How Power and Powerlessness Corrupt
Andy J. Yap
Submitted in partial fulfillment of the requirements for the degree of
Doctor of Philosophy under the Executive Committee
of the Graduate School of Arts and Sciences
COLUMBIA UNIVERSITY 2013
© 2013 Andy J. Yap
All rights reserved
ABSTRACT
How Power and Powerlessness Corrupt
Andy J. Yap
This dissertation examines how and when, both powerfulness and powerlessness, can
each lead to corrupt behavior. The first half of this dissertation (Chapters 2 to 5) focuses on the
link between power and corrupt behavior. Building on previous work that expansive posture
induces a state of power, four studies tested whether expansive posture incidentally imposed by
our environment lead to increases in dishonest behavior. Chapters 2 to 4 present three
experiments, which found that powerful individuals were more likely to steal money, cheat on a
test, and commit traffic violations in a driving simulation. Results suggested that participants’
self-reported sense of power mediated the link between postural expansiveness and dishonesty.
In an observational field study, Chapter 5 revealed that automobiles with more expansive
driver’s seats were more likely to be illegally parked on New York City streets.
The second part of the dissertation examines if powerlessness can lead to corrupt
behavior. Chapters 6 to 10 present a new theoretical model that comprehensively integrates
theories on power and regulatory focus. This model reveals that both powerfulness and
powerlessness can each lead to corrupt behavior, but through different routes. Three experiments
in Chapters 7 to 9 found that prevention-powerlessness and promotion-powerfulness produce
more corrupt behavior than promotion-powerlessness and prevention-powerfulness, as evident in
individuals’ tendency to exploit others, aggression, and dishonest behavior. I also found evidence
for the affective manifestations that accompany these effects. Indeed, a meta-analysis on the
data suggests that prevention-powerlessness and promotion-powerfulness significantly produced
more corrupt behavior than prevention-powerfulness and promotion-powerlessness. These
findings have important theoretical implications for power and regulatory focus, and explicate
how powerlessness can lead to taking action and even corruption.
i
TABLE OF CONTENTS
Page
LIST OF TABLES iv
LIST OF FIGURES v
ACKNOWLEDGEMENTS vi
CHAPTER ONE
INTRODUCTION 1
The Ergonomics of Dishonesty 2
Powerful Postures 4
Power and Dishonest Behavior 5
The Focus of the Current Research on The Ergonomics of Dishonesty 6
CHAPTER TWO
FIELD EXPERIMENT: STEALING BY OMMISSION
Method 8
Results & Discussion 8
CHAPTER THREE
LABORATORY EXPERIMENT: CHEATING ON A TEST
Methods 10
Results & Discussion 11
CHAPTER FOUR
LABORATORY EXPERIMENT: HIT AND RUN IN A DRIVING SIMULATION
Methods 12
Results & Discussion 13
CHAPTER FIVE
OBSERVATIONAL FIELD STUDY: PARKING VIOLATIONS
Methods 14
Results & Discussion 15
ii
General Discussion On The Ergonomics of Dishonesty 16
CHAPTER SIX
THE INTEGRATIVE EFFECTS OF POWER AND REGULATORY FOCUS 19
Regulatory Focus Theory 20
Theoretical and Empirical Implications 23
Overview Of Studies 25
CHAPTER SEVEN
EXPERIMENT: EXPLOITING OTHERS
Methods 26
Results & Discussion 27
CHAPTER EIGHT
EXPERIMENT: AGGRESSION
Methods 29
Results & Discussion 30
CHAPTER NINE
EXPERIMENT: CHEATING ON A TEST
Methods 31
Results 32
Meta-analysis 32
Discussion of the Integrative Effects of Regulatory Focus and Power 33
CHAPTER TEN
GENERAL DISCUSSION AND CONCLUSIONS 34
Theoretical Contributions 34
Practical Implications 38
Conclusion 41
iii
FOOTNOTES 42
REFERENCES 44
TABLES 56
FIGURES 57
iv
LIST OF TABLES
Page
Table 1:
Table 1: Mean Level of Corrupt Behavior and Effect Sizes of Contrast 56
v
LIST OF FIGURES
Page
Figure 1: Poses employed in Field Experiment: Stealing by Omission 57
Figure 2: How the money was presented in Field Experiment: Stealing by Omission 58
Figure 3: Desk-space configurations for Laboratory Experiment: Cheating on a Test 59
Figure 4: Driver’s seat configurations for Laboratory Experiment: Driving Simulation 60
Figure 5: Laboratory Experiment: Driving Simulation: Mediation Analysis 61
Figure 6: Observational Field Study: Vehicle Size Computation 62
Figure 7: Exploiting others 63
Figure 8: Aggression 64
Figure 9: Cheating on a Test 65
Figure 10: Negative Affective States 66
vi
ACKNOWLEDGEMENTS
I am eternally indebted to a number of people who have contributed significantly to the
completion of this dissertation. I have the most amazing graduate advisors and mentors that
anyone can hope for. They are brilliant, distinguished scholars in their fields, and most
importantly, exceptionally devoted to me and my training.
I would like to thank Dana Carney for her generous support, mentorship and friendship.
Dana provided the type of support and nurturance that a fledgling academic can only dream
about. She taught me how to be brave and diligent when my research moves into uncharted
territory, and figure out which questions are important to pursue in my research.
I especially want to thank Tory Higgins for inspiring and supporting me. Tory has an
amazing ability to cultivate a sense of excitement and positivity, and an appreciation for big-
picture ideas in science. He is a model of the kind of academic I would someday like to be, and I
count myself a true honor to be his student.
I am very grateful to Adam Galinsky for his incomparable mentorship, patience, and keen
insight. Adam is the busiest and most available person at the same time. I thank him for always
making time to listen, offer advice, and inspire me with his enthusiasm to do good science.
Daniel Ames, shared with me his brilliance, diligence and grit to tackle difficult but
important questions. Daniel taught me how to effectively communicate my work with clarity and
finesse. Above all, Daniel is affable and brilliant, the perfect combination that makes him a great
mentor.
Joel Brockner taught me how to talk through problems collaboratively and keep
recursively question the answers to problems until a clear understanding materializes. Joel was
never afraid to challenge me and push my thinking while offering constant intellectual support.
vii
I would also like to thank my other faculty collaborators, particularly Malia Mason,
Michael Morris, and Leonard Lee. In their own special way, they have inspired me, supported
me, shaped my thinking, and challenged me to be a good scientist.
My sincere thanks also go to my student collaborators and friends who have enriched my
graduate school experience with fun and true friendship. I want to thank Lara Aknin, Jeremy
Yip, Geoff Ho, Sunny Kim, Kenneth Tai, Martin Schweinsberg, Zhang Shu, Abbie Wazlawek,
Brian Lucas, Liu Zhi, Claudius Hildebrand, Liza Wiley, former students of Columbia
particularly Aurelia Mok, Roy Chua, and Canny Zou, the PPIG family, and the Higgins Lab
family, for supporting me and filling my graduate school experience with fine companionship.
Finally, I would like to thank Charlene Chen for always being there for me, encouraging
me, and supporting me for the last fifteen years. And of course Chewie for his warm and fuzzy
hugs everyday throughout my time in graduate school. I also want to thank my brother and my
parents for the unconditional love and support they have continually showered on me.
1
CHAPTER 1
INTRODUCTION
Across America and the world, mainstream news abounds with accounts of powerful
people behaving badly. As business leaders, politicians, and even religious leaders seem to fall
from power and grace, the evidence seems clear—at least anecdotally—that power indeed
corrupts. However, science has only begun to empirically investigate if power really corrupts,
how and when it corrupts, and exactly what types of corrupt behavior power could lead to.
Similarly, an equally important question is: can powerlessness also corrupt?
This dissertation examines how and when, both powerfulness and powerlessness, can
each lead to corrupt behavior. In the first half of this dissertation (Chapters 2 to 5) I focus on the
link between power and corrupt behavior. Previous research has examined how power—acquired
through one’s leadership role (Kipnis, 1972; Lammers, Stoker, Jordan, Pollmann, & Stapel,
2011), semantic and experiential priming (Gruenfeld, Inesi, Magee, & Galinsky, 2008; Lammers,
Galinsky, Gordijn, & Otten, 2008), or social-economic class (Piff, Stancato, Cote, Mendoza-
Denton, & Keltner, 2012)—influences a myriad of corrupt behaviors, including the tendency to
manipulate, bully, stereotype, cheat and objectify others (Fiske, 1993; Gruenfeld, et al., 2008;
Howard, Blumstein, & Schwartz, 1986; Keltner, Capps, Kring, Young, & Heerey, 2001; Kipnis,
1972; Kraus, Piff, & Keltner, 2009). Recent research has shown that expansive nonverbal
postures can activate a psychological and a physiological trajectory resembling a powerful state
(Bohns & Wiltermuth, 2012; Carney, Cuddy, & Yap, 2010; Carney et al., 2013; Huang,
Galinsky, Gruenfeld, & Guillory, 2011). Importantly, research also shows that these expansive
postures seems to also influence a number of behavioral outcomes that are associated with
having power, such as risk-taking and action tendencies. Building on this work, this dissertation
2
investigates if expansive postures that are incidentally induced by the ergonomics of our ordinary
working and living environments can induce a state of power, and correspondingly lead to
increases in dishonest and corrupt behavior. This investigation is critical, because if expansive
postures do insidiously lead to corrupt action, it would suggest that dishonesty could be lurking
in our everyday environment, such as our cars, workstations, and offices. Importantly, it could
also provide valuable insights into how we can use our living and working spaces to mitigate
corrupt behavior.
Another important question that this dissertation aims to address is: Can powerlessness
corrupt? Most of previous research has focused on the link between powerfulness and corrupt
behavior, and have largely neglected the psychological state of powerlessness. Can
powerlessness also corrupt? If so, how and when? In the second half of this dissertation
(Chapters 6 to 10), I present and test a new theoretical model, which articulates how and when
both powerfulness and powerlessness can each lead to corrupt behavior. Specifically, I integrate
theories on power and regulatory focus (Higgins, 1997, 1998), and delineate the combinations
that would lead to corrupt acts such as the tendency to exploit others, act aggressively, and
behave dishonestly. This model also delineates the conditions when powerfulness and
powerlessness do not lead to corrupt behavior. Understanding how power interacts with
regulatory focus not only illuminates when power and powerlessness corrupt, it also provides
important insights on how we can mitigate corrupt behavior among the powerful and the
powerless.
THE ERGONOMICS OF DISHONESTY
The ergonomics and physical geography of our everyday environments are powerful.
They determine our social networks and relationships (Werner, Altman, & Brown, 1992),
3
personal and interpersonal functioning (Altman, Taylor, & Wheeler, 1971), our workplace
productivity (Knight & Haslam, 2010), and our subjective well-being (Kaplan & Kaplan, 2009;
Leonard, 2012). This dissertation examines the impact of our environment on an important social
behavior—dishonesty. Each day, our bodies are continually stretched and contracted by our
working and living environments—by the seats and levers positioned in our cars, and by the
furniture and workspaces in our homes and offices. Although we may pay very little attention to
ordinary and seemingly innocuous shifts in our bodily posture, these subtle postural shifts can
have tremendous impact on our thoughts, feelings and behavior (Damasio, 1994; Niedenthal,
2007; Niedenthal, Barsalou, Winkielman, Krauth-Gruber, & Ric, 2005). Most central to the
current research is the finding that expansive body postures lead to a psychological state of
power (Bohns & Wiltermuth, 2012; Carney, et al., 2010; Fischer, Fischer, Englich, Aydin, &
Frey, 2011; Huang, et al., 2011). And power—caused by myriad laboratory manipulations and
real-world structural features—appears to be linked to increases in a wide range of dishonest
behaviors (Boles, Croson, & Murnighan, 2000; Guinote, 2007; Keltner, Gruenfeld, & Anderson,
2003; Lammers, Stapel, & Galinsky, 2010; Lammers, et al., 2011). Is it possible that expansive
postures incidentally shaped by our environment could lead to dishonest behavior? This question
was the focus of the research in the first half of this dissertation.
