Close
About
FAQ
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
The angry brain: neural correlates of interpersonal provocation, directed rumination, trait direct aggression, and trait displaced aggression
(USC Thesis Other)
The angry brain: neural correlates of interpersonal provocation, directed rumination, trait direct aggression, and trait displaced aggression
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
THE ANGRY BRAIN: NEURAL CORRELATES OF INTERPERSONAL
PROVOCATION, DIRECTED RUMINATION, TRAIT DIRECT AGGRESSION,
AND TRAIT DISPLACED AGGRESSION
by
Thomas Frederick Denson
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(PSYCHOLOGY)
May 2007
Copyright 2007 Thomas Frederick Denson
ii
Dedication
To Nida, Mom, and Dad for all of your love, support, and believing in me.
Without you, none of this would have been possible.
iii
Acknowledgements
Thank you to my dissertation committee (in alphabetical order): Brian Lickel,
Zhong-Lin Lu, Norman Miller (chair), Elahe Nezami, and Stephen J. Read. Thank
you to Jaclyn Ronquillo, Marija Spanovic, and Eduardo A. Vasquez for help with
data collection.
This work was supported by a grant from the David and Dana Dornsife
Cognitive Neuroscience Imaging Center at USC and a grant from California State
University, Long Beach.
iv
Table of Contents
Dedication ii
Acknowledgements iii
List of Tables vii
List of Figures viii
Abbreviations ix
Abstract x
Chapter 1: Introduction 1
Overview 1
Organization of the Introduction 2
Provocation, Trait Direct Aggression, and Trait Displaced Aggression 2
Provocation 2
Trait Measures of Direct and Displaced Aggression 3
Rumination 6
Neuroscience of Personality 8
Neural Correlates of Anger and Aggression 10
Neural Correlates of Rumination 16
Prefrontal Asymmetry 20
The Current Research 21
Objective #1: Identify the brain regions associated with an anger-
inducing interpersonal provocation 22
Objective #2: Demonstrate that activity in the anterior cingulate
is associated with subjective reports of anger 22
Objective #3: Identify the brain regions associated with
subsequent rumination 22
Objective #4: Examine the extent to which hippocampus activity
following the provocation correlates with self-reported
rumination during the rumination tasks 23
Objective #5: Demonstrate unique patterns of brain activity
for those high in trait direct aggression (viz., the ACC) vs. those
high in trait displaced aggression (viz., the medial PFC) 23
Objective #6: Demonstrate that there is frontal asymmetry
following the provocation and during rumination and that the
degree of asymmetry is related to individual differences in
approach motivation 24
v
Chapter 2: Method 25
Participants 25
Personality Questionnaires 25
Displaced Aggression Questionnaire 25
Aggression Questionnaire 26
Self-Awareness 26
Neuroticism 26
Behavioral Approach and Inhibition 27
Rejection Sensitivity 27
Positive and Negative Affect Schedule 28
Procedures and Dependent Measures 29
Provocation Manipulation 29
Directed Rumination Manipulations 30
Dependent Measures 31
Image Acquisition 34
Analysis of Imaging Data 34
Chapter 3: Results 37
A Short Summary of the Novel Findings 37
Data Analysis Procedures 38
Manipulation Checks, Order Effects, and Gender 39
Provocation Manipulation Check 39
Directed Rumination Manipulation Check 39
Order Effects 40
Gender Differences in Personality 40
Objective #1: Brain Regions Active Following an Interpersonal
Provocation 41
Gender Differences in Brain Activity 41
Objective #2: The Relationship of the Anterior Cingulate with Self-
reported Angry Feelings in Response to the Provocation 49
Objective #3: Brain Regions Active During Rumination 49
Provocation-focused and Self-focused rumination > Distraction 50
Gender Differences in Brain Activity 51
Objective #4: Hippocampus Activity is Correlated with Self-reported
Rumination 56
Objective #5: Unique Patterns of Brain Activity Associated with Trait
Direct Aggression and Trait Displaced Aggression 57
Objective #6: Frontal Asymmetry and the Behavioral Approach System 59
Additional Findings 63
Brain Activity Following the Provocation is Correlated with
Additional Personality Traits 63
Brain Activity During Rumination is Correlated with Additional
Personality Traits 68
vi
State Emotional Reactions to the Provocation are Correlated with
Brain Activity 73
Neural Correlates During Rumination Task of Self-reported
Rumination 76
Neural Correlates of Experimentally-induced Displaced
Aggression 76
Separate Analyses of the Six Primary Objectives by Gender 77
Objective #1: Brain Regions Active Following an Interpersonal
Provocation 77
Objective #2: The Relationship of the Anterior Cingulate with
Self-reported Angry Feelings in Response to the Provocation 82
Objective #3: Brain Regions Active During Rumination 82
Objective #4: Hippocampus Activity is Correlated with Self-
reported Rumination 88
Objective #5: Unique Patterns of Brain Activity Associated with
Trait Direct Aggression and Trait Displaced Aggression 88
Objective #6: Frontal Asymmetry and the Behavioral Approach
System 90
Functional Signal Loss 90
Chapter 4: Discussion 91
Summary 91
The Neural Correlates of Provocation 92
The Neural Correlates of Rumination 94
The Role of Personality 96
Prefrontal Asymmetry 99
Limitations 100
Future Research 102
Conclusion 105
Bibliography 106
vii
List of Tables
Table 1- Descriptive statistics of the personality measures, mean differences
between genders, and differences in the variances between genders 42
Table 2- Brain regions active after exposure to a verbal interpersonal
provocation 45
Table 3- Brain regions active during rumination relative to distraction
(Provocation-focused and Self-focused > Distraction contrast) 52
Table 4- Correlations between personality and BOLD response in brain regions
active after exposure to a verbal interpersonal provocation. For scales
where there were significant mean differences between men and women,
partial correlations controlling for gender are presented first, followed by
zero-order correlations in parentheses. 64
Table 5- Correlations between personality and BOLD response in brain regions
active during directed rumination relative to distraction. 69
Table 6- Correlations between BOLD response in brain regions active after
exposure to a verbal interpersonal provocation and state mood measures 74
Table 7- Brain regions active after exposure to a verbal interpersonal provocation
for men (N = 7) 78
Table 8- Brain regions active after exposure to a verbal interpersonal provocation
for women (N = 9) 80
Table 9- Brain regions active during rumination relative to distraction
(Provocation-focused and Self-focused > Distraction contrast) for men
(N = 8) 83
Table 10- Brain regions active during rumination relative to distraction
(Provocation-focused and Self-focused > Distraction contrast) for women
(N = 12) 85
viii
List of Figures
Figure 1- Brain regions of interest. MPFC = medial prefrontal cortex; dACC =
dorsal anterior cingulate cortex; PCC = posterior cingulate cortex;
VMPFC = ventromedial prefrontal cortex; rACC = rostral anterior
cingulate cortex; lateral PFC = lateral prefrontal cortex 9
Figure 2- Hippocampus activity following provocation. This activity was
correlated with degree of self-reported rumination during the subsequent
rumination task 56
Figure 3- Brain activity following provocation. Panel A illustrates activity in
the left dorsal anterior cingulate cortex, which was positively associated
with individual differences in general aggression and angry affect
following the provocation. Panel B illustrates activity in the right medial
prefrontal cortex, which was positively associated with individual
differences in displaced aggression. 58
Figure 4- BOLD response in the left medial prefrontal cortex during rumination.
Activity in this region was positively associated with individual
differences in displaced aggression, neuroticism, and negative affect. 59
Figure 5- Asymmetry scores following the provocation. Higher scores represent
greater relative left activity in the middle frontal gyrus (i.e., lateral PFC) 61
Figure 6- Asymmetry scores during rumination. Higher scores represent greater
relative left activity in the middle frontal gyrus (i.e., lateral PFC) 62
ix
Abbreviations
ACC = anterior cingulate cortex
PCC = posterior cingulate cortex
PFC = prefrontal cortex
AQ = Aggression Questionnaire
DAQ = Displaced Aggression Questionnaire
PET = positron emission tomography
fMRI = functional magnetic resonance imaging
BOLD = blood oxygen level-dependent
x
Abstract
The present study investigated the neural correlates of interpersonal provocation (i.e.,
an insult), subsequent rumination, aggressive personality traits, and subjective anger
experience. In the current study, the experimenter provoked 20 undergraduates and
subsequent fMRI data was acquired. Participants then engaged in directed
rumination tasks as well as a distraction task during functional image acquisition. As
expected, the following areas were active in response to the provocation: the medial
prefrontal cortex, the anterior and posterior cingulate cortex, the lateral prefrontal
cortex, the insula, hippocampus, and thalamus. Also, consistent with hypotheses, the
following regions were active during rumination: anterior and posterior cingulate,
insula, medial prefrontal cortex, lateral prefrontal cortex, precuneus, and thalamus.
Activity in the dorsal anterior cingulate was correlated with subjective reports of
anger. Hippocampus activity following the provocation was correlated with the
degree of subsequent self-reported rumination. Unique patterns of brain activity were
observed for those high in trait general aggression versus those high in trait displaced
aggression. Activity was greater in the left lateral prefrontal cortex than the right
among men following the provocation, and among both men and women during
rumination. This asymmetry was related to individual differences in behavioral
approach, thus supporting the motivational direction (approach-withdrawal) model of
prefrontal asymmetry.
1
CHAPTER 1: INTRODUCTION
Overview
In recent decades, scientists have made much progress uncovering the neural
correlates of emotions and feelings. Animal studies and evidence from patients with
brain lesions provide support for the localization of specific brain regions associated
with emotional life (Canli, Desmon, Zhao, Glover, & Gabrieli, 1998; Damasio, 1994;
Damasio, Grabrowski, Frank, Galaburda, & Damasio, 2005; LeDoux, 1996). The
advent of functional imaging technology such as positron emission tomography
(PET) and functional magnetic resonance imaging (fMRI), has allowed researchers
to explore the neural correlates of emotions in all participants, healthy or otherwise
(e.g., Damasio et al., 2000; Phan, Wager, Taylor, & Liberzon, 2002). Moreover, a
subset of these studies has investigated patterns of brain activity associated with
individual differences in personality traits (Canli et al., 2001, 2004; Eisenberger,
Lieberman, & Sapute, 2005; Johnson et al., in press; Kumari, Ffytche, Williams, &
Gray, 2004; Ray et al., 2005; Siegle, Steinhauser, Thase, Stenger, & Carter, 2002).
The current study adds to this body of knowledge by exploring neural correlates
associated with: (a) an experimentally manipulated anger-inducing provocation, (b)
experimental manipulations of directed rumination, and (c) individual differences in
trait general aggression and trait displaced aggression (Buss & Perry, 1992; Denson,
Pedersen, & Miller, 2006a). Specifically, participants were provoked by an
experimenter and subsequently engaged in two types of rumination as well as a
distraction task while functional images were acquired. It was expected that brain
2
activity in regions associated with emotional experience and self-awareness of
emotional experiences would be correlated with aggressive personality dimensions
as well as subjective responses to the provocation and directed rumination tasks.
Organization of the Introduction
In what follows, I will first discuss the concept of interpersonal provocation
at the behavioral level and two dimensions of aggressive personality (viz., trait direct
aggression and trait displaced aggression). Next, I will discuss the role of rumination
in augmenting aggressive behavior, followed by an overview of the neuroscience of
personality traits related to emotion. Subsequently, I shall review what is known
about the neural correlates of anger, aggression, and rumination, followed by a
discussion on the relationship between lateral prefrontal hemispheric asymmetry and
anger. Finally, I will conclude by outlining six specific objectives of the current
research.
Provocation, Trait Direct Aggression, and Trait Displaced Aggression
Provocation
Interpersonal provocation is one of the strongest elicitors of aggressive
behavior in humans (Bettencourt & Miller, 1996; Bettencourt, Talley, Benjamin, &
Valentine, 2006). Therefore, identifying the neural correlates following exposure to
a provocation will increase our understanding of the neural responses to this
important precursor to aggression.
3
Trait Measures of Direct and Displaced Aggression
Direct aggression is a widely studied phenomenon that refers to behaviors
intended to directly harm the source of provocation (Anderson & Bushman, 2002;
Geen, 1990). In the present study, individual differences in direct aggression (also
referred to as general aggression) were measured with the Aggression Questionnaire
(Buss & Perry, 1992), which assesses four aspects of general aggression: anger,
hostility, acts of verbal aggression, and acts of physical aggression. Consistent with
social-cognitive models of personality (e.g., Mischel & Shoda, 1995), when
provoked, individuals high in general aggression are likely to display increased
activity in brain regions associated with angry-hostile affect and aggression relative
to those individuals low in general aggression.
In contrast to direct aggression, displaced aggression – a less frequently
studied phenomenon – occurs when a person is provoked, is unwilling or unable to
retaliate against the original provocateur (such as a boss), and subsequently
aggresses against a seemingly innocent target (Dollard, Doob, Miller, Mowrer, &
Sears, 1939; Hovland & Sears, 1940; Marcus-Newhall, Pedersen, Carlson, & Miller,
2000). For instance, a man assaults his spouse for no apparent reason after having
been berated previously by his boss. In the city of Los Angeles, the police
department reported a total of 1,450 cases of spousal or child abuse during 2005
(http://www.lapdonline.org). Road rage and workplace aggression are also chronic
problems. It is highly likely that such terrible acts are often instances of displaced
aggression in which angry or frustrated individuals aggress against undeserving
4
others (e.g., spouses, children, coworkers). Further, the transfer (i.e., misattribution)
of physiological arousal from the initial provocation does not explain the occurrence
of displaced aggression (Bushman et al., 2005; Miller et al., 2003, pp 91-93).
Trait displaced aggression describes individual differences in the tendency to
aggress against innocent others following a provocation (Denson et al., 2006a;
Denson et al., 2006b). In other words, individuals high in trait displaced aggression
are likely to “take it out” on undeserving others. When provoked, they tend to
ruminate about the provocation and dedicate much cognitive activity to planning
revenge. I suggest that the chronic aggressive priming effects associated with
frequent rumination strengthens neural connections among areas associated with
self-relevant cognition and the self-awareness of angry affect. As such, I expect that
following a provocation such repeated activation should lead to increased activity in
areas associated with the self-awareness of negative emotional experience (e.g.,
rumination) among those high in trait displaced aggression.
Recently, my colleagues and I developed an individual difference measure of
trait displaced aggression (Denson et al., 2006a). This 31-item self-report personality
measure – termed the Displaced Aggression Questionnaire (DAQ) – consists of three
subscales: an affective component (angry rumination; e.g., “When angry, I tend to
focus on my thoughts and feelings for a long period of time.”), a cognitive
component (revenge planning, e.g., “I have long living fantasies of revenge after the
conflict is over.”), and a behavioral component (behavioral displaced aggression,
e.g., “I take my anger out on innocent others.”). The scale is structured such that the
5
three subscales comprise the higher-order trait displaced aggression construct
(Denson et al., 2006a), similar in manner to how trait direct aggression is often
conceptualized as having four components (i.e., anger, hostility, verbal aggression,
and physical aggression; Buss & Perry, 1992).
Understanding individual differences in brain activity associated with trait
general aggression and trait displaced aggression is likely to lead to benefits at the
societal level. There is meta-analytic evidence that individuals high in trait general
aggression behave more aggressively in the laboratory than those with lower levels
of the trait (Bettencourt et al., 2006). Moreover, a large number of studies also
demonstrate a link between individual differences in general aggression and actual
aggressive behavior in the real world (e.g., Archer & Webb, 2006; Barnwell,
Borders, & Earleywine, 2006; Bushman & Wells, 1998).
In terms of trait displaced aggression, DAQ scores are related to self-reported
road rage and domestic violence (Denson et al., 2006a). Those high in trait displaced
aggression are more likely to be arrested and have the police visit their house for
domestic disturbances than those low in the trait (Denson et al., 2006b). Scores on
the DAQ also predicted physical aggression against an innocent participant in two
laboratory experiments (Denson et al., 2006a). Thus, empirical evidence suggests
that following a provocation, individuals high in trait displaced aggression are likely
to aggress against undeserving others such as family members, fellow drivers, and/or
coworkers.
6
Rumination
When confronted with a provocation, there are a number of emotional self-
regulation strategies that individuals use to cope with the aversive event.
Provocation-focused rumination is one such strategy. Provocation-focused
rumination refers to perseverative cognition and extended negative affect resulting
from a provocation (e.g., an insult). Recent work has investigated the role of this
type of rumination in augmenting direct and displaced aggression when individuals
are confronted with an interpersonal provocation (Bushman, 2002; Bushman et al.,
2005; Denson et al., 2006a).
Theory and data suggest that while angry feelings usually dissipate rather
quickly (Fridhandler & Averill, 1982; Tyson, 1998), models of aggressive behavior
with associative network components (e.g., Anderson & Bushman, 2002; Berkowitz,
1989; 1990; 1993; Huesmann, 1998; Miller et al., 2003) suggest that rumination
maintains aggressive affect and cognition by keeping a cognitive representation of
the provocation highly accessible. Experimental work has supported this (Bushman
et al., 2005; Kross, Ayduk, & Mischel, 2005; Rusting & Nolen-Hoeksema, 1998).
Anecdotally, there are many real-world instances of displaced aggression wherein
individuals aggress against an innocent person days or even weeks after the initial
provocation.
Provocation-focused rumination can be experimentally manipulated or
considered a personality variable (Caprara, 1986; Denson et al., 2006a; Pedersen,
Denson, Goss, Vasquez, & Miller, 2006; Sukhodolsky, 2001). Provocation-focused
7
rumination increases the strength of association between aggression-related concepts
in the network. The more a person thinks about or re-lives a provoking incident, the
more accessible become its aggression-related components. Such rumination-
induced priming has been implicated in marital conflict (Kachadourian, Fincham, &
Davila, 2005) and increased laboratory aggression (Bushman, 2002). Similarly,
Koneni (1974) found that preventing participants from ruminating decreased direct
aggression. A series of three triggered displaced aggression (TDA) studies
demonstrated that rumination about a provocation increased aggression toward a
somewhat annoying bogus participant (viz., the triggering agent in the TDA
paradigm; Bushman et al., 2005; see also Denson et al., 2006a, Experiment 2).
In addition to provocation-focused rumination, there is substantial work on
another type of rumination referred to as self-focused rumination. This type of
rumination is often described as consisting of “self-focused attention” or directing
attention inward on the self, particularly on one’s own negative emotions
(Lyubomirsky & Nolen-Hoeksema, 1995; Nolen-Hoeksema & Morrow, 1991, 1993;
Trapnell & Campbell, 1999). Theoretical work on self-focused rumination asserts
that negative affect occurs when one becomes aware of discrepancies between the
actual and desired self (e.g., Duval & Wicklund, 1972). Much research indicates that
self-focused rumination exacerbates anger, stress, anxiety, and worry (Morrison &
O’Connor, 2005; Muris, Roelofs, Meesters, & Boomsma, 2004; Rusting & Nolen-
Hoeksema, 1998; Segerstrom, Tsao, Aldern, & Craske, 2000; Watkins, 2004). Self-
focused rumination also increases both depressive symptoms and the length of
8
episodes of depressed mood (e.g., Carver, Scheier, & Weintraub, 1989; Nolen-
Hoeksema, Morrow, & Fredrickson, 1993). Individuals who tend to frequently
ruminate are at risk for decreased life satisfaction (Brown & Phillips, 2005; Chang,
2004; McCullough, Bellah, Kilpatrick, & Johnson, 2001).
Individuals high in trait displaced aggression are especially sensitive to
experimental manipulations of both provocation- and self-focused rumination after
being confronted with an interpersonal provocation (Denson et al., 2006a,
Experiment 2). Specifically, following a provocation, individuals high in trait
displaced aggression behave more aggressively toward undeserving others when
engaging both types of rumination relative to a distracting task (Denson et al., 2006a,
Experiment 2). Further, although the two types of rumination differ in terms of
content, it remains to be seen whether they recruit the same neural substrates.
Neuroscience of Personality
Figure 1 is presented as an anatomical guide to the brain regions of interest
that are discussed in the following sections.
9
Figure 1. Brain regions of interest. MPFC = medial prefrontal cortex; dACC = dorsal anterior
cingulate cortex; PCC = posterior cingulate cortex; VMPFC = ventromedial prefrontal cortex; rACC
= rostral anterior cingulate cortex; lateral PFC = lateral prefrontal cortex.
Although the modern neuroscience of emotion has emerged only in the past
couple of decades, scientists interested in the mind have been aware for quite some
time that the brain plays a crucial role in personality and emotion. The infamous
case of Phineas Gage illustrates this quite well. After suffering injury to his
ventromedial prefrontal cortex (PFC), Gage evidenced extreme personality changes
and labile affect (Damasio, 1994; Damasio et al., 2005). Reports of patients with
lesions have since identified this area as crucial to emotional behavior and therefore,
personality (Bechara, Damasio, & Damasio, 2000). A recent review and two meta-
analyses of PET and fMRI studies suggests that the following areas of the brain are
commonly involved in emotion: the ACC, medial PFC, lateral PFC, insula,
10
ventromedial PFC, amygdala, and thalamus, among others (Murphy, Nimmo-Smith,
& Lawrence, 2003; Phan, Wager, Taylor, & Liberzon, 2002). These areas include
both cortical and subcortical structures (e.g., the thalamus).
