two most surprising findings from the first
two datasets detailed in the previous chapters. First, in terms of the primary
drivers of gratitude in gift-exchange, need and effort, both correlate with grati-
tude, but the felt need for the gift is more highly correlated with gratitude than
the perceived effort. How will this translate to the brain? How will different
ratings of effort and need alter the patterns of brain activity? Second, the differ-
entiability of gratitude is important. That is, it may not be a case where gratitude
needs to be high or amplified in every situation. Rather, data from individuals
high in perspective-taking and vagal tone indicate that there is some deftness
to be used when evaluating the context of the situation to create a correct or
differentiated amount of gratitude. Finally, what can we expect from the brain:
How will the brain handle the varying degrees of gratitude? Will the differences
in activity relative to the level of gratitude change in terms of their own level of
activity, or will different patterns of brain activity emerge for different levels of
gratitude?
To the best of my knowledge, there are no prior direct investigations of grati-
tude or gift-exchange using functional magnetic resonance imaging (fMRI), though
others have discussed gratitude and brain function (Emmons, 2008; Zahn et al.,
63
64 the brain’s virtuous cycle: mapping the neural correlates of gratitude
2008a). Gratitude has been studied indirectly however, in comparison to other
moral emotions Zahn et al. (2008b) and individual differences in gratitude are
correlated with increased morphological volume in the posteromedial cortex. In
this study, the objective is to elucidate the neural activity patterns correlated
with the experience of gratitude. A study of the neural basis of gratitude will
reveal how gratitude unfolds in the brain at the neural systems level. This is
critical to the goal of linking gratitude’s origin in human physiology to its en-
hancement in everyday life.
Forming predictions for how the brain generates gratitude requires examin-
ing prior studies of affect, decision-making and social cognitive processes and
translating these findings to form predictions applicable to gratitude specifically.
I hypothesize the following:1a. Ratings of gifts classified as high-effort and low-
need will be associated with activity in regions of the brain associated with
affective components of pain (Singer et al.,2004): the anterior insular cortex and
the anterior cingulate cortex (ACC). 1b. Gifts rated as low-effort and high-need
will be correlated with activity in brain regions associated with reward, such
as the nucleus accumbens and orbitofrontal cortex, as well as with activity in
regions associated with perspective taking and value estimation in prefrontal
cortices (?Saxe, 2006). 1c. Stimuli rated as high-effort and high-need will be
associated with activity in brain regions associated with social cognition and
reward, such as the medial prefrontal and orbitofrontal cortices (Bechara et al.,
2000; Harbaugh et al., 2007), as well as to activity in regions of the brain asso-
ciated with compassion and admiration, such as the posteromedial and insular
cortices (Immordino-Yang et al., 2009). 2. Stimuli eliciting the highest ratings
of gratitude will largely overlap with those found in the high-effort/high-need
ratings, however we predict gratitude will also elicit activity in regions known
to be involved in oxytocin function and social bonding, such as the subgenual
cingulate cortex (Hsu et al., 2008).
5.1 method 65
5.1 method
5.1.1 Participants
The study consists of data from 26 participants (13 female); three participants
had to be removed from the analysis, two due to computer malfunctions, and
one for falling asleep during the experiment, leaving a final sample of 23 par-
ticipants (12 female); average age 21 (min = 18;max = 28;sd = 2.21). The par-
ticipants were recruited using USC’s psychology subject pool as well as posted
fliers and advertisements on USC’s University Park Campus. All research partic-
ipants were given informed consent, and all activities were done in accordance
with and approval from USC’s Institutional Review Board policies on human
subjects research.
5.1.2 Procedure
The paradigm employed in this experiment is largely identical to that of the
psychophysiology experiment. Some differences were made however, to adjust
for the scanning environment. The stimulus presentation was done using a slow
event-related design in which trials were presented in random order and sep-
arated by a rest condition. The design of the experiment in general, and justi-
fication for the approach is detailed in chapters 1 and 2. In short, participants
viewed 48 scenarios, divided into four runs consisting of 12 stimuli each. The
scenarios were derived from stories given by survivors of the Holocaust, found
in the USC Shoah Foundation Institute’s Visual History Archive. The scenarios
averaged30 words in length. For each stimulus, participants were asked to read
the stimulus as text on a screen and to take the perspective of the survivor and
imagine as much as possible how they would have felt if they were in the same
66 the brain’s virtuous cycle: mapping the neural correlates of gratitude
circumstance. Participants were given ten seconds to read the stimulus, after
which they were presented with a blank light-blue screen, and during this time
participants were asked to reflect on the gift that they were given and to feel
how they would have felt if it had just happened. The time given for reflection
in response to an event was 12 seconds. In the event that the participant’s reac-
tion began to build before the10 second duration was reached, they were told to
press a button to advance the trial manually. After viewing the stimulus, partic-
ipants were shown a blank light blue “reflection” screen. The participants were
told that this was the screen in which they should experience their reaction to
the event itself, to imagine themselves in the situation presented, and to form
as deep and realistic of a reaction as they could. Following the reflection period,
participant rated on a scale of 1-4 how much gratitude they felt in response
to the event. Participants were told to scale their gratitude such that a 1-rating
would be associated with a small amount of gratitude, as in receiving lunch
from a friend, to a 4-rating for events that overwhelmed them with gratitude.
Following the their rating of the stimulus, a fixation cross appeared and they
were told to rest, and to put everything about the prior stimulus out of their
mind. The time given for the rest condition averaged 12 seconds and was a nor-
mally distributed jittered range of values from8 to16 seconds. Participants were
told to treat each stimulus, or event, as an independent event in and of itself and
to rate it as honestly as possible according to how they felt while reflecting on
the gift (See Figure 16).
After participants completed all four phases of the experiment, they were dis-
connected from the psychophysiology equipment and then they re-visited each
stimulus in the same order as during the experiment, however, in revisiting the
stimuli they rated each event according to its effort, need and gratitude and they
did so on a scale of1-9. Participants were instructed to use the lowest ratings for
need in response to gifts in which they felt their physiological, psychological, or
basic needs were only barely met or improved, whereas the highest ratings were
5.1 method 67
to be given to things in which they perceived a need was met great enough that
their life was greatly impacted for the better or was saved altogether as a result
of the gift. Low ratings on the effort scale were to be given for gifts in which the
participant judged the donor’s sum effort to be small, or that their sacrifice was
such that their life was not impacted or put at risk as a result of the gift. The
highest ratings of effort, the participants were told, should then only be used
for instances in which the donor’s way of life was severely impacted, or in some
cases sacrificed to provide the gift. Participants were again told to rate each
event as an independent event in and of itself, and not to compare their ratings
from one gift to another. These guidelines were given to help provide differen-
tiation in the ratings, to help the participants judge the stimuli, and to prevent
a ceiling effect in which every gift was rated using only the highest values in
the range. The participants’ ratings of need and effort for each gift were used to
correlate to their physiological measurements during the main experiment and
to their ratings of gratitude. After they re-rated the stimuli, participants filled
out a set of questionnaires such that their responses could be correlated to their
behavior during the experiment.
5.1.3 Image Acquisition
Brain imaging was performed at the Dana and David Dornsife Cognitive Neu-
roscience Imaging Center at USC. Functional and structural magnetic resonance
imaging (MRI) was employed using a Siemens 3T scanner. Four functional runs,
one anatomical MPRAGE, and one T2 weighted image was acquired for each
subject.
Prior to performing the functional scans, structural images were collected in
radiological convention with 176 slices, dimensions: 224256176 and then
resampled with voxel dimensions 111 mm, TR 1950 ms. For functional
68 the brain’s virtuous cycle: mapping the neural correlates of gratitude
Figure 16: fMRI stimulus presentation paradigm. This sequence was repeated 12 times
per run, over a total of four runs.
scans, 250 volumes were acquired, with 37 slices per volume. The TR used was
2000 ms, with an interslice time of54 and a TE of30 ms. Inplane resolution was
6464. Voxel resolution was3.53.53.5mm, with no slice gap, the flip angle
was90