The idea that the human body has the ability to shape the mind has piqued the interest of
scholars for centuries. Darwin (1872/1904) and the father of experimental psychology, William
James (1884), were among the first to theorize about mind-body connections. But it wasn’t until
the 1970s that the bi-directional connection between bodily displays and psychological states
was empirically demonstrated (Duclos et al., 1989; Laird, 1974; Rhodewalt & Comer, 1979;
Riskind, 1983; Riskind & Gotay, 1982; Strack, Martin, & Stepper, 1988; Wells & Petty, 1980).
4
For example, Laird (1974) hooked participants up to facial EMG and asked them to “furrow
eyebrows” (i.e., frown) or “clench teeth” (i.e., smile). When participants’ teeth were clenched
they reported more happiness and humor. Strack, Martin and Stepper (1988) later replicated and
extended this work. Similarly, Wells and Petty (1980) demonstrated that participants who
nodded their heads (in an agreement motion) while listening to messages found the messages to
be more persuasive than those who shook their heads (in a disagreement motion).
POWERFUL POSTURES
Across humans and animals, power and dominance are expressed through expansive,
open-bodied postures (spreading out and occupying more space), whereas powerlessness and
subordination are expressed through relatively more contractive, closed-bodied postures (Carney,
Hall, & LeBeau, 2005; Darwin, 1872/1904; de Waal, 1998; Ellyson & Dovidio, 1985; Hall,
Coats, & Smith LeBeau, 2005; Tiedens & Fragale, 2003; Weisfeld & Beresford, 1982). Research
also shows these expansive nonverbal “power poses” may activate mental concepts and feelings
associated with power and may go so far as to initiate a physiological trajectory resembling a
powerful state (Bohns & Wiltermuth, 2012; Carney, et al., 2010; Carney, et al., 2013; Huang, et
al., 2011). These findings demonstrated that when men and women engaged in expansive (vs.
contractive) postures, they felt more powerful, became more approach-oriented and risk-seeking,
and appeared to evidence a physiological pain and stress-buffer. Similarly, Riskind and Gotay
(1982) demonstrated that slumped and constricted postures induced a state of learned
helplessness and feelings of stress (versus upright/confident postures). Finally, Harmon-Jones
and Peterson (2009) found that supine (i.e., lying down) versus upright body posture reduces
approach motivation.
5
POWER AND DISHONEST BEHAVIOR
Regardless of how power is manipulated or observed in the lab or field, power is
consistently related to dishonesty. For example, power is associated with cheating to improve
odds-of-winning (Lammers, et al., 2010), lying (Boles, et al., 2000), lying more easily (Carney,
et al., 2013), hypocrisy (Lammers, et al., 2010), and infidelity (Lammers, et al., 2011).
According to Keltner, Gruenfeld and Anderson (2003), power activates the Behavioral Approach
System, which causes powerful individuals to focus on rewards and act on their own self-
interests and goals (Galinsky, Gruenfeld, & Magee, 2003; Guinote, 2007; Inesi, 2010). Power
also leads to overconfidence (Fast, Sivanathan, Mayer, & Galinsky, 2012), which makes power-
holders more likely to gamble and take risks (Anderson & Galinsky, 2006; Carney, et al., 2010).
Overall, these studies offer evidence that a psychological state of power increases the likelihood
that individuals would focus on their own desires and ignore the consequences for others.
Similarly, they suggest the powerful individuals could have a heightened tendency to behave
dishonestly and not be deterred by the risk of getting caught because they are thoroughly focused
on achieving their goals.
More recent research also proposes that the link between power and corrupt behavior
could have a physiological basis. Carney et al. (2013) offered the first evidence that power could
lead to corruption by buffering powerful individuals from the psychological and physiological
stress associated corrupt acts. Corrupt behaviors, such as lying, can be physiologically and
psychologically stressful. In an experiment modeled after a mock crime paradigm, they had
participants either behave honestly or steal $100 and then interrogated them. They found that
power buffered the stress experienced during the videotaped interrogation about the theft they
had just committed. High-power liars were able to lie about the theft with ease—their power
6
inoculated them from the emotional, cognitive, nonverbal and physiological stress of engaging in
deception. In fact, the high-power liars experienced as little stress as the truth-tellers in their
study. By contrast, low-power liars were the epitome of stress.
Another physiological response that is related to having power and corruption, is elevated
levels of the dominance hormone testosterone (Mazur, 1976). High levels of testosterone has
been linked to antisocial and adversarial behaviors (Dabbs & Morris, 1990; Mazur & Booth,
1998), violent crimes (Dabbs, Carr, Frady, & Riad, 1995), endorsing murder (Carney & Mason,
2010), financial risk-taking (Apicella et al., 2008; Coates & Herbert, 2008), and aggression
(Mehta & Beer, 2010). All these studies suggest that power confers to its bearer immunity from
stress, coupled with a motivation to contend, take risks and focus on one’s desires.
If expansive postures can lead to a state of power, and power can lead to dishonest
behavior, this suggests something of real concern—the ordinary expanded (vs. contracted)
nonverbal postures forced upon us by our environments, which we happen or choose, could
impact our decisions and actions in ways that render us less (or more) honest.
THE FOCUS OF THE CURRENT RESEARCH ON THE ERGONOMICS OF
DISHONESTY
Chapters 2 to 5 tested the hypothesis that expansive postures would lead to dishonest
behaviors in four studies conducted in the field and the laboratory. Chapter 2 presents a field
experiment that examined whether expansive (vs. contractive) postures, as employed in previous
research (Bohns & Wiltermuth, 2012; Carney, et al., 2010; Huang, et al., 2011), would lead to
stealing in an “overpayment” paradigm. Chapter 3 presents a laboratory experiment that
manipulated the expansiveness of workspaces in the lab and tested whether incidentally
7
expanded bodies (shaped organically by one’s environment) led to more dishonesty on a test.
Chapter 4 presents another laboratory experiment that examined if participants in a more
expansive driver’s seat would be more likely to “hit and run” when incentivized to go fast in a
video-game driving simulation. I also tested the mediating role of sense of power in these effects.
Finally, to extend results to a real-world context, Chapter 5 presents an observational field study
tested the ecological validity of the effect by examining whether automobile drivers’ seat size
predicted the violation of parking laws in New York City. Consistent with recommendations
from Simmons, Nelson, and Simonsohn (2012), I report how we determined our sample size, all
data exclusions, all manipulations, and all measures in the studies.
8
CHAPTER 2
FIELD EXPERIMENT: STEALING BY OMMISSION
Method
Participants and Procedure. Eighty-eight1 (31 women) community members were recruited
from South Station in Boston, MA and outside a library at the city campus of Columbia
University to participate in a study that ostensibly examined the relationship between stretching
and impression formation. Participants were told they would receive $4 for participation.
Postural expansiveness was manipulated using a procedure similar to Carney et al. (2010). Here,
I used a cover story about the effects of stretching on impression formation, participants were
randomly assigned to hold either an expansive or a contractive pose (Figure 1) for one minute
while they formed impressions of faces shown to them by the experimenter. Next, in order to
bolster the cover story, participants indicated their impressions of a best friend2. Finally, though
participants believed they would receive $4 payment as they were initially told, the experimenter
handed them $8, which was comprised of $1, $1, $5, and $1 bills, fanned out (Figure 2) and
presented such that participants noticed the “accidental” overpayment. The dependent measure
was whether or not the participant kept the overpayment. The experimenter coded for whether
participants checked the money after they had received it.3
Results and Discussion
Consistent with my theorizing, a χ2 analysis found that participants who performed the
expansive pose were significantly more likely to keep the overpayment (i.e. “steal by omission”),
χ2(1, N = 78) = 13.0, p < .001, Φ = .41. Seventy-eight percent of the expanded-posture
participants kept the overpayment, compared to 38% of contracted-posture participants.
9
This experiment found initial evidence that expansive postures can lead to dishonest
behavior. Participants in this study were explicitly instructed to assume a specific pose, yet the
main focus of the current research is posture imposed by the ergonomics of the environment.
With this first experiment establishing the link between posture and dishonesty, Chapters 3-5
investigated the impact of incidentally induced expansive (vs. contractive) postures on dishonest
behavior. Participants in these studies were not explicitly instructed to assume specific poses, nor
were they made explicitly aware that their posture was being manipulated. Instead, posture was
naturally shaped by ordinary chairs and workspaces. This offers a key methodological
contribution to the research on embodied power because this would suggest that mere
expansiveness of one’s posture can adequately induce a psychological state of power and do not
require the specific postural configurations that were employed in previous studies (Bohns &
Wiltermuth, 2012; Carney, et al., 2010; Fischer, et al., 2011; Huang, et al., 2011).
10
CHAPTER 3
LABORATORY EXPERIMENT: CHEATING ON A TEST
Method
Participants and Procedure. Thirty-four university students4 (20 women) from Columbia
University participated in a study for monetary compensation that supposedly examined how
Feng Shui influences creativity.
Participants worked in individual cubicles at a desk set up with either a large (24” by 38”)
or a small (12” by 19”) desk pad (Figure 3). Participants saw only their own workspace and not
that of other participants. They were then instructed to complete two creativity tasks.
First was an anagram test on which unbeknownst to participants, they would later have an
opportunity to cheat. This cheating paradigm was adopted from Ruedy and Schweitzer (2010).
Participants received a packet of materials contained in a manila folder and were allotted four
minutes to unscramble 15 anagrams that were printed on the first page. They were incentivized
by the experimenter’s promise of one dollar for every anagram solved. When time was called,
participants were instructed to detach and retain the first page and return the folder and its
remaining contents to the experimenter. Participants were unaware that an imprint of their test
answers were created by a sheet of carbonless copy paper hidden at the back of the folder.
Incidental posture was manipulated in the next task, which ostensibly measured inductive
creativity. Participants were allotted seven minutes to create a collage using materials that were
placed around the edges of their desk pad. During the task, participants were only allowed to use
the space on the desk pad. Posture was incidentally manipulated by the size of participants’ desk
pads. The large desk pad arrangement forced participants to stretch and reach for materials, thus
incidentally imposing expansive postures. These participants also had chairs that were high
11
enough to help them reach for the materials. In contrast, the small desk pad arrangement
constrained participants arm extensions as materials were within close reach, thus incidentally
imposing contractive postures.
At the completion of the collage task, the experimenter, appearing very busy, rushed to
each cubicle and handed participants the answer key for the anagram test. The experimenter
explained that he had to manage another study in the adjacent lab and asked that the participant
grade his/her own test. Participants were thus given an opportunity to alter their original answers
in private. I used participants’ number of altered answers as a measure of cheating, which we
identified by comparing their self-graded test to the carbon copy containing their original
answers.
Results and Discussion
I hypothesized that expanded-posture participants would alter more answers, which
would earn them more bonus money. As predicted, a one-way ANOVA revealed that expanded-
posture participants altered more answers (M = 1.20, SD = 1.70) than contracted-posture
participants (M = .27, SD = .59), F(1, 29) = 4.04, p = .05, d = 0.73.5
The experiments in Chapters 2 and 3 found consistent evidence that power induced
through expansive postures, whether posed or incidentally imposed, lead to more dishonest
behavior. In the third experiment, I examined whether drivers’ seat expansiveness can lead to
more traffic violations in a driving simulation. Importantly, I also tested the mechanism of this
effect. If expansive posture leads to a state of power, and power leads to increases in dishonest
behavior, then the link between expansive posture and dishonest behavior should be mediated by
participants’ sense of power.