A number of researchers have reported that emotion-laden personality traits
moderate the effects of experimental manipulations on brain activity. The most
commonly studied traits are extraversion and neuroticism because they are
associated with the frequent experience of positive and negative affect, respectively
(Canli et al., 2001, 2004; Eisenberger et al., 2005; Kumari et al., 2004). Neuroticism
is especially relevant to the current study because trait displaced aggression is
moderately correlated with neuroticism and both traits share a fundamental
emotional reactivity component (Denson et al., 2006a). For instance, neuroticism
was positively correlated with activity in the left middle frontal and temporal gyrus
and negatively correlated with the right middle frontal gyrus when participants were
exposed to negative affect-inducing pictures (Canli et al., 2001).
Neural Correlates of Anger and Aggression
(1) No previous neuroimaging study has examined brain activity in response
to an anger-inducing provocation, but (2) functional imaging studies have
investigated activity in brain areas related to anger and aggression. Indirect evidence
about anger and aggression comes from two PET experiments in which Raine and
colleagues (Raine, Buchsbaum, & LaCasse, 1997; Raine, Meloy, Bihrle, Stoddard, &
LaCasse, 1998) identified decreased prefrontal activity and increased subcortical
(e.g., amygdala) activation in impulsive murderers relative to matched control
11
participants during a sustained-attention task. These findings suggest a failure in
prefrontal executive control of activity in subcortical limbic activity among
murderers. Furthermore, patients who have undergone the removal of portions of the
anterior cingulate cortex (ACC; i.e., a cingulotomy) demonstrate decreased anger
(Cohen et al., 2001).
Most functional imaging studies of anger in nonclinical samples have
examined visual cues such as angry faces (Adams, Gordon, Baird, Ambady, &
Kleck, 2003; Blair, Morris, Frith, Perrett, & Dolan, 1999; Sprengelmeyer, Rausch,
Eysel, & Przuntek, 1998; Whalen, Shin, McInerney, Fischer, Wright, & Rauch,
2001) as well as re-experiencing anger-inducing life episodes (Damasio et al., 2000;
Dougherty et al., 1999; Kimbrell et al., 1999). For example, using PET technology,
Damasio et al. (2000) asked participants to recall four basic emotions. During the
recall of a personal anger experience, activity occurred in the insula, orbitofrontal
cortex, and the anterior and PCC, as well as subcortical regions associated with
maintaining homeostasis (e.g., the hypothalamus).
Two recent meta-analyses of PET and fMRI studies revealed that in the
studies that included an anger-related manipulation, some of the most prominent
areas of brain reactivity were the medial PFC, ventromedial PFC, ACC and PCC,
lateral PFC, and thalamus (Murphy, Nimmo-Smith, & Lawrence, 2003; Phan,
Wager, Taylor, & Liberzon, 2002). Murphy et al. (2003) concluded that anger (as
well as fear and disgust) evidenced discrete patterns of brain activity. Across all
studies, emotional tasks that involved cognitive demands were strongly associated
12
with activity in the ACC and the insula (Phan et al., 2002). In addition, method of
induction (i.e., recall vs. visual stimuli) produced potentially artifactual activity due
to the induction method. For instance, the occipital cortex was activated when visual
stimuli were used as the emotion induction method. In light of these findings,
examining patterns of brain activity in response to an actual provocation without
visual or cognitive features may help clarify the functional anatomy of anger.
Although no functional imaging study to our knowledge has specifically
examined in vivo an actual anger-inducing interpersonal insult, one study did
investigate brain activity in 14 men high and low in psychopathy during their
performance in a modified Taylor (1967) aggression paradigm (Lotze, Veit, Anders,
& Birbaumer, 2007). Of primary interest, activity in the medial PFC was correlated
with the intensity of electric shocks participants selected to be delivered to the
confederate (Lotze et al., 2007).
Additional studies have investigated frustration and social exclusion, which
are known antecedents of aggression. Frustration is generally defined as the
blocking of a goal (Berkowitz, 1989; Dollard et al., 1939). In an fMRI study, while
being scanned, participants were told that they could win money based on their
responses to a button-pressing task (Abler, Henrik, & Erk, 2005). In reality,
frustration was induced on some trials by not paying the participants despite correct
responses. During these frustrating trials, results showed increased activity in the
right insula, right ventral lateral PFC, and ACC. Due to the similarities between
frustration and provocation in eliciting negative affect (Berkowtiz, 1989), I expected
13
that these same areas reported to be active following frustration would also be active
following an interpersonal provocation.
The insula, ventral PFC, and ACC are also associated with the experience of
social exclusion. For instance, in one study, participants were led to believe that they
were playing a ball-tossing video game with two other participants (Eisenberger,
Lieberman, & Williams, 2003). Social exclusion was manipulated when the two
bogus players stopped tossing the ball to the participant for the majority of the
experiment. Results showed that the anterior insula, right ventral PFC, and the
dorsal ACC were active during social exclusion. Self-reported feelings of exclusion
and social distress were correlated with right ventral PFC and dorsal ACC activity.
Although social exclusion does not necessarily lead to anger, it may produce a
number of negative emotions with shared neural substrates, especially the dorsal
ACC. Indeed, the ACC is associated with the affective component of physical pain
as well as a wide range of negative emotions (Eisenberger et al., 2003; Murphy et al.,
2003; Phan et al., 2002). Thus it appears likely that in response to the interpersonal
provocation of the current study, the ACC will be active following the anger-
inducing provocation. In another fMRI study that manipulated social exclusion, a
composite measure of trait anger and hostility was moderately positively correlated
with reactivity in the dorsal ACC (Eisenberger, Way, Taylor, Welch, & Lieberman,
in press).
Cues to brain regions active during anger and aggression come from the only
two fMRI studies that have investigated the effects of’ violent media on brain
14
activity (Murray et al., 2006; Weber, Ritterfeld, & Mathiak, 2006). In one study, in
an attempt to identify children most responsive to violent media, those who showed
heart rate acceleration to violent media on a pretest were exposed to violent (i.e.,
boxing) and nonviolent (i.e., animal) scenes while functional images were acquired
(Murray et al., 2006). The strongest activity was located in the right PCC and the
right precuneus (in the medial posterior parietal cortex). Because of its role in
memory, these authors speculated that activity in the PCC might represent activation
of stored aggressive scripts. If so, individual differences in the tendency to use
physical aggression should be correlated with activity in the precuneus and PCC.
Another recent study investigated brain activity during violent video game
play (Weber et al., 2006). In a sophisticated frame-by-frame analysis of violent
game play, these authors demonstrated that activity in the dorsal ACC, which has
been termed a neural “alarm system” for its role in discrepancy detection (e.g.,
Eisenberger & Lieberman, 2004), preceded suppression in the rostral ACC (involved
in affective activity) and the amygdala during aggressive “search and destroy”
sequences. These effects were all bilateral. Moreover, when participants were in
danger, under attack, or using a weapon, the dorsal ACC was more active than when
passive or safe, whereas the rostral ACC and amygdala were less active when in
danger, under attack, or using a weapon (regardless of whether defensively or
offensively). Two meta-analyses of anger studies, supports the lack of amygdala
activity (Murphy et al., 2003; Phan et al., 2002).
15
Although not previously mentioned, it is likely that the hippocampus will
also be active following a provocation. Although the current study uses a mild
provocation – for our ancestors and us – real-world provocations can literally signal
life or death. Thus, I used a provocation manipulation that closely resembles a real-
world provocation. Because the current study manipulates anger by inducing an
actual provocation rather than recalling angry experiences or showing angry faces, it
is likely that participants will be motivated to take notice of the provocation. This
suggests that an interpersonal provocation is likely to elicit a strong orienting
response. An orienting response refers to attending to significant or novel events in
one’s environment. The hippocampus plays a substantial role in this process
(Williams et al., 2000); however, the hippocampus habituates rapidly to novel
stimuli (Yamaguchi, Hale, D’Esposito, & Knight, 2006). One fMRI experiment
reported that in response to visual stimuli, the hippocampus habituated after 4 trials
(8 seconds), but it is unknown how long the orienting response may last in response
to significant social stimuli such as a provocation. The hippocampus is also involved
in a second process, namely memory encoding (e.g., Kensinger, Clarke, & Corkin,
2003). Thus, greater hippocampus activity following the provocation may reflect a
combination of orienting response and memory encoding. It is therefore likely that
the degree of hippocampus activity will be related to the extent to which participants
will continue to think about the provocation (i.e., ruminate), because (1) they have
taken more notice of the provocation, and (2) they are better able to remember the
provocation. Although the causal direction remains to be explored, it is likely that
16
hippocampus activity in response to the provocation should then correlate with self-
reported rumination about the provocation.
In summary, these studies suggest that in response to provocation, the areas
most likely to exhibit activity changes from baseline following a provocation are the
lateral PFC, insula, medial PFC, ventromedial PFC, ACC and PCC, thalamus, and
hippocampus.
Neural Correlates of Rumination
To my knowledge, no functional imaging study has experimentally
manipulated rumination, but two studies investigated the functional neuroanatomy of
trait rumination. Ray et al. (2005) used a composite of trait rumination measures
including one measure of angry rumination (Sukhodolsky et al., 2001) from which
the Angry Rumination subscale of the DAQ is derived. These researchers found that
trait rumination was correlated with activity in the amygdala and ventral lateral PFC
in response to negative pictures. Moreover, when participants were asked to decrease
their negative affective responses to the aversive stimuli, trait rumination was
correlated with ACC and medial PFC activity (Ray et al., 2005). The medial PFC is
associated with self-awareness of emotions and self-relevant cognition (e.g., Lane,
Fink, Chau, & Dolan, 1997; Macrae, Moran, Heatherton, Banfield, & Kelley, 2004).
Thus, ruminators were likely “checking in” on their current emotional state when
prompted to self-regulate. The medial PFC is active when participants are asked to
reflect on their feelings (Ochsner et al., 2004) and when participants are asked to
reappraise their responses to distressing visual stimuli (Ochsner, Bunge, Gross, &
17
Gabrieli, 2002). Because rumination involves thinking about one’s affective state,
the medial PFC is expected be activated during experimentally-induced rumination
in the current research. A second study that investigated trait differences in
rumination found moderate correlations between a number of trait rumination
measures and extended amygdala activity in response to negatively-valenced,
personally relevant, emotional words (Siegle et al., 2002).
Although no other studies have directly assessed rumination, some studies
provide clues as to which brain systems might be active during rumination. Both the
ACC and medial PFC are activated when attending to one’s negative emotional state
(Lane et al., 1997; Paradiso et al., 1999). These same regions were active when
judging the self-relevance of personality traits (Macrae et al., 2004). Indeed a
number of studies demonstrate that the medial PFC is associated with self-relevant
cognition and affect (Zimmer, 2005). Even though provocation-focused rumination
may be more likely to induce a primarily outward focus (e.g., thinking about the
provocateur), it is also likely to consist of substantial self-focus on one’s negative
mood (i.e., anger). Thus, the medial PFC is especially likely to be active during
rumination.
Perhaps not surprisingly, the medial PFC also appears related to the
personality trait of self-awareness. Self-awareness is a broad personality construct,
and the medial PFC may be involved in some aspects of self-awareness such as self-
monitoring, self-evaluation, and even the Buddhist notion of the “observer self”
(Austin, 1998). Eisenberger et al. (2005) reported that this personality trait was
18
correlated with activity in the medial PFC and precuneus (in the medial posterior
parietal cortex) during an oddball task. Although we do not have data on the
correlation between the DAQ and self-awareness per se, we do know that the DAQ
is related to the tendency to focus on one’s negative mood as assessed by the
Rumination subscale of the Rumination-Reflection Questionnaire (Denson et al.,
2006a; Trapnell & Campbell, 1999). Scores on the DAQ should therefore be related
to the magnitude of activity in the medial PFC in the present study. The current
study also included a measure of trait self-awareness to examine both traits in the
context of directed rumination manipulations.
There is also evidence that the ACC and PCC might be active during
rumination. Johnson et al. (in press) had participants think about self-referential
promotion-focused (i.e., one’s goals and aspirations), prevention-focused (i.e., one’s
duties and obligations), and non-personally relevant (i.e., distraction) statements.
Self-relevant cognition was associated with activity in the ACC and PCC.
Moreover, relative to distraction, activity in the medial PFC was greater during
promotion focus, while activity in the PCC was greater during prevention focus.
Interestingly, these authors speculated that the medial PFC may be associated with
an inward orientation, while the PCC may be associated with an outward social
orientation. Because provocation-focused and self-focused rumination differ to some
extent in their orientation (i.e., inward vs. outward), these results might be of interest
to the current study. For instance, the medial PFC should display greater activity
during self-focused rumination (inward focus) whereas the PCC should display
19
greater activity during provocation-focused rumination (outward focus). However,
their hypothesis is complicated; given that both promotion and prevention focus
necessarily require placing the self (inward focus) in a social context (outward
focus). The PCC is also active when reflecting on one’s abilities, traits, and attitudes
(Johnson et al., 2002).
A final area that appears likely to be active during rumination is the right
anterior insula (also referred to as the insular cortex). In general, the insula has been
implicated in incorporating internal physiological states. This region is involved in
constructing a constantly updated sense of bodily state. This representation
translates into a subjective sense of “feeling” which forms the basis for the sense of
“self” (Craig, 2003; Damasio, 1994; Damasio, 2003). For example, patients with
damage to the right anterior insula have an impoverished sense of self (Damasio,
1994). The insula is richly connected to other brain regions involved in emotional
life such as the ACC, amygdala, hypothalamus, and orbitofrontal cortex (Craig,
2003). Activation in the right insula is associated with subjective perceptions of pain
and temperature and the awareness of the timing of one’s own heartbeat (Brooks,
Nurmikko, Bimson, Singh, & Roberts, 2002; Craig, Chen, Brandy, & Reiman, 2000;
Critchley, Wiens, Rotshtein, Öhman, & Dolan, 2004). In addition, the insula was
activated when participants were asked to reflect on their feelings in response to
aversive photographs (Ochsner et al., 2002).
20
In summary, the above studies suggest the following areas are likely to be
activated during rumination: medial PFC, lateral PFC, precuneus, ACC, PCC, the
insula, and thalamus (because of its association with anger).
Prefrontal Asymmetry
The current experiment will enhance understanding of prefrontal asymmetry
and the behavioral approach system (for a review, see Coan & Allen, 2003) using
neuroimaging technology. There is some debate as to whether asymmetry in lateral
prefrontal activity reflects affective valence (positive-negative) or motivational
direction (approach-withdrawal). Proponents of the affective valence model posit
that greater relative left activity is associated with positive, approach-related affect
(Davidson, 1998), while proponents of the motivational direction model posit that
greater left activity is associated with approach-related motivational tendencies
(Harmon-Jones, 2003).
Anger has proven an informative case because it is a negatively valenced
emotion associated with approach tendencies. Harmon-Jones and Sigelman (2001)
reported that a laboratory provocation increased relative left lateral prefrontal
electroencephalographic (EEG) activity in male participants and this asymmetry was
related to state anger. Thus, at least for men, we expect to replicate this effect with
neuroimaging technology. Such a finding would lend further support to the
motivational direction account of frontal asymmetry. Indeed, a meta-analysis of
emotions found support for the motivational direction model such that relatively
greater left activity is associated with approach-related emotions (Murphy et al.,
21
2003). However, many of these studies were limited to anger recall and the
presentation of angry faces, and therefore did not involve an actual provocation. The
current study will also be the first to examine prefrontal asymmetry during angry
rumination using fMRI. I expect that the degree of asymmetry should be moderated
by trait behavioral approach (but not trait positive affect), such that greater relative
left asymmetry will be observed among those high in behavioral approach
tendencies.
The Current Research
In the current study, participants first completed a number of personality
questionnaires related to aggression and multiple personality traits. On a subsequent
visit approximately two weeks later, they were provoked while in the scanner. They
then completed the directed provocation-focused rumination, self-focused
rumination, and distraction tasks in counterbalanced order while being scanned.
Following the scanning session, they were given the opportunity to aggress against
an innocent participant (i.e., a physical displaced aggression measure). Finally, they
completed self-report measures of their affective reaction to the provocation, the
extent to which they ruminated about the provocation, and current mood. Thus, the
design is constant provocation followed by three fully counterbalanced, within-
participant block design for the rumination conditions (provocation-focused, self-
focused, and distraction control). The study had six main objectives:
22
Objective #1: Identify the brain regions associated with an anger-inducing
interpersonal provocation.
This is the first neuroimaging experiment to manipulate an anger-inducing
interpersonal provocation. Thus, the goal is to identify areas active following a
provocation. Based on my review of the existing neuroimaging studies that used
anger recall experiences and exposure to angry faces, the activated areas should
include the lateral PFC, medial PFC, ventromedial PFC, ACC and PCC, insula, and
thalamus. Moreover, because humans are likely “hard wired” to take notice of
interpersonal provocations, the hippocampus should also be active. Presumably, the
hippocampus activity reflects the orienting response and memory encoding of the
provocation.
Objective #2: Demonstrate that activity in the anterior cingulate is associated with
subjective reports of anger.
Because the ACC is related to the subjective experience of negative affective
states such as social rejection and physical pain (e.g., Eisenberger & Lieberman,
2004), I expected that activity in this region would correlate with self-reported angry
feelings.
Objective #3: Identify the brain regions associated with subsequent rumination.
This is the first experiment to manipulate rumination following an anger-
inducing provocation. Thus, the goal is to identify areas active during rumination
following an anger episode. Based on my review of the literature, the activated areas
23
should include the medial PFC, lateral PFC, precuneus, ACC and PCC, the insula,
and thalamus.
Objective #4: Examine the extent to which hippocampus activity following the
provocation correlates with self-reported rumination during the rumination tasks.
Because the hippocampus is associated with the orienting response and
encoding of memories, I expected that activity in this region would correlate with
subsequent self-reported rumination. In other words, those who paid attention to and
encoded the provocation to a greater extent should engage in more rumination about
the provocation than those who did not strongly notice and encode the provocation
(e.g., “brushed it off”).
Objective #5: Demonstrate unique patterns of brain activity for those high in trait
direct aggression (viz., the ACC) vs. those high in trait displaced aggression (viz.,
the medial PFC).
Individuals high in trait direct aggression tend to become angered
immediately when they are provoked and directly confront the provocateur. Because
the ACC is likely to be associated with self-reported anger, I expected that individual
differences in trait direct aggression would be correlated with activity in this region.
In contrast, individuals high in trait displaced aggression do not immediately
confront the provocateur and tend to focus on their negative affective state. Because
the medial PFC is associated with both self-awareness of emotions and self-relevant
cognition (e.g., Lane et al., 1997; Macrae et al., 2004; Ochsner et al., 2004), I
24
therefore expected that individual differences in trait displaced aggression would be
associated with activity in this region.
Objective #6: Demonstrate that there is frontal asymmetry following the provocation
and during rumination and that the degree of asymmetry is related to individual
differences in approach motivation.
Finally, I expect to replicate Harmon-Jones and Sigelman’s (2001) EEG
findings with fMRI technology by showing that frontal asymmetry is associated with
an anger-inducing provocation in males. I also expect to demonstrate for the first
time that rumination following a provocation will be associated with frontal
asymmetry. Moreover, I expect that this asymmetry in activity would be correlated
with individual differences in behavioral approach, because mounting evidence
implicates frontal asymmetry as indicative of approach motivation (and not affective
valence). This would be the first such neuroimaging confirmation of this
phenomenon.
25
CHAPTER 2: METHOD
Participants
Twenty (8 men and 12 women; M age = 18.68, SD = 0.75 years, range 18 to
20 years; 14 Whites, 4 Asians, 1 African American, and 1 Latino) right-handed
volunteers from the University of Southern California psychology department
subject pool participated for extra course credit after giving informed consent.
Approximately two weeks prior to the scanning session, they completed a packet of
personality measures and safety screening during an experimental overview session.
All the scales were administered in the order presented below. In order to reduce
suspicion, participants were told that they were participating in an experiment on
cognitive ability and mental imagery. They were also told that they would receive a
picture of their brain, which was emailed to them u
1
pon completion of the study.
Personality Questionnaires
Displaced Aggression Questionnaire. As discussed previously, the DAQ is a
31-item self-report measure of trait displaced aggression ( = .96, M = 2.80, SD =
1.80). The scale consists of three subscales: Angry Rumination (ex., “I keep thinking
about events that angered me for a long time”), Revenge Planning (ex., “When
someone makes me angry, I can’t stop thinking about how to get back at this
person”) and Behavioral Displaced Aggression (ex., “I take my anger out on
innocent others”; Denson et al., 2006a). The response options range from 1
(extremely uncharacteristic of me) to 7 (extremely characteristic of me). Only the
1
Descriptive statistics and reliability coefficients reported are those obtained in the current sample,
not from the original sources.
26
scale endpoints were described. The scale has good internal consistency, test-retest
reliability, convergent validity, and discriminant validity (Denson et al., 2006a). In
addition, by comparison with the AQ, the DAQ is a stronger predictor of laboratory
displaced aggression and real-world indicators of displaced aggression such as
domestic abuse and road rage (Denson et al., 2006a; Denson et al., 2006b).
