ESSAYGWIZARD
AGAIN, THE DISSERTATION ATTACHED IS THE PROFESSORS EXAMPLE THAT IS JUST FOR A FEW STRUCTURAL ELEMENTS IT DOES NOT NEED TO BE EVEN MORE THAN A HANDFUL OF PAGES.
(ESSAYGURU Id like my topic to be on making the decision of taking a new job while taking a difficult classif you can make that work, it would be perfect as the professor knows my life vs. school struggles at the moment.)
ASSIGNMENT:
The science of decision making is a young one. There is still much to discover about how/why people make the decisions they do. For this assignment I want you to perform a decision making experiment that will expand our knowledge of how people make decisions. This assignment should include:
1.) A well written description of the type of decision you are studying and what we already know about decisions of this sort, i.e. research of this type of decision.
2.) A good enough description of your experiment that I could replicate it if I wanted to.
3.) A thoughtful analysis of the results of your experiment. E.g. a graph or table that makes it clear what you found.
4.) A discussion of what your results mean and what they tell us about how people make decisions.
5.) A list of future studies that could further our understanding of Decision Making based on what you discovered or improvements to your methods.
Example:
One student did an experiment that looked at weather people at a dog park picked up the dog poo more when other people were around or if they were by themselves (the student apartment was overlooking the park). In this experiment they demonstrated the importance of social decision making and how people’s willingness to follow rules changed in different circumstances.
See my dissertation (attached below)
for general structural guidance
. My report will obviously be more in depth but hopefully you can get guidance from the
structure
and how I laid out the past research (
Introduction
), how I did the experiments (
Methods
), what I found (
Results
) and what these results mean (Disscussion and future suggestions). 1
A TALE OF TWO SYSTEMS: EXECUTIVE FUNCTION IN ULTIMATUM GAME DECISIONS
by
Aaron Daniel Kuechler Tesch
_______________
A Dissertation Submitted to the Faculty of the
DEPARTMENT OF PSYCHOLOGY
In Partial Fulfillment of the Requirements
For the Degree of
DOCTOR OF PHILOSOPHY
In the Graduate College
THE UNIVERSITY OF ARIZONA
2009
2
THE UNIVERSITY OF ARIZONA
GRADUATE COLLEGE
As members of the Dissertation Committee, we certify that we have read the dissertation
prepared by Aaron Daniel Kuechler Tesch
entitled A Tale of Two Systems: Executive function in ultimatum game decisions
and recommend that it be accepted as fulfilling the dissertation requirement for the Degree
of Doctor of Philosophy
____________________________________________________________Date: 11/11/09
Alan Sanfey
____________________________________________________________Date: 11/11/09
William Jake Jacobs
___________________________________________________________Date: 11/11/09
Anouk Scheres
___________________________________________________________Date: 11/11/09
Lee Ryan
Final approval and acceptance of this dissertation is contingent upon the candidate’s
submission of the final copies of the dissertation to the Graduate College.
I hereby certify that I have read this dissertation prepared under my direction and
recommend that it be accepted as fulfilling the dissertation requirement.
____________________________________________________________Date: 11/11/09
Dissertation Director: Alan Sanfey
3
STATEMENT BY AUTHOR
This dissertation has been submitted in partial fulfillment of requirements for an
advanced degree at the University of Arizona and is deposited in the University Library
to be made available to borrowers under rules of the Library.
Brief quotations from this dissertation are allowable without special permission, provided
that accurate acknowledgment of source is made. Requests for permission for extended
quotation from or reproduction of this manuscript is whole or in part may be granted by the
head of the major department of the Dean of the Graduate College when in his or her
judgement the proposed use of the material is in the interests of scholarship. In all other
instances, however, permission must be obtained from the author.