12
CHAPTER 4
LABORATORY EXPERIMENT: HIT AND RUN IN A DRIVING SIMULATION
Method
Participants and Procedure. Seventy-one students6 (48 women) from the University of
California, Berkeley were recruited to participate in a study ostensibly about physiology and
video games. A realistic driving simulator was set up with a Playstation 3 and a Logitech driving
force GT racing wheel, which included a steering wheel and foot-pedals. Participants were
randomly assigned to sit in an expansive or contractive driver’s seat (See Figure 4 for visual
display of the setup) 7. Participants played the game “Need for Speed: Hot Pursuit,” which
challenges players to race to the finishing line as fast as possible. Participants were allotted one
initial practice race to become accustomed to the game controls. They were then offered a chance
to win $10 if they could complete the same race within five minutes. Importantly, we
implemented a rule that participants must stop and count-to-ten after a collision with any object
in the race. Violation of this rule would shorten participants’ total race time, and thus help them
to win the bonus money. Rule violation, specifically the number of times a participant hit an
object and did not stop, served as my measure of cheating. Races were video recorded and coded
by two research assistants for the number of “hit and runs.” Inter-rater reliability was determined
by having the two coders rate the same subset of videos (10%). Once inter-rater reliability was
established (r = .95), the remaining videos were divided equally between coders. After the race,
participants reported how powerful they felt on a 1 (Not at all) to 5 (Extremely) Likert-type
scale.
13
Results and Discussion
Consistent with my theorizing, being seated in an expansive seat lead participants to drive
somewhat more recklessly (Mean number of objects hit = 7.11, SD = 8.51) than being seated in a
contractive seat (M = 4.33, SD = 3.60), F(1,67) = 3.02, p = .087. Importantly, participants in the
expansive seat (M = 6.31, SD = 8.45) were more likely to “hit and run” than those in the
contractive seat (M = 2.94, SD = 2.61) after controlling for the number of objects hit, F(1, 66) =
4.12, p = .046, d = .54. The effect was significant when number of objects hit was not included
as a covariate, F(1, 67) = 4.81, p = .032. 8
I also predicted that participants’ sense of power would mediate this effect. Bootstrapping
analyses (Preacher & Hayes, 2004) based on 5000 bootstraps were conducted for estimating
direct and indirect effects. The total effect of expansive posture on incidence of hit and run (total
effect = 3.37, p = .03) became nonsignificant when sense of power was included in the model
(direct effect of expansive posture = 2.65, p = .09). Additionally, the total indirect effect (i.e., the
difference between the total and direct effects) of expansive posture on hit and run through sense
of power was significant (point estimate = .72, bias-corrected bootstrap 95% confidence interval
.0197, 2.775)—zero fell outside this interval, indicating a significant mediation effect 9 (Figure
5).
Three experiments found consistent evidence that expansive posture, whether posed, or
shaped incidentally by one’s desk space or driver’s seat can lead to dishonest behavior. While the
emergence of these effects in the lab may be intriguing, to understand their generalizability and
pervasiveness, I examined whether the same pattern of results would occur naturally in the real
world in Chapter 5.
14
CHAPTER 5
OBSERVATIONAL FIELD STUDY: PARKING VIOLATIONS
With three lab experiments in hand, I thought it was critical to test the real world
generalizability of the incidental posture effect. Thus, the study in Chapter 5 used observational
field-study methods, to investigate whether drivers in expansive automobile seats were more
likely to commit parking violations, an established measure of corrupt behavior in the economics
literature (Fisman & Miguel, 2007). Specifically, I focused on double-parking—the parking of a
car in an open lane such that adjacent vehicles are blocked in and active driving space is partially
obstructed, which forces other drivers to maneuver through tighter spaces.
Method
Participants and Design. Two hypothesis-blind research assistants recorded instances of
double-parking on East-West streets between 116th and 102nd street in New York City from
12pm to 7pm on weekdays. The research assistants recorded information about each double-
parked vehicle as well as information about the legally parked adjacent vehicle (in the event that
more than one legally parked vehicle was blocked in by the double-parked vehicle, information
about the legally parked vehicle that overlapped most with the double parked car was recorded).
The legally parked vehicles served as our control sample. A total of 126 automobiles were
recorded.
Measure of driver’s seat (space) size. As an index of the expansiveness of the each automobile’s
driver’s seat, we calculated the volume of the space using information posted on respective car
manufacturers’ websites. Volume was computed by halving the product of the wheelbase (length
between the front wheels and the back wheels), height, and width of the car (Figure 6).
15
Measure of status of automobile brand. Because social status has been found to predict
unethical behavior (Piff, et al., 2012), we controlled for the status of vehicle brands by including
it as a covariate in our analyses. To create an index of status, we did a stimulus-rating study of
each of the observed vehicle brands (participants were N = 95 Americans). The status of each
vehicle brand was rated using a scale of 1 (Extremely low status) to 7 (Extremely high status).
Responses were averaged to form a measure of vehicle status for each specific brand.
Results and Discussion
Consistent with theorizing, a binary logistic regression controlling for status10 of cars
indicated that vehicles with larger driver’s seats were more likely to be double-parked (B = .020,
SE = .005, p < .001). For a standard deviation increase in driver’s seats size from the mean, the
probability that the vehicle would be double-parked increases from 51% to 71%.
To account for the fact that drivers of lengthy cars might be more likely to double park
due to increased difficulty of finding large enough parking spots in a congested city like New
York, we controlled for status and car length in another regression. The relationship remained
marginally significant (B = .015, SE = .009, p = .087) despite the fact that length was very
strongly correlated with driver’s seat size (r = .83, p < .001).
Although the results of this study provide some insights on the ecological validity of this
phenomenon, the methodology has clear limitations (as is often the case with observational
work). For example, we were unable to ascertain driver demographics, such as gender or body
size, and drivers could not be randomly assigned to the conditions. Without professional
appraisal of each car in our sample, we were also unable to accurately determine present value.
However, when taken together with the 3 experiments, the package offers a more complete
16
picture. Importantly, the experiment in Chapter 4 offsets the limitations of this observational
study because in the Experiment, participants were (1) randomly assigned to expansive or
contractive driver’s seats, and (2) vehicle attributes like length and price were not an issue
because expansiveness (vs. contractiveness) of driver’s seat was the only variable manipulated
across conditions.
GENERAL DISCUSSION ON THE ERGONOMICS OF DISHONESTY
Together, these four studies provide multi-method evidence from both the lab and field
that power through expansive postures, whether posed or incidentally shaped by our environment
can lead to dishonesty. The first three studies provided consistent evidence for the causal
relationship between postural expansiveness and dishonest behavior. The use of different
participant populations and real-world parking data suggest the external and ecological validity
of this effect.
While researchers in design and human factors (Stokols, 1978; Werner, et al., 1992)
would not be surprised with our findings, very little research in psychology has ventured into the
domain of ergonomics and social behavior. The current research may suggest that dishonesty
could be lurking in our ordinary, everyday environments—such as our cars, workstations, and
offices. Our bodies are perpetually enslaved by the structure of our physical spaces, and the
current findings suggest that when our bodily postures are incidentally expanded by these spaces,
we could be lured into behaving dishonestly. These studies demonstrate that the mere
expansiveness of one’s posture can induce a state of power and do not require the specific
postural configurations that were employed in previous studies (Bohns & Wiltermuth, 2012;
Carney, et al., 2010; Fischer, et al., 2011; Huang, et al., 2011). These findings also challenge the
traditional theorizing that power has to be socially and structurally acquired (French, 1956;
17
French & Raven, 1959). They suggest that one’s sense of power can be fleeting, dynamic, and
self-induced simply by spreading out and taking up more space.
One prescriptive point that could be offered from this work is that we may need to
consider the science of ethics more holistically—taking into consideration not only the
sometimes toxic effect of power itself, but also the nefarious impact of incentivizing the wrong
things. Finally, the very ways in which offices and furniture are designed also need examination
and consideration. Future research could explore ways in which we could capitalize on even the
simplest features of our physical environments toward the goal of promoting ethical, prosocial,
and healthy workplace behaviors.
Does power always lead to nefarious outcomes?
The current research proposes that expansive postures could lead to corrupt behavior, but
other research show that these postures could also produce beneficial effects like resilience from
pain and stress, and bolster executive functioning much like the research on social power has
shown (Bohns & Wiltermuth, 2012; Carney, et al., 2013; Smith, Jostmann, Galinsky, & van
Dijk, 2008). The theoretical argument in Carney et al. (2013) is one in which power renders a
physiological system more willing and able to engage with all acts—whether honest or
dishonest. Consistent with this idea, power does seem to promote ethical and socially responsible
behaviors under certain conditions (e.g. Chen, Lee-Chai, & Bargh, 2001; DeCelles, DeRue,
Margolis, & Ceranic, 2012). How do we reconcile these differences? There are some additionally
useful theoretical ideas to consider.
Hirsh, Galinsky and Zhong (2011) proposed that power could be a catalyst that reveals
the person. Recent research has also found that power enhances moral awareness among
individuals with high moral identity, but decreases moral awareness in those with low moral
18
identity (DeCelles, et al., 2012). Similarly, individuals with a communal relationship-orientation
are more socially responsible than those with an exchange relationship-orientation, because
power amplifies the dominant dispositional cues (Chen, et al., 2001).
Power can also shape the person by amplifying the dominant situational cue (Hirsh, et al.,
2011). Powerful individuals tend to focus on any contextually activated goals (Guinote, 2007).
They are more likely to cheat and take risks when the rewards are attractive like those in the
current research (Anderson & Galinsky, 2006; Galinsky, Magee, Inesi, & Gruenfeld, 2006; Inesi,
2010; Lammers, et al., 2010). However, when the most dominant contextual cue is to be
cooperative, power would correspondingly promote more other-focused behaviors and less self-
interested behaviors (Handgraaf, Van Dijk, Vermunt, Wilke, & De Dreu, 2008). Therefore, it
seems that although power and expansive posture could lead to self-focused and dishonest
behaviors, they do also lead to prosocial and socially responsible outcomes if the situational cues
for such goals are salient.
The next half of this dissertation shifts the focus from the powerful to the powerless.
Specifically, I examined if powerlessness can also lead to corrupt behavior. In addition, I
investigated the conditions when powerfulness does not corrupt. Toward this goal, I took a
holistic and integrative approach in addressing these questions, and examined if powerfulness
and powerlessness would interact with motivational states, particularly regulatory focus
(Higgins, 1997, 1998), and investigated its interactive effects on corrupt behavior. This approach
allowed me to articulate when both powerfulness and powerlessness can each lead to corrupt
action, and when corrupt behavior is attenuated.
19
CHAPTER 6
THE INTEGRATIVE EFFECTS OF POWER AND REGULATORY FOCUS
Can powerlessness corrupt? Although past research and the studies in Chapters 2 to 5
demonstrate that there isn’t a link between powerlessness and corrupt behavior, there are several
real world examples that allude to the possibility that powerlessness can indeed lead to nefarious
behaviors. One example is the recent “occupy” demonstrations that had swept across America
and several parts of the world. Although the protest demonstrations by ordinary civilians,
students, lower/middle-class citizens, and homeless people were for a worthy cause, there were
numerous cases of crime, anti-social behavior, and blatant disregard for civil law committed by
these protestors (Buckley & Flegenheimer, 2011). Crime is also prevalent in shantytowns and
poor communities. Organizational research also show that lower-ranked employees who are
unfairly treated also have a tendency to steal from the organization (Greenberg, 2002; Greenberg
& Scott, 1996). Overall, these examples suggest that being in a state of powerlessness, including
low SES, or low hierarchical rank, could sometimes propel one to behave unethically. Thus, can
powerlessness corrupt? And if so, when does powerlessness corrupt? Most of past research has
mainly examined the relationship between powerfulness and corrupt behavior, but have
neglected the psychological state of powerlessness, and the possibility that powerlessness can
lead to corrupt behavior.