Aggression Questionnaire. The Aggression Questionnaire (Buss & Perry,
1992) is a 29-item measure of general trait aggression ( = .93, M = 3.28, SD =
1.04). It consists of four subscales: hostility, anger, physical aggression, and verbal
aggression. The scale is reliable and has proven useful in predicting laboratory and
real world aggression (Bushman & Wells, 1998; Buss & Perry, 1992). Participants
responded on a scale ranging from 1 (extremely uncharacteristic of me) to 7
(extremely characteristic of me). Only the scale endpoints were described.
Self-awareness. Fenigstein, Scheier, and Buss’ (1975) measure of trait self-
awareness (i.e., self-consciousness) assesses individual differences in the tendency to
focus on one’s internal states such as thoughts and emotions (e.g., “I’m alert to
changes in my mood”) ( = .81, M = 4.79, SD = 0.86). The scale has been in wide
use for over three decades and possesses adequate internal consistency and test-retest
reliability. Participants responded on a scale ranging from 1 (extremely
uncharacteristic of me) to 7 (extremely characteristic of me). Only the scale
endpoints were described.
Neuroticism. Neuroticism is one of the Big Five personality factors. It
represents a general trait tendency toward emotional instability and the frequent
27
experience of negative affect. I measured this dimension with Goldberg et al.’s
(2006; International Personality Item Pool) 10-item scale ( = .89, M = 3.21, SD =
1.01). The scale has good construct validity and internal consistency (Gow,
Whiteman, Pattie, & Deary, 2005). Participants responded on a scale ranging from 1
(extremely uncharacteristic of me) to 7 (extremely characteristic of me). Only the
scale endpoints were described.
Behavioral Approach and Inhibition. I included the Behavioral Approach (
= .78, M = 5.18, SD = 0.62) and Inhibition Scales ( = .80, M = 4.94, SD = 0.98;
BIS/BAS; Carver & White, 1994) to assess whether individual differences in these
behavioral orientations are associated with brain activity in the current study. The
scales have good internal consistency and have demonstrated predictive validity.
Participants responded on a scale ranging from 1 (extremely uncharacteristic of me)
to 7 (extremely characteristic of me). Only the scale endpoints were described. The
DAQ is moderately positively correlated with the BIS.
Rejection Sensitivity. To some, an interpersonal provocation may be
construed as a type of social rejection (e.g., Leary, Twenge, & Quinlivan, 2006).
Because individuals differ in the extent to which they are affected by social rejection,
I also included a self-report measure of social rejection sensitivity ( = .67, M =
8.87, SD = 2.27; Downey & Feldman, 1996). This 18-item measure assesses the
extent to which individuals are concerned by and expect social rejection across of
variety of interpersonal situations. Participants responded on 6-point scales regarding
how concerned (1 = very unconcerned, 6 = very concerned) and how much they
28
would expect social rejection (1 = very unlikely, 6 = very likely). Only the scale
endpoints were described. I multiplied these two scores for each question and
averaged them into a total index of social rejection sensitivity (see Downey &
Feldman, 1996).
Positive and Negative Affect Schedule. Because the potential moderating
effects of numerous emotional traits in response to provocation and rumination have
not yet been investigated, I included the Positive and Negative Affect Schedule –
Expanded Version (PANAS-X; Watson & Clark, 1994). The PANAS-X has 15
subscales that assess individual differences in the extent to which people generally
experience positive emotion ( = .75, M = 3.69, SD = 0.49), negative emotion ( =
.85, M = 2.10, SD = 0.59), fear ( = .81, M = 2.04, SD = 0.63), hostility ( = .55, M
= 2.18, SD = 0.48), guilt ( = .85, M = 1.75, SD = 0.64), sadness ( = .86, M = 2.02,
SD = 0.72), joviality ( = .92, M = 3.68, SD = 0.73), self-assurance ( = .68, M =
3.26, SD = 0.60), attentiveness ( = .21, M = 3.74, SD = 0.46), shyness ( = .89, M =
2.29, SD = 0.95), fatigue ( = .76, M = 2.95, SD = 0.81), serenity ( = .90, M = 3.35,
SD = 1.01), surprise ( = .88, M = 2.33, SD = 0.96), basic positive affect ( = .41, M
= 3.56, SD = 0.41), and basic negative affect ( = .77, M = 2.00, SD = 0.48).
Because this was the trait version, participants responded on 5-point scales regarding
the way they generally feel (1 = very slightly or not at all, 2 = a little, 3 =
moderately, 4 = quite a bit, 5 = extremely). This is in contrast to the state version,
which was used as the provocation manipulation check (see below). There are
between 3 and 23 items per subscale. The basic affect and general emotion subscales
29
consist of items from the individual positive and negative emotions. For example, the
basic positive affect subscale consists of the joviality, self-assurance, and
attentiveness subscales, whereas the general positive emotion subscale is comprised
of more higher-order emotions (e.g., inspired, strong, determined). The basic
negative affect subscale is comprised of the sadness, guilt, hostility, and fear
subscales, whereas the basic negative emotion subscale consists of the following
items: afraid, scared, nervous, jittery, irritable, hostile, guilty, ashamed, upset, and
distressed. Note that there is some overlap among the “basic” and “general”
emotions subscales. Of particular interest is the hostility subscale, which consists of
the following items: irritable, disgusted, scornful, excited, hostile, and loathing.
Procedures and Dependent Measures
Provocation manipulation. I personally greeted participants at the Dornsife
Neuroimaging Center. First, I collected 2-minutes of baseline functional imaging
data during which participants were asked to stare at a green fixation point in the
center of the screen. Instructions for the anagram task were then presented on the
screen. Using provocation manipulation adapted from previous research (e.g.,
Pedersen et al., 2000), participants were presented with 4 easy and 8 difficult
anagrams on a computer monitor in the scanner. Participants were asked to solve the
anagrams by stating their answer aloud. Each anagram was presented for 15
seconds. At 3 points during the anagram task, the experimenter (Tom Denson)
interrupted the participant from the control room. Approximately 1 min into the task,
the experimenter said in a normal tone of voice, “Look I can barely hear you. I need
30
you to speak louder please.” Two min into the task, the experimenter said in a
normal tone of voice, “Hey, I need you to speak louder.” Finally, 3 min into the task,
the experimenter stated in a frustrated voice, “Look this is the 3
rd
time I have had to
say this! Can’t you follow directions?” In fact, all participants did speak louder when
instructed to do so. Because the insinuation was that participants were not intelligent
enough to follow simple instructions, the provocation manipulation represented the
delivery of an unjustified insult. Indeed, during debriefing, all participants reported
experiencing some degree of anger and annoyance at the experimenter for this
unjustified comment. Further, this provocation manipulation has successfully
angered participants in prior research (Pedersen et al., 2000). Two-minutes of
functional data were acquired immediately (< 500ms) following the delivery of this
last statement.
Directed rumination manipulations. Next, participants were assigned to the
provocation-focused rumination, self-focused rumination, or distraction conditions in
counterbalanced order via a Graeco-Latin square design (e.g., 1 2 3; 2 3 1; 3 1 2; 1 3
2; etc.). Because there were 3 conditions and 20 participants, 6 participants received
the provocation-focused task first, 7 received the self-focused first, and 7 received
the distraction task first. In the provocation-focused rumination condition,
participants were presented with a series of statements on the monitor and asked to
think about each statement for 15 seconds each (e.g., “Think about whom you have
interacted with in the experiment up to this point”, “Think about exactly what you
have done from the start of the study until now”). We have used these procedures in
31
previous research (Pedersen et al., 2006). Statements from the self-focused
rumination and distraction conditions were taken from Rusting and Nolen-Hoeksema
(1998; also used in Bushman et al., 2005). In the self-focused rumination condition,
participants were asked to think about a series of self-referential statements that did
not mention anger or other emotions (e.g., “Think about why people treat you the
way they do”, “Think about why you react the way you do”). In the distraction
condition, participants were asked to think about a series of affectively neutral
statements (e.g., “Think about the layout of the local post office”, “Think about a
double-decker bus driving down the street”). In each of the three conditions, 12
statements were presented for 15s each, such that each condition took 3 min (180s)
to complete. The conditions were separated by 16s rest periods. Functional EPI
whole brain images were taken during the entire directed rumination period.
Dependent measures. Baseline mood and arousal data were collected with the
short version of the Profile of Mood States (POMS; Shacham, 1983). There are 37
items that comprise the following 6 subscales: anger/hostility ( = .70 , M = 1.21, SD
= 0.29), tension/anxiety ( = .80 M = 1.89, SD = 0.67), depression/dejection ( =
.92, M = 1.32, SD = 0.53), vigor/activity ( = .90, M = 3.01, SD = 0.83),
fatigue/inertia ( = .82, M = 2.43, SD = 0.80), and confusion/bewilderment ( = .53,
M = 2.08, SD = 0.50). The subscales possess adequate internal consistency and
construct validity. Participants rated how they felt right now (1 = not at all, 2 = a
little, 3 = moderately, 4 = quite a bit, 5 = extremely).
32
Upon completion of image acquisition, a second experimenter (Jaclyn
Ronquillo or Marija Spanovic) took participants out of the scanner and brought them
to another experimental room to complete a few additional questionnaires. A second
experimenter was included so that participants would not have further interaction
with the provoking experimenter (i.e., myself). These self-report measures included a
measure of physical displaced aggression, state affective responses to the
provocation (i.e., a second PANAS-X), state rumination, and a second POMS, all of
which are described next.
For credibility purposes, the experimenter informed participants that in the
future she planned to have participants complete the rumination task under physical
distraction as part of an unrelated study on brain activity and physical distraction.
However, because it was supposedly under development, she needed to know how
much distraction would be appropriate. Participants would thus need to indicate how
much distraction they felt would be appropriate since they had just completed the
task. Specifically, participants were asked to recommend a length of time for the
next participant to emerge his hand in painfully cold ice water (10º C) on a scale
ranging from “0 seconds, no distraction” to “60 seconds, very strong distraction”
(see Vasquez et al., 2005 for details). Before assigning a recommended length of
distraction, participants were first instructed to immerse their hand in the bucket for
five seconds, ostensibly in order to help them determine an appropriate amount of
physical distraction. Our prior work indicates that this is long enough for participants
to determine that the task is indeed painful.
33
Because affective reactions to interpersonal provocation may include a
number of other emotions besides anger (e.g., fear, guilt, sadness), I also included the
state version of the PANAS-X and modified it so that it referred to the provocation.
Specifically, participants rated the degree to which they experienced each of 65
emotions related to the provocation (i.e., “Indicate to what extent you felt during or
immediately after the anagram task.”) Ratings were made on 5-point scales (1 =
very slightly or not at all, 2 = a little, 3 = moderately, 4 = quite a bit, 5 = extremely)
for each of the 15 subscales: positive emotion ( = .58, M = 2.14, SD = 0.58),
negative emotion ( = .87, M = 2.32, SD = 0.72), fear ( = .90, M = 1.85, SD =
0.82), hostility ( = .81, M = 2.17, SD = 0.76), guilt ( = .90, M = 2.28, SD = 1.01),
sadness ( = .87, M = 1.74, SD = 0.78), joviality ( = .91, M = 1.61, SD = 0.62),
self-assurance ( = .27, M = 1.98, SD = 0.48), attentiveness ( = .75, M = 3.23, SD =
0.78), shyness ( = .91, M = 1.94, SD = 0.90), fatigue ( = .86, M = 2.48, SD =
1.00), serenity ( = .88, M = 2.04, SD = 0.92), surprise ( = .84, M = 2.08, SD =
0.90), basic positive affect ( = .43, M = 2.28, SD = 0.44), and basic negative affect
( = .79, M = 1.98, SD = 0.67).
The measures of state rumination consisted of 6 items assessing how often
and how strongly (1 = not at all, 7 = very often) participants thought about their
performance on the anagram task during each of the 3 experimental conditions (i.e.,
2 items for each condition). Only the endpoints were described. These items were
designed to assess state rumination about the provocation without arousing
suspicion. All participants were then fully debriefed by Tom Denson, informed of
34
the importance of their research participation, thanked, and provided an electronic
copy of a structural slice of their brain. All participants appeared to leave in a
positive mood following debriefing.
Image Acquisition
Participants viewed the experimental tasks through mirrors, which were
presented on a high-resolution monitor placed at the end of a Siemens Magnetom 3T
scanner. Padded foam head constraints controlled participant movement. Once
participants were situated in the scanner, a localizer scan was conducted to ensure
image acquisition of our regions of interest. Next, I acquired three-dimensional
structural images (MPRAGE, 192 slices, FOV 256mm, Thickness 1mm, TR 2070ms,
TE 4.14ms). Prior to beginning the functional scan, I visually inspected a dummy
echo-planar image scan (8s) to ensure the quality of the functional data. Whole-
brain functional images were acquired with interleaved echo-planar (EPI) pulse
sequence (29 axial slices, slice thickness = 4mm, FOV= 24cm, TE=68 ms, TR= 2000
ms, 90º flip angle). T1-weighted structural images (FOV 256mm, Thickness 3.5mm,
TR 580ms,TE 17ms) were taken last.
Analysis of Imaging Data
All analyses were conducted with Brain Voyager QX (Brain Innovation, B.
V.). Data were preprocessed using motion and scan-time corrections, and smoothed
with a Gaussian temporal filter. Brains were normalized via Talairach
transformation (Talairach & Tournoux, 1988). Because there are large individual
differences in brain structure, the Talairach transformation provides a standardized
35
format that allows one to compare findings between individuals and entire studies.
Brain regions were identified using the Talairach Daemon (Lancaster, Summerln,
Rainey, Freitas, & Fox, 1997), which is an electronic database of Talairach
coordinates. Functional images were coregistered with the normalized structural
images. Automated segmentation routines were used to create separate cortical and
subcortical anatomical masks, which were then used in subsequent analyses. As in
all fMRI studies, the main dependent measure was blood oxygen level-dependent
(BOLD) responses to the experimental conditions. This measure indicates blood flow
to brain regions and is highly correlated with neural activity. All BOLD responses
are expressed in percent signal change. For comparisons between conditions, whole-
brain random-effects general linear model (GLM) group analyses were conducted
with participant specified as the random factor. The 2 min following the provocation
are presented in the Results section detailing the reaction to the provocation. For the
rumination data, I used block contrasts to determine differences between the
conditions, which are presented in the Results section on rumination. Block
contrasts, analogous to planned linear contrasts, refer to comparisons between
conditions (e.g., self-focused rumination vs. distraction).
Because fMRI data analysis consists of comparing activity in thousands of
voxels, some form of Type I error control is necessary. I controlled Type I error due
to multiple comparisons with the false discovery rate set at q(FDR) < .05 (q refers to
the maximum FDR that a researcher will tolerate), voxelwise p < .005 (Genovese,
Lazar, & Nichols, 2002). FDR is the proportion of discovered active voxels that are
36
false positives. The primary advantage of this method is that the FDR controls for the
expected proportion of the rejected hypotheses that are falsely rejected. Simulation
and actual fMRI data analyses testify to the accuracy of this method (e.g., Genovese
et al., 2002). For correlations between the individual difference measures and brain
activity, the BOLD percent signal change for each participant was exported to R
(http://www.r-project.org) for statistical analysis. When means and standard
deviations for brain activity are presented, these represent percent signal change.
37
CHAPTER 3: RESULTS
A Short Summary of the Novel Findings
As hypothesized, regions active following the provocation included the
lateral PFC, medial PFC, ventromedial PFC, ACC, PCC, insula, thalamus, and
hippocampus (Objective #1). Activity in the left dorsal ACC following the
provocation was correlated with self-reported feelings of anger, r = .56, p < .05
(Objective #2). Also as hypothesized, the areas active during rumination included
the medial PFC, lateral PFC, precuneus, ACC, PCC, insula, and thalamus (Objective
#3). Activity in the hippocampus following the provocation, which presumably
reflected the orienting response and memory encoding, also correlated with self-
reported rumination during the provocation-focused rumination task such that greater
hippocampus activity was associated with increased rumination, r = .51, p < .05
(Objective #4). Unique patterns of brain activity for the two aggressive personality
dimensions were also observed. Direct aggression was correlated with increased
activity following the provocation in the dorsal ACC, r = .61, p < .05, whereas
displaced aggression was correlated with increased activity in the medial PFC both
following the provocation, r = .57, p < .05, and during the directed rumination task, r
= .55, p < .05 (Objective #5). Finally, men showed greater relative left asymmetry in
the lateral PFC following the provocation, t(6) = 2.04, p = .04, and both genders
showed greater left asymmetry in this same region during rumination, t(19) = 2.53, p
= .01. This activity was correlated with individual differences in behavioral
38
approach orientation both among men in response to the provocation, r = .72, p =
.035, and in both genders during rumination, r = .58, p = .004 (Objective #6).
Data Analysis Procedures
Participants reported no suspicion of the experimental procedures or study
hypotheses. Due to a change in the imaging protocol, the provocation data from the
first 4 participants were not analyzed.
2
This left a total of 16 participants with
complete data from the provocation manipulation and a total of 20 for the directed
rumination manipulations. The first set of analyses consisted of random-effects GLM
analyses on those brain regions hypothesized to be responsive to the experimental
manipulations. A second exploratory analysis was then conducted to identify
additional areas that might be active, but yet unexpected, in response to provocation
and rumination. I used a stricter significance threshold for these supplementary
analyses to reduce the likelihood of Type I error, q(FDR) < .01, voxelwise p < .001.
Because of the relatively small number of participants, no data were excluded.
However, scatterplots were investigated to ensure that outliers did not drive the
observed correlations. Moreover, in such small samples, outliers often work against
finding significant effects through decreased statistical power due to the increased
variability associated with outliers (e.g., Wilcox, 1998).
2
Originally, the protocol consisted of analyzing 1 min of functional data following the provocation.
After initial inspection of these first four participants, I decided to increase the scanning time to 2 min
in order to obtain increased statistical power (i.e., 60 TR for 2 min vs. 30 TR for 1 min). Thus, the
functional data following the provocation from the first 4 participants could not be reasonably
compared with the latter 16 participants, and was therefore eliminated from analyses.
39
Manipulation Checks, Order Effects, and Gender
Provocation Manipulation Check
A paired t-test indicated that participants reported an increase in angry
feelings from baseline in response to the provocation (M
pre
= 1.21, SD
pre
= .29; M
post
= 2.17, SD
post
= .76), t(15) = 5.32, p < .001. This post-provocation value
corresponds to a “mild-to-moderate” degree of angry affect. Thus, the provocation
manipulation was successful. This provocation manipulation has also been used
successfully in prior research (Pedersen et al., 2000). Moreover, as I describe in
detail below, the brain regions activated in response to the provocation were largely
consistent with the findings of two recent meta-analyses that examined studies using
angry faces and anger recall episodes (Murphy et al., 2003; Phan et al., 2002).
Directed Rumination Manipulation Check
Recall that two items assessed the extent to which participants reported
ruminating about the provocation during each of the three rumination conditions:
provocation-focused ( = .58), self-focused ( = .76), and distraction ( = .94). As
expected, paired comparisons revealed that participants thought more about the
provocation in the provocation-focused rumination condition, M = 4.78, SD = 1.26,
than in the self-focused rumination, M = 3.20, SD = 1.45, t(19) = 4.68, p < .001, or
distraction conditions, M = 3.20, SD = 1.92, t(19) = 3.44, p = .003. Also as expected,
because participants were instructed to focus on themselves during the self-focused
rumination condition, the distraction and self-focused conditions did not differ in the
extent to which participants reported ruminating about the provocation, t(19) = 0.00,
40
ns. This is consistent with our prior work (Pedersen et al., 2006). Overall, these
results suggest a successful implementation of the directed rumination
manipulations.
Order Effects
The order of rumination manipulations had only one marginal effect on
BOLD activity, this being in the left middle temporal gyrus, F(2,17) = 3.52, p < .06.
Post hoc tests revealed that those who received the self-focused rumination
manipulation first showed a trend for more activity in this region than those who
received the provocation-focused manipulation first, p < .09. This is consistent with
a recent study that found greater activity in this region when participants viewed
pictures of their own faces than the familiar face of a fraternity brother (Platek et al.,
2006), suggesting a role for the processing of self-relevant information. Thus, when
given the self-focus condition first, this region appeared to be activated slightly more
than following the provocation-focused condition. However, these results should be
interpreted cautiously as they were only marginally significant.
Gender Differences in Personality
Although in my prior work, no gender differences were observed between
men and women on levels of trait displaced aggression, in this sample, men rated
themselves higher in trait displaced aggression (i.e., the DAQ) than women, t(18) =
3.39, p =. 003. Men also rated themselves higher in trait general aggression (i.e., the
AQ), t(18) = 3.48, p =. 003, and self-assurance than women, t(18) = 3.47, p =. 003.
Although not ubiquitous, gender differences in direct aggression are not uncommon
41
(Buss & Perry, 1992, p. 455). Table 1 displays the descriptive statistics for all of the
personality measures. I also tested whether the variances between the genders were
significantly different from each other, using the statistical method outlined by
Wilcox (2003, pp. 264-265). None of these effects were significant.
In order to provide a clearer picture of the correlations between the
personality variables and BOLD response, partial correlations between these three
personality variables, including their subscales, and brain activity that control for
gender are presented in the relevant tables. Thus, gender was used as a covariate.