SIGNED: Aaron Daniel Kuechler Tesch
4
Table of Contents
Figure Index………………………………………………………………………………………………………….5
Table Index…………………………………………………………………………………………………………..6
Abstract………………………………………………………………………………………………………………..7
Introduction…………………………………………………………………………………………………………..8
Greek rationality…………………………………………………………………………………………………….9
Problems with purely rational theories………………………………………………………………………10
Emergence of non-rational theories…………………………………………………………………………..12
Two-system models…………………………………………………………………………………………………13
Neurological two-system models………………………………………………………………………………14
Possible Problems with Two-system models………………………………………………………………18
The Somatic Marker Hypothesis………………………………………………………………………………19
Can low-level systems be separated from high-level cognitive systems?……………………….24
A two-system model for second player ultimatum game decisions………………………………..25
Experiment 1 – Working Memory Easy Study…………………………………………………………….27
Experiment 1 – Methods………………………………………………………………………………………28
Experiment 1 – Results………………………………………………………………………………………..30
Experiment 1 – Discussion…………………………………………………………………………………..32
Experiment 2 – Working Memory Hard Study……………………………………………………………34
Experiment 2 – Methods………………………………………………………………………………………35
Experiment 2 – Results………………………………………………………………………………………..36
Experiment 2 – Discussion…………………………………………………………………………………..37
Experiment 3 – Within Subject Variable Memory Study………………………………………………39
Experiment 3 – Methods………………………………………………………………………………………41
Experiment 3 – Results………………………………………………………………………………………..48
Experiment 3 – Discussion……………………………………………………………………………………61
Experiment 4 – Incongruent (Incongruent) Feedback Manipulation Study……………………..65
Experiment 4 – Methods………………………………………………………………………………………66
Experiment 4 – Results………………………………………………………………………………………..67
Experiment 4 – Discussion……………………………………………………………………………………71
Experiment 5 – Priming vs. Loading Study…………………………………………………………………72
Experiment 5 – Methods……………………………………………………………………………………….73
Experiment 5 – Results…………………………………………………………………………………………75
Experiment 5 – Discussion…………………………………………………………………………………….85
Overall Results………………………………………………………………………………………………………..88
Overall Discussion…………………………………………………………………………………………………..89
Appendix A…………………………………………………………………………………………………………….97
References………………………………………………………………………………………………………………98
5
Figure Index
Figure-1. Time line for Experiment 1 & 2………………………………………………………………….31
Figure-2. The difference between Easy Working Memory and Control conditions………….31
Figure-3. Correlation of ultimatum game acceptance rates and working
memory task error rates…………………………………………………………………………………………..36
Figure-4. Memory pattern entry key………………………………………………………………………….42
Figure-5. Experiment 3 timeline……………………………………………………………………………….43
Figure-6. N-back instructions…………………………………………………………………………………..44
Figure-7. Example Global-Local Stimuli…………………………………………………………………..46
Figure-8. Memory ability vs. memory difficulty in Experiment 3…………………………………48
Figure-9. Ultimatum game acceptance rates vs. feedback condition………………………………50
Figure-10. Ultimatum game acceptance rates vs memory difficulty level……………………….51
Figure-11. The relationship between perceived memory difficulty and
acceptance rates………………………………………………………………………………………………………52
Figure-12. The relationship between performance on N-back tasks and ultimatum game acceptance
rates……………………………………………………………………………………………………………………….56
Figure-13. The relationship between visual digit span performance and
acceptance rate in the Correct Feedback condition……………………………………………………….56
Figure-14. Order effects for Experiment one and the Easy Working Memory………………….60
Figure-15. Memory performance vs. memory difficulty for the Incongruent
Feedback condition………………………………………………………………………………………………….67
Figure-16. Correlations between % correct and % accepted vs. conditions……………………..68
Figure-17. Ultimatum game acceptance rates vs. all feedback conditions……………………….70
Figure-18. Memory performance vs. memory difficulty separated by
Priming and Loading conditions………………………………………………………………………………..75
Figure-19. Bar graph of memory performance and condition vs. ultimatum
game acceptance rate………………………………………………………………………………………………..77
Figure-20. Memory performance and memory difficulty vs. ultimatum game
acceptance rates……………………………………………………………………………………………………….78
Figure-21. Memory difficulty and offer level vs. ultimatum game
acceptance rates……………………………………………………………………………………………………….80
Figure-22. Bar graph of SSRT defectors and compliers vs. ultimatum game