.
5.1.4 Analysis
The brain imaging data are primarily analyzed using the FSL (Smith et al.,2010)
software package. FMRI data processing was carried out using FEAT (FMRI Ex-
pert Analysis Tool), part of FSL (FMRIB’s Software Library, www.fmrib.ox.ac.uk/fsl).
Registration to high resolution structural and/or standard space images was car-
ried out using FLIRT (Jenkinson and Smith,2001; Jenkinson et al.,2002). The fol-
lowing pre-statistics processing was applied; motion correction using MCFLIRT
(Jenkinson et al., 2002); slice-timing correction using Fourier-space time-series
phase-shifting; non-brain removal was done using BET (Smith, 2002); spatial
5.1 method 69
smoothing using a Gaussian kernel of FWHM 5.0mm; grand-mean intensity
normalization of the entire4D dataset by a single multiplicative factor; highpass
temporal filtering (Gaussian-weighted least-squares straight line fitting, with
sigma=50.0s). Time-series statistical analysis was carried out using FILM with
local autocorrelation correction (Woolrich et al., 2001). Z (Gaussianised T/F)
statistic images were thresholded using clusters determined by Z > 2.3 and
a (corrected) cluster significance threshold ofP =0.05 (Worsley, 2001).
Task conditions are modeled with separate regressors that model the timing
of each trial type (as designated by our a priori ratings, the ratings of the par-
ticipant, or the ratings given by participants in chapter two) convolved with a
double gamma hemodynamic response function. A General Linear Model was
used to fit these regressors to the fMRI data.
This investigation is primarily focused on determining the neural correlates
of gratitude. To identify the neural correlates of gratitude, a design matrix was
created with four predictor functions. We included the prime, reflect, and probe
conditions as well as a parametrically varying predictor for the reflect time pe-
riod whose height was determined by the level of gratitude reported for each
trial. Ratings were included after being mean corrected for each subject. We
included the standard predictor for the reflection time in addition to the mod-
ulated predictor function to examine how variance in blood oxygenation level
dependent (BOLD) signal during the reflection period was explained by the rat-
ing itself. This same analysis was also performed using the post-experiment
effort and need ratings to examine how these ratings would explain variance in
the BOLD signal.
To determine how effort and need differentially played a role in explaining
brain activity during the experiment, trials were classified into separate condi-
tions using the ratings of effort and need from the participants in the heart rate
experiment (chapter two). This classification of the stimuli provides an indepen-
dent way of sorting and classifying the trials into unique categories, which is
70 the brain’s virtuous cycle: mapping the neural correlates of gratitude
likely better than my own sorting of the stimuli, and is more nuanced–due to
the sample size in the physiology sample. For each individual stimulus, the rat-
ing of need was subtracted from the rating of effort and then that difference was
averaged across all participants. As an example, consider a stimulus in which
a participant rates it as a 9 for effort and a 3 for need, this subject’s difference
score for this stimulus would be6; if the participant rated the stimulus as a3 for
effort and a9 for need, the difference score would be -6. The difference scores for
each stimulus were averaged across all the participants and then sorted by value.
The stimuli were then sorted from high to low, giving an array that stretched
from stimuli more highly rated for their effort to stimuli rated more highly for
need. The top 16 stimuli in which effort was rated more highly than need were
placed into a condition called High-Effort/Low-Need. The middle 16 stimuli
were those in which effort and need were rated in roughly the same value, and
are called Effort=Need. The stimuli rated higher for need than effort were placed
into a condition called High-Need/Low-Effort. These conditions formed the ba-
sis for the contrast analysis, and they were compared by subtracting the BOLD
signal activity for each condition from each other, e.g. High-Need/Low-Effort
minus Effort=Need.
Certain regions (such as the insula and the posteromedial cortex (PMC)) pre-
dicted to be involved in processing the gifts were not found to be active by the
whole brain analyses. Yet it is still of interest to determine how these regions
were behaving during the experiment. To do this, an exploratory task-state cor-
relation analysis was performed. This analysis involves performing correlations
using the BOLD signal retrieved from an region of interest (ROI) with every
other ROI, i.e. the timeseries data from the inferior posteromedial cortex (Inf PMC)
is correlated to the data from secondary somatosensory cortex (S2). The ROI
method is detailed below. Once the correlations are performed at the individual
subject level, the overall task-state correlation matrix for all the subjects’ high-
rated gratitude trials can be compared the matrix for all the subjects’ low-rated
5.1 method 71
gratitude trials. This comparison can be done to compare low, medium and high
levels and for the need, effort and gratitude ratings, respectively or to compare
individual differences using groups of subjects.
In each subject and each ROI, event-related selective averaging was performed
using PEATE (http://www.jonaskaplan.com/peate/). Each of the18 ROI’s time
series activity was correlated to each other ROI using a Pearson correlation yield-
ing an r value for each pair of ROIs for each subject. The r value was stored for
each pair of ROI’s in an18x18 matrix that represented, for that subject, how each
ROI’s time series data correlated to each of the other ROI’s. These matrices were
then averaged across subjects, yielding an average correlation matrix. These ma-
trices were then compared using a Kilmogorov-Smirnov test (KS) statistical test,
which is used when one can not make assumptions of normality in the data, to
determine if these distributions were likely to be drawn from the same distribu-
tion. A significance level ofp< .05 was chosen to determine if the distributions
significantly differed from each other.
The task-state correlations were calculated using 18 total ROIs (9 per hemi-
sphere). The regions included were the: VMPFC, ACC, nucleus accumbens (Nuc Acc),
temporal pole (Temp Pole), anterior insula (AI), posterior insula (PI), Inf PMC, superior
posteromedial cortex (Sup PMC), and S2. A 5mm sphere was placed in each of
these locations, centered on specific anatomical landmarks found in each partic-
ipant’s anatomical brain scan. See Table 7 in the appendix for the specifications
of each ROI’s anatomical landmarks. The ROI size (5mm) was chosen because
it is not so large that it might extend into other brain regions, and not too small
such that its summary activity is potentially biased by noise from a small subset
of voxels.
In addition to providing a way of exploring task-state correlations for the low,
medium, and high ratings of gifts, task-state correlations can also analyze differ-
ences between groups of subjects. This approach was used to analyze individ-
ual differences based on scores and ratings from the GQ-6, the IRI Perspective-
72 the brain’s virtuous cycle: mapping the neural correlates of gratitude
Taking sub-scale, and on the participants based on the standard deviation of
their gratitude ratings. For each of these measures, participants were drawn
into sub-groups by selecting participants whose scores placed them in the top
or bottom quartiles respectively. The correlation matrices for these groups of
participants, e.g. High GQ-6 vs Low GQ-6, were averaged, and the resulting cor-
relation matrices were compared using a KS test. A significance level of p < .05
was chosen to determine if the distributions significantly differed from each
other. The justification to compare these measures in particular, as opposed to
the many others filled out by participants, comes from the studies described
earlier in my thesis in which these measures revealed the most stark individual
differences related to how the gifts were processed. In addition, traditional re-
gression models are not well suited in this dataset because of the small sample
size, rendering the tests underpowered. In addition, tests were selected in such
a way as to draw unique groups of subjects into comparison, in other words,
subjects high on the GQ-6 scale were not for the most part the same individu-
als as those high on the IRI scale. It is important to note that this technique is
exploratory and is also based upon a small subset of individuals, thus interpre-
tations of its results should be treated with caution.
To improve the visualization of differences between the comparisons, the task-
state correlation matrices were plotted using the Brain Connectivity Toolbox in
MATLAB (Rubinov and Sporns,2010). The Brain Connectivity Toolbox provides
a function for sorting the rows and columns according to each cell’s correla-
tion strength, moving the strongest correlations to the center of the graph. This
means that the order of the regions as they appear on the x- and y-axes of the
graph are sorted to make visualizing the differences between matrices as clear
as possible. This is why the order of the rows and columns differs between
Figure 19 and Figure 20.
When analyzing the task-state correlations of the low, medium, and high rat-
ings, the ACC and medial prefrontal cortex (MPFC) were not included in the
5.2 results 73
analyses. These regions were shown to be modulated by the ratings in the whole-
brain analysis, and as such will skew statistical comparison between the regions
that were not shown to be modulated by the ratings. However, as the analysis
of individual differences was done to examine differences between groups of
subjects, the ROIs placed in the MPFC and the ACC were included because of the
interest in determining how these regions may function differently between the
groups of under consideration.
5.2 results
The primary focus of this investigation was to determine the neural correlates
of gratitude. The results showed at the whole-brain level that the rating of grati-
tude correlated with brain activity in the MPFC of both hemispheres, extending
posteriorly from near to the frontal pole to the peri-genual ACC. (see Figure 17
and Table 4). Separate analyses in which the need and effort ratings were used
as covariates in the design matrix did not yield significant results.
Figure 17: Neural correlates of gratitude; regions in medial prefrontal cortex correlate
with ratings of gratitude during the experience of receiving gifts.
The design of the experiment also allowed the investigation of how a gift’s felt-
need and perceived effort modulate brain activity. In this analysis, the trials were
74 the brain’s virtuous cycle: mapping the neural correlates of gratitude
peak voxel
Analysis Region k p Z x y z
Correlates of Gratitude L & R MPFC 816 0.00919 3.48 -12 40 4
High Need > High Effort L & R ACC, MPFC 866 0.00141 3.88 4 54 -6
High Need > Effort = Need L STS 463 0.0179 4.07 -64 -28 -2
Effort = Need > High Effort L & R MPFC 621 0.018 3.35 -6 26 14
Table 4: Summary of two approaches to analyzing the data. Row one summarizes brain