Objective #1: Brain Regions Active Following an Interpersonal Provocation
The first analyses examined the brain regions active during the 2 min
following the delivery of the provocation (see Table 2). In response to provocation,
there was bilateral activation of the rostral and dorsal ACC, as well as the PCC,
insula, lateral middle frontal gyrus (i.e., lateral PFC), medial frontal gyrus (i.e.,
medial PFC), hippocampus, and left thalamus. There was also activation in the left
ventromedial PFC. Unexpected areas included bilateral activation in the precentral
gyri, lateral cerebral sulci, sulci calcarinus, medial longitudinal fasciculi, and left
medial occipital gyrus.
Gender Differences in Brain Activity
Men showed more activity than women in the left middle frontal gyrus, M
men
= 1.15, SE
men
= .14, M
women
= .49, SE
women
= .20, t(14) = 2.50, p = .03, and right
medial longitudinal fasciculus, M
men
= .76, SE
men
= .08, M
women
= .42, SE
women
= .12,
t(14) = 2.19, p = .05. No other gender differences emerged.
42
Table 1. Descriptive statistics of the personality measures, mean differences between genders, and differences in the variances between genders.
Measure M SD skew kurtosis M
men
SD
men
M
women
SD
men
_______________________________________________________________________________________________________
AQ total score 3.28 1.04 -0.35 0.87 4.06* 0.54 2.75* 0.96
Anger subscale 3.19 1.14 0.18 0.97 3.80* 1.01 2.77* 1.06
Hostility subscale 3.10 1.20 0.12 1.06* 3.31 0.83 2.96 1.42
Physical aggression subscale 2.89 1.42 0.17 1.26* 4.32* 0.72 1.94* 0.82
Verbal aggression subscale 3.92 1.33 -0.34 1.07 4.80* 1.00 3.33* 1.22
DAQ total score 2.80 1.05 0.25 0.83 3.58* 0.95 2.27* 0.77
Angry Rumination subscale 3.71 1.34 -0.35 1.18 4.30 1.27 3.32 1.29
Revenge planning subscale 2.11 1.02 0.88 0.82 2.94* 0.93 1.56* 0.65
Behavioral displaced aggression subscale 2.63 1.23 0.51 1.00 3.55* 1.28 2.02* 0.74
Neuroticism 3.21 1.01 0.05 0.83 2.80 0.95 3.48 0.99
43
Table 1, Continued.
Self-awareness 4.79 0.86 -0.55 0.69 5.06 0.71 4.61 0.93
Behavioral inhibition 4.94 0.99 0.33 0.82 4.50 0.99 5.24 0.89
Behavioral approach 5.19 0.62 0.70 0.48 5.18 0.58 5.20 0.67
Rejection sensitivity 8.87 2.27 0.21 1.84 8.53 2.35 9.09 2.29
PANAS Subscales
Positive emotion 3.69 0.49 -0.39 0.38 3.80 0.42 3.61 0.54
Negative emotion 2.10 0.59 -0.35 0.49 2.11 0.68 2.09 0.55
Fear 2.04 0.63 0.25 0.51 2.04 0.58 2.04 0.68
Hostility 2.18 0.48 0.05 0.39 2.27 0.53 2.11 0.46
Guilt 1.75 0.64 0.76 0.50 1.79 0.73 1.72 0.60
Sadness 2.02 0.72 0.04 0.56 1.95 0.67 2.07 0.77
Joviality 3.68 0.73 -0.45 0.52 3.70 0.58 3.66 0.85
44
Table 1, Continued.
Self-assurance 3.26 0.60 0.04 0.56 3.71* 0.56 2.96* 0.41
Attentiveness 3.74 0.46 -0.18 0.36 3.91 0.42 3.63 0.47
Shyness 2.29 0.95 0.11 0.81 2.25 0.82 2.31 1.06
Fatigue 2.95 0.81 0.02 0.65 3.13 0.61 2.83 0.93
Serenity 3.35 1.01 -0.33 0.78 3.88 0.73 3.00 1.03
Surprise 2.33 0.96 0.39 0.83 2.42 0.87 2.28 1.05
Basic positive affect 3.56 0.41 -0.45 0.31 3.77 0.31 3.41 0.42
Basic negative affect 2.00 0.48 -0.21 0.37 2.01 0.53 1.99 0.47
NOTE: In all cases, * p < .05. An asterisk (*) following M
men
or M
women
indicates a significant mean difference (t-test) between men and women.
45
Table 2. Brain regions active after exposure to a verbal interpersonal provocation.
Region Talairach No. of voxels M(SE) % Significance Test
coordinates signal change
x y z
Anterior cingulate (dorsal)
Right 8 24 34 787 0.62 (0.07) t(15) = 9.23, p < .00001
Left -7 22 33 731 0.59 (0.12) t(15) = 4.79, p < .001
Anterior cingulate (rostral)
Right
region1
5 32 15 446 0.62 (0.13) t(15) = 4.85, p < .001
Right
region2
4 30 -7 719 1.06 (0.29) t(15) = 3.63, p < .001
Left -3 33 -8 590 1.25 (0.24) t(15) = 5.14, p < .001
Insula
Right 37 -2 7 637 0.50 (0.11) t(15) = 4.66, p < .001
Left -37 4 15 767 0.54 (0.09) t(15) = 6.13, p < .0001
46
Table 2, Continued.
Posterior cingulate
Right 5 -52 21 796 0.61 (0.08) t(15) = 7.48, p < .0001
Left
region1
-7 -44 23 240 0.50 (0.07) t(15) = 7.02, p < .00001
Left
region2
-2 -21 28 276 0.59 (0.12) t(15) = 5.01, p < .001
Medial frontal gyrus (i.e., medial PFC)
Right
region1
6 47 13 764 0.72 (0.10) t(15) = 7.20, p < .0001
Right
region2
5 45 19 562 0.59 (0.07) t(15) = 8.65, p < .00001
Medial frontal gyrus (i.e., ventromedial PFC)
Left -5 32 -11 555 1.39 (0.28) t(15) = 4.92, p < .001
Lateral middle frontal gyrus (i.e., lateral PFC)
Right 33 47 7 504 0.76 (0.11) t(15) = 6.68, p < .0001
Left -32 47 9 617 0.78 (0.15) t(15) = 5.10, p < .001
Hippocampus
Right 30 -31 -3 1,013 0.49 (0.07) t(15) = 6.73, p < .00001
Left -30 -31 -3 934 0.60 (0.08) t(15) = 7.22, p < .00001
47
Table 2, Continued.
Thalamus
Left -13 -10 3 675 0.60 (0.12) t(15) = 5.02, p < .001
†Precentral gyrus
Right 34 12 35 134 0.48 (0.08) t(15) = 6.02, p < .0001
Left -36 9 35 436 0.46 (0.07) t(15) = 6.42, p < .0001
†Lateral cerebral sulcus
Right 46 -11 18 649 0.66 (0.08) t(15) = 8.02, p < .00001
Left -45 -12 17 269 0.67 (0.09) t(15) = 7.36, p < .00001
†Sulcus calcarinus
Right 17 -86 5 194 0.80 (0.12) t(15) = 6.67, p < .00001
Left -17 -88 5 110 0.58 (0.07) t(15) = 8.36, p < .000001
†Medial longitudinal fasciculus
Right 38 -57 11 181 0.57 (0.08) t(15) = 6.74, p < .00001
Left -42 -55 13 277 0.58 (0.08) t(15) = 7.62, p < .00001
48
Table 2, Continued.
†Medial occipital gyrus
Left -27 -80 20 57 0.60 (0.11) t(15) = 5.63, p < .0001
† = exploratory analysis
49
Objective #2: The Relationship of the Anterior Cingulate with Self-reported Angry
Feelings in Response to the Provocation
As expected, the left dorsal ACC was uniquely associated with self-reported
state anger as assessed by the PANAS-X Hostility subscale, r = .56, p < .05. Men
and women did not differ in self-reported angry feelings. Further none of the other
14 emotion subscales were related to activity in this region. This provides strong
support for the role of the dorsal ACC in generating the subjective sense of anger in
the context of an interpersonal provocation.
Objective #3: Brain Regions Active During Rumination
I first contrasted the provocation-focused rumination condition against the
self-focused rumination condition (i.e., provocation-focused > self-focused) to
determine whether any differences in brain activity existed between the two types of
rumination. Using the conservative statistical threshold, q(FDR) < .05, voxelwise p <
.005, no differences in activity emerged between the two rumination conditions.
However, with a more relaxed threshold, q(FDR) < .20, voxelwise p < .01, three
small differences did emerge in the expected regions of interest. Relative to self-
focused rumination, provocation-focused rumination elicited slightly more activity in
the right middle frontal gyrus, M = .042, SE = .012, t(19) = 3.58, p = .002, right PCC,
M = .036, SE = .008, t(19) = 4.35, p < .001, and left precuneus, M = .027, SE = .007,
t(19) = 3.73, p = .001. Recall, that based on Johnson et al.’s (in press) experiment, I
had hypothesized that the PCC should display greater activity during provocation-
focused rumination than during self-focused rumination due to the outward focus
50
associated with provocation-focused rumination. This hypothesis was supported,
although the effect was quite small in magnitude.
Provocation-focused and Self-focused rumination > Distraction
Because only small differences between the provocation-focused and self-
focused rumination conditions were discovered outside of my a priori statistical
threshold, I contrasted the provocation focused (+1) and self-focused conditions (+1)
with the distraction condition (-2) to determine the effects of rumination on brain
activity relative to distraction. Although this may obscure some unique variance
(e.g., the small differences between provocation-focused and self-focused
rumination), it provides increased statistical power to detect differences between
rumination and distraction.
Table 3 displays the regions active during the directed rumination task
relative to distraction. As expected, rumination increased activity in the dorsal ACC,
rostral ACC, PCC, medial frontal gyrus, left precuneus, lateral middle frontal gyrus,
left lateral inferior frontal gyrus, and left thalamus. Unexpected bilateral activity
occurred in the superior frontal gyrus (although this appeared to be primarily an
extension of medial PFC activity), precentral gyri, lateral cerebral sulci, central
cerebral sulci, middle temporal gyri, lingual gyri, sulci calcarinus, inferior
longitudinal fasciculus, and inferior parietal lobules.
Together, these results suggest that following a provocation both types of
rumination recruit highly similar neural substrates relative to distraction. Even
though the content of the two types of rumination is different, they both produce
51
aversive, negative emotional states as found in our prior work (Bushman et al., 2005;
Pedersen et al., 2006). This negative emotional response and the self-awareness of
this negative state (i.e., rumination) are reflected in my functional imaging data,
specifically the dorsal ACC and the medial PFC. Had I included state measures of
affect during rumination, they would have likely been correlated with activity in the
dorsal ACC and/or the medial PFC (as was the case with the relationship between
self-reported anger and dorsal ACC activity following the provocation – see
Objective #2).
Gender Differences in Brain Activity
In response to the directed rumination tasks, women showed greater activity
than men in two of the hypothesized regions: the right medial frontal gryus, M
men
=
.35, SE
men
= .08, M
women
= .90, SE
women
= .15, t(18) = 2.75, p = .01, and the right PCC,
M
men
= .28, SE
men
= .16, M
women
= .77, SE
women
= .12, t(18) = 2.43, p = .03. This
increased activity in the medial PFC and PCC suggests that women focused more on
the self and their negative affective state than men (e.g., Eisenberger et al., 2005;
Lane et al., 1997). Additional regions in which women showed greater activity than
men included the right postcentral gyrus, M
men
= .62, SE
men
= .18, M
women
= 1.32,
SE
women
= .12, t(18) = 3.36, p = .004, right precentral gyrus, M
men
= .63, SE
men
= .19,
M
women
= 1.04, SE
women
= .10, t(18) = 2.12, p < .05, right lateral cerebral sulcus, M
men
= .52, SE
men
= .09, M
women
= 1.11, SE
women
= .19, t(18) = 2.34, p = .03, and right
central cerebral sulcus, M
men
= .67, SE
men
= .20, M
women
= 1.12, SE
women
= .10, t(18) =
2.18, p = .04.
52
Table 3. Brain regions active during rumination relative to distraction (Provocation-focused and Self-focused > Distraction contrast).
Region Talairach No. of voxels M(SE) % Significance test
coordinates signal change
x y z
Anterior cingulate (dorsal)
Right 7 15 35 401 0.71 (0.14) t(19) = 5.13, p < .001
Left -7 15 35 618 0.72 (0.11) t(19) = 6.39, p < .00001
Anterior cingulate (rostral)
Right 3 35 9 736 0.82 (0.15) t(19) = 5.67, p < .0001
Left -9 37 13 402 0.69 (0.13) t(19) = 5.20, p < .0001
Insula
Right 38 -3 7 740 0.68 (0.10) t(19) = 6.50, p < .00001
Left -38 -3 7 827 0.66 (0.11) t(19) = 6.27, p < .00001
Posterior cingulate
Right
region1
6 -16 39 422 0.59 (0.10) t(19) = 5.79, p < .0001
Right
region2
6 -53 25 377 0.60 (0.11) t(19) = 5.34, p < .0001
53
Table 3, Continued.
Left
region1
-6 -15 35 333 0.64 (0.11) t(19) = 5.84, p < .0001
Left
region2
-6 -58 22 488 0.57 (0.11) t(19) = 5.00, p < .0001
Medial frontal gyrus (i.e., medial PFC)
Right 9 42 15 451 0.69 (0.12) t(19) =6.02, p < .00001
Left -9 50 19 339 0.52 (0.13) t(19) = 3.89, p < .001
Superior frontal gyrus
Right 7 48 31 509 0.69 (0.14) t(19) = 5.01, p < .0001
Left -9 46 33 601 0.82 (0.11) t(19) = 7.81, p < .000001
Precuneus
Left -9 -53 34 455 0.63 (0.10) t(19) =6.20, p < .00001
Lateral middle frontal gyrus (i.e., Lateral PFC)
Right 36 45 15 684 0.74 (0.13) t(19) = 5.83, p < .0001
Left -37 46 16 368 1.31 (0.27) t(19) = 4.83, p < .001
Lateral inferior frontal gyrus (i.e., Lateral PFC)
Left -50 23 15 330 0.75 (0.13) t(19) = 5.80, p < .0001
54
Table 3, Continued.
Thalamus
Left -13 -21 11 160 0.60 (0.11) t(19) = 5.31, p < .0001
†Precentral gyrus
Right 55 -5 24 676 0.87 (0.10) t(19) = 8.32, p < .000001
Left -52 -9 30 375 0.88 (0.11) t(19) = 7.94, p < .000001
†Lateral cerebral sulcus
Right 59 -34 22 416 0.87 (0.14) t(19) = 6.18, p < .00001
Left -59 -33 21 401 0.70 (0.10) t(19) = 7.37, p < .00001
†Central cerebral sulcus
Right 57 -9 22 885 0.94 (0.11) t(19) = 8.34, p < .000001
†Middle temporal gyrus
Right 41 -64 15 414 0.69 (0.11) t(19) = 6.30, p < .00001
Left -52 -33 1 206 0.66 (0.09) t(19) = 7.26, p < .00001
55
Table 3, Continued.
†Lingual gyrus
Right 10 -81 -7 109 0.82 (0.17) t(19) = 4.67, p < .001
Left -19 -81 -6 521 0.66 (0.09) t(19) = 7.57, p < .000001
†Sulcus calcarinus
Right 6 -82 7 512 0.78 (0.13) t(19) = 5.94, p < .0001
Left -7 -84 10 467 0.79 (0.13) t(19) = 6.12, p < .00001
†Inferior longitudinal fasciculus
Right 33 -65 2 219 0.57 (0.08) t(19) = 7.17, p < .00001
Left -39 -65 4 90 0.69 (0.12) t(19) = 5.88, p < .0001
†Inferior parietal lobule
Right 39 -64 22 345 0.64 (0.11) t(19) = 6.01, p < .00001
Left -39 -64 20 447 0.71 (0.11) t(19) = 6.23, p < .00001
† = exploratory analysis
56
Objective #4: Hippocampus Activity is Correlated with Self-reported Rumination
I also wished to determine whether the degree of brain activity experienced
following the provocation manipulation (especially in the hippocampus) would be
associated with the degree of self-reported rumination about the provocation during
the directed rumination task. This was indeed the case. In the provocation-focused
rumination condition, activity in the hippocampus following provocation was
correlated with self-reported rumination, r = .51, p < .05 (see Figure 2) suggesting
that those who paid much attention to and deeply encoded the provocation also
tended to ruminate about the provocation during the subsequent directed rumination
task.
Figure 2. Hippocampus activity following provocation. This activity was correlated with degree of
self-reported rumination during the subsequent rumination task.
Also of interest was that activity in the right insula – a region known to be
involved in the processing of internal states – was correlated with the degree of self-
57
reported rumination, r = .54, p < .05. Further, two regions commonly involved in
negative emotional experience – the right rostral ACC and left PCC – was correlated
with self-reported rumination, rs = .60 and .52, ps < .05. These findings are
especially noteworthy given that brain activity following the provocation temporally
preceded self-reported rumination.
Objective #5: Unique Patterns of Brain Activity Associated with Trait Direct
Aggression and Trait Displaced Aggression
In support of my hypotheses, individual differences in direct aggression (i.e.,
the AQ) and displaced aggression (i.e., the DAQ) were associated with different
patterns of brain activity when confronted with the provocation (see Figure 3). As
expected, trait general aggression was primarily uniquely associated with activity in
the left dorsal ACC (the same region associated with self-reported anger), r = .61, p
< .05, whereas trait displaced aggression was primarily associated with activity in the
right medial frontal gyrus (i.e., medial PFC), r = .57, p < .05.
58
Figure 3. Brain activity following provocation. Panel A illustrates activity in the left dorsal anterior
cingulate cortex, which was positively associated with individual differences in general aggression
and angry affect following the provocation. Panel B illustrates activity in the right medial prefrontal
cortex, which was positively associated with individual differences in displaced aggression.
Analysis of the AQ and DAQ subscales indicated that the Physical
Aggression subscale of the AQ was associated with activity in the left PCC, r = .68,
p < .01, left hippocampus, r = .52, p < .05, and left thalamus, r = .64, p < .01. The
hippocampus activity suggests that those who report frequent acts of physical
aggression may encode provocations more strongly than those who do not engage in
such aggressive acts, whereas the PCC activity may represent the increased
accessibility of aggressive scripts, consistent with Murray et al.’s (2006) hypothesis.
During the rumination task, also as expected, the DAQ subscale was
positively associated with activity in the left medial frontal gyrus, r = .55, p < .05
(see Figure 4), thus further supporting the notion that individuals high in trait
displaced aggression tend to ruminate to greater extent following a provocation than
59
those low on the trait. The trait Anger subscale of the AQ was associated with
decreased activity in the left precuneus, r = -.42, p < .05.
Figure 4. BOLD response in the left medial prefrontal cortex during rumination. Activity in this
region was positively associated with individual differences in displaced aggression, neuroticism, and
negative affect.
Objective #6: Frontal Asymmetry and the Behavioral Approach System
During the two minutes following the provocation, activity in the left lateral
PFC (i.e., middle frontal gyrus) was not greater than activity in the right hemisphere,
t(15) = -0.09, ns.
3
However, since men exhibited greater left PFC activity than
women following the provocation, and since asymmetry effects have only been
found in men following provocation (Harmon-Jones & Sigelman, 2001), I conducted
separate analyses for each gender. Men showed greater left lateral PFC activity than
right lateral PFC, t(6) = 2.04, p = .04, but women did not, t(8) –1.58, ns. Consistent
with EEG asymmetry research investigating anger following a laboratory
provocation (e.g., Harmon-Jones & Sigelman, 2001), I created difference scores for
3
Due to the small number of participants and strong a priori justification (Coan & Allen, 2003;
Harmon-Jones & Sigelman, 2003), statistical tests on frontal asymmetry were one-tailed.
60
the provocation and rumination data with higher values representing greater relative
left PFC activity. For men, frontal asymmetry following the provocation was
positively correlated with BAS scores, r = .72, p = .035, but not for women, r = .02,
ns. Moreover, this difference score was unrelated to positive affect, negative affect,
and BIS scores. The lack of relationship between BIS and frontal asymmetry is
consistent with prior experimental work with EEG (Coan & Allen, 2003) and further
highlights the role of frontal asymmetry as being specific to behavioral approach
motivation. Scatterplots for both genders are presented in Figure 5.
61
Figure 5. Asymmetry scores following the provocation. Higher scores represent greater relative left
activity in the middle frontal gyrus (i.e., lateral PFC).
Women
Behavioral approach scores
7.0 6.5 6.0 5.5 5.0 4.5 4.0
Asymmetry score
1.0
.5
0.0
-.5
-1.0
-1.5
-2.0
Men
Behavioral approach scores
5.6 5.4 5.2 5.0 4.8 4.6 4.4
Asymmetry scores
1.5
1.0
.5
0.0
-.5
Since there were no gender differences in lateral PFC activity during the
rumination task, men and women were analyzed together. There was significantly
greater activity in the left lateral PFC than the right lateral PFC, t(19) = 2.53, p = .01.