Just as powerlessness can sometimes corrupt, powerfulness can sometimes not corrupt.
For example, non-human alpha primates care and act benevolently towards lower-power
members of their colony. In humans, some research suggest that the powerful are much more
socially and emotionally sensitive than the less powerful (Hall, Rosip, Smith LeBeau, Horgan, &
Carter, 2006). This is consistent with the notion of “Noblesse Oblige”, which literally means that
20
the more powerful have the obligation to help others—especially those less powerful. To this
end, a critical question to also consider is when does power corrupt? And when does power not
corrupt?
In this half of the dissertation (Chapters 6 to 10), I present evidence for a new theoretical
model that articulates how and when both powerfulness and powerlessness can each lead to
corrupt behavior. Specifically, I integrate theories on power and regulatory focus (Higgins, 1997,
1998), and delineate the combinations that would lead to corrupt acts such as the tendency to
exploit others, acting aggressively, and dishonest behavior. Understanding how power interacts
with regulatory focus not only elucidates when power and powerlessness can lead to corrupt acts,
it also provides important insights on how we can curtail corruption among those with and
without power.
REGULATORY FOCUS THEORY
Regulatory focus theory postulates two motivational orientations: promotion and
prevention (Higgins, 1997, 1998), which determines how goals are being framed and pursued.
Although regulatory focus theory played a role in shaping the leading theory on power (Keltner,
et al., 2003), it has never been fully investigated together with power. According to this theory,
when individuals are in a promotion focus, they are concerned with growth and advancement.
They view their goals as ideals (hopes and aspirations) and focus on achieving gains and positive
outcomes. Individuals with a promotion focus would usually seek the desired gains (i.e. “+1”) by
employing eager and risky strategies (Crowe & Higgins, 1997; Liberman, Molden, Idson, &
Higgins, 2001). In contrast, when individuals are in a prevention focus, they are concerned with
safety and security (i.e. “0”). They perceive their goals as oughts (duties and obligations), and
focus on preventing losses and negative states (i.e. “-1”). Individuals with a prevention focus are
21
especially sensitive to negative states and seek to avoid these states by employing vigilant
strategies. Regulatory focus can be chronic (personality variable) or momentary (induced by
situations).
Prevention Focus and Powerlessness
A prevention focus makes individuals especially sensitive to negative states, and
motivates them to avoid these states. Powerlessness is an aversive negative-valence (“-1”) state,
which could propel individuals to escape from it (Keltner, et al., 2003; Rucker & Galinsky,
2008). When prevention individuals feel powerless, I postulate that they would be motivated to
regulate this “-1” state and return to safety “0”. This is particularly true for prevention
individuals and not promotion individuals because a prevention focus makes the experience of a
negative-valence state, such as powerlessness, intensely unpleasant (Scholer, Stroessner, &
Higgins, 2008; Scholer, Zou, Fujita, Stroessner, & Higgins, 2010). This is consistent with past
research showing that when prevention individuals experience a negative state, like a sense of
loss, they would experience negative affect marked by feelings of agitation (Idson, Liberman, &
Higgins, 2000, 2004; Molden, Lee, & Higgins, 2008). Thus, I predict that when prevention
individuals feel powerless, they would do whatever it takes, even risky and unethical ones, in an
attempt to return to status quo security.
Previous research provides support for this account. Although risky tactics are usually
employed by promotion individuals in the service of their eager strategies, prevention individuals
have been found to also utilize risky tactics in the service of their vigilant strategies, particularly
when they are currently under a state of loss “-1”, and when these tactics provide the possibility
of returning to a secure state-of-affairs “0” (Scholer, et al., 2010). Losses (e.g., powerlessness)
are negative states and are unacceptable to prevention individuals. They will take whatever
22
action is necessary to feel secure again.
Prevention Focus and Powerfulness
Contrasting with prevention-powerless, when prevention individuals are powerful (a
positive state), they do not experience any extreme negative affect. They would not be motivated
to take any risky or corrupt action that could potentially result in losses; such actions would be
suppressed or inhibited. Hence, I predict that prevention-powerful individuals would be less
motivated than prevention-powerless individuals to behave corruptly.
Promotion Focus and Powerlessness
A promotion focus causes one to focus on gains and advancement. Thus, unlike
prevention individuals, promotion individuals are not sensitive to negative states and are less
concerned about these states (Crowe & Higgins, 1997; Liberman, et al., 2001; Scholer, et al.,
2010). Therefore, when promotion individuals are powerless, they do not experience much
negative affect as they are mainly focused on achieving a gain “+1”. Unlike prevention
individuals, promotion individuals perceive negative “-1” states and status quo safety “0” to be
the equivalent, and therefore are not motivated to regulate it (Scholer, et al., 2010). To this end, I
predict that promotion-powerless individuals would be less motivated than prevention-powerless
individuals to behave unethically.
Promotion Focus and Powerfulness
However, when promotion individuals are powerful, they move from safety “0” to “+1”,
which would make them become more eager (Higgins, 1987; Molden, et al., 2008). This high-
intensity motivational state strengthens their action engagement (Higgins, 2000), making them
even more reward-seeking, which would lead them to focus on the gains that could result from
23
behaving unethically.
Hence, given the above analysis of the interaction between regulatory focus and power, I
posit that Prevention-Powerless will behave unethically to minimize pain, while Promotion-
Powerful will behave unethically to maximize gain. Prevention-Powerful and Promotion-
Powerless will be less motivated to behave unethically. Accordingly, I hypothesized that: (1)
Prevention-Powerless and Promotion-Powerful individuals will exhibit relatively stronger
tendencies to behave unethically. (2) Prevention-Powerful and Promotion-Powerless individuals
will exhibit relatively weak tendencies to behave unethically.
THEORETICAL AND EMPIRICAL IMPLICATIONS
This research aims to build on and extend current literature in three primary ways. First,
this research aims to revisit Gray’s (1982; 1990) influential model of approach and avoidance.
According to Gray, the behavioral approach system (BAS) reacts to two indications of reward—
the presence of positive reward (e.g., food) and the absence of punishment (e.g., safety). This
correspondingly results in behavioral approach towards the positive reward (approach reward)
and towards safety (approach safety). Most of previous research has focused mainly on
behavioral approach towards a positive reward (e.g., gain food), which dovetails with promotion
focus. Promotion individuals tend to employ risky and approach tactics to acquire positive
reward or gain. By contrast, the tendency to approach safety relates more to prevention focus.
Prevention individuals would approach safety (“0”) when they are currently not safe (i.e. “-1”)
(Scholer, et al., 2008; Scholer, et al., 2010). It is important to emphasize that in this case,
prevention individuals are not exhibiting behavioral avoidance, such as inhibiting risky or
careless behaviors, but in fact are exhibiting behavioral approach towards safety, such that the
24
risky approach tactic is employed in the service of being vigilant to reach or restore safety. The
literature has emphasized mainly approaching gains and avoiding pains. The aspect of Gray’s
model that is concerned with approaching safety has been largely overlooked. My theoretical
model builds on this postulation that “-1” conditions for prevention will lead to strong approach
action by showing that when prevention individuals experience powerlessness (a “-1” state), they
will act, take risks, and do whatever it takes—even if it is unethical—to feel safe and secure.
Second, this research aims to delineate and elucidate when powerlessness leads to
avoidance/inhibition, and when it leads to approach self-regulation. Although some research
reveal that powerlessness leads to inhibition (Galinsky, et al., 2003; Keltner, et al., 2003),
research in consumer decision-making suggests that powerlessness could also lead to active
attempts to self-regulate (Inesi, Botti, Dubois, Rucker, & Galinsky, 2011; Rucker & Galinsky,
2008). For example, Rucker and Galinsky (2008) found that powerless individuals have greater
desire to acquire high-status products in an attempt to enhance their sense of power. My findings
would complement both accounts—we predict that powerlessness would produce inhibitive
tendencies when combined with a promotion focus but would produce self-regulatory approach
tendencies when combined with a prevention focus (because a powerless “-1” state-of-affairs is
unacceptable to individuals in a prevention focus and taking action is necessary).
The third aim is to distinguish the conceptual differences between power and regulatory
focus. One might consider powerfulness and promotion to be similar because they both generally
produce more approach, and powerlessness and prevention to be similar because they generally
produce more careful and cautious responding. However, the present research emphasizes how
these motivational inclinations are theoretically and empirically separate (see Higgins, 1997).
25
This distinction has received relatively little attention in the literature because power and
regulatory focus have not been examined together.
OVERVIEW OF STUDIES
In Chapters 7 to 9, I tested the hypotheses in three experiments, all of which
experimentally manipulated regulatory focus and power orthogonally. In all three experiments, I
first administered a standard method for manipulating regulatory focus—priming individuals’
ideals or oughts. I then employed the classic power manipulations—semantic (Chapter 7) and
experiential (Chapter 8, and 9) primes. According to the theorizing underlying the hypotheses, it
is important to first manipulate regulatory focus (promotion or prevention motivational
orientation) before power (“+1” or “-1” current state). Chapter 7 tested the interaction effect of
focus and power on the likelihood of exploiting others. Chapter 8 tested the interaction effect on
aggression. Finally, Chapter 9 tested the interaction effect on cheating behavior. Chapter 9 also
examined the experience of negative affect across the four focus by power conditions, which
would provide some insights into the mechanism. According to my theorizing, I expect
Prevention-Powerless individuals to experience significantly more negative affect, compared to
Prevention-Powerful, Promotion-Powerless, and Promotion-Powerful.
26
CHAPTER 7
EXPERIMENT: EXPLOITING OTHERS
In this study, I tested the interaction between regulatory focus and power on the tendency
to harm and exploit others (Kipnis, 1972; Lee-Chai, Chen, & Chartrand, 2001).
Method
Participants and Design. One hundred and twenty-five participants (72 female; mean age 32.3)
were recruited via Amazon Mechanical Turk, a national online sample of participants with a
wide range of income, education and backgrounds (Buhrmester, Kwang, & Gosling, 2011). The
experiment had two between-subjects factors: Regulatory Focus (promotion, prevention) and
Power (powerful, powerless).
Procedure. Participants completed the manipulations before answering the dependent measure.
Regulatory focus manipulation. I employed the standard procedure used in previous studies
(Freitas & Higgins, 2002; Higgins, Roney, Crowe, & Hymes, 1994). Participants in the
promotion condition were instructed to list three hopes or aspirations (ideals) they would ideally
like to accomplish,
Please think about three things that you ideally would like to do. In other words, think
about three hopes or aspirations that you currently have. Please list the hopes or
aspirations below.
Participants in the prevention condition were instructed to list three duties or obligations (oughts)
they currently have,
Please think about three things that you think you ought to do. In other words, think about
three duties or obligations that you currently have. Please list the duties or obligations
below.
27
Power manipulation. I manipulated power using a scrambled sentences priming task (Smith &
Trope, 2006), which consisted of 16 items. Each item contains five words, and participants were
told to use four of the words to make a grammatical sentence. For the high-power prime, 8 of the
16 sets of words contained a word related to having power (i.e., authority, captain, commands,
controls, dominates, executive, influenced, privileged). For the low-power prime, those same 8
sets contained a word related to lacking power (i.e., complied, janitor, obey, passive, servant,
submits, subordinate, yield).