regions that correlate with the rating of gratitude. Rows two through four sum-
marize brain responses to the stimuli as belonging to separate conditions. ‘High-
Need > High Effort’ refers to comparison of stimuli categorized by ratings that
are simultaneously High-Need and Low-Effort to stimuli that are High-Effort
and Low-Need. Conditions with stimuli judged as equal for effort and need
are ‘Effort = Need.’ Acronyms: MPFC, ACC, STS. L = Left Hemisphere; R = Right
Hemisphere
classified using participants’ ratings from the previously described psychophys-
iological study in chapters 3 and 4. The trials were classified based upon the
difference in their need and effort ratings by subtracting the need score from the
effort score. Trials in the High-Need/Low-Effort condition, compared to trials
rated High-Effort/Low-Need condition, elicited activity the MPFC. When com-
paring High-Need/Low-Effort with Effort = Need, the High-Need/Low-Effort
trials elicited activity in the left middle temporal gyrus (MTG) and superior tem-
poral sulcus (STS). The Effort = Need trials, compared to High-Effort/Low-Need,
elicited activity in the ACC, extending into a small portion of the MPFC. (see Fig-
ure 18 and Table 4). No other comparison yielded significant data.
The data were also analyzed using a task-state correlation. This method al-
lowed us to examine brain activity in regions that were a part of the a priori
regions of interest but were not found to be active or modulated in the whole
brain analyses. Task-state correlations were calculated by categorizing the tri-
als in three separate ways according to the need, effort and gratitude ratings,
respectively. Need, effort and gratitude were each then compared according to
the low, medium, and high ratings. The matrices for need did not differ for any
comparison between the ratings. The matrices for effort were different when
5.2 results 75
Need Effort Need Effort>
Need Effort> Need Effort=
> Need Effort Need Effort=
Figure 18: Overlay of regions active based on effort and need. The conditions are sep-
arated using ratings from a separate subject sample. The conditions are de-
fined as being ‘High-Need and Low-Effort’ when the rating of need is higher
than the rating of effort, represented in the figure legend as" Need# Effort.
‘High-Effort and Low-Need’ represents trials in which the rating of effort is
higher than the rating of need, represented in the figure legend as" Effort
# Need. ‘Effort = Need’ refers to conditions in which the rating of effort is
roughly equal to the rating of need.
comparing the medium rated trials to the high rated trials, and for the low rated
trials compared to the medium rated trials. The matrices for gratitude differed
in comparing the low and high rated trials as well as between the medium and
high rated trials. (see Table 5 and Figure 19)
This approach was also applied to explore individual differences among the
study participants. Participants were selected into groups based upon their quar-
tile rank score on the GQ-6 scale, the IRI-Perspective Taking sub-scale, and based
on the standard deviation of their gratitude ratings, respectively. The highest
and lowest quartile of participants were compared on each of these measures.
Low compared to high GQ-6 participants showed differences for each type of
rating at each level except for medium rated effort trials. Comparing low IRI-
76 the brain’s virtuous cycle: mapping the neural correlates of gratitude
A
C
Figure 19: Task-state Correlation for Differing Levels of Gratitude. Each square repre-
sents the average r value of the correlation between two brain regions, as
identified on the x- and y-axes. Panel A represents trials rated as eliciting high
levels of gratitude, panel B represents trials eliciting low levels of gratitude
and panel C represents trials rated as eliciting a medium level of gratitude.
The task-state correlation differs between the levels of reported gratitude.
5.2 results 77
Analysis of correlation matrix distributions
Condition Sample 1 Sample 2 p KS Score
Need Low Rated Trials High Rated Trials 0.374 0.124
Need Medium Rated Trials High Rated Trials 0.115 0.162
Need Low Rated Trials Medium Rated Trials 0.91 0.076
Effort Low Rated Trials High Rated Trials 0.374 0.124
Effort Medium Rated Trials High Rated Trials 0.004 0.238
Effort Low Rated Trials Medium Rated Trials 0.001 0.314
Gratitude Low Rated Trials High Rated Trials 0.001 0.333
Gratitude Medium Rated Trials High Rated Trials 0.038 0.19
Gratitude Low Rated Trials Medium Rated Trials 0.115 0.162
Table 5: Summary of functional connectivity comparisons between trial ratings sorted
by condition, i.e. need, effort and gratitude. This provides a summary statistical
comparison of how connectivity differed between each level of each rating.
Perspective Taking scores to high IRI-Perspective Taking scores showed differ-
ences in every comparison except medium and high need, and in low gratitude.
Comparing participants based upon the standard deviation of their gratitude
ratings, the differences were significant in every comparison except high need
and low effort. (see Table 6 and Figure 19).
78 the brain’s virtuous cycle: mapping the neural correlates of gratitude
Comparison of task-state correlation between personality types
Sample 1 Sample 2 Condition Rating Level p KS Score
Low GQ-6 High GQ-6 Need Low 0.001 0.431
Low GQ-6 High GQ-6 Need Med 0.001 0.294
Low GQ-6 High GQ-6 Need High 0.001 0.373
Low GQ-6 High GQ-6 Effort Low 0.001 0.379
Low GQ-6 High GQ-6 Effort Med 0.285 0.111
Low GQ-6 High GQ-6 Effort High 0.001 0.49
Low GQ-6 High GQ-6 Gratitude Low 0.001 0.529
Low GQ-6 High GQ-6 Gratitude Med 0.002 0.209
Low GQ-6 High GQ-6 Gratitude High 0.001 0.32
Low IRI-PT High IRI-PT Need Low 0.001 0.379
Low IRI-PT High IRI-PT Need Med 0.103 0.137
Low IRI-PT High IRI-PT Need High 0.717 0.078
Low IRI-PT High IRI-PT Effort Low 0.001 0.281
Low IRI-PT High IRI-PT Effort Med 0.01 0.183
Low IRI-PT High IRI-PT Effort High 0.001 0.281
Low IRI-PT High IRI-PT Gratitude Low 0.81 0.072
Low IRI-PT High IRI-PT Gratitude Med 0.002 0.209
Low IRI-PT High IRI-PT Gratitude High 0.001 0.294
Low Std Dev High Std Dev Need Low 0.003 0.203
Low Std Dev High Std Dev Need Med 0.005 0.196
Low Std Dev High Std Dev Need High 0.077 0.144
Low Std Dev High Std Dev Effort Low 0.62 0.085
Low Std Dev High Std Dev Effort Med 0.001 0.301
Low Std Dev High Std Dev Effort High 0.041 0.157
Low Std Dev High Std Dev Gratitude Low 0.002 0.209
Low Std Dev High Std Dev Gratitude Med 0.021 0.17
Low Std Dev High Std Dev Gratitude High 0.001 0.353
Table 6: Participants were selected for having a trait in the highest or lowest quartile of
scores. High or Low Std Dev refers to participants’ variability in how they rated
their felt gratitude in the experiment as measured by the standard deviation of
their ratings, i.e. participants with a higher standard deviation of their gratitude
ratings have more variety in how they rate the gratitude of the trials.
5.2 results 79
Low Trait Gratitude
High Trait Gratitude
Figure 20: Task-state Correlation Matrices of High and Low Trait-Gratitude Participants. Overall, greater patterns of task-state correlation are
found in the participants with lower scores of trait gratitude.
80 the brain’s virtuous cycle: mapping the neural correlates of gratitude
5.3 discussion
This investigation sought to determine the neural correlates of gratitude, and
to investigate how altering parameters in how gifts are perceived will modulate
the brain networks involved in the processing of gifts. It was hypothesized that
ratings of gratitude would correlate with regions involved in social cognition
(ACC, Temporal Pole), reward, and basic emotion (VMPFC, Nucleus Accumbens,
Insula). The data supported part of this hypothesis. Ratings of gratitude corre-
lated with activity in a cluster that includes much of the MPFC, including the
peri-genual ACC. Insofar as candidate regions to correlate with gratitude, this
region is a logical candidate, as it is involved in myriad social and emotional
tasks, including mind-reading and mentalizing (Saxe, 2006), fairness and eco-
nomic decision making (Tabibnia and Lieberman, 2007; Weber et al., 2009), as
well as emotion processes such as the formation of somatic markers and self-
referential processing (Araujo et al., 2013; Damasio, 1996).
The ratings of need and effort did not significantly correlate with any regions.
This indicates that the ratings of gratitude, and the experience of gratitude itself
is the primary factor for explaining variance in brain activity in this experiment.
This finding is the first of its kind that I know of to identify regions of the brain
that correlate with gratitude, but it is somewhat disappointing that these results
limit the ability to differentiate need and effort as well, as they independently
do not appear to correlate with variance in brain activity.
The gifts in the study were designed to vary according to how much they were
needed and how much perceived effort was involved, and it was predicted that
these factors would also be related to modulations in brain activity. Specifically,
the independent modulation of need and effort was hypothesized to modulate
brain activity as a product of the difference between these ratings. For instance,
we often receive gifts that come at great effort, but that are not needed (High-
5.3 discussion 81
Effort/Low-Need). Conversely, we often receive gifts that fulfill a serious need,
but did not come at significant effort from the donor (High-Need/Low-Effort).
The stimuli were thus analyzed by subtracting the need rating from the effort
rating, and this provides a measure of the difference between need and effort.
The stimuli were then compared by classifying the stimuli using the participant
responses from chapter 2 into the following categories: High-Effort/Low-Need,
High-Need/Low-Effort or Effort=Need. Brain activity for each of these condi-
tions was then contrasted with each other.
Stimuli classified as High-Need/Low-Effort, when compared to the High-
Effort/Low-Need, elicited a pattern of activity similar to the regions found to
correlate with the ratings of gratitude, with the exception being that the pat-
tern did not include the ACC as it did in the gratitude correlation analysis. The
similarity in the patterns of activity between the gratitude correlation, and the
High-Need/Low-Effort stimuli fits the story that the need for a gift is more
predictive of gratitude, as gifts that are high-need are processed in overlapping
parts of the brain with those associated with gratitude.
The High-Need/Low-Effort, contrasted with trials where need and effort were
equal, elicited activity in the STS. This region is not part of the a priori regions,
but its presence in this contrast makes sense given the context of the experi-
ment. Previous reports have shown that this region’s activity is correlated with
ratings of intensity in narrative stories (Wallentin et al., 2011), and in the recall
of emotionally intense autobiographical memories (St Jacques et al., 2011). In
the current experiment, participants were faced with processing emotionally in-
tense stimuli, and part of this intensity can be drawn from large differences in
the need and effort, as these trials may induce greater emotional conflict. Thus
stimuli in which there was disparity in the components of need and effort would
elicit more activity in regions coding for intensity when compared to stimuli in