The difference score from the rumination task revealed a significant positive
relationship with BAS scores, r = .58, p = .004, but was unrelated to positive affect,
negative affect, and BIS scores. This is especially notable, given that lateral PFC
activity by itself was correlated with both BAS and positive emotion. Together,
62
these results support the growing evidence that greater left frontal asymmetry is
associated with approach motivation rather than positive affective valence. Separate
analyses by gender revealed fairly strong relationships between BAS and asymmetry
for men, r = .67, p < .02, and women, r = .49, p = .22, but the association was only
significant for women. Scatterplots for both genders are presented in Figure 6.
Figure 6. Asymmetry scores during rumination. Higher scores represent greater relative left activity in
the middle frontal gyrus (i.e., lateral PFC).
Women
Behavioral approach scores
7.0 6.5 6.0 5.5 5.0 4.5 4.0
Asymmetry scores
3.5
3.0
2.5
2.0
1.5
1.0
.5
0.0
-.5
Men
Behavioral approach scores
6.5 6.0 5.5 5.0 4.5
Asymmetry scores
2.5
2.0
1.5
1.0
.5
0.0
-.5
-1.0
-1.5
63
Finally, in order to assure that the relationship between BAS and the
asymmetry was not accounted for by aggressive personality, I computed the
correlations between BAS and the aggressive personality dimensions separately by
gender. For women, BAS was unrelated to both the AQ, r = .31, p = .33, and the
DAQ, r = .07, p = .83. For men, BAS was also unrelated to the AQ, r = -.04, p = .92,
and DAQ, r = -.27, p = .52. Thus, the aggressive personality dimensions did not
account for the relationship between BAS and asymmetry.
Additional Findings
This section describes additional results that were not part of the six
objectives, but do provide some further insight into the neural correlates of
provocation, rumination, state measures of affect, and personality. These results
should be interpreted cautiously due to the large number of statistical tests conducted
and the exploratory nature of the analyses.
Brain Activity Following the Provocation is Correlated with Additional Personality
Traits
Although the primary personality variables of interest were direct and
displaced aggression, I also conducted additional analyses examining neuroticism,
behavioral approach and inhibition, rejection sensitivity, self-awareness, and a broad
range of affective traits (as assessed by the PANAS-X trait version). Table 4
displays the correlations among the personality measures and BOLD response in the
primary brain regions of interest following the provocation.
64
Table 4. Correlations between personality and BOLD response in brain regions active after exposure to a verbal interpersonal provocation. For scales where
there were significant mean differences between men and women, partial correlations controlling for gender are presented first, followed by zero-order
correlations in parentheses.
Personality ACC
(dorsal)
R L
ACC
(rostral)
R
1
R
2
L
Insula
R L
PCC
R L
1
L
2
Medial
frontal
gyrus
R
1
R
2
Medial
frontal
gyrus
(ventral)
L
Lateral
middle
frontal
gyrus
R L
Hippo-
campus
R L
Thalamus
L
AQ total score .35 .61*
(.41 .54*)
.51* .06 .15
(.60* -.17 .18)
.22 -.03
(.15 -.19)
.24 .11 .13
(.40 .15 .15)
.24 .37
(.25 .28)
.05
(.05)
.05 -.22
(-.14 .30)
.21 .21
(-.07 .03)
.30
(.37)
Anger subscale .22 .43
(.32 .44
+
)
.42 -.08 .01
(.54* -.25 .08)
.03 -.02
(.03 -.18)
.30 .12 -.06
(.27 .20 .16)
.10 .40
(.16 .32)
-.08
(-.04)
-.16 -.14
(-.28 .32)
.37 .02
(.06 -.10)
.12
(.11)
Hostility
subscale
.44 .62* .57* .19 .31 .26 .06 .25 .14 .05 .12 .28 .22 .13 -.34 .16 .19 .26
Physical
aggression
subscale
.11 .32
(.26 .32)
.27 -.05 -.14
(.45
+
-.28 .03)
.56* .26
(.27 -.09)
.34 .33 .68**
(.14 .24 .66**)
.42 .05
(.30 .07)
-.14
(-.05)
.28 -.36
(-.09 .36)
.23 .52*
(-.14 .10)
.64**
(.35)
65
Table 4, Continued.
Verbal
aggression
subscale
.25 .45
+
(.33 .48
+
)
.31 .07 .17
(.45
+
-.10 .19)
-.03 -.25
(-.02 -.33)
-.02 -.12 -.08
(-03 .17 -.04)
.17 .36
(.21 .33)
.05
(.05)
-.05 .05
(-.16 .33)
-.11 .05
(-.23 -.04)
.24
(.24)
DAQ total score .07 .24
(.23 .31)
.45
+
-.12 -.12
(.56* -.29 -.01)
.23 .15
(.16 -.08)
.48
+
.07 .17
(.43
+
.30 .13)
.02 .57*
(.11 .42)
-.18
(-.10)
.14 .09
(-.09 .47
+
)
.06 .40
(-.16 .15)
.33
(.43)
Angry
Rumination
subscale
.15 .35 .59* -.01 .13 .25 .06 .30 .15 .08 .27 .51* .05 .12 .15 .00 .35 .36
Revenge
planning
subscale
.05 .25
(.21 .32)
.18 -.13 -.17
(.39 -.29 -.05)
.12 -.05
(.10 -.20)
.50
+
-.07 .27
(.48
+
.33 .03)
.12 .45
+
(.01 .35)
-.20
(-.13)
.24 -.10
(-.01 .35)
-.23 .18
(-.35 .01)
.28
(.24)
Behavioral
displaced
aggression
subscale
-.05 -.06
(.14 .11)
.27 -.18 -.34
(.45
+
-.32 -.15)
.18 .38
(.14 .08)
.41 .06 .10
(.38 .27 .12)
-.17 .45
+
(-.02 .34)
-.33
(-.21)
-.03 .15
(-.19 .49*)
.13 .46
+
(-.11 .19)
.41
(.33)
Neuroticism -.01 .02 .16 .04 -.08 .13 .26 .29 -.18 -.06 -.14 .50* -.08 .19 .11 .09 .29 .26
Self-awareness -.27 .14 .23 -.09 .24 -.26 -.67** -.08 -.04 .31 .27 -.20 .25 -.20 -.01 -.37 -.34 -.34
66
Table 4, Continued.
Behavioral inhibition -.27 -.27 -.25 -.16 -.42 .07 .23 -.03 -.33 -.13 -.31 -.21 -.38 -.00 -.07 -.24 .30 .26
Behavioral approach -.28 -.18 -.19 -.24 -.18 -.32 -.46
+
-.28 -.25 -.53* .11 -.03 -.23 -.05 .08 -.47
+
-.11 .04
Rejection sensitivity -.09 -.44
+
.12 .08 .24 -.31 -.01 -.33 -.43
+
-.15 -.20 -.03 .26 -.19 .32 .24 -.12 -.34
PANAS Subscales
Positive emotion -.29 .14 .03 -.31 -.21 .02 -.30 .17 .43
+
.08 .32 -.42 -.29 -.12 .13 -.12 .20 .19
Negative
emotion
-.01 .34 .51* -.43
+
-.26 .37 .15 .41 .14 .49
+
.19 .01 -.36 -.21 .10 .17 .29 .33
Fear .11 .35 .49
+
-.34 -.20 .44
+
.29 .51* .16 .59* .12 -.03 -.27 -.10 .13 .18 .38 .28
Hostility -.16 .37 .15 -.35 -.39 .31 .14 .41 .37 .36 .38 -.08 -.45
+
-.23 -.07 .33 .24 .34
Guilt -.14 .21 .55* -.26 .14 -.19 -.46
+
-.03 .06 -.02 .31 .06 -.03 -.39 .28 .11 -.16 -.08
Sadness .16 .26 .58* -.24 .01 .33 .14 .26 -.17 .30 .33 .30 -.07 .02 -.06 .05 .17 .09
67
Table 4, Continued.
Joviality -.42 -.11 -.17 -.40 -.43
+
-.14 -.25 .17 .27 -.06 -.03 -.45
+
-.46
+
-.24 -.01 -.16 .03 .02
Self-assurance -.09 .24
(-.04 .23)
.23 -.27 .14
(.41 -.25 .25)
-.26 -.67**
(-.19 -.64**)
-.08 .31 -.04
(-.07 .49
+
-.04)
-.27 -.20
(.29 -.12)
.09 -.19 -.01
(-.30 .36)
-.14 -.34
(-.28 -.35)
-.34
(-.21)
Attentiveness .36 .42 -.18 .20 -.15 .43
+
.11 -.13 .18 .32 .42
+
-.25 .14 .38 .20 -.31 .28 .41
Shyness .13 .40 .06 .01 -.10 .19 .23 .59* .43
+
.35 -.30 -.09 -.12 -.10 .02 .45
+
.39 .36
Fatigue .02 .30 .66* -.31 .00 .28 .05 .41 .14 .43
+
-.01 .07 -.13 -.16 .24 .18 .36 .27
Serenity .30 .47
+
.61* -.17 .13 -.01 -.26 .48
+
.55* .22 .18 .40 .01 -.19 -.01 -.01 -.02 -.07
Surprise -.41 .03 .16 -.39 -.28 -.20 -.36 .17 .26 -.12 .09 -.07 -.38 -.35 .21 .06 -.01 .10
Basic positive
affect
-.16 .18 -.01 -.29 -.12 -.01 -.39 .03 .45
+
.10 .26 -.44
+
-.21 -.14 .22 -.34 -.03 .08
Basic negative
affect
.00 .39 .61* -.39 -.12 .28 .02 .36 .10 .39 .39 .11 -.25 -.22 .09 .20 .19 .19
** p < .01, *p < .05,
+
p < .10.
68
The medial frontal gyrus was associated with neuroticism (as were individual
difference in trait displaced aggression). Activity in the right ACC was less anger-
specific and was related to a broad range of negative emotional traits including
general aggression, angry rumination, fear, guilt, sadness, fatigue, and composite
variables of negative affect. The PCC was also related to a broad range of traits
including behavioral approach, fear, and shyness. Surprisingly, activity in both the
ACC and PCC was correlated with the PANAS Serenity subscale, suggestive of
common neural mechanisms for calming and angry-arousing affective traits.
Brain Activity During Rumination is Correlated with Additional Personality Traits
Table 5 displays the correlations between the personality measures and brain
activity during rumination. Recall that during the rumination task, brain activity in
the left medial frontal gyrus was positively associated with individual differences in
trait displaced aggression. Activity in this region was also associated with a number
of affective traits, including positive associations with neuroticism, fear, and
negative affect composite variables, and negative associations with individual
differences in self-assurance and serenity. Activity in the ACC was also related to a
number of personality traits. More specifically, the left dorsal ACC was associated
with self-awareness and attentiveness. This is not surprising given, the relationship
of the dorsal ACC with discrepancy detection (Botvinick, Cohen, & Carter, 2004).
The right dorsal ACC was associated with behavioral approach. The right rostral
ACC was primarily associated with positive affective traits such as behavioral
approach, joviality, and positive affect composite variables. Social rejection
69
Table 5. Correlations between personality and BOLD response in brain regions active during directed rumination relative to distraction.
Personality ACC
(dorsal)
R L
ACC
(rostral)
R L
Insula
R L
PCC
R
1
R
2
L
1
L
2
Medial
frontal
gyrus
R L
Superior
frontal
gyrus
R L
Pre-
cuneus
L
Lateral
middle
frontal
gyrus
R L
Lateral
inferior
frontal
gyrus
L
Thal-
amus
L
AQ total score .10 .20 -.11 .16 .02 -.30 .16 -.04 -.10 -.03 .00 .22 -.07 .18 -.24 -.15 -.02 -.15 -.11
Anger subscale -.07 .00 -.13 .02 -.05 -.33 .03 -.23 -.16 -.03 -.01 .39 .07 .05 -.42* -.25 -.26 .04 -.32
Hostility
subscale
-.10 .05 -.30 .05 .03 -.21 .10 -.16 -.20 -.08 -.20 .24 -.17 -.05 -.38 -.27 -.17 .07 -.09
Physical
aggression
subscale
.10 .23 .12 .36 .11 -.15 .05 .18 .19 .10 .19 .22 .10 .42
+
.13 .26 .20 .30 -.14
Verbal
aggression
subscale
.37 .35 .05 .15 -.01 -.26 .29 .14 -.06 -.04 .10 -.12 -.15 .24 .01 -.08 .22 .10 .16
70
Table 5, Continued.
DAQ total score .16 .16 -.02 .11 .08 -.16 .11 .02 .07 .08 .23 .55* .21 .00 .10 -.06 -.20 -.06 -.24
Angry
Rumination
subscale
.10 .15 -.07 .12 .03 -.23 .11 -.06 -.04 .06 .23 .54* .21 .13 -.02 -.07 -.22 -.06 -.16
Revenge
planning
subscale
.19 .25 -.01 .11 .24 -.08 .14 .18 .14 .11 .20 .10 .09 .11 .17 .04 .09 .26 -.08
Behavioral
displaced
aggression
subscale
.14 .02 .05 .04 -.02 -.06 .03 -.03 .09 .05 .14 .66** .22 -.26 .15 -.12 -.34 -.30 -.37
Neuroticism .23 .17 -.06 -.00 .01 -.18 .16 .15 .16 .28 .39
+
.56* .25 -.11 .12 .01 -.13 .11 -.11
Self-awareness .36 .47* .21 -.09 .02 -.27 .15 .10 .14 .06 .11 -.32 -.22 .22 .16 .22 .32 .04 .17
Behavioral inhibition .42
+
.19 .30 -.26 -.22 -.30 .08 .26 .26 -.07 .28 .30 -.11 -.30 .21 .16 .11 .13 -.30
Behavioral approach .44* .32 .48* .16 -.06 -.33 .18 .21 .07 .17 .42
+
-.41
+
-.34 .30 .13 .15 .54* .29 .12
71
Table 5, Continued.
Rejection sensitivity -.14 -.37 -.19 -.19 -.70**-.61** -.39
+
-.27 -.30 .12 .08 .12 -.04 -.10 -.23 -.24 -.31 -.20 -.08
PANAS Subscales
Positive
emotion
.29 .42
+
.60** .09 .25 .00 .18 .19 .32 -.18 .05 -.15 -.19 .27 .18 .43
+
.50* -.01 -.22
Negative
emotion
-.11 .06 .04 -.09 .02 -.31 -.08 -.12 -.03 -.31 -.10 .57** .05 -.06 -.20 -.05 -.30 -.20 -.58**
Fear -.10 .05 -.02 -.13 -.16 -.39
+
-.22 -.17 -.04 -.29 -.09 .51* .05 -.17 -.17 -.01 -.29 -.19 -.55*
Hostility -.02 .36 .33 .31 .52* .23 .22 .05 .28 -.18 .01 .43
+
.17 .22 .08 .17 -.02 .03 -.35
Guilt .00 .02 .13 -.03 .11 -.46* -.05 -.19 -.13 -.17 -.00 .28 -.03 .22 -.38
+
-.01 -.07 -.07 -.34
Sadness -.25 -.10 -.19 .07 -.07 -.24 -.18 -.18 -.36 -.26 .02 .42
+
.08 -.10 -.21 -.08 -.33 -.16 -.29
Joviality .20 .16 .56* -.16 .15 -.03 .09 .10 .30 -.23 -.01 -.01 -.19 -.03 -.04 .28 .34 -.08 -.47*
Self-assurance -.06 .13 .20 .16 .09 -.12 -.17 .04 -.02 -.15 -.03 -.56* -.06 .43 -.03 .41
+
.41
+
.40
+
.21
Attentiveness .35 .52* .15 .02 .26 .25 .31 .32 .30 -.04 -.11 -.31 -.12 .24 .39
+
.28 .39
+
-.00 .24
72
Table 5, Continued.
Shyness .04 .23 -.03 .03 .19 .13 .17 -.07 .24 -.01 .05 .36 .28 -.09 .13 -.11 -.25 -.04 -.04
Fatigue -.30 .03 -.04 .00 .22 -.15 -.07 -.10 -.20 -.22 -.32 -.01 -.19 -.14 -.04 -.22 -.27 -.10 -.12
Serenity -.34 -.10 -.13 .35 .08 .09 -.25 -.21 -.38
+
-.29 -.44* -.48* -.03 .18 -.04 -.07 -.07 -.06 -.10
Surprise .17 .19 .42
+
-.02 .36 .00 .29 .08 .26 -.03 .06 .26 -.00 .13 -.04 .06 .12 -.05 -.34
Basic positive
affect
.14 .30 .45* -.01 .22 .03 .01 .04 .20 -.28 -.22 -.38 -.25 .25 .06 .32 .38 -.02 -.17
Basic negative
affect
-.13 .08 .05 .05 .01 -.31 -.10 -.17 -.12 -.29 -.02 .52* -.08 .04 -.24 .01 -.25 -.14 -.49*
** p < .01, *p < .05,
+
p < .10.
73
sensitivity was related to decreased activity in the insula bilaterally, whereas hostility
was associated with increased activity in the right insula.
State Emotional Reactions to the Provocation are Correlated with Brain Activity
No gender differences emerged in any of the 15 PANAS-X subscales
assessing emotional reactions to the provocation. The right dACC was uniquely
associated with feelings of guilt following the provocation. Other interesting results
included the correlation of the left insula – a region associated with sensitivity to
internal states (Critchley et al., 2004) – with feelings of attentiveness and fatigue.
Activity in the medial frontal gyrus was associated with increasing feelings of self-
assurance and activity in the right lateral middle frontal gyrus was associated with
decreasing feelings of serenity. Activity in the right hippocampus was associated
with increased positive emotion. Table 6 presents all of the correlations between the
reaction ratings and BOLD response.
4
4
One may find it seemingly contradictory that the PANAS Hostility trait measure did not display the
same pattern of results as the AQ trait Hostility subscale. By contrast, when used as a state measure
of the subjective reaction to the provocation, the PANAS Hostility subscale was associated with
activity in the ACC – the predicted region of interest known to be associated with anger (e.g., Murphy
et al., 2003; Phan et al., 2002). I conducted additional correlational analyses to reconcile these
findings. The PANAS Hostility trait measure was not significantly correlated with any of the AQ or
DAQ subscales or total scores. This suggests that the PANAS Hostility subscale lacks convergent
validity when used as a trait measure. This is consistent with prior work demonstrating that the
PANAS-X trait Hostility scale correlated approximately equally well with state measures of
depression and anxiety as it did with state measures of anger and hostility (Watson & Clark, 1992).
The authors themselves concluded “the convergent correlations among the anger/hostility measures
were only moderate” (p. 492). By contrast, the state PANAS Hostility measure was positively
correlated with the AQ total score, r = .65, p = .002, AQ Hostility subscale, r = .63, p = .003, AQ
Anger subscale, r = .66, p = .002, AQ Verbal Aggression subscale, r = .61, p = .004, DAQ total score,
r = .48, p = .03, and DAQ Angry Rumination subscale, r = .55, p = .01. These different patterns of
results indicate that in the current study, the PANAS-X Hostility subscale was a valid indicator of
hostile affect when used as a state measure, but lacked construct validity as an indicator of trait
hostility. This is not surprising given the similarity between the PANAS-X state Hostility subscale
and related state measures such as the POMS (McNair et al., 1992) and Mood Adjective Checklist
(Nowlis, 1965).
74
Table 6. Correlations between BOLD response in brain regions active after exposure to a verbal interpersonal provocation and state mood measures.
State Mood
Measure
ACC
(dorsal)
R L
ACC
(rostral)
R
1
R
2
L
Insula
R L
PCC
R L
1
L
2
Medial
frontal
gyrus
R
1
R
2
Medial
frontal
gyrus
(ventral)
L
Lateral
middle
frontal
gyrus
R L
Hippo-
campus
R L
Thal-
amus
L
PANAS
Subscales
Positive
emotion
-.07 .21 -.09 .36 .24 -.06 .03 -.03 .08 .07 .31 .14 .22 -.04 .17 .60** .06 .03
Negative
emotion
.44
+
.36 .30 .05 .08 .37 .19 .20 .41 -.06 .00 .39 .03 .18 -.21 .01 .27 .34
Fear .23 .13 . 33 .14 .25 .25 .14 .12 .38 .02 -.23 .17 .22 .11 -.18 .19 .19 .13
Hostility .40 .56* .27 -.17 -.05 .23 -.08 .08 .21 .14 .30 .41 -.15 -.11 -.05 -.06 -.05 .34
Guilt .58* .42 .26 .11 .04 .48
+
.27 .34 .49
+
.14 .13 .39 .03 .35 -.37 .01 .42 .43
+
75
Table 6, Continued.