acceptance rates……………………………………………………………………………………………………….82
Figure-23. Ultimatum game acceptance rates compared for all conditions………………………88
6
Table Index
Table-1. Example table of % accepted and % correct for each memory
difficulty level…………………………………………………………………………………………………………54
Table-2. The correlations between executive function for the Correct
Feedback condition………………………………………………………………………………………………….57
Table-3. The correlations between executive function for the No Feedback
condition………………………………………………………………………………………………………………..58
Table-4. Priming vs. loading timeline………………………………………………………………………..74
7
Abstract
Theories that formally describe decision-making have traditionally posited that decisions are
made by rational actors. However, it is generally accepted that humans often make irrational decisions
because of quick emotional judgements. In order to reconcile these two inconsistent ideas psychologists
have developed two-system theories that hypothesize decisions are made by two opposing cognitive
systems, representing the rational and emotional processing of decisions. Evidence for a two-system
model of decision-making can be observed in ultimatum game responder decisions. It is thought that
rational processing of these choices will produce acceptance of unfair offers and emotional processing
will encourage rejection of unfair offers. Emotional priming has been shown to decrease ultimatum
game acceptances and trans-cranial magnetic stimulation of rational brain areas, i.e. DLPFC, show
increases in ultimatum game acceptances. This study investigated the possibility of using behavioral
tasks that are known to activate rational brain areas to promote/disrupt ultimatum game acceptances.
The possible relationship between ultimatum game acceptances and executive functions was also
examined. Although there were promising indications that working memory loading may increase
ultimatum game acceptances in between-subject experiments, a within-subject investigation found little
support for this method of promoting/disrupting rational ultimatum game decisions. There were also no
relationships found between switching or inhibition executive functions and ultimatum game responder
decisions. A moderate positive relationship was found between updating executive function and
ultimatum game acceptance rates but this relationship was dependent on working memory task
feedback, a within-subject design and active loading of the working memory system. However, its
possible that these findings only apply to within-subject paradigms and future between-subject studies
are advised.
8
Introduction
Classical economic theory posits that the universe followed rational laws and that human
thought should therefore be rational. However, this premise is challenged every day by the irrationality
and emotionality of human decisions and behavior. One common explanation for this imbalance
between what we should do and what we actually do is to propose that we have within us two
diametrically opposed processes. One process behaves rationally and the other behaves instinctually.
This type of theory is called a two-system model theory. This paper will examine the plausibility of
promoting/disrupting the rational system with a working memory task while people are dealing with
unfairness. Decision processes that assess unfairness are responsible for many if not most of our
decisions we make every day, and thus discovering how they work should be one of the pillars of
decision-making theory.
9
Greek rationality
Socrates believed that all behavior is directed by a rational pursuit of a person’s goal. Or more
elegantly said, “nobody acts against his [or her] better knowledge” (Frede, & Striker, 1996 p. 7).
Socrates’ intellectual children, Plato and Aristotle later modified this thinking to include elements of the
soul that were free to be irrational. These later theories can be thought of as the first two-system
models. Plato believed in a purely rational spirit world where our souls inter-mingled with pure ideas,
i.e. Forms, only to be corrupted by the non-spirit world. For example, in Plato’s story of Meno, a slave
boy spontaneously discovers a geometric principle when pressed by Socrates (Klein, 1965). Plato
makes the argument that this geometric principle as well as principles like virtue, are imbedded in us
and we only need to be reminded for our souls to recall their pure spirit world “Form”. Plato believed
that this spirit world is absolutely analytical where pure rationality, i.e., mathematics, reigns supreme.