which the need and effort were largely equal to each other. In other words, there
is a higher component of “conflict” for stimuli in which need and effort differ
82 the brain’s virtuous cycle: mapping the neural correlates of gratitude
from each other, as they require greater independent processing for each com-
ponent, and this drives up activity in regions associated with intensity.
The Effort=Need condition elicited greater activity in the ACC when compared
to stimuli in the High-Effort/Low-Need condition. This part of the ACC is asso-
ciated with processing of empathy (Singer et al., 2006) and in processing the
pain of others (Fox et al.,2013). This finding is counterintuitive. The ACC is asso-
ciated with conflict, and as such it would seem that stimuli in which the effort
outweighed the need would create a sense of conflict, or even awkwardness in
the participants. This analysis however, does not take into account the level of
the ratings, so in fact it could be that the stimuli in the Effort=Need condition
are still perhaps associated with gifts that were not overall associated with grat-
itude and were thus seen as disappointing in some way–which could feasibly
map onto the other purported roles of the ACC, specifically those associated with
fairness in social exchanges (Guroglu et al., 2010, 2011)
There were other regions (such as the insula and PMC) that were hypothesized
to be involved in that were not significantly active in the correlation with grati-
tude or in the analysis of need and effort. However, it can be assumed that while
these regions did not correlate with gratitude, given their role in related emotion
processes, it is still of interest to explore their behavior during the experiment.
There are a number of ways in which these regions’ behavior could be examined,
such as using a lower whole-brain threshold, or comparing their activity using
an ROI analysis and performing ANOVA tests to compare their levels of activ-
ity between conditions. This approach could yield a rather coarse knowledge of
what these regions were doing in comparison to each other, but would not tell
us how these regions were working in relation to each other. For this reason, I
performed an exploratory analysis of task-state correlation by creating matrices
revealing how each regions’ activity correlated with each other region. Such an
analysis provides a summary glance at how these regions may be modulated by
the stimuli in terms of need, effort and gratitude, or by individual differences
5.3 discussion 83
on personality measures. Correlations found using this exploratory technique
can also be used to form hypotheses for future studies.
The first analysis of task-state correlation consisted of comparing correlations
among varying levels (e.g. low, medium, high) of need, effort, and gratitude.
Comparing the matrices between each level of each condition provides a glimpse
of how these regions may be correlated, and this gives us a glimpse into the
brain circuitry underlying the perception of the gifts at various levels within the
conditions.
The data indicated no differences in the distribution of correlations related to
the rating of the level of need. Insofar as this measure of task-state correlation
is concerned, need does not seem to be associated with the modulation of how
these regions may work in correlation to one another. This is an interesting
difference from the other analyses where need and gratitude seem to explain
variance in the brain data in a similar way. Perhaps then it is in these regions
that, because need is not important in modulating their activity, this may be
related to why they do not show up at the whole brain level.
The data showed differences in the Effort ratings between the medium and
high-rated trials, as well as between the low and medium rated trials. Overall,
the average values of the matrices in the effort trials showed more similarity
between the high and low ratings–each showing higher correlation values than
the medium rated trials. It may be that there is some sort of u-shaped function
involved in coding the effort of a gift, wherein the high and low effort are coded
similarly, perhaps based upon salience or conflict, whereas the medium effort
trials are less emotionally provocative, and so do not explain variance in the
brain regions of interest.
The data showed differences between the low and high rated trials, as well
as between the medium and high rated trials. The overall task-state correlation
matrix for gratitude showed a more linear relationship, with each level of rating
sponsoring an overall increase in the value of the correlation between the brain
84 the brain’s virtuous cycle: mapping the neural correlates of gratitude
regions (see Figure 19). Specifically, the increases between the levels seems to
be most driven by correlations between PMC, AI, and S2, as is apparent in the
graphs. These regions are all involved in the processing of the interoceptive
processing of stimuli, so these correlations may be related to the processing of
the interoceptive value of the gifts themselves.
The analysis of task-state correlation can also be applied to identifying differ-
ences between measurements of individual differences on personality measures.
This approach provides an interesting glance at individual differences, as it pro-
vides information not only to which regions are used differently between groups
of people, but also can identify how individuals in each group use their brains
differently. The personality types that were investigated were selected based
upon the results of the previous chapters, and they included the GQ-6 trait-
gratitude questionnaire, the IRI-perspective taking scale, and the measurement
of each participant’s standard deviation of their gratitude ratings.
In each trait comparison, participants from the lower group showed higher
task-state correlation than the participants in the higher group. As we have seen
in previous chapters, it stands to reason that individuals who are adept at hav-
ing an accurate amount of gratitude may react to these stimuli with less of an
attempt at generating a strong response. Specifically, in the comparison between
high and low trait-gratitude individuals, high trait-gratitude individuals had a
stepwise increase in task-state correlation for each level of rating of the stimuli,
whereas low trait-gratitude individuals show something of a u-shaped function,
with low and high levels of gratitude both showing higher levels of task-state
correlation than the medium ratings. As a point of greater detail, it appears
that high trait-gratitude, in the high-gratitude rated trials, shows distinctly less
task-state correlation between the left and right nucleus accumbens than low
trait-gratitude individuals. Although this is an exploratory analysis, given the
nucleus accumbens’ role in reward, perhaps these regions function differently
5.4 conclusion 85
in processing of gifts and social contexts in high trait-gratitude individuals com-
pared to other individuals, and this may be an avenue for future research.
The study intended to capture the brain activity during the experience of
gratitude. To the extent that participants honestly reported their level of felt
gratitude, the study was effective in eliciting gratitude and in capturing brain
activity during its experience.
The study focused on gratitude as it falls on a continuum, allowing for an
understanding of the range of gratitude experiences. This is important if we
are to understand, quite literally, what are the neural correlates of gratitude.
As an innovation in emotion research, this provides a more map-like quality to
the understanding of emotion as opposed to envisioning emotion as all-or-none
event.
5.4 conclusion
This investigation has largely met its goals and supported its primary hypothe-
ses. The neural correlates of gratitude were localized to the MPFC, an important
region for self-referential cognition and reward. In addition, altering levels of
need and effort also resulted in altered patterns of brain activity, pointing fu-
ture investigations to the promise of creating a brain map of gift-space.
We also see support for two ideas from the previous investigations. First, in
this study, need tended to better predict gratitude, and this was also reflected at
the level of the brain, as stimuli associated with high-need elicited activity that
closely resembled the patterns of activity elicited by the correlates of gratitude it-
self. Second, in the analysis of individual differences associated with processing
these stimuli, individuals who presented with higher scores on measurements
of trait gratitude and perspective-taking showed different patterns of task-state
correlation, indicating that there seems to be a ‘skill’ associated with processing
86 the brain’s virtuous cycle: mapping the neural correlates of gratitude
these stimuli; grateful individuals may indeed use their brains differently than
less grateful individuals.
Identifying the neural correlates of gratitude also plays a deeper role. The
experience of gratitude is rich and complex. Capturing it as it is experienced
by individuals poses many difficulties, but presents an opportunity to take this
healing and critical emotion and localize it to a specific set of regions. This
finding reinforces a notion of human goodness as it comes from science.
6
C O N C L U S I O N : T H E G R AT I T U D E P R E S C R I P T I O N
How much better it is, with gratitude for the pleasures one has enjoyed, not to calculate
the years allotted to other people, but to set a generous value on one’s own and count
them as gain. ‘God judged me worthy of this: this is enough; he could have given me
more, but even this is a benefit.’ Let us be grateful to the gods, grateful to men, grateful
to those who have done something for us, grateful to those who have done something for
those dear to us.
— Seneca (On Benefits, 5.17.5)
6.1 summary of findings
Gratitude is known to have powerful effects on our relationships, ourselves, and
our well-being. In the research that has been conducted to examine gratitude,
we have seen that gratitude emerges through the recognition of a gift that is
needed by the recipient and comes at effort by the donor. In clinical applications,
gratitude interventions can lead to powerful benefits.
However, previous research on gratitude has yet to investigate crucial ques-
tions for understanding gratitude’s true nature. First, it is yet unclear what pre-
cisely are the necessary and sufficient conditions for gratitude to be generated,
and how does gratitude relate to other emotions and feelings? In addition, given
gratitude’s connection to health and well-being, what changes occur in the body
during the experience of gratitude that may underly these health benefits? Fi-
87
88 conclusion: the gratitude prescription
nally, how can we use these findings to generate and enhance gratitude in our
daily lives?
To determine what are the necessary and sufficient criteria for gratitude, I
created the gift-space model of gratitude. The results indicated that gratitude
resulted from receiving a gift that came at effort from the benefactor and fulfilled
a need for the recipient. These data represented the first, that I know of, that put
need, effort, and gratitude on a 3-d axis to examine their relationship in detail.