Sadness -.01 .19 .39 -.23 -.10 .16 .01 .41 .25 .06 -.14 .33 -.18 -.13 -.40 .31 .22 .18
Joviality -.08 .10 .21 .20 .25 -.24 -.17 .12 -.03 .20 .23 .35 .20 -.11 .36 .40 .03 -.09
Self-
assurance
-.11 .28 .09 -.03 .01 .05 -.03 .01 .22 .18 .54* .03 .11
(.09)
-.23 .20 .45
+
-.03 .07
Attentiveness .17 .18 -.37 .43
+
.01 .34 .56* .01 .14 -.26 -.02 .19 .09 .32 -.17 .32 .34 .37
Shyness -.03 .05 .22 -.19 -.18 .22 .11 .44
+
.19 .09 -.25 .47
+
-.22 -.12 -.15 -.07 .30 .28
Fatigue -.26 .04 .17 -.15 .19 -.32 -.59* -.04 -.18 .23 .15 -.26 .08 -.19 .40 -.43
+
-.21 -.34
Serenity -.17 -.07 .02 -.08 .01 -.48
+
-.22 -.11 -.37 .03 -.12 -.07 -.02 -.52* .28 .18 -.39 -.36
Surprise -.06 .23 .02 -.08 -.01 -.29 -.10 .21 .37 .46
+
.22 .05 -.06 .14 -.08 .03 .16 .27
Basic positive
affect
.01 .26 -.07 .34 .14 .09 .21 .08 .15 .03 .31 .30 .13 .03 .16 .56
*
.19 .19
Basic
negative
affect
.41 .42
+
.39 -.03 .05 .37 .13 .31 .44
+
.12 .02 .42 -.02 .10 -.33 .14 .27 .36
** p < .01, *p < .05,
+
p < .10.
76
Neural Correlates During Rumination Task of Self-reported Rumination
In the self-focused rumination condition, self-reported rumination was
marginally related to activity in the left medial frontal gyrus, r = .42, p = .06,
consistent with the role of the medial PFC in attention to self-relevant cognition and
negative emotions, and inversely so in the left thalamus, r = -.41, p = .08. In the
provocation-focused rumination condition, there was a trend for self-reported
rumination about the provocation to be negatively related to activity in the right
PCC, r = -.41, p = .07. As expected, self-reported rumination in the distraction
condition was unrelated to brain activity.
Neural Correlates of Experimentally-induced Displaced Aggression
Activity in the right medial frontal gyrus was significantly related to the
laboratory physical displaced aggression measure (i.e., the cold pressor task), r = -
.50, p = .02, and marginally correlated with activity in the left rostral ACC, r = -.38,
p < .10, the left insula, r = -.43, p = .06, and the right middle frontal gyrus, r = -.39, p
= .09. These results are difficult to interpret due to the within-participants design
(i.e., all participants received the rumination and distraction conditions). Prior
between-subjects experiments have demonstrated that distraction reduces acts of
displaced aggression (Bushman et al., 2005; Denson et al., 2006, Experiment 2).
A secondary analysis consisted of comparing correlations between brain
activity during rumination and displaced aggression separately for those who
ruminated last and those who engaged in the distraction task last. Presumably, the
effects of rumination should be most pronounced for those who ruminated last. For
77
those who ruminated last, activity in the left inferior frontal gyrus was negatively
related to displaced aggression, r = -.87, p = .02. No other relationships emerged.
Separate Analyses of the Six Primary Objectives by Gender
This section describes separate, exploratory analyses for men and women. All
correlations are zero-order coefficients. Results should be interpreted cautiously due
to unstable estimates associated with the relatively small sample sizes (i.e., 8 men
and 12 women).
Objective #1: Brain Regions Active Following an Interpersonal Provocation
Tables 7 and 8 present the data for the hypothesized regions active following
the provocation for men and women, respectively. If the exact coordinates active in
the whole sample analyses were not active, nearby voxels within the ROIs were
examined for activity. These revised coordinates are presented in the tables. It
appears that men (Table 7) failed to demonstrate activity in the left rostral ACC, left
ventromedial PFC, right hippocampus, and left thalamus. However, this is likely due
to lessened statistical power. Indeed, at a relaxed statistical threshold (p < .10), all of
the same regions were active for men and women. The results for the women (Table
8) replicated the results observed for the sample as a whole.
78
Table 7. Brain regions active after exposure to a verbal interpersonal provocation for men (N = 7).
Region Talairach No. of voxels M(SE) % Significance Test
coordinates signal change
x y z
Anterior cingulate (dorsal)
Right 8 24 34 536 0.80 (0.09) t(6) = 9.03, p < .001
Left -7 22 33 583 0.83 (0.10) t(6) = 8.48, p < .001
Anterior cingulate (rostral)
Right
region1
5 32 15 176 1.00 (0.20) t(6) = 5.13, p < .01
Insula
Right 42 2 10 55 0.59 (0.09) t(6) = 6.87, p < .001
Left -42 2 10 9 0.64 (0.14) t(6) = 4.55, p < .01
Posterior cingulate
Right 5 -52 21 706 0.80 (0.12) t(6) = 6.98, p < .001
Left
region1
-7 -44 23 520 0.66 (0.11) t(6) = 6.29, p < .001
Left
region2
-2 -21 28 25 0.82 (0.19) t(6) = 4.23, p < .01
79
Table 7, Continued.
Medial frontal gyrus (i.e., medial PFC)
Right
region1
6 44 25 464 0.74 (0.10) t(6) = 7.81, p < .001
Right
region2
-7 37 27 277 0.76 (0.09) t(6) = 8.53, p < .001
Lateral middle frontal gyrus (i.e., lateral PFC)
Right 39 33 25 472 0.70 (0.06) t(6) = 11.03, p < .0001
Left -32 47 9 376 1.19 (0.12) t(6) = 9.59, p < .0001
Hippocampus
Left -27 -37 2 215 0.78 (0.11) t(6) = 6.90, p < .001
80
Table 8. Brain regions active after exposure to a verbal interpersonal provocation for women (N = 9).
Region Talairach No. of voxels M(SE) % Significance Test
coordinates signal change
x y z
Anterior cingulate (dorsal)
Right 8 24 34 653 0.60 (0.07) t(8) = 8.73, p < .0001
Left -8 27 29 365 0.72 (0.11) t(8) = 6.72, p < .001
Anterior cingulate (rostral)
Right
region1
5 33 11 409 0.53 (0.06) t(8) = 8.93, p < .0001
Right
region2
4 30 -7 514 1.37 (0.18) t(8) = 7.56, p < .0001
Left -3 33 -8 607 1.31 (0.13) t(8) = 10.55, p < .00001
Insula
Right 37 -2 7 289 0.53 (0.08) t(8) = 6.37, p < .001
Left -37 4 15 645 0.68 (0.08) t(8) = 8.12, p < .0001
Posterior cingulate
Right 5 -52 21 202 0.76 (0.13) t(8) = 5.64, p < .001
81
Table 8, Continued.
Left
region1
-7 -44 23 118 0.52 (0.07) t(8) = 7.65, p < .0001
Left
region2
-2 -21 28 43 0.61 (0.09) t(8) = 6.73, p < .001
Medial frontal gyrus (i.e., medial PFC)
Right
region1
6 48 19 560 0.81 (0.11) t(8) = 7.40, p < .0001
Right
region2
5 45 19 296 0.80 (0.12) t(8) = 6.93, p < .001
Medial frontal gyrus (i.e., ventromedial PFC)
Left -5 32 -11 425 1.37 (0.12) t(8) = 11.25, p < .00001
Lateral middle frontal gyrus (i.e., lateral PFC)
Right 33 47 7 785 0.88 (0.06) t(8) = 13.84, p < .00001
Hippocampus
Right 30 -31 -3 411 0.59 (0.08) t(8) = 7.02, p < .001
Left -30 -31 -3 1,045 0.65 (0.08) t(8) = 8.66, p < .0001
Thalamus
Left -13 -10 3 77 0.62 (0.10) t(8) = 6.25, p < .001
82
Objective #2: The Relationship of the Anterior Cingulate with Self-reported Angry
Feelings in Response to the Provocation
The relationship between left dorsal ACC activity and state anger was
significant for women, r = .81, p < .01, but not for men, r = .04, p = .93. This might
be due to the restricted range in men’s left dorsal ACC activity (values ranged from
.24 to 1.28) relative to women (values ranged from -.49 to 1.44). Indeed the range of
women’s scores was nearly twice that of the men’s scores (range
women
= 1.93 vs.
range
men
= 1.04). Thus, although it appears that there was a very strong relationship
between state anger and dorsal ACC activity in women but not men, statistical
considerations preclude strong conclusions about gender differences.
Objective #3: Brain Regions Active During Rumination
Tables 9 and 10 present the data for the hypothesized regions active during
the rumination tasks relative to distraction for men and women, respectively. It
appears that men (Table 9) failed to demonstrate activity in the left rostral ACC,
lateral inferior frontal gyrus, and left thalamus. As was the case for brain regions
active following provocation, this also was likely due to decreased statistical power.
Indeed, at a relaxed statistical threshold (p < .10) men and women showed activity in
the same regions. As was the case for the provocation data, the results for the
women (Table 10) replicated the results observed for the sample as a whole.
83
Table 9. Brain regions active during rumination relative to distraction (Provocation-focused and Self-focused > Distraction contrast) for men (N = 8).
Region Talairach No. of voxels M(SE) % Significance test
coordinates signal change
x y z
Anterior cingulate (dorsal)
Right 7 18 31 30 0.58 (0.11) t(7) = 5.21, p < .01
Left -6 16 31 7 0.66 (0.14) t(7) = 4.78, p < .01
Anterior cingulate (rostral)
Right 3 35 10 9 0.65 (0.13) t(7) = 5.15, p < .01
Insula
Right 41 -2 11 124 0.73 (0.12) t(7) = 6.29, p < .001
Left -41 -3 10 87 0.85 (0.15) t(7) = 5.81, p < .001
Posterior cingulate
Right 6 -16 39 25 0.62 (0.10) t(7) = 6.15, p < .001
Left
region1
-6 -15 35 90 0.70 (0.14) t(7) = 4.93, p < .01
Left
region2
-6 -55 19 119 0.51 (0.09) t(7) = 5.44, p < .001
84
Table 9, Continued.
Medial frontal gyrus (i.e., medial PFC)
Right 9 50 14 276 1.01 (0.16) t(7) = 6.38, p < .001
Left -9 44 24 440 0.94 (0.10) t(7) = 9.39, p < .0001
Superior frontal gyrus
Right 7 48 31 109 0.64 (0.16) t(7) = 3.99, p < .01
Left -9 44 32 325 0.91 (0.13) t(7) = 7.03, p < .001
Precuneus
Left -9 -53 49 223 0.63 (0.09) t(7) = 7.21, p < .001
Lateral middle frontal gyrus (i.e., Lateral PFC)
Right 40 38 14 122 0.66 (0.06) t(7) = 11.70, p < .00001
Left -34 14 33 299 0.68 (0.09) t(7) = 7.74, p < .001
85
Table 10. Brain regions active during rumination relative to distraction (Provocation-focused and Self-focused > Distraction contrast) for women (N = 12).
Region Talairach No. of voxels M(SE) % Significance test
coordinates signal change
x y z
Anterior cingulate (dorsal)
Right 7 15 35 656 0.99 (0.12) t(11) = 8.11, p < .00001
Left -7 15 35 824 0.82 (0.14) t(11) = 5.84, p < .001
Anterior cingulate (rostral)
Right 3 35 9 728 0.90 (0.15) t(11) = 5.96, p < .00001
Left -6 34 10 556 0.69 (0.10) t(11) = 6.87, p < .0001
Insula
Right 38 -3 7 669 0.80 (0.13) t(11) = 6.09, p < .0001
Left -40 -2 1 604 0.73 (0.13) t(11) = 5.79, p < .001
Posterior cingulate
Right
region1
6 -16 39 922 0.84 (0.10) t(11) = 8.83, p < .00001
Right
region2
6 -53 25 505 0.76 (0.13) t(11) = 5.84, p < .001
Left
region1
-6 -15 35 578 0.78 (0.10) t(11) = 7.46, p < .0001
86
Table 10, Continued.
Left
region2
-6 -58 22 458 0.66 (0.13) t(11) = 5.09, p < .001
Medial frontal gyrus (i.e., medial PFC)
Right 9 42 15 969 0.94 (0.13) t(11) = 7.17, p < .0001
Left -9 50 19 104 0.64 (0.11) t(11) = 5.75, p < .001
Superior frontal gyrus
Right 7 48 31 169 0.91 (0.21) t(11) = 4.32, p < .01
Left -9 46 33 371 0.92 (0.18) t(11) = 5.06, p < .001
Precuneus
Left -9 -55 32 750 0.69 (0.11) t(11) = 6.28, p < .0001
Lateral middle frontal gyrus (i.e., Lateral PFC)
Right 36 45 15 492 0.89 (0.19) t(11) = 4.69, p < .001
Left -37 46 16 379 1.55 (0.34) t(11) = 4.53, p < .001
Lateral inferior frontal gyrus (i.e., Lateral PFC)
Left -50 23 15 689 0.98 (0.14) t(11) = 7.17, p < .0001
87
Table 10, Continued.
Thalamus
Left -9 -21 11 116 0.67 (0.15) t(11) = 4.64, p < .001
88
Objective #4: Hippocampus Activity is Correlated with Self-reported Rumination
When separate analyses were conducted by gender, no significant effects
emerged. Specifically, the correlation between the left hippocampus activity and
self-reported rumination was not significant for men, r = .65, p = .12, or women, r =
.35, p = .35, most likely due to loss of statistical power. Activity in the right insula
was correlated with the degree of self-reported rumination for women, r = .70, p <
.04, but not for men, r = .46, p = .30. Activity in the right rostral ACC and was
correlated with self-reported rumination roughly equally for women, r = .76, p < .02,
and men, r = .74, p < .06. Activity in the left PCC was marginally significant for
men, r = .73, p < .07, but not for women, r = -.09, p = .83. Thus, whereas both
genders likely have a common substrate for brain activity related to subsequent
rumination (i.e., the rostral ACC and possibly the right insula), women and men
showed unique activity in this regard, (viz. the left PCC for men). This latter result
could not be attributed to a restricted range of values.
Objective #5: Unique Patterns of Brain Activity Associated with Trait Direct
Aggression and Trait Displaced Aggression
Following the provocation, the AQ was primarily uniquely associated with
activity in the left dorsal ACC (the same region associated with self-reported anger)
for women, r = .76, p < .02, but not men, r = -.07, p = .89. As noted above, there
was a restricted range for men on activity in the left dorsal ACC. Moreover, the
range of scores on the AQ were much smaller for men (values from 3.36 to 4.86)
89
than women (values from 1.38 to 4.23). Indeed, the range of AQ scores was nearly
twice as large for women than men (range
women
= 2.85, range
men
= 1.51).
Following the provocation, the DAQ was primarily associated with activity in
the right medial frontal gyrus (i.e., medial PFC) for men, r = .95, p = .001, but not
women, r = .27, p = .48. Thus, there was a double-dissociation between personality
and gender for brain activity following the provocation. This may also be due to a
broader range of DAQ scores for men than women (range
men
= 3.19, range
women
=
2.10). The ranges of right medial PFC activity were comparable (.77 vs. 83 for men
and women, respectively). Alternatively, this might mean that men respond
differently to this provocation than do women.
During the rumination task, activity in the left medial frontal gyrus was
correlated approximately equally for men, r = .63, p = .09, and women, r = .53, p
=.08, but only marginally significant due to loss of statistical power. Also, during the
rumination task, the DAQ was positively associated with activity in the right medial
frontal gyrus for men, r = .75, p = .03, but not women, r = .08, p = .81. This latter
finding could also be attributable to a greatly restricted range of activity in the right
medial frontal gyrus for men relative to women (range
men
=.76, range
women
= 2.15).
As stated above, men also had a broader range of DAQ scores. Thus, these results
may reflect real gender differences or be the result of statistical artifact. Further
research is needed.
90
Objective #6: Frontal Asymmetry and the Behavioral Approach System
Separate analyses for men and women were presented in the preceding
section.
Functional Signal Loss
Numerous studies have demonstrated that the ventral PFC (including the
orbitofrontal cortex) is involved in anger recall and viewing angry faces (Murray et
al., 2003; Phan et al., 2002). The current study found activity only in the left
ventromedial PFC and not at all in the orbitofrontal cortex. Moreover, because my
functional imaging protocol consisted of axial slices, these slices may not have been
optimal for detecting activity in this region due to the ventral PFC’s extended
horizontal boundary. I therefore conducted a post hoc analysis where I examined the
degree of signal loss in the ventral PFC.
This analysis consisted of (1) creating anatomical ROI masks individually
from each subject from their 3D anatomical image, (2) overlaying the first whole-
brain functional volume (i.e., all 29 slices) from the baseline period, (3) determining
the mean functional signal intensity in the ventral PFC, and (4) determining the mean
functional signal intensity in a control region that does not have a boundary with
cerebrospinal fluid (i.e., an undistorted region), and (5) using the following formula
to calculate the approximate percentage of signal loss in the ventral PFC:
X
X X
d Undistorte
VentralPFC d Undistorte
91
This analysis resulted in a 41% mean signal loss (SD = 10%) in the ventral PFC
relative to undistorted brain regions. Thus, it is likely that the signal in the left
ventromedial PFC and orbitofrontal cortex may have been too weak to reach
statistical significance in the current experiment, i.e., a potential Type II error.
CHAPTER 4: DISCUSSION
Summary
This experiment was the first to investigate the neural correlates of
interpersonal provocation, directed rumination, trait direct aggression, and trait
displaced aggression. The current study revealed which brain regions are active in
response to an insult delivered by an experimenter as well as which brain regions are
active during subsequent rumination. Further, evidence for the role of particular
brain regions in phenomenological experience was obtained. Specifically, activity in
the left dorsal ACC following the provocation was correlated with self-reported
feelings of anger, and activity in the hippocampus following the provocation was
positively correlated with self-reported rumination during the subsequent
provocation-focused rumination task. The two aggressive personality dimensions
were also associated with distinct and theoretically predicted patterns of brain
activity. Trait direct aggression was correlated with increased activity following the
provocation in the dorsal ACC, whereas trait displaced aggression was correlated
with increased activity in the medial PFC both following the provocation and during
rumination. Finally, my data provided strong support for the motivational direction
model of frontal asymmetry. Specifically, men showed greater relative left
92
asymmetry in the lateral PFC following the provocation, and both genders showed
greater left asymmetry in this same region during rumination. Further, this activity
was correlated with individual differences in behavioral approach orientation both
among men in response to the provocation and in both genders during rumination.
The Neural Correlates of Provocation
The interpersonal provocation increased activity in the rostral and dorsal
ACC, PCC, insula, lateral PFC, medial PFC, ventromedial PFC, hippocampus, and
left thalamus. These findings are largely consistent with prior research on anger and
aggression and the findings of two recent meta-analyses that summarized the limited
number of neuroimaging studies on anger (e.g., Damasio et al., 2000; Murphy et al.,
2003; Murray et al., 2006; Phan et al., 2002). The current experiment represents a
meaningful methodological departure from prior neuroimaging experiments in that
previous studies of anger and aggression have relied on autobiographical episodic
recall (Damasio et al., 2000; Dougherty et al., 1999; Kimbrell et al., 1999), angry
faces as stimuli (Adams et al., 2003; Blair et al., 1999; Sprengelmeyer et al., 1998;
Whalen et al., 2001), or violent media in children (Murray et al., 2006; Ritterfeld et
al., 2006). By using a provocation that modeled a real-world anger-inducing
situation, the current experiment provides a relatively high level of ecological
validity, despite participants being in an MRI scanner. Specifically, participants were
unjustly led to believe that they were not intelligent enough to follow the simple
instruction to speak louder on the task. In fact, all participants did speak louder when
instructed to do so. Thus, the provocation represented the unjustified delivery of an
93
insult. An additional advantage of using an actual provocation is that artifactual
results due to stimuli presentation were primarily eliminated. For example, some
prior research had demonstrated that the orbitofrontal cortex (often referred to as the
lateral ventromedial PFC) is active during anger (e.g., Damasio et al., 2000; Phan et
al., 2002). I did not find this activity, suggesting that the orbitofrontal cortex may be
primarily elicited by anger recall or visual stimuli such as angry faces.
Brian activity following the provocation was also related to subjective
emotional reactions to the provocation. Perhaps the two most intriguing findings are
that activity in the left dorsal ACC was uniquely correlated with feelings of hostility
and activity in the right dorsal ACC was uniquely correlated with feelings of guilt.
As was the case for the trait measures, the correlations with the subjective measures
are consistent with literature demonstrating the role of the ACC in affective
phenomenology. For example, Eisenberger et al. (2003) reported that the dorsal
ACC was related to subjective feelings of emotional distress following social
exclusion. Also of interest were the relationships between various brain regions and
self-reported rumination during the provocation-focused rumination task.
Participants who demonstrated increased activity in the left hippocampus, right
insula, right rostral ACC, and left PCC following the provocation also reported that
they thought more about the anger-inducing experience during the anagram task
when directed to do so.
Taken together, the ACC responses revealed that the left dorsal ACC was
associated with feelings of hostility, and the right dorsal ACC was associated with
94
feelings of guilt and rumination. Feelings of guilt are preceded by the recognition of
one’s wrongdoing and self-blame (e.g., “I did a bad thing”) (Tangney, 2002). Thus,
activity found in the right dorsal ACC suggests that it may be associated with a self-
conscious feeling of having done something wrong which motivates perseverative
cognition (i.e., rumination) to resolve the issue. At first glance, the occurrence of
self-blame when one is the recipient of an unjustified provocation may seem
surprising. However, individuals often blame themselves for even terrible
occurrences outside of their control (Foa, Elhers, Clark, Tolin, & Orsillo, 1999).