Next, participants filled out an 18-item scale by Lee-Chai, Chen, & Chartrand (2001),
which captures people’s tendencies to use power to harm and exploit others. Some items reflect a
desire to hurt others (e.g. If I had the opportunity to sue another individual, I would sue for all
the money he or she was worth), others demonstrate flagrant indifference for the consequences
of exploiting others (e.g. One should always take advantage of any opportunity that comes one’s
way, regardless of the consequences for others). All items showed good internal consistency (α =
.79). Higher scores denote higher tendencies to exploit others. Although these items were
initially used to measure trait attributes, trait measures are also influenced by priming and
contextual manipulations (Smith & Bargh, 2008; Smith, Wigboldus, & Dijksterhuis, 2008).
Results and Discussion
As predicted, a two-way ANOVA revealed a significant interaction between regulatory
focus and power on exploiting others, F(1, 117) = 4.68, p = .033, η2 = .038 (see Figure 7)11.
Importantly, contrasts found that prevention-powerless and promotion-powerful led to a higher
tendency to exploit others than promotion-powerless and prevention-powerful (See Table 1 for
pair-wise contrasts). Prevention-powerless and promotion-powerful were not significantly
different from each other. 12
28
This study found evidence for the predicted interaction between power and regulatory
focus. Prevention-powerless and promotion-powerful participants reported a higher tendency to
exploit others than promotion-powerless and prevention-powerful. The next study tested if the
same interaction effect persists with a measure of aggressive behavior.
29
CHAPTER 8
EXPERIMENT: AGGRESSION
Recent work suggest that power holders become aggressive and demeaning especially
when they feel incompetent (Fast & Chen, 2009) and when they lack social status (Fast, Halevy,
& Galinsky, 2011). This study aims to test the predicted interactional pattern on aggression.
Specifically, I aim to examine if powerlessness leads to more aggression on a stranger under
prevention focus using a well-established measure of aggression (Bushman & Baumeister,
1998).
Method
Participants and Design. Two hundred and thirty-two participants (142 females; mean age
32.5) from Mechanical Turk participated for cash. The experiment had two between-subjects
factors: Regulatory Focus and Power.
Procedure. I used the same regulatory focus manipulation as Study 1 and manipulated power
using an experiential prime. Following Galinsky et al. (2003), participants wrote about a memory
where they either had power over another individual (powerful) or where someone else had
power over them (powerless). Participants assigned to the powerful condition were instructed,
Recall a particular incident in which you had power over another individual or
individuals. By power, we mean a situation in which you controlled the ability of another
person or persons to get something they wanted, or were in a position to evaluate those
individuals. Please describe this situation in which you had power—what happened, how
you felt, etc.
Participants assigned to the powerless condition were instructed,
30
Please recall a particular incident in which someone else had power over you. By power,
we mean a situation in which someone had control over your ability to get something you
wanted, or was in a position to evaluate you. Please describe this situation in which you
did not have power—what happened, how you felt, etc.
Participants then responded to the aggression measure, which was a variation of the well-
validated noise induction paradigm (Bushman & Baumeister, 1998; Fast & Chen, 2009). They
were told that subjects in an upcoming experiment would be asked to extract and recall
information from written passages. Those subjects would receive a 1-s sound blast from a horn
for every question they answered incorrectly. Participants were asked to select noise levels for
each of the 10 trials. The levels ranged from Level 1(10 dB) to Level 7(130 dB). These responses
were averaged to form our measure of aggression.
Results and Discussion
Consistent with the previous study, a two-way ANOVA revealed a significant interaction
between regulatory focus and power on aggression, F(1, 228) = 3.88, p = .05, η2 = .017 (See
Figure 8). Contrasts found that prevention-powerless and promotion-powerful led to more
aggression than promotion-powerless and prevention-powerful (See Table 1). Prevention-
powerless and promotion-powerful were not significantly different from each other.
In the next study, I tested the hypothesis with a behavioral measure of cheating.
Additionally, I also examined if prevention individuals, as compared to participants from the
other three conditions, experienced powerlessness more negatively and intensely. To this end, I
measured negative affect and predict that prevention-powerless individuals would experience
significantly more negative affect as compared to Prevention-Powerful, Promotion-Powerless,
and Promotion-Powerful.
31
CHAPTER 9
EXPERIMENT: CHEATING ON A TEST
Method
Participants and Design. One hundred and six participants (68 females; mean age 21.7 years)
were recruited from Columbia University. The experiment had two between-subjects factors:
Regulatory Focus and Power.
Procedure. The first task was an anagram test used in Chapter 3, which unbeknownst to the
participants, they would have the opportunity to cheat on later. Participants received a manila
folder and were told that they had four minutes to unscramble the 15 anagrams on the first page
of the folder. They were incentivized to solve as many as possible as they would earn one dollar
for every correct answer. At the end of the four minutes, participants detached and retained the
worksheet page and turned in the remaining packet—unaware that hidden at the back of the
folder was a sheet of carbonless copy paper, which recorded an imprint of their answers.
Next, participants completed the same manipulations from Chapter 8 and then reported
how distress, agitated, sad, discouraged, happy (reverse-scored), cheerful (reverse-scored), calm
(reverse-scored), and relaxed (reverse-scored) (adapted from Higgins, 1987) they felt on a 5-
point scale, anchored at 1 (not at all) and 5 (Extremely)(α = .76, M = 2.68, SD = .64). These
eight emotions were the most pertinent emotions to regulatory focus (Molden, et al., 2008).
After this task, the experimenter, who appeared very busy, handed the answer key for the
anagram test to the participant and requested that the participant grade the test him/herself. The
experimenter explained that she had to manage another study and would not have the time to
grade the anagram task. This created an opportunity for participants to alter their original
answers in private.
32
Results
As predicted, a two-way ANOVA revealed a significant interaction between regulatory
focus and power on cheating, F(1, 102) = 4.11, p = .045, η2 = .038 (see Figure 9). Similar to the
studies in Chapters 7 and 8, Contrasts found that prevention-powerless led to more cheating than
promotion-powerless and prevention-powerful (See Table 1). Prevention-powerless and
promotion-powerful were not significantly different from each other.
We also found support for the affective states we had predicted (see Figure 10).
Prevention-powerless (Mean = 3.00, SD = .63) reported feeling significantly more negative
affect than promotion-powerless (Mean = 2.61, SD = .70), t(102) = 2.28, p = .025, prevention-
powerful (Mean = 2.53, SD = .61), t(102) = 2.86, p =.005, and promotion-powerful (Mean =
2.53, SD = .53), t(102) = 3.01, p = .003. This is evidence that when prevention individuals feel
powerless, they do indeed experience significantly more negative affect than individuals in the
other conditions.
Meta-analysis
A meta-analysis was also undertaken to determine which effects remained stable and
reliable when the data from all three studies were considered together. Toward that goal, the
effect-size r coefficients were Fisher's z-transformed, weighted by sample-size, and then
averaged. The average rz was then converted back into r for presentation (see Table 1).
Main interaction. The regulatory focus X power interaction was significant, the weighted
average effect size r was .16 and the associated combined z-value was 6.14 and p < .001. The
main effects for both regulatory focus and power were not significant.
Contrasts with prevention-powerless. Prevention-powerless were significantly more corrupt than
33
promotion-powerless, the weighted average effect size r was .22 and the associated combined z-
value was 6.26 and p < .001. Prevention-powerless were also significantly more corrupt than
prevention-powerful, the weighted average effect size r was .18 and the associated combined z-
value was 4.83 and p < .001. Interestingly, prevention-powerless were significantly more corrupt
than promotion-powerful, the weighted average effect size r was .068 and the associated
combined z-value was 1.94 and p < .03.
Contrasts with promotion-powerful. Promotion-powerful were significantly more corrupt than
prevention-powerful, the weighted average effect size r was .10 and the associated combined z-
value was 2.62 and p < .005. Promotion-powerful were also significantly more corrupt than
Promotion-powerless, the weighted average effect size r was .15 and the associated combined z-
value was 3.97 and p < .001.
Discussion of the Integrative Effects of Regulatory Focus and Power
The three studies in Chapters 7 to 9 tested a new theoretical model that integrates theories
on regulatory focus and power. This model articulates how both powerfulness and powerlessness
can each lead to corrupt behavior. Employing different power manipulations and measures of
corrupt behavior, we found that powerlessness combined with a prevention focus, and
powerfulness combined with a promotion focus, produces the highest propensity to exploit
others, display aggression and dishonest behavior. In contrast, individuals in the promotion-
powerless and prevention-powerful conditions displayed relatively less corrupt behavior.
Importantly, we also found some evidence for the emotional manifestations that accompany
these effects. Prevention-powerless individuals reported feeling more negative affect than the
other conditions. This is evidence that powerlessness leads to a more intense negative state in
prevention individuals than in promotion individuals.
34
CHAPTER 10
GENERAL DISCUSSION & CONCLUSIONS
Theoretical Contributions
I highlight five main theoretical contributions of this research. Firstly, this is the first
research that examines the effects of power and regulatory focus together. It is important to
understand how these constructs interact with each other because both power and regulatory
focus has been found to influence an array of imperative social and organizational outcomes
independently (Brockner & Higgins, 2001; Brockner, Higgins, & Low, 2004; Gino & Margolis,
2011; Magee & Galinsky, 2008). Importantly, I demonstrated that the same behaviors (i.e.
unethical behavior) could be a consequence of both high and low power, depending on the
mindsets people are in. This adds to the relatively scarce but important literature on moderators
of power on behavioral outcomes (Jordan, Sivanathan, & Galinsky, 2011; Lammers, et al.,
2008). For instance, Lammers, et al. (2008) found that the link between power and approach is
broken when the power relationship is illegitimate, and that the powerless will approach,
especially when the power is illegitimate. Similarly, Jordan, et al. (2011) found that the stability
of the hierarchy moderates the relationship between power and risk-taking such that the
powerless are more likely to take risks when the hierarchy is unstable. Going beyond structural
and legitimacy of the hierarchy, my research extends this literature by showing that
powerlessness leads to corrupt behavior when it is coupled with a prevention focus but not with a
promotion focus.
It is also plausible that power and regulatory focus interactively influence a myriad of
approach behaviors other than corrupt behavior since both power and regulatory focus have been
found to independently influence approach tendencies (Anderson & Berdahl, 2002; Higgins,
35
1998; Keltner, et al., 2003). One of the most important approach behaviors exhibited by powerful
individuals is the likelihood to engage in competitive interpersonal exchange (Huang, et al.,
2011; Keltner, et al., 2003; Magee, Galinsky, & Gruenfeld, 2007). As a pilot test, I ran a study
that examined this interaction with a measure of the propensity to negotiate (Huang, et al., 2011;
Keltner, et al., 2003; Magee, et al., 2007). In this study (N = 240 Americans), I manipulated
power and regulatory focus using the manipulations from Chapters 8 and 9 and then had
participants respond to a purchase scenario adapted from Magee et al. (2007): “You are buying a
new car. How likely would you be to negotiate the price?” This was measured on a 7-point scale
anchored at 1 (not at all likely) and 7 (very likely). Consistent with the three studies in Chapters 7
to 9, results show a significant 2-way interaction between regulatory focus and power on
propensity to negotiate, F(1, 232) = 6.07, p = .014, η2 = .025. The contrasts also show that
prevention-powerless and promotion-powerful led to a higher propensity to negotiate as
compared to the prevention-powerful and promotion-powerless, which appeared to be exhibiting
relatively more inhibitive tendencies. Future research should examine if this interactive effects
can influence other approach behaviors.