This approach yielded a novel finding: that the felt need for a gift was a better
predictor of gratitude than the perceived effort. Insofar as what are the necessary
and sufficient criteria for gratitude, the data indicated that while need and effort
together correlate with gratitude, both need not be present.
I next sought to investigate how the experience of gratitude would influence
heart rate, yielding new insight into gratitude and physiology. Higher levels of
gratitude were marked by overall lower patterns of heart rate when compared
to lower ratings of gratitude. This indicates that gratitude is likely related to
activity in the parasympathetic system, which may explain the lowered heart
rate in the trials in which gratitude was deeply experienced.
To investigate individual differences in the experience of gratitude as they
relate to physiology, participants’ vagal tone (CVT) was correlated to the partic-
ipants’ gratitude ratings. The results indicated a negative correlation with the
participants’ average gratitude ratings, but a positive correlation to the standard
deviation of the ratings of gratitude. On the assumption that it is better to have
a higher vagal tone, this finding indicates that the level of gratitude we should
feel is to be differentiated based on circumstance, experienced at an appropriate
level, and not self-driven to the highest possible value.
Finally, I measured brain activity during the experience of gratitude. Ratings
of gratitude significantly correlated with activity in the medial prefrontal cortex,
a region associated with social cognition, reward, and self-processing. The par-
ticipants’ ratings of need and effort were also analyzed. In this analysis, need
6.2 seneca’s oversight 89
was associated with brain activity in largely the same region as the regions that
correlated with gratitude. This finding reinforces the importance of focusing on
self-processing in the experience of gratitude. It also fulfills a larger purpose in
localizing the rich and complex emotion of gratitude to a part of the brain that
literally connects ourselves to our surroundings and in so doing, our deeper
humanity.
6.2 seneca’s oversight
I began my thesis by describing Seneca’s philosophy of benefits. As a way to fur-
ther contextualize my results, I would like to reexamine some of his statements
keeping my findings in mind. How does history’s greatest thinker on gratitude
square with the results from my thesis? I would say that Seneca’s wisdom has
successfully bounded two millennia, but he has committed two important over-
sights.
Seneca describes in some terms the fact that the joy in a gift is some combina-
tion of the need for a benefit and the effort involved in creating it:
“Two things must combine to produce a benefit. First comes the ques-
tion of size; some things do not measure up to the name. Who has
ever called a morsel of bread a benefit, or a coin tossed as alms or
allowing someone to kindle his fire from ours? There are occasions
when these things are more helpful than the greatest gifts, and that
their cheapness, even when they become necessities because of the
needs of the moment, reduces their value. The second and most im-
portant requirement is that I must act for the sake of the person who
is the intended recipient, and that I must judge him worthy and give
willingly, deriving joy from my gift (4.29.2).”
90 conclusion: the gratitude prescription
Seneca notices that there are two components to a gift, but is he mistaken when
he says that the intention is more important?
Seneca was not speaking in a vacuum: his was a life of privilege, of all basic
needs being met. Seneca would not likely be grateful for benefits that provide
a basic need, since these were constantly being met. Instead I would predict
that he would be most grateful by donors working to address his psychological
needs; these acts are best understood through their intention. In my studies,
need was more predictive of gratitude, but that also did not occur in a vacuum:
in the tragedy faced by survivors of the Holocaust, their deficits were indeed so
great that any need fulfilled was a critical need. In a wintertime death march, the
source of the new shoes and the intent behind them becomes secondary to the
need they serve. Let us not limit our conception of the rich and complex emotion
of gratitude by pointing to need or effort as the primary driver of gratitude.
Given changing contexts, either dimension may come to the fore. The fact is,
that when we say “it’s the thought that counts” what we are really noticing is
the way that someone’s thoughtfulness is benefitting our psychological needs–
intentionally or not. We may be so lucky as to consider only others’ effort in
determining our gratitude for a benefit.
There was another surprising result in my data: that individuals high in
personality measures of perspective-taking (chapter 1), high on cardiac vagal
tone (chapter 3), and high in personality measures of trait-gratitude (chapter 4),
showed interesting differences in how they responded to the gifts. These unique
participants did not rate the gifts using higher ratings of gratitude as we might
expect, but instead they used greater differentiability in the rating the gifts. That
we should seek greater differentiability in our perceptions of the gifts we receive
relates to Seneca’s other oversight. The data indicate that we should find the ap-
propriate amount of gratitude for a given context, and not just more overall. We
can apply this finding to Seneca’s widely panned discussion of whether or not
a slave could be grateful to a master, or vice versa.
6.2 seneca’s oversight 91
Seneca is technically not wrong to say that a slave could be grateful to a master.
As a matter of probability, it is feasible that some slaves did have instances of
gratitude toward their masters. The question however, is whether or not it is
appropriate for a slave to feel grateful to a master. In the lens of modern ethics,
this at first seems a disgusting notion, but there may be some that would say
that we should always feel gratitude, that gratitude is transcendent and that
it is up to us to generate gratitude in every situation. Indeed, on this matter,
Seneca seems to think that yes, at times it would be appropriate. Recall this
quote: “Just as an employee gives a benefit when he provides more than what
he has contracted for, so too does a slave: when the good will he feels for his
master goes beyond the boundaries set by his position in life and when he aims
at something higher, something that would do credit even to someone of luckier
birth, and so surpasses the expectations of his master, then we have found a case
of benefit given within the household (2.22.1).” This is where we see Seneca’s
oversight in the light of my findings. Learning from those in my study sample
population who are high in perspective-taking, vagal tone and trait-gratitude,
we would predict that those who are skilled in social situations would have no
gratitude to a master, and this is the appropriate level of gratitude. But why?
Let us consider that a primary function of gratitude is as a way to maintain
social relationships. Maintaining relationships is critical to enhance our daily
well-being, but we cannot be cheated into feeling gratitude as a way to maintain
abusive relationships. In the master-slave relationship, one party has freedom,
the other obviously does not. Maintaining this relationship through gratitude
will only perpetuate the abuse. In the case of abusive relationship, the induction
of gratitude is cheating the brain’s moral drive to kindle positive relationships.
Thus, this is a relationship where gratitude is indeed inappropriate, and we can
learn from the data in my thesis that we should seek the right level of gratitude
for a given situation, even and especially if the right level of gratitude is no
gratitude at all.
92 conclusion: the gratitude prescription
6.3 how to be grateful.
The data from these studies has revealed new insight into gratitude. The fact
that we can, with the help of modern brain imaging, take the rich and complex
emotion of gratitude, and pin it to a specific region in the brain provides a
deep connection between our biology and our humanity. In finding that cardiac
vagal tone correlates with ratings of gratitude, we see that the experience of
something as seemingly subjective as gratitude takes its root in our deepest
physiology. On their own, my findings provide a fascinating insight into what
makes us grateful and what happens in our body when we feel gratitude. But
how can we connect these findings to our desire to enhance gratitude in our
daily lives? How can we use these data to generate gratitude on a day that
we are struggling to maintain well-being? The data give us two preliminary
prescriptions for increasing gratitude in a given moment (the examination of
which will form my next investigations).
As we saw in chapter 1, the felt need for a gift is more important to grati-
tude than effort. That tells us: think first of how the things others have done to
benefit us fulfill our needs at deeper levels. When we receive a gift, we should
consider it for its ability to fulfill our psychological need to feel cared for and
beloved. This is different from the folk wisdom telling us that “it’s the thought
that counts.” While in my studies, the perceived effort in a gift did indeed cor-
relate with the level of felt gratitude, need was a better predictor. In my exam-
inations of the physiological bases of gratitude, effort did not explain variance
in heart rate and brain activity nearly as well as need, telling us that the felt
need for a gift is closer to gratitude itself. I think the reason to focus on need
instead of effort stems from the fact that our accuracy in reading other people’s
minds will vary considerably. If your ability to read someone’s true intention is
what you use to generate gratitude, the amount of gratitude you feel may not be
6.3 how to be grateful. 93
accurately linked to the situation. In this, it seems that it is better to rely on the
dimension that we can know well: ourselves and our needs. We should focus on
how the things we have fulfill the diverse needs we had.
The next lesson to be gleaned from these data emerged from my analysis of
individual differences. In three separate datasets, I saw that individuals high on
a number of traits that we deem advantageous (perspective-taking, trait grati-
tude, and cardiac vagal tone) did not rate the gifts as eliciting higher levels of
gratitude overall. Instead, these individuals saw the gradations in the stimuli.
On the large assumption that we can treat these individuals as some sort of
example, the take home message is that we should not try to feel maximum
gratitude at all times. Instead, we should try to see each situation and gift we
receive as necessitating an accurate, appropriate amount of gratitude.
In closing, if there is an emotion capable of promoting good human conduct
and respect between people, that emotion is gratitude. Gratitude arms us against
despair and discrimination, fortifies us against trauma, and impels us to help
our neighbors. A neuroscience account of good human conduct unfolds a story
told by the body, revealing volumes about the necessary and sufficient condi-
tions for the flowering of human goodness at the level of the person and at the
level of society.