Further, self-blame appeared to be a reasonable response given the nature of the
provocation used in the current experiment, whereby a high status experimenter
communicated (albeit rudely) that the participant was not speaking loud enough.
Thus, some participants may have inferred that their actions were invalidating a quite
important and expensive experiment.
The Neural Correlates of Rumination
Rumination increased activity in the dorsal and rostral ACC, PCC, medial
PFC, precuneus, insula, lateral PFC, and left thalamus. With a relaxed statistical
threshold, only small differences between the two types of rumination appeared in
the right middle frontal gyrus, left precuneus, and right PCC. Interestingly,
consistent with Johnson et al.’s (in press) hypothesis about the inward versus
outward focus of self-relevant social cognition, provocation-focused rumination
produced more activity in the right PCC, possibly due to its outward social focus.
Thus, although provocation-focused and self-focused rumination are distinct in terms
95
of content, the two types of rumination appear to share the overlapping neural
substrates. This notion was further supported by the rumination manipulation
checks. Participants did indeed report thinking about the provocation more during
the provocation-focused rumination task than either the self-focused rumination
(when they presumably thought more about themselves) or distraction tasks. Thus,
although the content of the rumination differed, the same regions were active with
very similar intensity. This high degree of similarity between both types of
rumination is consistent with behavioral experiments demonstrating that both
provocation-focused and self-focused rumination increase negative affect and
aggression when preceded by a provocation (Bushman, 2002; Bushman et al., 2005;
Denson et al., 2006, Experiment 2; Pedersen et al., 2006).
The role of the medial PFC during rumination is largely consistent with a
growing body of evidence demonstrating that this region is associated with social
cognition. The medial PFC is active when thinking about the self or others relative
to objects (for a review, see Amodio & Frith, 2006). Broadly relevant to the current
study, the medial PFC is active during impression formation (Mason & Macrae,
2004), thinking about one’s emotional state (Ochsner et al., 2004), and making
attributions and determining the mental state of others (Gallagher & Frith, 2003;
Harris, Todorov, & Fiske, 2005), all of which can occur during rumination. These
results are also consistent with the only study to investigate individual differences in
rumination (Ray et al., 2005). These authors reported that a composite variable of
rumination was correlated with brain activity in the ACC and medial PFC in
96
response to negatively valenced photos. In the current study, activity in the medial
PFC was marginally related (p = .06) to increased self-reported rumination.
The activity in the insula during rumination is also of interest. This region is
constantly incorporating physiological information from the body, which researchers
have suggested forms the neural substrate for the subjective sense of self and feeling
states (Craig, 2003; Critchley et al., 2004; Damasio, 1994; Damasio, 2003). The
thalamus has similarly been implicated (Damasio, 2003), thus it is not surprising that
both the thalamus and insula were active both following the provocation and during
rumination.
The Role of Personality
In addition to the main effects of the provocation and rumination
manipulations on brain activity, the present analyses demonstrated that trait direct
aggression and trait displaced aggression exacerbate responses in some brain regions,
but not others. Specifically, whereas trait direct aggression was primarily associated
with increased activity in the left dorsal ACC, trait displaced aggression was
primarily associated with increased activity in the right medial PFC (as was
neuroticism) following the provocation. This pattern of results is consistent with my
prior research demonstrating that individuals high in trait displaced aggression
behave differently than those high in trait direct aggression following provocation
(e.g., Denson et al., 2006a). Those high in trait direct aggression react with
immediate anger as evidenced by the dorsal ACC activity, whereas those high in trait
displaced aggression ruminate about the provocation as evidenced by the medial PFC
97
activity. In this regard, when confronted with a provocation, individuals high in trait
displaced aggression appear to be more similar to individuals high in neuroticism
than those high in trait direct aggression. Indeed, both neuroticism and trait
displaced aggression are associated with an inward focus and sensitivity to one’s
negative mood state.
There were some particularly novel findings related to the Physical
Aggression subscale of the Aggression Questionnaire (Buss & Perry, 1992).
Individuals who score high on this subscale tend to engage frequently in physical
acts of aggression. This subscale was associated with increased activity in the PCC,
insula, hippocampus, and thalamus. Recall that Murray et al. (2006) found PCC
activity in children who viewed boxing clips. These authors suggested that PCC
activity represents the activation of aggressive scripts stored in memory. My findings
are consistent with their theorizing as well as connectionist models of aggression
(e.g., Berkowitz, 1993), whereby provocation activates aggressive concepts in
memory. For instance, when individuals high in physical aggression encounter an
interpersonal provocation, this may prime knowledge structures and memories
related to physical aggression to a greater extent than those low in physical
aggression. Meta-analytic evidence demonstrated that the left PCC also plays a role
in the recall of autobiographical memories (Svoboda, McKinnon, & Levine, 2006),
suggesting that those high in physical aggression may activate physical aggression-
related autobiographical memories and scripts. In other words, these results provide
98
evidence consistent with a neurophysiological basis for priming physical aggression
(e.g., Berkowitz & LePage, 1967; Todorov & Bargh, 2002).
Another interesting finding related to trait physical aggression was its
positive relationship with the hippocampus. Because the hippocampus is involved in
orienting responses (e.g., Yamaguchi et al., 2006; Williams et al., 2000) and memory
encoding (e.g., Kensinger, Clarke, & Corkin, 2003), it appears that those high in
physical aggression may have paid greater attention to the provocation (i.e., an
orienting response) and encoded the provocation to a greater extent than those low in
physical aggression. Indeed, one would expect that those high in physical aggression
would orient themselves toward the provocation because it may have greater
meaning for them than those low in trait physical aggression. Such individuals
possess a repertoire and history of aggressive responding to provocations. This
pattern of hippocampal activity following a provocation may produce a cycle of
aggression such that those high in physical aggression pay much attention to and
remember provocations quite vividly, which increases their accessibility, further
increasing the likelihood of physical aggression.
There were also other notable relationships between other affective traits and
brain activity following the provcation. Specifically, whereas activity in the left
dorsal ACC was primarily associated with trait general aggression (especially trait
hostility), activity in the right rostral ACC was associated with a number of emotion-
related traits including guilt, sadness, fatigue, serenity, and general negative affect.
Activity in the ACC was not related to any positive emotional traits. This is
99
consistent with a large body of evidence demonstrating the role of the ACC in a wide
range negative emotions (Murphy et al., 2003; Phan et al., 2002) as well as one study
that found a correlation between a composite variable consisting of trait anger-
hostility and dorsal ACC reactivity following social exclusion (Eisenberger et al., in
press).
As was the case with brain activity following the provocation, a number of
personality variables moderated the effect of directed rumination on neural
responses. As expected, trait displaced aggression was related to exacerbated
activity in the medial PFC. This same region of the left medial PFC also
demonstrated increased reactivity to a number of other negatively valenced
personality factors including neuroticism, fear, and negative affect composite
measures, and was related to decreased activity among those high in self-assurance
and serenity. In contrast to the negatively valenced emotional traits, activity in the
left lateral PFC was associated with the positive emotion composite and behavioral
approach.
Prefrontal Asymmetry
The current data are also consistent with work on the motivational direction
model of frontal asymmetry. Specifically, the present study used neuroimaging
technology to replicate the EEG asymmetry finding that greater left PFC activity was
observed in men following a provocation (Harmon-Jones & Sigelman, 2001). It is
notable that the experiment by Harmon-Jones and Sigelman (2001) contained only
male participants. Moreover, this difference in activation was also associated with
100
individual differences in behavioral approach motivation. Together, these data
indicate a neurophysiological difference between men and women following a
provocation, which could be related to gender differences in anger and aggressive
behavior. In other words, men may be “hard wired” to respond immediately with
approach tendencies when confronted with an interpersonal provocation. However,
during rumination, both genders displayed relatively greater left activity, suggesting
that rumination increased approach motivation following a provocation in both men
and women. Moreover, this asymmetry was related exclusively to individual
differences in behavioral approach, offering conceptual supporting evidence for the
relationship between frontal EEG asymmetry and the personality dimension of
behavioral approach (Coan & Allen, 2003).
Limitations
As with any experiment, there were a number of limitations associated with
the current study. Due to practical considerations, the directed rumination task was
conducted within-participants rather than between-participants. While ideal in many
ways for identifying the neural substrates of rumination, the inclusion of a distraction
condition for all participants and the counterbalanced order led to a problematic
interpretation of the relationship of brain activity with the physical displaced
aggression measure (i.e., the cold pressor).
Another shortcoming of the current experiment was the temporal placement
of the state mood measures that assessed the reaction to the provocation at the end of
the experiment. This was done for two reasons. First, I wanted to assess a wide
101
variety of emotional reactions in the current study. To do so, would have required an
excessive burden on participants if they were asked to rate their mood state on all 65
items in the scanner. Second, I did not want to arouse suspicion of the experimental
hypotheses by inquiring about the participants’ mood immediately following the
provocation. Lindsay and Anderson (2000) have argued that such temporal
placement is necessary in aggression research due to pronounced order effects when
variables such as angry affect or cognition are assessed prior to the main dependent
variable of interest (in this case, brain activity). Nonetheless, the state measures
appeared to possess good construct validity (see footnote 3) and others and myself
have found successful results with this placement of the mood measures in the past
(Denson et al., 2006a; Denson et al., in press; Lindsay & Anderson, 2000; Vasquez
et al., 2005).
Yet another shortcoming concerns the lack of manipulation checks relating to
self-focus. Although the manipulation checks concerning rumination about the
provocation revealed the hypothesized pattern of results, one cannot be sure if
participants did actually engage in self-focused rumination during the self-focus task
since I did not ask about the degree of self-focus. However, we know that during the
self-focused rumination task, participants did not think about the provocation to a
greater extent than during the distraction task. Moreover, and as expected,
participants thought about the provocation to a significantly lesser extent than during
the provocation-focused rumination task.
102
Another limitation is that the present data do not allow firm conclusions
about gender differences due to (a) the small sample size for each gender, and (b)
potentially misleading correlations due to the restricted range of scores on brain
activity and the aggressive personality measures. Future research with larger groups
of men and women is required to determine whether the relationship between brain
activity and aggressive personality actually differs for each gender when the data in
both groups possess comparable statistical properties. Further research could also
address how phenomenological experience varies as a function of personality,
gender, and brain activity.
Finally, a considerable degree of signal loss reduced sensitivity for detecting
activity in the ventral PFC and orbitofrontal cortex. Although there was activity in
the left ventromedial PFC following the provocation, it is not certain whether
additional activity would have been detected if the amount of signal loss in these
regions were minimized. Future work employing coronal slices and shorter TE
durations may yet uncover activity in these regions as a result of anger and
rumination.
Future Research
In light of the findings from the current study, there appear to be quite a few
promising research avenues to explore. I identified brain regions that are active
following an interpersonal provocation and demonstrated that activity in the dorsal
ACC is related to subjective feelings of hostility and guilt. Future research could
link activity in these areas with real world phenomenology. For instance, using
103
longitudinal diary methods, activity in the dorsal ACC could be used to
prosepectively predict subsequent distress following daily hassles in real world
settings. Such findings would lend ecological validity to the current experiment’s
results.
Another fruitful avenue to explore would be to connect brain reactivity to
provocations with other physiological responses that constitute the body’s response
to stress. The neural correlates of cognitive and affective activity during rumination
produce physically observable outcomes in the body proper. These effects must
have their origin in the brain. According to the reactivity hypothesis of disease,
individuals who tend to demonstrate high levels of cardiovascular reactivity to
stressful experiences may be at increased risk for negative cardiovascular outcomes
(Lepore, 1998). Therefore, this extended activation of the stress response is likely to
lead to poor cardiovascular health (e.g., Sapolsky, 1998). Understanding the role of
the central nervous system (i.e., the brain) during rumination may therefore have
considerable implications for improving physical health.
A smaller group of studies has documented the effects of rumination on
additional peripheral measures. Laboratory manipulations of rumination increased
cardiovascular arousal (Pedersen et al., 2005; Denson et al., 2006b). Compared to
people who reported low levels of rumination, those who ruminate more reported
more physical health deficits (Thomsen et al., 2004a), increased blood pressure
(Hogan & Linden, 2004), dysregulation of immune function (Thomsen et al., 2004b),
and decreased sleep quality (Thomsen, Mehlsen, Christensen, & Zachariae, 2003).
104
Some have suggested that rumination may be a risk factor for coronary artery disease
(Kubazansky, Davidson, & Rozanski, 2005).
Brosschot, Gerin, and Thayer (2006) reviewed the only six published
psychoneuroimmunology studies to contain a state or trait measure of worry or
rumination. With the exception of one study with a failed manipulation, the
remaining five studies found that worry and rumination were associated with
increased cortisol and altered immune parameters. Moreover, a larger meta-analysis
of laboratory stressors found that rumination exacerbated hypothalamic-pituitary-
adrenal (HPA) axis activity (i.e., cortisol) and multiple measures of immune
reactivity (Denson, Spanovic, & Miller, 2006c). Unfortunately, I did not find
activity in the hypothalamus in response to the provocation or during the directed
rumination task in the current study and at least one experiment has demonstrated
decreased cortisol was associated with anger during a laboratory stress manipulation
(Lerner, Gonzalez, Dahl, Hariri, & Taylor, 2005). Future experimental work is
needed to explore these issues.
Another noteworthy area of research is related to the fairly powerful
moderating effects of personality on brain reactivity observed in the present study.
Our moderate to large correlations between personality and brain activity is
consistent with the growing body of literature that has investigated these
relationships (e.g., Canli et al., 2001, 2004; Eisenberger et al., 2005; Eisenberger et
al., in press; Johnson et al., in press; Kumari et al., 2004; Ray et al., 2005; Siegle et
al., 2002). Longitudinal functional imaging studies that incorporate genetic and
105
social risk factors could potentially determine how children and adolescents develop
aggressive personalities. Such insight could eventually lead to interventions
designed to decrease the development of aggressive personality in at risk individuals.
Identifying these patterns of brain activity may eventually aid in reducing the harm
caused by those high in general and trait displaced aggression. Although speculative
at this point, developing interventions based on methods that reduce brain activity in
the areas related to angry affect and rumination identified in the current study may
lead to decreased displaced aggression among those high in the trait. Perhaps, in
conjunction with additional neuroimaging studies that investigate manipulations of
meditation or relaxation training, such interventions could soon become reality.
Conclusion
Although previous research has investigated the neural correlates of anger
indirectly through the use of recall or facial stimuli, the current experiment was the
first to provide evidence into the neural reactivity to an interpersonal insult and link
this activity to the subjective experience of anger. This was also the first experiment
was to reveal brain activity during rumination following a provocation. Moreover,
this is first demonstration that individual differences associated with direct and
displaced aggression are differentially associated with exacerbated reactivity to areas
associated with negative affective experience and self-relevant cognition. Together,
these findings increase our understanding of the neural processes associated with the
risk for aggressive behavior.
106
Bibliography
Abler, B., Walter, H., & Erk, S. (2005). Neural correlates of frustration.
NeuroReport, 16, 669-672.
Adams Jr., R.B., Gordon, A.A., Ambady, N., & Kleck, RE. (2003). Effect of gaze on
amygdala sensitivity to anger and fear faces. Science, 300, 1536.
Amodio, D. M., & Frith, C. D. (2006). Meeting of minds: The medial frontal cortex
and social cognition. Nature Reviews Neuroscience, 7, 268-277.
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of
Psychology, 53, 27-51.
Archer, J., & Webb, I. A. (2006). The relation between scores on the Buss-Perry
Aggression Questionnaire and aggressive acts, impulsiveness,
competitiveness, dominance, and sexual jealousy. Aggressive Behavior, 32,
464-473.
Austin, J. H. (1998). Zen and the brain: Toward an understanding of meditation and
consciousness. Cambridge, MA, US: MIT Press.
Aviles, F., Earleywine, M., Pollock, V., Stratton, J., & Miller, N. (2005). Alcohol’s
effect on triggered displaced aggression. Psychology of Addictive Behaviors,
19, 108-111.
Axelrod, R. (1984). The evolution of cooperation. New York, NY: Basic Books.
Barnwell, S. S., Borders, A., & Earleywine, M. (2006). Alcohol-aggression
expectancies and dispositional aggression moderate the relationship between
alcohol consumption and alcohol-related violence. Aggressive Behavior, 32,
517-525.
Bechara, A., Damasio, H., & Damasio, A.R. (2000). Emotion, decision making and
the orbitofrontal cortex. Cerebral Cortex, 295-307.
Berkowitz, L. (1989). Frustration-aggression hypothesis: Examination and
reformulation. Psychological Bulletin, 106, 59-73.
Berkowitz, L. (1990). On the formation and regulation of anger and aggression: A
cognitive-neoassociationistic analysis. American Psychologist, 45, 494-503.
Berkowitz, L. (1993). Aggression: Its causes, consequences, and control. New York:
McGraw-Hill.
107
Berkowitz, L., & LePage, A. (1967). Weapons as aggression-eliciting stimuli.
Journal of Personality and Social Psychology, 7, 202-207.
Bettencourt, B. A., Talley, A., Benjamin, A. J., & Valentine, J. (2006). Personality
and aggressive behavior under provoking and neutral conditions: A meta-
analytic review. Psychological Bulletin, 132, 751-777.
Blair, R. J. R., Morris, J. S., Frith, C. C., Perrett, D. I., & Dolan, R. J. (1999).
Dissociable neural responses to facial expressions of sadness and anger.
Brain: A Journal of Neurology, 122, 883-893.
Botvinick, M.M., Cohen, J. D., & Carter, C. S. (2004) Conflict monitoring and
anterior cingulate cortex: An update. TRENDS in Cognitive Sciences, 8, 539-
546.
Brooks, J. C., Nurmikko, T. J., Bimson, W. E., Singh, K. D., & Roberts, N. (2002).
fMRI of thermal pain: Effects of stimulus laterality and attention.
Neuroimage, 16, 909-919.
Brown, R. P., & Phillips, A. (2005). Letting bygones be bygones: Further evidence
for the validity of the Tendency to Forgive scale. Personality and Individual
Differences, 38, 627-638.
Brosschot, J. F., Gerin, W., & Thayer, J. F. (2006). The perseverative cognition
hypothesis: A review of worry, prolonged stress-related physiological
activation, and health. Journal of Psychosomatic Research, 60, 113-124.
Bushman, B. J. (2002). Does venting anger feed or extinguish the flame? Catharsis,
rumination, distraction, anger and aggressive responding. Personality and
Social Psychology Bulletin, 28, 724-731.
Bushman, B. J., Bonacci, A. M., Pedersen, W. C., Vasquez, E. A., & Miller, N.
(2005). Chewing on it can chew you up: Effects of rumination on triggered
displaced aggression. Journal of Personality and Social Psychology, 88, 969-
983.
Bushman, B. J., & Wells, G. L. (1998). Trait aggressiveness and hockey penalties:
Predicting hot tempers on the ice. Journal of Applied Psychology, 83, 969-
974.
Buss, A. H. & Perry, M. (1992). The Aggression Questionnaire. Journal of
Personality & Social Psychology, 63, 452-459.
108
Canli, T., Desmond, J. E., Zhao, Z., Glover, G., & Gabrieli, J. D. E. (1998).
Hemispheric asymmetry for emotional stimuli detected with fMRI.
NeuroReport, 9, 3233-3239.
Canli, T., Zhao, Z., Desmond, J. E., Kang, E., Gross J., & Gabrieli, J. D. E. (2001).
An fMRI study of personality influences on brain reactivity to emotional
stimuli. Behavioral Neuroscience, 115, 33-42.
Canli, T., Amin, Z., Haas, B., Omura, K., & Constable, R.T. (2004). A double
dissociation between mood states and personality traits in the anterior
cingulate. Behavioral Neuroscience, 118, 897-904.
Caprara, G. V. (1986). Indicators of aggression: The Dissipation-rumination scale.
Personality and Individual Differences, 7, 763-769.
Carver, C. S., & White, T. L. (1994). Behavioral inhibition, behavioral activation,
and affective responses to impending reward and punishment: The BIS/BAS
Scales. Journal of Personality and Social Psychology, 67, 319-333.
Carver, C. S., Scheier, M. F., & Weintraub, J. K. (1989). Assessing coping strategies:
A theoretically based approach. Journal of Personality & Social Psychology,
56, 267-283.
Chang, E. C. (2004). Distinguishing between ruminative and distractive responses in
dysphoric college students: Does indication of past depression make a
difference? Personality and Individual Differences, 36, 845-855.
Coan, J. A., & Allen, J. B. (2003). Frontal EEG asymmetry and the behavioral
activation and inhibition systems. Psychophysiology, 40, 106-114.
Coan, J. A., & Allen, J. B. (2003). The state and trait nature of frontal EEG
asymmetry in emotion. In K. Hugdahl & R. J. Davidson (Eds.), The
asymmetrical brain, (pp. 565-615). Cambridge, MA, US: MIT Press.
Cohen, R. A., Paul, R., Zawacki, T. M., Moser, D. J., Sweet, L., & Wilkinson, H.
(2001). Emotional and personality changes following cingulotomy. Emotion,
1, 38-50.
Craig, A D. (2003). Interoception: The sense of the physiological condition of the
body. Current Opinion in Neurobiology, 13, 500-505.