In the current research, regulatory focus was manipulated with the most classic primes of
promotion and prevention. Future research should also examine if promotion and prevention
tasks could also elicit the same interactive effects. For instance, tasks that encourage the pursuing
of gains (promotion) versus tasks that encourages the avoidance of mistakes (prevention) could
interact with the amount of power one possesses while working on those tasks. Future research
could also examine if reversing the order of the manipulations (i.e. power à regulatory focus)
would generate the same results. My current theorizing presupposes that individuals need to be
in a regulatory mindset first before experiencing power. Reversing the order of the manipulation
36
could induce a different psychological experience. Future research could extend the current
theorizing and examine if the same interactive effects could be observed.
This research builds on Gray’s (1982; 1990) influential model of approach and
avoidance, and reemphasizes that according to Gray, individuals would not only approach
reward, but would also approach safety. Most of the extant literature has focused on approaching
reward and avoiding punishment. Approach towards a positive reward, is analogous to a
promotion focus, where individuals tend to employ risky and approach tactics to acquire that
reward or gain. Avoiding punishment is akin to a prevention focus, where individuals employ
vigilant tactics to avoid punishment. However, the current research proposes that prevention
individuals would employ approach tactics to move towards safety (i.e. “0”) when they are
currently powerless (a “-1” state). This is consistent with recent research, which found that
prevention individuals would adopt risky approach strategies in the service of fulfilling their
vigilant goal of returning to safety when they are in a loss frame (Scholer, et al., 2008; Scholer,
et al., 2010).
Third, this research delineates and elucidates when powerlessness would lead to
inhibition and when it would lead to self-regulation by acting out. Although some research reveal
that powerlessness leads to behavioral inhibition (i.e. less action, less risk-seeking) (Galinsky, et
al., 2003; Keltner, et al., 2003), research in consumer decision-making suggests that
powerlessness could also lead to active attempts to self-regulate (Inesi, et al., 2011; Rucker &
Galinsky, 2008). For example, Rucker and Galinsky (2008) found that powerless individuals
have greater desire to acquire high-status products in an attempt to enhance their sense of power.
My findings would complement both accounts—I found that powerlessness would produce
37
inhibitive tendencies when combined with a promotion focus but would produce self-regulatory
corrupt tendencies when combined with a prevention focus.
My research also clarifies the conceptual difference between power and regulatory focus.
Although one might consider powerfulness and promotion to be related because both generally
produce more approach-related tendencies, and powerlessness and prevention to be related since
they produce cautious responding, the findings indicate that these psychological dimensions need
to be distinguished. Power functions as the means and resources that would allow one to achieve
one’s goals. In contrast, promotion ideals and prevention oughts are concerns with desired end-
states to be attained or maintained—promotion being concerned with advancing to achieve a
gain and prevention being concerned with maintaining or restoring safety. These motivational
concerns interact with the current conditions created by power. For individuals with prevention
concerns, being in a powerless state induces high levels of negative affect, which motivates them
to use approach tactics and do whatever is necessary, including taking chances and behaving
unethically, to restore a secure state. For individuals with promotion concerns, being in a
powerful state, make them eager for additional power that they gain through approach tactics.
Lastly, the results of my studies address the question of whether power combined with
focus creates regulatory fit effects. Regulatory fit occurs when there is a match between the
manner in which one pursues a goal and one’s goal orientation (Avnet & Higgins, 2006). When
regulatory fit occurs, people feel “right” about their response and their response intensifies,
without the valence of the response being altered (Cesario, Grant, & Higgins, 2004; Higgins,
2006; Higgins, Cesario, Hagiwara, Spiegel, & Pittman, 2010). The current results are consistent
with the possibility that powerfulness “fits” with promotion because the corrupt tendency was
augmented when these two factors were combined and both promotion and powerfulness
38
generally lead to approach. However, the data is not consistent with powerlessness being a “fit”
with prevention. If there were simply a regulatory fit between prevention and powerlessness, we
should expect corrupt tendencies to decrease significantly since both prevention and
powerlessness generally lead to avoidance-related behaviors, such as being careful and not
taking chances. If all that were happening was a fit between prevention and powerlessness, it is
these avoidance responses that would be intensified, making individuals more careful and less
willing to take any chances. Instead, the current studies found a robust pattern whereby
prevention combined with powerlessness produced greater approach-related corrupt tendencies.
This obtained pattern of findings makes sense if powerlessness is a negative valence
condition because there is recent evidence that prevention-focused individuals in a negative or
loss condition will engage in approach tactics, even risky tactics, if such tactics have the
potential to restore safety and security (see Scholer et al., 2010). And, indeed, there was evidence
(Chapter 9) that prevention-focused individuals in a current powerless state did experience
negative affect especially, which is consistent with powerlessness having negative valence
especially for prevention-focused individuals. Taken together, then, the pattern of results overall
does not support a standard regulatory fit account but it does support an account where
powerfulness as a “+1” current state intensifies promotion approach tactics and powerlessness as
a “-1” current state intensifies prevention approach tactics.
Practical Implications
This research provides an explanation for why powerless individuals will sometimes take
action in the real world. Consider the recent uprisings that have engulfed the Arab world—
ordinary individuals taking to the streets and overthrowing the all-powerful regimes of Tunisia,
Egypt, and Libya. Or the ongoing “Occupy” demonstrations occurring all across the world,
39
where thousands of students, lower/middle-class citizens and homeless people, protest against
the powerful capitalists that control each nation’s economy. Throughout history, the world has
witnessed numerous similar uprisings of the powerless. From the Spartacus slave revolt of
ancient Rome, to the recent Arab Spring, these events illustrate that the powerless will act in
ways that are characteristic of the powerful—they marched, they yelled, and they fought back.
Importantly, a critical note about these powerless individuals who participate in social-political
movements is that they are often focused on prevention-related oughts, such as “Equal gender
and racial employment rights amendment” or “Bring American elections up to international
standards of a paper ballot precinct counted and recounted in front of an independent and party
observers system” (Hart, 2011) 13. This attention to oughts is a natural prime that activates a
prevention focus state (and as our studies have shown, prevention focus can be induced simply
by asking participants to think about their oughts). This orientation makes them especially
sensitive to losses and when coupled with a sense of powerlessness (a state of loss), they become
motivated to move from this state of loss to a state of safety and security. And as observed in
social movements, the only way to do so is to act, negotiate and fight back—sometimes with
violence and aggression if that is perceived as being necessary.
My findings also have implications for the link between power and corrupt behavior.
They suggest that the influence of powerfulness on corruption, namely exploiting others,
aggression and dishonest behavior, is reduced when powerful individuals are in a prevention
state. Hence, one practical strategy to reduce the corruptive effects of power is to instill a
prevention focus orientation on leaders and decision-makers (think about one’s duties and
obligations to the organization and employees), especially in situations when opportunities to act
selfishly and unethically are present. The current research would suggest that this focus on one’s
40
oughts and duties could undermine excessive risk-taking and reward-focused tendencies.
Importantly, however, my data also suggests that powerlessness can also produce corrupt
action when it is combined with a prevention focus. This is the first empirical demonstration that
powerlessness can also lead to corruption. I think that this is an important direction that future
research should seek to investigate further. Although the extant literature and the studies from
Chapters 2 to 5 have clearly documented an empirical link between powerfulness and corruption
(Kipnis, 1972; Lammers, et al., 2010; Lammers, et al., 2011), it has paid less attention to the
possibility that powerlessness could also lead to corruption. Future research should examine
whether powerlessness combined with prevention could influence other types of corrupt
behavior such as lying, infidelity and other ethical improprieties.
41
CONCLUSION
In sum, this dissertation demonstrated that bodily configurations, or postures, incidentally
imposed by our environment could induce a psychological state of power and correspondingly
lead to increases in dishonest behavior. Four studies that were conducted in the laboratory and
the field found consistent evidence that individuals who engaged in expansive postures (either
explicitly or inadvertently) were more likely to steal money, cheat on a test, and commit traffic
violations. Indeed, participants’ self-reported sense of power mediated the link between postural
expansiveness and dishonesty. This suggests that power need not always be socially and
structurally acquired (French, 1956; French & Raven, 1959). Instead, one’s psychological sense
of power can be fleeting, dynamic, and self-induced simply by spreading out and taking up more
space. Importantly, such postural shifts can inadvertently render us more or less honest.
This dissertation also found evidence that powerlessness could also corrupt, particularly
under a prevention focus. Importantly, the research also found that powerfulness does not always
corrupt, specifically under a prevention focus. Three studies provided empirical support for a
new theoretical model that delineates the regulatory focus conditions when powerfulness and
powerlessness can lead to corrupt action. All three studies found that the highest levels of corrupt
behavior were observed among individuals who were promotion-powerful and prevention-
powerless, and lower levels of corrupt behavior among individuals who were prevention-
powerful and promotion-powerless. These findings not only illuminate when power and
powerlessness corrupt, it also provides important insights on how we can mitigate corrupt
behavior among the powerful and the powerless.
42
FOOTNOTES
1This study includes two samples. Sample size was not predetermined but data-analysis was
conducted after completion of each data collection period. Both samples were subject to the
exact same procedure with the exception that participants in one sample were administered the
Regulatory Focus Questionnaire (Higgins et al., 2001) before the posture manipulation. We
followed Schimmack’s (2012) recommendation to combine these replications into a single
analysis. A meta-analytic approach to combining the samples was also undertaken to verify that
our effect was as strong as it seemed when the raw data were combined. Toward that goal, the Φ
effect-size coefficients (which are exactly equivalent to effect size r in a 2x2 chi-squared case)
were Fisher's z-transformed, weighted by sample-size, and then averaged. The average rz was
then converted back into r (and this Φ in this 2x2 chi-squared case) for presentation. The
weighted average effect size Φ was .41 and the associated combined z-value was 5.03 and p <
.001.
2This survey was administered as part of our cover story. The data were not analyzed.
3Eight participants did not count the money and two were aware of our dishonesty measure. We
made an a priori decision to exclude these participants from our analysis. Including these
participants yielded, χ2(1, N=88)=7.28, p=.007.
4We aimed to recruit 40 participants but due to logistical laboratory issues (i.e. an initially small
subject population which was further reduced by competition for participants with two other
researchers using the same dishonesty paradigm) we were only able to recruit 34 participants
during the study time-frame.
5Debriefing checks revealed that three participants were aware of our dishonesty paradigm, and
43
were excluded from analyses. One of them also altered the workspace layout without permission.
One outlier, more than 3 SDs above the overall mean, was also excluded. Including these
participants yielded, F(1,33)=.29, p=ns.
6We aimed to recruit between 70 to 80 participants, but stopped recruitment at 71 because the
study time-frame ended.
7We asked participants if the task was difficult (on a 7-point Likert-type scale) and we found no
significant difference between conditions.
8From the video-recording, two participants had problems maneuvering the car, which resulted in
them repeatedly crashing into objects throughout the race. We made an a priori decision to
exclude these participants. Including them yielded, F(1,69)=.50, p=ns.
9Bootstraping analyses considering “hit & run” as a mediator between posture and sense of
power as the outcome is marginally significant. However, further analyses revealed that “hit &
run” did not significantly predict sense of power for both expansive and contractive participants
when analyzed separately.
10There was no effect of status, (B=.45, SE=.34, p=.18). When status was not included as a
covariate, the effect was significant; (B=.019, SE=.005, p=.001).
11There were no significant findings as a function of gender in all studies.
12 Data from three outliers more than 2 SDs above the mean and one participant who failed an
attention reading check were excluded.