So Plato would explain non-rational behavior as part of the mental corruption where the physical world
has strayed from true decisions. Later philosophical traditions like stoicism also adopted Socrates’
belief in a purely rational universe and thus postulated that to find truth they had to purge any physical
world corruption, i.e., emotion, from their thinking (Hadas, 1961). While these different Greek
philosophies differ on the extent that people act rationally they all agree that pure analytic rationality
represents the ideal of how we should make decisions. This belief continues to be at the core of all
normative economic theories.
10
Problems with purely rational theories
The belief in the purity and trueness of rational decisions lead to analytical, i.e., normative,
models of decision-making. In these models, people are thought to make decisions by calculating the
value of the possible outcomes and multiplying by the probability of each outcome and then rationally
choosing the option with the highest calculated value. For example, if someone had to pick between an
option of getting $1 or a 50% chance of getting $3 they would always pick the second option because it
has a calculated expected value of $1.50, which is higher than the $1 expected in the first option.
A major challenge to this purely rational theory of economic behavior was made by Daniel
Bernoulli (1738/1954) who pointed out that people do not fully equate money and value. He observed
that when given the chance to buy an option with infinite expected value, i.e., the amount to win
multiplied by the probability of winning is infinite, people often refuse. If people used purely rational
calculations of what option has the highest expected value and then choose the option with the highest
value this should never happen. To demonstrate the disconnect between what people actually decide
and what pure rationalist calculations would predict, Bernoulli described the St. Petersburg paradox. In
this paradox the reader is presented with a bet that would give a dollar, or an equivalent form of
currency, if a flip of a coin came up heads and for every subsequent heads, would be given an
additional 2n-1 dollars (where n is the number of heads) until the coin came up tails. So if the coin only
came up heads once you would get one dollar; if it came up heads two times you would get 1+2 2-1
dollars; if it came up three times you would get 1 + 2 2-1(2)+ 2 3-1 (4)… dollars and this pattern would
continue until the coin came up tails. This bet has an infinite calculated expected value because each
possible coin flip has a calculated expected value of a half dollar, and since there are endless possible
coin flips the total expected value is calculated to be infinite. However, it has been observed that
contrary to the prediction of the calculated expected value, people do not put an infinite value on this
coin-flipping gamble since they are rarely willing to place a value of over $25 (Martin, 2004).
11
Bernoulli explained that the reason people are unwilling to pay very much for this bet is that money is
not treated as an exact measure of value. Money can only buy so much happiness, i.e., value, so having
an infinite amount of money would not mean an infinite amount of happiness.
The unwillingness to pay for a bet with an infinite expected value demonstrates that contrary to
normative economic theory people do not treat money as an exact measure of value. Bernoullis theory
was formalized into new normative decision-making theories that posit a single information processing
system that computes decisions by logical analysis of all available information (Von Neumann &
Morgenstern, 1944). These theories are described as utility theories because they posit that people will
calculate the expected utility, i.e., how much it is worth to them, of the possible outcomes of a decision
and choose the option with the highest expected utility.
Utility theory has been challenged by the insight that humans have a limited ability to calculate
expected utility. For example, Simon (1956) pointed out that strict utility maximization couldnt be
used to describe actual decision-making behavior because humans have a limited ability to process all
of the factors that would be needed to fully evaluate all decision options. In other words the limited
capacity of human memory and cognitive processing power makes a strict utility calculation
impossible. To solve this conundrum it has been suggested that people avoid endless utility calculations
by using cognitive short cuts called heuristics (Gigerenzer, Todd & The ABC Research Group, 1999).
Tversky and Kahneman’s (1974) Prospect Theory also challenged the idea that people make decisions
based on strict utility calculation by suggesting that psychological weighting of probabilities and value
do not conform to utility theories. These challenges have weakened the dominance of economic
theories that postulate the strict rational calculation of utility to explain decisions. Nevertheless, strict
rationalist utility theories are often still lauded as the best way to predict behavior (Camerer, 2009).