Part IV
A P P E N D I X

appendix 97
Appendix A: Instructions to participants
We will now present you with descriptions of the four phases of the Holocaust in the
form of video montages. After each montage, you will be asked to rate a series of events,
as if you were personally experiencing them as a Holocaust victim and survivor. Please
imagine yourself as a persecuted person during the Holocaust, and imagine how you
would feel if you were in the situation that has been described and rate it accordingly.
Specifically, after you hear the description of each phase, we will ask you to respond to
a number of events. The questions will ask you to rate how much you needed something,
or to rate how much effort it took to give something to you. As a point of reference
for your ratings, we want you to consider something that directly led to saving your
life as the highest type of need, whereas something that cost the giver his or her life
would qualify as the highest amount of effort. Something that you did not need at all, for
instance, it did not help or hurt, would be low on the rating scale. A ”medium” rating
on the need scale would be an event that significantly helped your life, but you could
survive without it; a ”medium” rating on the effort scale would be an event where the
donor’s way of life is impacted, but he or she will still be able to survive after the act of
giving.
For further clarification, when we say ”need” we are referring to how much the item
fulfills your ability to live a complete and happy life; something that fulfills a psycho-
logical, physiological, or even basic need. When we ask you to rate the amount of effort,
or need, keep in mind that the highest rating should only be used for life-or-death situ-
ations. The ”events” that follow the phase description are not ordered in any particular
sequence. So, you may find events that do not have any relation to each other. Treat every
event as an independent event in and of itself.
Finally, every one of the events has been drawn from the Shoah Foundation’s Visual
History Archives. These are real events that happened to real victims and survivors of
the Holocaust.
98 appendix
Appendix B: Central point of hand-placed ROIs
Region of Interest Placed on slice: Location
MPFC Sagittal Halfway horizontally between the ACC ROI and the frontal pole
ACC Sagittal Halfway between the anterior-most point of the genu of the corpus callosum and the
cingulate sulcus
Nuc Acc Coronal Inferior to the striatum, where the caudate and putamen connect; at the center of the
nucleus accumbens itself
Temp Pole Coronal On the anterior-most slice where white matter is visible, halfway across the width of the
white matter
AI Sagittal On the sagittal slice in which the insula is the widest, centered between the two anterior-
most insular gyrii
PI Sagittal On the sagittal slice in which the insula is the widest, centered on the posterior-most
insular gyrus
Inf PMC Sagittal Halfway along the sub-parietal sulcus between the arcuate cingulate sulcus and the occip-
itoparietal sulcus
Sup PMC Sagittal Halfway between the Inf PMC and the edge of the brain, centered on the precuneus
S2 Sagittal On the operculum halfway between the point of intersection between the central and
sylvian sulcii
Table 7: Guidelines for the placement of ROIs in each individual’s anatomical brain image. ROIs were placed in both hemispheres. ROIs were
placed at the depth of the cortex such that the white matter was just visible; some ROIs along the mid-sagittal plane were placed
slightly deeper to prevent the sphere from extending into the opposite hemisphere. Acronyms: medial prefrontal cortex (MPFC),
anterior cingulate cortex (ACC), nucleus accumbens (Nuc Acc), temporal pole (Temp Pole), anterior insula (AI), posterior insula (PI),
inferior posteromedial cortex (Inf PMC), superior posteromedial cortex (Sup PMC), secondary somatosensory cortex (S2).
B I B L I O G R A P H Y
Abercrombie, H., Chambers, A., Greischar, L. and Monticelli, R. (2008), ‘Orient-
ing, emotion, and memory: phasic and tonic variation in heart rate predicts
memory for emotional pictures in men.’, Neurobiol Learn Mem 90(4), 644–650.
Algoe, S. B., Fredrickson, B. L. and Gable, S. L. (2013), ‘The social functions of
the emotion of gratitude via expression.’.
Algoe, S., Haidt, J. and Gable, S. (2008), ‘Beyond reciprocity: gratitude and rela-
tionships in everyday life.’, Emotion 8(3), 425–429.
Ames, D., Flynn, F. and Weber, E. (2004), ‘It’s the thought that counts: on perceiv-
ing how helpers decide to lend a hand.’, Pers Soc Psychol Bull 30(4), 461–474.
Araujo, H. F., Kaplan, J. and Damasio, A. (2013), ‘Cortical midline structures
and autobiographical-self processes: An activation-likelihood estimation meta-
analysis.’, Frontiers in human neuroscience 7, 548.
URL: http://www.ncbi.nlm.nih.gov/pubmed/24027520
Banks, S. (1979), ‘Gift-giving: A review and an interactive paradigm’, Advances
in consumer research 6(1), 319.
Bartlett, M. and DeSteno, D. (2006), ‘Gratitude and prosocial behavior: helping
when it costs you.’, Psychol Sci 17(4), 319–325.
Bechara, A., Damasio, H. and Damasio, A. R. (2000), ‘Emotion, decision making
and the orbitofrontal cortex’, Cerebral cortex 10(3), 295–307.
Bradley, M. (2009), ‘Natural selective attention: orienting and emotion.’, Psy-
chophysiology 46(1), 1–11.
Damasio, A. (1994), Descartes’ error: Emotion, reason, and the human brain, Quill
New York.
Damasio, A. (1996), ‘The somatic marker hypothesis and the possible functions
of the prefrontal cortex.’, Philos Trans R Soc Lond B Biol Sci 351(1346), 1413–
1420.
Damasio, A. (2011), ‘Neural basis of emotions’, Scholarpedia 6(3), 1804.
Davis, M. H. (1994), Empathy: A social psychological approach., Westview Press.
DeSteno, D., Bartlett, M., Baumann, J., Williams, L. and Dickens, L. (2010), ‘Grati-
tude as moral sentiment: emotion-guided cooperation in economic exchange.’,
Emotion 10(2), 289–293.
99
100 bibliography
Eisenberg, N., Fabes, R. A., Murphy, B., Karbon, M., Maszk, P ., Smith, M.,
O’Boyle, C. and Suh, K. (1994), ‘The relations of emotionality and regulation
to dispositional and situational empathy-related responding.’, Journal of per-
sonality and social psychology 66(4), 776.
Emmons, R. (2008), ‘Gratitude, subjective well-being and the brain’, The science
of subjective well-being p. 469.
Emmons, R. and McCullough, M. (2003), ‘Counting blessings versus burdens:
an experimental investigation of gratitude and subjective well-being in daily
life.’, J Pers Soc Psychol 84(2), 377–389.
Emmons, R. and McCullough, M. (2004), The psychology of gratitude, Oxford Uni-
versity Press, USA.
Fabes, R. A. and Eisenberg, N. (1997), ‘Regulatory control and adults’ stress-
related responses to daily life events.’, Journal of personality and social psychology
73(5), 1107.
Feinstein, J. S., Adolphs, R., Damasio, A. and Tranel, D. (2011), ‘The human
amygdala and the induction and experience of fear’, Current biology 21(1),34–
38.
Fox, G. R., Sobhani, M. and Aziz-Zadeh, L. (2013), ‘Witnessing hateful people
in pain modulates brain activity in regions associated with physical pain and
reward.’, Frontiers in psychology 4, 772.
URL: http://www.ncbi.nlm.nih.gov/pubmed/24167496
Fredrickson, B., Tugade, M., Waugh, C. and Larkin, G. (2003), ‘What good are
positive emotions in crises? a prospective study of resilience and emotions
following the terrorist attacks on the united states on september11th,2001.’, J
Pers Soc Psychol 84(2), 365–376.
Froh, J., Fan, J., Emmons, R., Bono, G., Huebner, E. and Watkins, P . (2011), ‘Mea-
suring gratitude in youth: Assessing the psychometric properties of adult grat-
itude scales in children and adolescents.’, Psychol Assess .
Geisler, F., Kubiak, T., Siewert, K. and Weber, H. (2013), ‘Cardiac vagal tone is
associated with social engagement and self-regulation’, Biological psychology
93(2), 279–286.
Geisler, F., Vennewald, N., Kubiak, T. and Weber, H. (2010), ‘The impact of heart
rate variability on subjective well-being is mediated by emotion regulation’,
Personality and Individual Differences 49(7), 723–728.
Griffin, M. T. (2013), Seneca on Society: A Guide to De Beneficiis, Oxford University
Press.
Gunther Moor, B., Crone, E. and van der Molen, M. (2010), ‘The heartbrake of
social rejection: heart rate deceleration in response to unexpected peer rejec-
tion.’, Psychol Sci 21(9), 1326–1333.
bibliography 101
Guroglu, B., van den Bos, W., Rombouts, S. and Crone, E. (2010), ‘Unfair? it de-
pends: Neural correlates of fairness in social context.’, Soc Cogn Affect Neurosci
.
Guroglu, B., van den Bos, W., van Dijk, E., Rombouts, S. and Crone, E. (2011),
‘Dissociable brain networks involved in development of fairness considera-
tions: Understanding intentionality behind unfairness.’, Neuroimage .
Harbaugh, W., Mayr, U. and Burghart, D. (2007), ‘Neural responses to taxa-
tion and voluntary giving reveal motives for charitable donations.’, Science
316(5831), 1622–1625.
Hsu, M., Anen, C. and Quartz, S. (2008), ‘The right and the good: distributive
justice and neural encoding of equity and efficiency.’, Science 320(5879), 1092–
1095.
Immordino-Yang, M., McColl, A., Damasio, H. and Damasio, A. (2009), ‘Neu-