13 There are no official demands for the occupy movements, but these illustrate some sentiments
of the participants of the occupy movement.
44
REFERENCE
Altman, I., Taylor, D. A., & Wheeler, L. (1971). Ecological Aspects of Group Behavior in Social
Isolation. Journal of Applied Social Psychology, 1, 76-100.
Anderson, C., & Berdahl, J. L. (2002). The experience of power: Examining the effects of power
on approach and inhibition tendencies. Journal of Personality and Social Psychology, 83,
1362-1377.
Anderson, C., & Galinsky, A. D. (2006). Power, optimism, and risk-taking. European Journal of
Social Psychology, 36, 511-536.
Apicella, C. L., Dreber, A., Campbell, B., Gray, P. B., Hoffman, M., & Little, A. (2008).
Testosterone and financial risk preferences. Evolution and Human Behavior, 29, 384-390.
Avnet, T., & Higgins, E. T. (2006). How regulatory fit affects value in consumer choices and
opinions. Journal of Marketing Research, 43, 1-10.
Bohns, V. K., & Wiltermuth, S. S. (2012). It Hurts When I Do this (or You Do that): Posture and
Pain Tolerance Journal of Experimental Social Psychology, 48, 341-345.
Boles, T. L., Croson, R. T. A., & Murnighan, J. K. (2000). Deception and retribution in repeated
ultimatum bargaining. Organizational Behavior and Human Decision Processes, 83, 235-
259.
Brockner, J., & Higgins, E. T. (2001). Regulatory focus theory: Implications for the study of
emotions at work. Organizational Behavior and Human Decision Processes, 86, 35-66.
Brockner, J., Higgins, E. T., & Low, M. B. (2004). Regulatory focus theory and the
entrepreneurial process. Journal of Business Venturing, 19, 203-220.
45
Buckley, C., & Flegenheimer, M. (2011). At Scene of Wall St. Protest, Rising Concerns About
Crime, 2013, from http://www.nytimes.com/2011/11/09/nyregion/at-occupy-wall-street-
protest-rising-concern-about-crime.html?_r=0
Buhrmester, M., Kwang, T., & Gosling, S. D. (2011). Amazon's Mechanical Turk: A New
Source of Inexpensive, Yet High-Quality, Data? Perspectives on Psychological Science,
6, 3-5.
Bushman, B. J., & Baumeister, R. F. (1998). Threatened egotism, narcissism, self-esteem, and
direct and displaced aggression: Does self-love or self-hate lead to violence? Journal of
Personality and Social Psychology, 75, 219-229.
Carney, D. R., Cuddy, A. J. C., & Yap, A. J. (2010). Power Posing: Brief Nonverbal Displays
Affect Neuroendocrine Levels and Risk Tolerance. Psychological Science, 21, 1363-
1368.
Carney, D. R., Hall, J. A., & LeBeau, L. S. (2005). Beliefs about the nonverbal expression of
social power. Journal of Nonverbal Behavior, 29, 105-123.
Carney, D. R., & Mason, M. F. (2010). Decision making and testosterone: When the ends justify
the means. Journal of Experimental Social Psychology, 46, 668-671.
Carney, D. R., Yap, A. J., Lucas, B. J., Mehta, P. H., McGee, J. A., & Wilmuth, C. (2013).
Power Buffers Stress – For Better and For Worse. . Unpublished Manuscript.
Cesario, J., Grant, H., & Higgins, E. T. (2004). Regulatory fit and persuasion: Transfer from
"feeling right". Journal of Personality and Social Psychology, 86, 388-404.
Chen, S., Lee-Chai, A. Y., & Bargh, J. A. (2001). Relationship orientation as a moderator of the
effects of social power. Journal of Personality and Social Psychology, 80, 173-187.
46
Coates, J. M., & Herbert, J. (2008). Endogenous steroids and financial risk taking on a London,
trading floor. Proceedings of the National Academy of Sciences of the United States of
America, 105, 6167-6172.
Crowe, E., & Higgins, E. T. (1997). Regulatory focus and strategic inclinations: Promotion and
prevention in decision-making. Organizational Behavior and Human Decision Processes,
69, 117-132.
Dabbs, J. M., Carr, T. S., Frady, R. L., & Riad, J. K. (1995). Testosterone, Crime, and
Misbehavior among 692 Male Prison-Inmates. Personality and Individual Differences,
18, 627-633.
Dabbs, J. M., & Morris, R. (1990). Testosterone, Social-Class, and Antisocial-Behavior in a
Sample of 4,462 Men. Psychological Science, 1, 209-211.
Damasio, A. (1994). Descartes' Error. New York: New York: Grosset/Putnam.
Darwin, C. (1872/1904). The expression of emotions in man and animals. London: Murray
(Original work published 1872).
de Waal, F. (1998). Chimpanzee politics: Power and sex among apes. . Baltimore, MD: Johns
Hopkins.
DeCelles, K. A., DeRue, D. A., Margolis, J. D., & Ceranic, T. L. (2012). Does power corrupt or
enable? When and why power facilitates self-interested behavior. Journal of Applied
Psychology, 97, 681-689.
Duclos, S. E., Laird, J. D., Schneider, E., Sexter, M., Stern, L., & Van Lighten, O. (1989).
Emotion-specific effects of facial expressions and postures on emotional experience.
Journal of Personality and Social Psychology, 57, 100-108.
47
Ellyson, S. L., & Dovidio, J. F. (1985). Power, dominance, and nonverbal behavior. New York:
Springer-Verlag.
Fast, N. J., & Chen, S. (2009). When the Boss Feels Inadequate: Power, Incompetence, and
Aggression. Psychological Science, 20, 1406-1413.
Fast, N. J., Halevy, N., & Galinsky, A. D. (2011). The destructive nature of power without status.
Journal of Experimental Social Psychology, 48, 391-394.
Fast, N. J., Sivanathan, N., Mayer, N. D., & Galinsky, A. D. (2012). Power and overconfident
decision-making. Organizational Behavior and Human Decision Processes, 117, 249-
260.
Fischer, J., Fischer, P., Englich, B., Aydin, N., & Frey, D. (2011). Empower my decisions: The
effects of power gestures on confirmatory information processing. Journal of
Experimental Social Psychology, 47, 1146-1154.
Fiske, S. T. (1993). Controlling Other People - the Impact of Power on Stereotyping. American
Psychologist, 48, 621-628.
Fisman, R., & Miguel, E. (2007). Corruption, norms, and legal enforcement: Evidence from
diplomatic parking tickets. Journal of Political Economy, 115, 1020-1048.
Freitas, A. L., & Higgins, E. T. (2002). Enjoying goal-directed action: The role of regulatory fit.
Psychological Science, 13, 1-6.
French, J. R. P. (1956). A Formal Theory of Social Power. Psychological Review, 63, 181-194.
French, J. R. P., & Raven, B. H. ( 1959). The bases of social power. . .Ann Arbor, MI: University
of Michigan Press.
Galinsky, A. D., Gruenfeld, D. H., & Magee, J. C. (2003). From power to action. Journal of
Personality and Social Psychology, 85, 453-466.
48
Galinsky, A. D., Magee, J. C., Inesi, M. E., & Gruenfeld, D. H. (2006). Power and perspectives
not taken. Psychological Science, 17, 1068-1074.
Gino, F., & Margolis, J. D. (2011). Bringing ethics into focus: How regulatory focus and risk
preferences influence (Un)ethical behavior. Organizational Behavior and Human
Decision Processes, 115, 145-156.
Gray, J. A. (1982). The neuropsychology of anxiety: An enquiry into the functions of the septo-
hippocampal system. . New York: Oxford University Press.
Gray, J. A. (1990). Brain systems that mediate both emotion and cognition. . Cognition and
Emotion, 4, 269-288.
Greenberg, J. (2002). Who stole the money, and when? Individual and situational determinants
of employee theft. Organizational Behavior and Human Decision Processes, 89, 985–
1003.
Greenberg, J., & Scott, K. S. (1996). Why do workers bite the hands that feed them? Employee
theft as a social exchange process. In B. M. Staw & L. L. Cummings (Eds.), Research in
organizational behavior, (Vol. Vol. 18, pp. 111–155). Greenwich, CT JAI Press.
Gruenfeld, D. H., Inesi, M. E., Magee, J. C., & Galinsky, A. D. (2008). Power and the
objectification of social targets. Journal of Personality and Social Psychology, 95, 111-
127.
Guinote, A. (2007). Power and goal pursuit. Personality and Social Psychology Bulletin, 33,
1076-1087.
Hall, J. A., Coats, E. J., & Smith LeBeau, L. (2005). Nonverbal behavior and the vertical
dimension of social relations: A meta-analysis. Psychological Bulletin, 131, 898-924.
49
Hall, J. A., Rosip, J. C., Smith LeBeau, L., Horgan, T. G., & Carter, J. D. (2006). Attributing the
sources of accuracy in unequal-power dyadic communication: Who is better and why?
Journal of Experimental Social Psychology, 42, 18-27.
Handgraaf, M. J. J., Van Dijk, E., Vermunt, R. C., Wilke, H. A. M., & De Dreu, C. K. W.
(2008). Less power or powerless? Egocentric empathy gaps and the irony of having little
versus no power in social decision making. . Journal of Personality and Social
Psychology, 95, 1136–1149.
Harmon-Jones, E., & Peterson, C. K. (2009). Supine Body Position Reduces Neural Response to
Anger Evocation. Psychological Science, 20, 1209-1210.
Hart, L. J. (2011). Forum Post: Proposed List Of Demands For Occupy Wall St Movement! ,
from http://occupywallst.org/forum/proposed-list-of-demands-for-occupy-wall-st-
moveme/
Higgins, E. T. (1987). Self-Discrepancy - a Theory Relating Self and Affect. Psychological
Review, 94, 319-340.
Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 52, 1280-1300.
Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a motivational principle.
Advances in Experimental Social Psychology, Vol 30, 30, 1-46.
Higgins, E. T. (2000). Making a good decision: Value from fit. American Psychologist, 55,
1217-1230.
Higgins, E. T. (2006). Value from hedonic experience and engagement. Psychological Review,
113, 439-460.
50
Higgins, E. T., Cesario, J., Hagiwara, N., Spiegel, S., & Pittman, T. (2010). Increasing or
Decreasing Interest in Activities: The Role of Regulatory Fit. Journal of Personality and
Social Psychology, 98, 559-572.
Higgins, E. T., Friedman, R. S., Harlow, R. E., Idson, L. C., Ayduk, O. N., & Taylor, A. (2001).
Achievement orientations from subjective histories of success: promotion pride versus
prevention pride. European Journal of Social Psychology, 31, 3-23.
Higgins, E. T., Roney, C. J. R., Crowe, E., & Hymes, C. (1994). Ideal Versus Ought
Predilections for Approach and Avoidance - Distinct Self-Regulatory Systems. Journal of
Personality and Social Psychology, 66, 276-286.
Hirsh, J. B., Galinsky, A. D., & Zhong, C. B. (2011). Drunk, Powerful, and in the Dark: How
General Processes of Disinhibition Produce Both Prosocial and Antisocial Behavior.
Perspectives on Psychological Science, 6, 415-427.
Howard, J. A., Blumstein, P., & Schwartz, P. (1986). Sex, Power, and Influence Tactics in
Intimate-Relationships. Journal of Personality and Social Psychology, 51, 102-109.
Huang, L., Galinsky, A. D., Gruenfeld, D. H., & Guillory, L. E. (2011). Powerful Postures
Versus Powerful Roles: Which Is the Proximate Correlate of Thought and Behavior?
Psychological Science, 22, 95-102.