12
Emergence of non-rational theories
The dominance of the theories that assumed rationality in our behavior and thinking was
probably most famously disputed by Sigmund Freud. Freud challenged the convention of assuming
rational decisions when he divided the mind into the canonical Id, Ego, and Superego (Freud, 1949).
This theory recognized that much of our behavior is directed by primitive non-rational, i.e., emotional,
motivations as well as the rational facets of the mind. This theory conflicted with to the philosophy of
stoic restraint that was promoted as the ideal for all behavior in the Victorian society of Freud’s time
(Gay, 1998). It should be noted that Freud’s reasoning relied mainly on distorted descriptions of case
studies and narcissistic self-reflection. However, it should also be noted that more methodological
psychologists such as William James had also considered primitive emotional instincts as important
motivators of behavior (Eysenck, 1986).
The idea that behavior was influenced by non-rational facets of the mind was exploited by
advertising pioneers like Walter Dill Scott. For example, in Scott’s advertising manifesto “The Theory
and Practice of Advertising” he details how advertisements should appeal to the non-rational
motivations of consumers (Scott, 1903). In other words, advertising should try to speak to costumers
primitive drives, i.e., sex sells.
13
Two-system models
The introduction of theories that focus on non-rational processing of decisions was an
important step towards the realization of theories that can predict behavior more accurately. However,
emotional non-rational theories like their purely rational counterparts also fail to predict decision-
making with much certainty. Therefore, in an attempt to improve these disparate theories hybrid
theories have been hypothesized. These hybrid theories postulate that decisions are made through the
competition of rational and emotional cognitive systems. The most common formulation of a hybrid
theory pits non-rational processes, i.e., emotional, against volitional, i.e., rational, processes (Posner &
Snyder, 1975; Schneider & Shiffrin, 1977; Metcalfe & Jacobs, 1996). Many versions of these
dichotomous decision-making processes have been proposed, but the general concept was well
described by Kahneman (2003), who stated that there are two evaluation systems termed System 1 and
System 2. In this conceptualization non-rational processes are represented by a System 1, which is fast,
emotional, automatic, uses heuristics, and can work in parallel with other systems. Volitional processes
are thought to be represented by a System 2, which is relatively slow, completely rational, unemotional,
can override System 1, and processes information serially. It should also be noted that some decision-
making theorists have suggested that the operation of these two systems can be better described as a
continuum of emotional and volitional processing rather than decisions being dominated by one or the
other (Cohen et al., 1990). An example of how these systems are thought to interact can be
demonstrated in the hypothetical reaction to being bumped in a hallway. An initial emotional reaction,
i.e., System 1, to being bumped might be to hit the person back, and for some people this is what
happens. However, a volitional reaction, i.e., System 2, often overrides this initial emotional reaction.
We often realize the bump may have been a mistake, and as a result we ignore it or say “excuse me”.
14
Neurological two-system models
Le Doux (1994) proposed one example of a neurological two-system model, which was based
on a theory that there are two distinct visual processes in the brain. One process, often called the high
road follows a route from the retina, through the lateral geniculate nucleus, i.e., visual thalamus, and
onto the visual cortex. The other process, often called the low road, follows a route from the retina to
the lateral geniculate nucleus then directly to emotional areas like the amygdala. hman and Soares
(1994) demonstrated the unconscious visual processing of the low road by presenting participants with
phobic stimuli, i.e., pictures of spiders, too quickly to be consciously perceived, and then detecting
emotional responses after the pictures were presented. We are most often unaware of the visual
processing that follows the low road because it bypasses conscious awareness of sight. This
unconscious emotional visual processing can be demonstrated in the classroom by placing a replica of a
bug at the entrance of the room. As people enter the room may get an emotional jolt before they realize
this bug is not real and thus does not pose a threat. This demonstration makes it clear that the bug is
being processed by two different neurological circuits. One circuit is fast, emotional, and unconscious
while the other is relatively slow and can override the emotional reaction once it is assessed that there
is no threat.