ral correlates of admiration and compassion.’, Proc Natl Acad Sci U S A
106(19), 8021–8026.
Jenkinson, M., Bannister, P ., Brady, M. and Smith, S. (2002), ‘Improved optimiza-
tion for the robust and accurate linear registration and motion correction of
brain images’, Neuroimage 17(2), 825–841.
Jenkinson, M. and Smith, S. (2001), ‘A global optimisation method for robust
affine registration of brain images’, Medical image analysis 5(2), 143–156.
Kashdan, T., Uswatte, G. and Julian, T. (2006), ‘Gratitude and hedonic and eu-
daimonic well-being in vietnam war veterans.’, Behav Res Ther 44(2), 177–199.
Kashdan, T., Uswatte, G., Steger, M. and Julian, T. (2006), ‘Fragile self-esteem
and affective instability in posttraumatic stress disorder.’, Behav Res Ther
44(11), 1609–1619.
Kleiman, E. M., Adams, L. M., Kashdan, T. B. and Riskind, J. H. (2013), ‘Grati-
tude and grit indirectly reduce risk of suicidal ideations by enhancing mean-
ing in life: Evidence for a mediated moderation model’, Journal of Research in
Personality .
Kok, B. E., Coffey, K. A., Cohn, M. A., Catalino, L. I., Vacharkulksemsuk, T.,
Algoe, S. B., Brantley, M. and Fredrickson, B. L. (2013), ‘How positive emotions
build physical health perceived positive social connections account for the
upward spiral between positive emotions and vagal tone’, Psychological science
24(7), 1123–1132.
Kok, B. E. and Fredrickson, B. L. (2010), ‘Upward spirals of the heart: Autonomic
flexibility, as indexed by vagal tone, reciprocally and prospectively predicts
positive emotions and social connectedness’, Biological psychology 85(3), 432–
436.
102 bibliography
Lambert, N., Fincham, F., Stillman, T. and Dean, L. (2009), ‘More gratitude, less
materialism: The mediating role of life satisfaction’, The Journal of Positive Psy-
chology 4(1), 32–42.
Lane, J. and Anderson, N. (1976), ‘Integration of intention and outcome in moral
judgment’, Memory and Cognition 4(1), 1–5.
Lester, D. (1990), ‘Maslow’s hierarchy of needs and personality’, Personality and
Individual Differences 11(11), 1187–1188.
McCraty, R., Atkinson, M. and Bradley, R. (2004), ‘Electrophysiological evidence
of intuition: Part 2. a system-wide process?’, The Journal of Alternative & Com-
plementary Medicine 10(2), 325–336.
McCraty, R., Atkinson, M., Tiller, W., Rein, G. and Watkins, A. (1995), ‘The ef-
fects of emotions on short term heart rate variability using power spectrum
analysis’, American Journal of Cardiology 76(14), 1089–1093.
McCraty, R., Atkinson, M., Tomasino, D. and Bradley, R. (2009), ‘The coherent
heart heart–brain interactions, psychophysiological coherence, and the emer-
gence of system-wide order’, Integral Review 5(2).
Mccullough, M., Emmons, R. and Tsang, J. (2002), ‘The grateful disposition: a
conceptual and empirical topography.’, J Pers Soc Psychol 82(1), 112–127.
McCullough, M., Kilpatrick, S., Emmons, R. and Larson, D. (2001), ‘Is gratitude
a moral affect?’, Psychol Bull 127(2), 249–266.
Moor, B. G., Crone, E. A. and van der Molen, M. W. (2010), ‘The heartbrake of so-
cial rejection heart rate deceleration in response to unexpected peer rejection’,
Psychological Science 21(9), 1326–1333.
Oveis, C. S. (2009), ‘Vagal tone, positive emotion, and pro-sociality’, Dissertation
.
Porges, S. (2007), ‘The polyvagal perspective.’, Biol Psychol 74(2), 116–143.
Rubinov, M. and Sporns, O. (2010), ‘Complex network measures of brain con-
nectivity: uses and interpretations’, Neuroimage 52(3), 1059–1069.
Ruth, Brunel, J., Otnes, F. and Cele (2004), ‘An investigation of the power of
emotions in relationship realignment:’, Psychology & Marketing; 21, 1;, 29.
Saxe, R. (2006), ‘Four brain regions for one theory of mind’, JT Ca .
Sherry, J. F. (1983), ‘Gift giving in anthropological perspective’, Journal of Con-
sumer Research 10(2), 157–168.
Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R. and Frith, C. (2004),
‘Empathy for pain involves the affective but not sensory components of pain.’,
Science 303(5661), 1157–1162.
bibliography 103
Singer, T., Seymour, B., O’Doherty, J. P ., Stephan, K. E., Dolan, R. J. and Frith,
C. D. (2006), ‘Empathic neural responses are modulated by the perceived fair-
ness of others.’, Nature 439(7075), 466–469.
URL: http://www.ncbi.nlm.nih.gov/pubmed/16421576
Smith, S. M. (2002), ‘Fast robust automated brain extraction’, Human brain map-
ping 17(3), 143–155.
Smith, S., Miller, K., Salimi-Khorshidi, G., Webster, M., Beckmann, C., Nichols,
T., Ramsey, J. and Woolrich, M. (2010), ‘Network modelling methods for fmri.’,
Neuroimage .
St Jacques, P . L., Botzung, A., Miles, A. and Rubin, D. C. (2011), ‘Functional
neuroimaging of emotionally intense autobiographical memories in post-
traumatic stress disorder.’, Journal of psychiatric research 45(5), 630–637.
URL: http://www.ncbi.nlm.nih.gov/pubmed/21109253
Stellar, J. E., Manzo, V . M., Kraus, M. W. and Keltner, D. (2012), ‘Class and
compassion: Socioeconomic factors predict responses to suffering.’, Emotion
12(3), 449.
Tabibnia, G. and Lieberman, M. D. (2007), ‘Fairness and cooperation are re-
warding: evidence from social cognitive neuroscience.’, Annals of the New York
Academy of Sciences 1118, 90–101.
URL: http://www.ncbi.nlm.nih.gov/pubmed/17717096
Tarvainen, M. P ., Niskanen, J.-P ., Lipponen, J., Ranta-Aho, P . and Karjalainen,
P . (2009), Kubios hrv—a software for advanced heart rate variability analysis,
in ‘4th European Conference of the International Federation for Medical and
Biological Engineering’, Springer, pp. 1022–1025.
Tarvainen, M. P ., Ranta-aho, P . O. and Karjalainen, P . A. (2002), ‘An advanced de-
trending method with application to hrv analysis’, IEEE Transactions on Biomed-
ical Engineering 49(2), 172–175.
Tesser, A., Gatewood, R. and Driver, M. (1968), ‘Some determinants of gratitude.’,
Journal of Personality and Social Psychology 9(3), 233–236.
Tsang, J. (2006), ‘The effects of helper intention on gratitude and indebtedness’,
Motivation and Emotion 30, 199–205.
Tully, E. C., Donohue, M. R. and Garcia, S. E. (2014), ‘Children’s empathy re-
sponses and their understanding of mother’s emotions.’, Cognition & emo-
tion .
URL: http://www.ncbi.nlm.nih.gov/pubmed/24650197
Vernon, L., Dillon, J. and Steiner, A. (2009), ‘Proactive coping, gratitude,
and posttraumatic stress disorder in college women.’, Anxiety Stress Coping
22(1), 117–127.
104 bibliography
Vögele, C., Sorg, S., Studtmann, M. and Weber, H. (2010), ‘Cardiac autonomic
regulation and anger coping in adolescents’, Biological psychology 85(3), 465–
471.
Wallentin, M., Nielsen, A. H., Vuust, P ., Dohn, A., Roepstorff, A. and Lund,
T. E. (2011), ‘Amygdala and heart rate variability responses from listening to
emotionally intense parts of a story.’, NeuroImage 58(3), 963–973.
URL: http://www.ncbi.nlm.nih.gov/pubmed/21749924
Watkins, P ., Woodward, K., Stone, T. and Kolts, R. (2003), ‘Gratitude and
happiness: Development of a measure of gratitude, and relationships with
subjective well-being’, Social Behavior and Personality: an international journal
31(5), 431–451.
Weber, B., Rangel, A., Wibral, M. and Falk, A. (2009), ‘The medial prefrontal
cortex exhibits money illusion.’, Proc Natl Acad Sci U S A .
Wood, A., Froh, J. and Geraghty, A. (2010), ‘Gratitude and well-being: A review
and theoretical integration’, Clinical Psychology Review .
Wood, A., Joseph, S. and Maltby, J. (2008), ‘Gratitude uniquely predicts satisfac-
tion with life: Incremental validity above the domains and facets of the five
factor model’, Personality and Individual Differences 45(1), 49–54.
Wood, A., Joseph, S. and Maltby, J. (2009), ‘Gratitude predicts psychological well-
being above the big five facets’, Personality and Individual Differences 46(4),443–
447.
Woolrich, M. W., Ripley, B. D., Brady, M. and Smith, S. M. (2001), ‘Temporal auto-
correlation in univariate linear modeling of fmri data’, Neuroimage 14(6),1370–
1386.
Worsley, K. (2001), ‘Statistical analysis of activation images.’, Ch 14, in Func-
tional MRI: An Introduction to Methods eds. P. Jezzard, P.M. Matthews and
S.M. Smith. OUP.
Zahn, R., Garrido, G., Moll, J. and Grafman, J. (2013), ‘Individual differences
in posterior cortical volume correlate with proneness to pride and gratitude’,
Social cognitive and affective neuroscience p. nst158.
Zahn, R., Moll, J., Paiva, M., Garrido, G., Krueger, F., Huey, E. and Grafman, J.
(2008a), ‘The neural basis of human social values: Evidence from functional
mri.’, Cereb Cortex .
Zahn, R., Moll, J., Paiva, M., Garrido, G., Krueger, F., Huey, E. and Grafman, J.
(2008b), ‘The neural basis of human social values: Evidence from functional
mri.’, Cereb Cortex .
colophon
This document was typeset using the typographical look-and-feelclassicthesis
developed by André Miede. The style was inspired by Robert Bringhurst’s sem-
inal book on typography “The Elements of Typographic Style”. classicthesis is
available for both L
A
T
E
X and L
Y
X:
http://code.google.com/p/classicthesis/
Happy users of classicthesis usually send a real postcard to the author, a
collection of postcards received so far is featured here:
http://postcards.miede.de/
Final Version as of November 4, 2014 (classicthesis Version 1).