Idson, L. C., Liberman, N., & Higgins, E. T. (2000). Distinguishing gains from nonlosses and
losses from nongains: A regulatory focus perspective on hedonic intensity. Journal of
Experimental Social Psychology, 36, 252-274.
Idson, L. C., Liberman, N., & Higgins, E. T. (2004). Imagining how you'd feel: The role of
motivational experiences from regulatory fit. Personality and Social Psychology Bulletin,
30, 926-937.
51
Inesi, M. E. (2010). Power and loss aversion. Organizational Behavior and Human Decision
Processes, 112, 58-69.
Inesi, M. E., Botti, S., Dubois, D., Rucker, D. D., & Galinsky, A. D. (2011). Power and Choice:
Their Dynamic Interplay in Quenching the Thirst for Personal Control. Psychological
Science, 22, 1042-1048.
James, W. (1884). What is an emotion? . Mind, 9, 188-205.
Jordan, J., Sivanathan, N., & Galinsky, A. D. (2011). Something to Lose and Nothing to Gain:
The Role of Stress in the Interactive Effect of Power and Stability on Risk Taking.
Administrative Science Quarterly, 56 530-558.
Kaplan, S., & Kaplan, R. (2009). Creating a larger role for environmental psychology: The
Reasonable Person Model as an integrative framework. Journal of Environmental
Psychology, 29, 329–339.
Keltner, D., Capps, L. M., Kring, A. M., Young, R. C., & Heerey, E. A. (2001). Just teasing: A
conceptual analysis and empirical review. Psychological Bulletin, 127, 229–248.
Keltner, D., Gruenfeld, D. H., & Anderson, C. (2003). Power, approach, and inhibition.
Psychological Review, 110, 265-284.
Kipnis, D. (1972). Does Power Corrupt? Journal of Personality and Social Psychology, 24, 33-
41.
Knight, C., & Haslam, S. A. (2010). The relative merits of lean, enriched, and empowered
offices: an experimental examination of the impact of workspace management strategies
on well-being and productivity. Journal of Experimental Psychology: Applied, 16, 158-
172.
52
Kraus, M. W., Piff, P. K., & Keltner, D. (2009). Social Class, Sense of Control, and Social
Explanation. Journal of Personality and Social Psychology, 97, 992-1004.
Laird, J. D. (1974). Self-attribution of emotion: The effects of expressive behavior on the quality
of emotional experience. Journal of Personality and Social Psychology, 29, 475-486.
Lammers, J., Galinsky, A. D., Gordijn, E. H., & Otten, S. (2008). Illegitimacy moderates the
effects of power on approach. Psychological Science, 19, 558-564.
Lammers, J., Stapel, D. A., & Galinsky, A. D. (2010). Power Increases Hypocrisy: Moralizing in
Reasoning, Immorality in Behavior. Psychological Science, 21, 737-744.
Lammers, J., Stoker, J. I., Jordan, J., Pollmann, M., & Stapel, D. A. (2011). Power Increases
Infidelity Among Men and Women. Psychological Science, 22, 1191-1197.
Lee-Chai, A. Y., Chen, S., & Chartrand, T. L. (2001). From Moses to Marcos: Individual
differences in the use and abuse of power. In A. Y. Lee-Chai & J. A. Bargh (Eds.), The
use and abuse of power: Multiple perspectives on the causes of corruption (pp. 57-74).
Philadelphia, PA: Psychology Press.
Leonard, P. (2012). Changing Organisational Space: Green? Or Lean and Mean? Sociology.
Liberman, N., Molden, D. C., Idson, L. C., & Higgins, E. T. (2001). Promotion and prevention
focus on alternative hypotheses: implications for attributional functions. Journal of
Personality and Social Psychology, 80, 5-18.
Magee, J. C., & Galinsky, A. D. (2008). Social Hierarchy: The Self-Reinforcing Nature of Power
and Status. Academy of Management Annals, 2, 351-398.
Magee, J. C., Galinsky, A. D., & Gruenfeld, D. H. (2007). Power, propensity to negotiate, and
moving first in competitive interactions. Personality and Social Psychology Bulletin, 33,
200-212.
53
Mazur, A. (1976). Effects of testosterone on status in primate groups. Folia Primatologica;
International Journal of Primatology, 26, 214-226.
Mazur, A., & Booth, A. (1998). Testosterone and dominance in men. Behavioral and Brain
Sciences, 21, 353-363.
Mehta, P. H., & Beer, J. S. (2010). Neural mechanisms of the testosterone-aggression relation:
The role of orbitofrontal cortex. Journal of Cognitive Neuroscience, 22, 2357-2368.
Molden, D. C., Lee, A. Y., & Higgins, E. T. (2008). Motivations for promotion and prevention.
New York: Guilford Press.
Niedenthal, P. M. (2007). Embodying Emotion. Science, 316, 1002-1005.
Niedenthal, P. M., Barsalou, L. W., Winkielman, P., Krauth-Gruber, S., & Ric, F. (2005).
Embodiment in attitudes, social perception, and emotion. Personality and Social
Psychology Review, 9, 184-211.
Piff, P. K., Stancato, D. M., Cote, S., Mendoza-Denton, R., & Keltner, D. (2012). Higher social
class predicts increased unethical behavior. Proceedings of the National Academy of
Sciences of the United States of America, 109, 4086-4091.
Preacher, K. J., & Hayes, A. F. (2004). SPSS and SAS procedures for estimating indirect
effects in simple mediation models. Behavior Research Methods, Instruments, &
Computers, 36, 717−731.
Rhodewalt, F., & Comer, R. (1979). Induced-compliance attitude change: Once more with
feeling. Journal of Experimental Social Psychology, 15, 35-47.
Riskind, J. H. (1983). Nonverbal expressions and the accessibility of life experience memories:
A congruence hypothesis. Social Cognition, 2, 62-86.
Riskind, J. H., & Gotay, C. C. (1982). Physical posture: Could it have
54
regulatory or feedback effects on motivation and emotion? Motivation and Emotion, 6, 273-298.
Rucker, D. D., & Galinsky, A. D. (2008). Desire to acquire: Powerlessness and compensatory
consumption. Journal of Consumer Research, 35, 257-267.
Ruedy, N. E., & Schweitzer, M. E. (2010). In the Moment: The Effect of Mindfulness on Ethical
Decision Making. Journal of Business Ethics, 95, 73-87.
Schimmack, U. (2012). The Ironic Effect of Significant Results on the Credibility of Multiple-
Study Articles. . Psychological Methods.
Scholer, A. A., Stroessner, S. J., & Higgins, E. T. (2008). Responding to negativity: How a risky
tactic can serve a vigilant strategy. Journal of Experimental Social Psychology, 44, 767-
774.
Scholer, A. A., Zou, X., Fujita, K., Stroessner, S. J., & Higgins, E. T. (2010). When Risk
Seeking Becomes a Motivational Necessity. Journal of Personality and Social
Psychology, 99, 215-231.
Simmons, J. P., Nelson, L. D., & Simonsohn, U. (2012). A 21 Word Solution Available at SSRN:
http://ssrn.com/abstract=2160588 or http://dx.doi.org/10.2139/ssrn.2160588.
Smith, P. K., & Bargh, J. A. (2008). Nonconscious effects of power on basic approach and
avoidance tendencies. Social Cognition, 26, 1-24.
Smith, P. K., Jostmann, N. B., Galinsky, A. D., & van Dijk, W. W. (2008). Lacking power
impairs executive functions. . Psychological Science, 19, 441-447.
Smith, P. K., Wigboldus, D. H., & Dijksterhuis, A. (2008). Abstract thinking increases one's
sense of power. Journal of Experimental Social Psychology, 44, 378-385.
Stokols, D. (1978). Environmental Psychology. Annual Review of Psychology, 29, 253-295.
55
Strack, F., Martin, L. L., & Stepper, S. (1988). Inhibiting and facilitating conditions of the
human smile: A nonobtrusive test of the facial feedback hypothesis. Journal of
Personality and Social Psychology, 54, 768–777.
Tiedens, L. Z., & Fragale, A. R. (2003). Power Moves: Complementarity in Dominant and
Submissive Nonverbal Behavior. Journal of Personality and Social Psychology, 84, 558-
568.
Weisfeld, G. E., & Beresford, J. M. (1982). Erectness of posture as an indicator of dominance or
success in humans. Motivation and Emotion, 6, 113–129.
Wells, G. L., & Petty, R. E. (1980). The effects of overt head movements on persuasion:
Compatibility and incompatibility of responses. Basic and Applied Social Psychology, 1,
219–230.
Werner, C. M., Altman, I., & Brown, B. B. (1992). A Transactional Approach to Interpersonal
Relations: Physical Environment, Social Context and Temporal Qualities. Journal of
Social and Personal Relationships, 9, 297.
56
Table 1: Mean Level of Corrupt Behavior and Effect Sizes of Contrasts
Mean corrupt behavior Effect size and significance of contrast comparison Powerful Powerless Promotion-Powerful Prevention-Powerless
N Promotion Prevention Promotion Prevention Prevention Powerful
Promotion Powerless
Prevention Powerful
Promotion Powerless
Experiment 1: Semantic prime, DV = Exploiting others
121 2.73 2.56 2.55 2.92 t = .97 t = .102 t = 2.06 t = 2.09 (.77) (.70) (.71) (.59) p = .33 p = .31 p = .042 p =.039 r = .115 r = .121 r = .268 r = .273
Experiment 2: Recall experiential prime, DV = Aggression 232 3.25 2.98 2.84 3.37 t = .90 t = 1.46 t = 1.33 t = 1.95
(1.53) (1.48) (1.51) (1.58) p = .37 p = .15 p = .19 p = .05 r = .089 r = .134 r = .126 r = .169
Experiment 3: Recall experiential prime,DV = Cheating on a Test
106 0.83 0.5 0.23 1.03 t = .83 t = 1.49 t = 1.37 t = 2.03 (1.81) (.98) (.43) (1.72) p = .41 p = .14 p = .17 p = .045 r = .113 r = .222 r = .186 r = .304
Meta-analysis
Regulatory focus X Power interaction Contrast comparisons Weight average effect size r = .16 r = .1 r = .15 r = .18 r = .22 Combined z = 6.14 z = 2.62 z = 3.97 z = 4.83 z = 6.26 p < .001 p < .004 p < .001 p < .001 p < .001
Note. Standard deviations are given in parentheses. DV = dependent variable.
57
Figure 1: Poses employed in Field Experiment: Stealing by Omission.
58
Figure 2: How the money was presented in Field Experiment: Stealing by Omission.
59
Figure 3: Desk-space configurations for Laboratory Experiment: Cheating on a Test.
60
Figure 4: Driver’s seat configurations for Laboratory Experiment: Hit and Run in a Driving
Simulation
61
Figure 5: Results from Laboratory Experiment: Hit and Run in a Driving Simulation: mediation analysis predicting “hit and run”. The numbers alongside the arrows are unstandardized regression coefficients; coefficients in parentheses are the values obtained when both Posture and Sense of Power were included as predictors of “hit and run”. Asterisks indicate values, *p = .058, **p < .05.
62
Figure 6: Observational Field Study: Parking Violations: Dimensions of the automobile
considered in the size computation.
63
Figure 7. Exploiting others. Error bars represent standard errors of the mean.
64
Figure 8. Aggression. Error bars represent standard errors of the mean.
65
Figure 9. Cheating on a Test. Error bars represent standard errors of the mean.
66
Figure 10. Negative Affective States. Error bars represent standard errors of the mean.
2
2.2
2.4
2.6
2.8
3
3.2
Promotion-‐Powerful Promotion-‐Powerless Prevention-‐Powerful Prevention-‐Powerless
Negative Affect