The separation of conscious and unconscious visual processing is also be supported by the
phenomenon known as blindsight. Hints that there might be sight without the visual cortex can be
demonstrated by removing the visual cortex, i.e., occipital lobe, of monkeys and then observing the
monkeys respond to visual stimuli (Kluver, 1949; Blackmore, 2004). Human neuropsychological
patients were also discovered who claimed they can not see but can also identify objects, color and
movement using visual information despite their conscious blindness (Weiskrantz, 1986). Like the
high and low road visual processing theory the existence of blindsighted individuals suggests that there
is more to vision then a single unified neurological system and implies that there are both conscious
15
and unconscious vision systems.
A similar dichotomy can be found in the expressions of the face. Duchenne (1855) discovered
humans have specialized facial muscles that activate during different emotional states. These muscles
have been linked to separate emotional and volitional systems (Damasio, 1994). For example, when a
photographer tells little kids to smile they often will produce a forced smile, but if they tell the kids a
silly joke, i.e., “say monkey butts”, the kids will produce a natural smile. Two distinct cognitive
systems seem to be responsible for facial expressions. The first system is unconscious and responds to
emotional stimuli and the second is a conscious volitional system. It should be noted that humans can
and do train their volitional expression system to override emotional expressions. However, this
volitional expression system is relatively slow so emotional expressions, i.e., micro-expressions, often
are produced before being overridden (Ekman, 2003).
Evidence for two-system models involved in decision-making has also been mounting. In
numerous neuroimaging studies brain areas whose activation seems to be correlated with both the
emotional and volitional systems have been identified. For example, McClure et al. (2004) found that
frontal brain areas ( areas) were more active than a set of limbic areas ( areas) during decisions to
delay payment to get a larger reward. However, there were no differences between the activation of
these brain areas when participants were making decisions to take smaller rewards sooner rather than
waiting for a greater reward. This study demonstrated that neural activation in brain areas associated
with rational volitional deliberation correlated with decisions to delay reward. This finding supports a
two-system decision-making theory where dichotomous cognitive systems compete to dominate
decisions.
This neurological two-system model framework may also help to explain the inconsistent
decisions participants make during delay discounting paradigms (Kirby & Marakovic, 1995). For
example, some participants would take an immediate reward ($1 now) over a delayed reward ($2 in a
16
month) but would not take a relatively short delayed reward ($1 in a month) over a long delayed reward
($2 in two months). Both of these sets of choices have the same delay and reward differences between
the options. Differences in the emotionality of decisions and therefore the relative activation of areas
may explain the decision inconsistencies observed by Kirby and Marakovic (1995). The reduced
activation of the areas when participants refuse to wait for a larger reward suggests that these choices
may result from less self control or a relatively weak System 2.
A two-system model may also help explain decisions we make about moral behavior. Greene et
al. (2001) found that when participants were making deontological decisions, i.e., concerned most with
rightness vs. wrongness, a set of brain areas associated with emotion were relatively more active then
areas associated with rationality. These deontological decisions happened most often when participants
were making personal moral judgments, i.e., being in direct contact with a person your decision would
kill. However, when participants were making utilitarian decisions, i.e., concerned with the best
outcome, areas of the brain associated with rationality and working memory were more active. These
utilitarian decisions are most common when making impersonal moral decisions, i.e., pulling a lever
that will result in a distant person’s death. These findings are supported by the observation that priming
disgust can increase deontological reactions to moral problems, because disgust activates emotional
brain areas (Schnall, Haidt, Clore, & Jordan, 2008; Sanfey et al., 2003). It has also been found that
damage to emotional processing brain areas, i.e., ventromedial prefrontal cortex, increases utilitarian
moral decisions further supporting the idea that there is an emotional component to our moral decision-
making (Koenigs et al., 2005). These studies of moral decision-making seem to demonstrate that some
moral decisions can be explained by a two-system model.
The neurological studies described above demonstrate that