D E C L A R AT I O N
I declare that the attached dissertation is my own, original work undertaken in
partial fulfillment of my degree. The work presented here has not previously
been accepted in substance for any degree and is not being concurrently submit-
ted in candidature for any degree. I have made no use of sources, materials or
assistance other than those which have been openly and fully acknowledged in
the text. Any direct quotation or source of ideas has been identified in the text
by author, date and page number(s) immediately after such an item, and full
details are provided in a reference list at the end of the text. I consent that this
thesis, if accepted, to be available to others, and for the title and summary to be
made available to outside organizations.
Los Angeles, December, 2014
Glenn R. Fox

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Contribution of fundamental brain networks to individual differences in social cognition

PDF

Individual differences in heart rate response and expressive behavior during social emotions: effects of resting cardiac vagal tone and culture, and relation to the default mode network

PDF

Psychopathic traits and the fronto-amygdala circuit in a community sample: the role of trait anxiety

PDF

Engagement of the action observation network through functional magnetic resonance imaging with implications for stroke rehabilitation

PDF

Mapping the neural architecture of concepts

PDF

Spatial and temporal patterns of brain activity associated with emotions in music

PDF

Functional models of fMRI BOLD signal in the visual cortex

PDF

Experience modulates neural activity during action understanding: exploring sensorimotor and social cognitive interactions

PDF

Decoding information about human-agent negotiations from brain patterns

PDF

The neuroscience of ambivalent and ambiguous feelings

PDF

Neurobiological correlates of skilled and disabled reading

PDF

Majoring in music: how conservatory training changes the brain

PDF

Signal processing methods for brain connectivity

PDF

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

PDF

Computational models and model-based fMRI studies in motor learning

PDF

Functional magnetic resonance imaging characterization of peripheral form vision

PDF

Behavioral and neural influences of interoception and alexithymia on emotional empathy in autism spectrum disorder

PDF

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

PDF

Schema architecture for language-vision interactions: a computational cognitive neuroscience model of language use

PDF

Neural and behavioral correlates of fear processing in first-time fathers

Asset Metadata

Creator Fox, Glenn R. (author)

Core Title The brain's virtuous cycle: an investigation of gratitude and good human conduct

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Neuroscience

Publication Date 11/07/2014

Defense Date 07/10/2014

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

Tag emotion,fMRI,Gratitude,Neuroscience,OAI-PMH Harvest

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Tjan, Bosco S. (committee chair), Aziz-Zadeh, Lisa (committee member), Damasio, Antonio (committee member), Damasio, Hanna (committee member)

Creator Email glennfox@usc.edu,glennrfox@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c3-515043

Unique identifier UC11297542

Identifier etd-FoxGlennR-3069.pdf (filename),usctheses-c3-515043 (legacy record id)

Legacy Identifier etd-FoxGlennR-3069.pdf

Dmrecord 515043

Document Type Dissertation

Format application/pdf (imt)

Rights Fox, Glenn R.

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