Craig, A. D., Chen, K., Brandy, D., & Reiman, E. M. (2000). Thermosensory
activation of the insular cortex. Nature Neuroscience, 3, 184-190.
109
Critchley, H. D., Wiens, S., Rotshtein, P., Öhman, A., & Dolan, R. J. (2004). Neural
systems supporting interoceptive awareness. Nature Neuroscience, 7, 189-
195.
Damasio, A. R. (1994). Descartes' error: Emotion, reason, and the human brain.
New York, NY: Harper Collins.
Damasio, A. R. (2003). Mental self: The person within. Nature, 423, 227.
Damasio, A.R., Grabowski, T.J., Bechara, A., Damasio, H., Ponto, L.L.B., Parvizi,
J., & Hichway, R.D. (2000). Subcortical and cortical brain activity during the
feeling of self-generated emotions. Nature Neuroscience, 3, 1049-1056.
Damasio, H., Grabowski, T., Frank, R., Galaburda, A. M., & Damasio, A. R. (2005).
The return of Phineas Gage: Clues about the brain from the skull of a famous
patient. In J. t. Cacioppo & G. G. Berntson (Eds.). Social neuroscience: Key
readings (pp. 21-28). New York, NY: Psychology Press.
Davidson, R. J. (1998). Anterior electrophysiological asymmetries, emotion, and
depression: Conceptual and methodological conundrums. Psychophysiology,
35, 607-614.
Davis, K. D. (2000). The neural circuitry of pain as explored with functional MRI.
Neurological Research, 22, 313-317.
Denson, T. F., Aviles, F., Pollock, V., Earleywine, M., Vasquez, E. A., & Miller, N.
(in press). The effects of alcohol and the salience of aggressive cues on
triggered displaced aggression. Aggressive Behavior.
Denson, T. F., Pedersen, W. C., & Miller, N. (2006a). The displaced aggression
questionnaire. Journal of Personality and Social Psychology, 90, 1032-1051.
Denson, T. F., Pedersen, W. C., & Miller, N. (2006b). Trait displaced aggression,
rumination, and health outcomes. Manuscript submitted for publication.
Denson, T. F., Spanovic, M., & Miller, N. (2006c). Cognitive appraisals and
emotions predict cortisol and immune responses: A Meta-analysis of acute
laboratory social stressors and emotion inductions. Manuscript in
preparation.
Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., & Sears, R. R. (1939).
Frustration and aggression. Oxford, England: Yale Univ. Press.
110
Dougherty, D. D., Shin, L. M., Alpert, N. M., Pitman, R. K., Orr, S. P., Lasko, M.,
Macklin, M. L., Fishman, A. J., & Rauch, S. L. (1999). Anger in healthy men:
A PET study using script-driven imagery. Biological Psychiatry, 46, 466-472.
Downey, G., & Feldman, S. I. (1996). Implications of rejection sensitivity for
intimate relationships. Journal of Personality and Social Psychology, 70,
1327-1343.
Duval, S., & Wicklund, R. A. (1972). A theory of objective self awareness. Oxford,
UK: Academic Press.
Eisenberger, N. I., & Lieberman, M. D. (2004). Why rejection hurts: a common
neural alarm system for physical and social pain. Trends in Cognitive
Sciences, 8, 294-300.
Eisenberger, N. I., Lieberman, M. D., & Williams, K. D. (2005). Does rejection hurt?
An fMRI study of social exclusion. Science, 302, 290-292.
Eisenberger, N. I., Lieberman, M. D., & Satpute, A. B. (2005). Personality from a
controlled processing perspective: An fMRI study of neuroticism,
extraversion, and self-consciousness. Cognitive, Affective, and Behavioral
Neuroscience, 5, 169-181.
Eisenberger, N. I., Way, Baldwin, M., Taylor, S. E., Welch, W. T., & Lieberman, M.
D. (in press). Understanding genetic risk for aggression: Clues from the
brain’s response to social exclusion. Biological Psychiatry.
Farrell, M. J., Laird, A. R., & Egan, G. F. (2005). Brain activity associated with
painfully hot stimuli applied to the upper limb: A meta-analysis. Human
Brain Mapping, 25, 129-139.
Foa, E. B., Ehlers, A., Clark, D. M., Tolin, D. F., & Orsillo, S. M. (1999). The
Posttraumatic Cognitions Inventory (PTCI): Development and validation.
Psychological Assessment, 11, 303-314.
Fridhandler, B. & Averill, J. R. (1982). Temporal dimensions of anger: An
exploration of time and emotion. In J. R. Averill (Ed.), Anger and
aggression: An essay on emotion (pp. 253-280). New York: Springer-Verlag.
Gallagher, H. L., & Frith, C. D. (2002). Functional imaging of ‘theory of mind’.
Trends in Cognitive Sciences, 7, 77-83.
Geen, R. G. (1990). Human aggression. Belmont, CA, US: Thomson Brooks/Cole
Publishing.
111
Genovese, C. R., Lazar, N. A., & Nichols, T. (2002). Thresholding of statistical
maps in functional neuroimaging using the false discovery rate. NeuroImage,
15, 870-878.
Goldberg, L. R., Johnson, J. A., Eber, H. W., Hogan, R., Ashton, M. C., Cloninger,
C. R., & Gough, H. C. (2006). The International Personality Item Pool and
the future of public-domain personality measures. Journal of Research in
Personality, 40, 84-96.
Gouldner, A. W. (1960). The norm of reciprocity. American Sociological Review,
25, 161-178.
Gow, A. J., Whiteman, M. C., Pattie, A., & Deary, I. J. (2005). Goldberg’s ‘IPIP’
Big-Five factor markers: Internal consistency and concurrent validation in
Scotland. Personality and Individual Differences, 39, 317-329.
Harmon-Jones, E. (2003). Clarifying the emotive functions of asymmetrical frontal
cortical activity. Psychophysiology, 40, 838-848.
Harris, L. T., Todorov, A., & Fiske, S. T. (2005). Attributions on the brain: Neuro-
imaging dispositional inferences, beyond theory of mind. NeuroImage, 28,
763-769.
Hovland, C., & Sears, R. (1940). Minor studies in aggression: VI. Correlation of
lynchings with economic indices. Journal of Psychology, 9, 301-310.
Huesmann, L. R. (1998). The role of social information processing and cognitive
schema in the acquisition and maintenance of habitual aggressive behavior. In
R. G. Geen and E. Donnerstiein (Eds.) Human aggression: Theories,
research, and implications for social policy. (pp. 73-109). San Diego, CA:
Academic Press, Inc.
International Personality Item Pool: A Scientific Collaboratory for the Development
of Advanced Measures of Personality Traits and Other Individual Differences
(http://ipip.ori.org/). Internet Web Site.
Jackson, P. L., Brunet, E., Meltzoff, A. N., & Decety, J. (2006). Empathy examined
through the neural mechanisms involved in imagining how I feel versus how
you feel pain. Neuropsychologia, 44, 752-761.
Johnson, M. K., Raye, C. L., Mitchell, K. J., Touryan, S. R., Greene, E. J., & Nolen-
Hoeksema, S. (in press). Dissociating medial frontal and posterior cingulate
activity during self-reflection. Social Cognitive and Affective Neuroscience.
112
Johnson, S. C., Baxter, L. C., Wilder, L. S., Pipe, J. G., Heiserman, J. E., &
Prigatano, G. P. (2002). Neural correlates of self-reflection. Brain, 125, 1808-
1814.
Kachadourian, L. K., Fincham, F., & Davila (2005). Attitudinal ambivalence,
rumination, and forgiveness of partner transgressions in marriage. Personality
and Social Psychology Bulletin, 31, 334-342.
Kensinger, E. A., Clarke, R. J., & Corkin, S. (2003). What neural correlates underlie
successful encoding and retrieval? A functional magnetic resonance imaging
study using a divided attention paradigm. Journal of Neuroscience, 23, 2407-
2415.
Kimbrell, T. A., George, M. S., Parekh, P. I., Ketter, T. A., Podell, D. M., Danielson,
A. L., Repella, J. D., Benson, B. E., Willis, M. W., Herscovitch, P., & Post,
R. M. (1999). Regional brain activity during transient self-induced anxiety
and anger in healthy adults. Biological Psychiatry, 46, 454-465.
Konecni, V. J. (1974). Self-arousal, dissipation of anger, and aggression. Personality
and Social Psychology Bulletin, 1, 192-194.
Kross, E., Ayduk, O., & Mischel, W. (2005). When asking "why" does not hurt:
Distinguishing rumination from reflective processing of negative emotions.
Psychological Science, 16, 709-715.
Kumari, V., Ffytche, D. H., Williams, S. C. R., & Gray, J. A. (2004). Personality
predicts brain responses to cognitive demands. The Journal of Neuroscience,
24, 10636-10641.
Kubazansky, L. D., Davidson, K. W., & Rozanski, A. (2005). The clinical impact of
negative psychological states: Expanding the spectrum of risk for coronary
artery disease. Psychosomatic Medicine, 67, S10-S14.
Lancaster, J. L., Summerln, J. L., Rainey, L., Freitas, C.S., Fox, P.T. (1997), The
Talairach Daemon, a database server for Talairach Atlas Labels. Neuroimage
5, S633.
Lane R. D., Fink, G. R., Chau, P. M-L., & Dolan, R. J. (1997). Neural activation
during selective attention to subjective emotional responses. NeuroReport, 8,
3969-3972.
LeDoux, J. (1996). The emotional brain: The mysterious underpinnings of emotional
life. New York, NY: Touchstone.
113
Lepore, S. J. (1998). Problems and prospects for the social support-reactivity
hypothesis. Annals of Behavioral Medicine, 20, 257-269.
Lerner, J. S., Gonzalez, R. M., Dahl, R. E., Hariri, A. R., & Taylor, S. E. (2005).
Facial expressions of emotion reveal neuroendocrine and cardiovascular
stress responses. Biological Psychiatry, 58, 743-750.
Lindsay, J. J., & Anderson, C. A. (2000). From antecedent conditions to violent
actions: A general affective aggression model. Personality and Social
Psychology Bulletin, 26, 533-547.
Lotze, M., Veit, R., Anders, S. & Birbaumer, N. (2007). Evidence for a different role
of the ventral and dorsal medial PFC for social reactive aggression: An
interactive fMRI study. NeuroImage, 34, 470-478.
Lyubomirsky, S., & Nolen-Hoeksema, S. (1995). Effects of self-focused rumination
on negative thinking and interpersonal problem solving. Journal of
Personality and Social Psychology, 69, 176-190.
Macrae, C. N., Moran, J. M., Heatherton, T. F., Banfield, J. F., & Kelley, W. M.
(2004). Medial prefrontal activity predicts memory for self. Cerebral Cortex,
14, 647-654.
Marcus-Newhall, A., Pedersen, W. C., Carlson, M., & Miller, N. (2000). Displaced
aggression: A meta-analytic review. Journal of Personality and Social
Psychology, 78, 670-689.
Mason, M. F., & Macrae, C. N. (2004). Categorizing and individuating others: The
neural substrates of person perception. Journal of Cognitive Neuroscience,
16, 1785-1795.
McCullough, M. E., Bellah, C. G., Kilpatrick, S. D., & Johnson, J. L. (2001).
Vengefulness: Relationships with forgiveness, rumination, well-being, and
the Big Five. Personality and Social Psychology Bulletin, 27, 601-610.
McNair, D. M., Lorr, M., & Droppelman, L. F. (1992). EDITS manual for the Profile
of Mood States. San Diego, CA: Educational and Testing Service.
Miller, N., Pedersen, W. C., Earleywine, M., & Pollock, V. E. (2003). A theoretical
model of triggered displaced aggression. Personality and Social Psychology
Review, 7, 75-97.
114
Mischel, W., & Shoda, Y. (1995). A cognitive-affective system theory of
personality: Reconceptualizing situations, dispositions, dynamics, and
invariance in personality structure. Psychological Review, 102, 246-268.
Morrison, R., & O’Connor, R. C. (2005). Predicting psychological distress in college
students: The role of rumination and stress. Journal of Clinical Psychology,
61, 447-460.
Muris, P. Roelofs, J., Meesters, C., & Boomsma, P. (2004). Rumination and worry in
nonclinical adolescents. Cognitive Therapy and Research, 28, 539-554.
Murphy, F. C., Nimmo-Smith, I., & Lawrence, A. D. (2003). Functional
neuroanatomy of emotions: A meta-analysis. Cognitive, Affective, and
Behavioral Neuroscience, 3, 207-233.
Murray, J.P., Liotti, M., Ingmundson, P.T., Mayberg, H.S., Pu, Y., Zamarripa, F.,
Liu, Y., Woldorff, M.G., Gao, J., & Fox, P.T. (2006). Children’s brain
activations while viewing televised violence revealed by fMRI. Media
Psychology, 8, 25-37.
Nolen-Hoeksema, S., & Morrow, J. (1991). A prospective study of depression and
posttraumatic stress symptoms after a natural disaster: The 1989 Loma Prieta
earthquake. Journal of Personality and Social Psychology, 61, 115-121.
Nolen-Hoeksema, S., & Morrow, J. (1993). Effects of rumination and distraction on
naturally occurring depressed mood. Cognition and Emotion, 7, 561-570.
Nolen-Hoeksema, S., Morrow, J., & Fredrickson, B. L. (1993). Response styles and
the duration of episodes of depressed mood. Journal of Abnormal
Psychology, 102, 20-28.
Nowlis, V. (1965). Research with the mood adjective checklist. In S. S. Tompkins &
C. S. Izard (Eds.) Affect, cognition, and personality: Empirical studies.
Oxford, UK: Springer Verlag.
Ochsner, K.N., Bunge, S.A., Gross, J.J., & Gabrieli, J.D.E. (2002). Rethinking
feelings: An fMRI study of the cognitive regulation of emotion. Journal of
Cognitive Neuroscience, 14, 1215-1229.
Ochsner, K.N., Knierim, K., Ludlow, D.H., Hanelin, J., Ramachandran, T., Glover,
G., & Mackey, S.C. (2004). Reflection upon feelings: An fMRI study of
neural systems supporting the attribution of emotion to self and other. Journal
of Cognitive Neuroscience, 16, 1746-1772.
115
Paradiso, S., Johnson, D. L., Andreason, N. C., O’Leary, D. S., Watkins, G. L.,
Ponto, L. L. B., & Hichwa, R. D. (1999). Cerebral blood flow changes
associated with attribution of emotional valence to pleasant, unpleasant, and
neutral visual stimuli in a PET study of normal subjects. American Journal of
Psychiatry, 156, 1618-1629.
Pedersen, W. C., Denson, T. F., Goss, R. J., Vasquez, E. A., & Miller, N. (2006).
The impact of rumination on aggressive thoughts, feelings, behavior, and
arousal. Manuscript submitted for publication.
Pedersen, W. C., Gonzales, C., & Miller, N. (2000). The moderating effect of trivial
triggering provocation on displaced aggression. Journal of Personality and
Social Psychology, 78, 913-927.
Phan, K.L., Wager, T., Taylor, S.F., Liberzon, I. (2002). Functional neuroanatomy of
emotion: A meta-analysis of emotion activation studies in PET and fMRI.
NeuroImage, 16, 331-348.
Phan, K. L., Wager, T.D., Taylor, S.F., & Liberzon, I. (2004). Functional
neuroimaging studies of human emotions. CNS Spectrums, 9, 258-266.
Phan, K. L., Taylor, S. F., Welsh, R. C., Decker, L. R., Noll, D. C., Nichols, T. E.,
Britton, J. C., & Liberzon, I. (2003). Activation of the medial PFC and
extended amygdala by individual ratings of emotional arousal: A fMRI study.
Biological Psychiatry, 53, 211-215.
Raine, A., Buchsbaum, M., & LaCasse, L. (1997). Brain abnormalities in murderers
indicated by positron emission tomography. Biological Psychiatry, 42, 495-
508.
Raine, A., Meloy, J. R., Bihrle, S., Stoddard, J., LaCasse, L., & Buchsbaum, M. S.
(1998). Reduced prefrontal and increased subcortical brain functioning
assessed using positron emission tomography in predatory and affective
murderers. Behavioral Sciences and the Law, 16, 319-332.
Ray, R.D., Ochesner, K.N., Cooper, J.C., Roberston, E.R., Gabrieli, J.D.E., & Gross,
J.J. (2005) Individual differences in trait rumination and the neural systems
supporting cognitive reappraisal. Cognitive, Affective, & Behavioral
Neuroscience, 5, 156-168.
Rusting, C. L., & Nolen-Hoeksema, S. (1998). Regulating responses to anger:
Effects of rumination and distraction on angry mood. Journal of Personality
and Social Psychology, 74, 790-803.
116
Sapolsky, R. M. (1998). Why zebras don’t get ulcers: An updated guide to stress,
stress-related diseases, and coping. New York, NY, US: Freeman and
Company.
Segerstrom, S. C., Tsao, J. C., Alden, L. E., & Craske, M. G. (2000). Worry and
rumination: Repetitive thought as a concomitant and predictor of negative
mood. Cognitive Therapy and Research, 24, 671-688.
Shacham, S. (1983). A shortened version of the Profile of Mood States. Journal of
Personality Assessment, 47, 305-306.
Siegle, G. J., Steinhauer, S. R., Thase, M. E., Stenger, A., & Carter, C. S. (2002).
Can’t shake that feeling: Event-related fMRI assessment of sustained
amygdala activity in response to emotional information in depressed
individuals. Biological Psychiatry, 51, 693-707.
Singer, T., Seymour, B., O’Doherty, J., Kaube, H., Dolan, R. J., & Frith, C. D.
(2004). Empathy for pain involves the affective but not sensory components
of pain. Science, 303, 1157-1162.
Sprengelmeyer, R., Rausch, M., Eysel, U. T., & Przuntek, H. (1998). Neural
structure associated with recognitionof facial expressions of basic emotions.
Proceedings of the Royal Society. London, Series B, 265, 1927–1931.
Sukhodolsky, D. G., Golub, A., & Cromwell, E. N. (2001). Development and
validation of the anger rumination scale. Personality and Individual
Differences, 31, 689-700.
Talairach, J., & Tournoux, P. (1988). Co-planar stereotaxic atlas of the human
brain. New York, NY, US: Thieme.
Tangney, J. P. (2002). Self-conscious emotions: The self as a moral guide. In A.
Tesser, D. A. Stapel, & J. V. Wood (Eds.) Emerging psychological
perspectives (pp. 97-117). Washington, DC, US: American Psychological
Association.
Thomsen, D. K., Mehlsen, M. Y., Hokland, M., Viidik, A., Olesen, F., & Avlund, K.
et al. (2004b). Negative thoughts and health: Associations among rumination,
immunity, and health care utilization in a young and elderly sample.
Psychosomatic Medicine, 66, 363-371.
117
Thomsen, D. K., Mehlsen, M. Y., Olesen, F., Hokland, M., Viidik, A., & Avlund, K.
et al. (2004a). Is there an association between rumination and self-reported
physical health? A one-year follow-up in a young and an elderly sample.
Journal of Behavioral Medicine, 27, 215-231.
Thomsen, D. K., Mehlsen, M. Y., Christensen, S., & Zachariae, R. (2003).
Rumination--relationship with negative mood and sleep quality. Personality
and Individual Differences, 34, 1293-1301.
Todorov, A., & Bargh, J. A. (2002). Automatic sources of aggression. Aggression
and Violent Behavior, 7, 53-68.
Trapnell, P. D., & Campbell, J. D. (1999). Private self-consciousness and the five-
factor model of personality: Distinguishing rumination from reflection.
Journal of Personality and Social Psychology, 76, 284-304.
Tyson, P. D. (1998). Physiological arousal, reactive aggression, and the induction of
an incompatible relaxation response. Aggression and Violent Behavior, 3,
143-158.
Yamaguchi, S., Hale, L. A., D’Esposito, M., & Knight, R. T. (2006). Rapid
prefrontal-hippocampal habituation to novel events. Journal of Neuroscience,
24, 5356-5363.
Vasquez, E.A., Denson, T. F., Pedersen, W. C., Stenstrom, D. M., & Miller, N.
(2005). The moderating effect of trigger intensity on triggered displaced
aggression. Journal of Experimental Social Psychology, 41, 61-67.
Watkins, E. (2004). Adaptive and maladaptive ruminative self-focus during
emotional processing. Behaviour Research and Therapy, 42, 1037-1052.
Watson, D. & Clark, L. A. (1994). The PANAS-X: manual for the positive and
negative affect schedule – expanded form.
Weber, R., Ritterfield, U., & Mathiak, K. (2006). Does playing violent video games
induce aggression? Empirical evidence of a functional magnetic resonance
imaging study. Media Psychology, 8, 39-60.
Whalen, P.J., Shin, L.M., McInerney S.C., Fischer, H., Wright, C. I., & Rauch, S. L.
(2001). A functional MRI study of human amygdala responses to facial
expressions of fear versus anger. Emotion, 1, 71-83.
Wilcox, R. R. (2003). Applying contemporary statistical techniques. San Diego, CA,
US: Academic Press.
118
Wilcox, R. R. (1998). How many discoveries have been lost by ignoring modern
statistical methods? American Psychologist, 53, 300-314.
Williams, L. M., Brammer, M. J., Skerrett, D., Lagopolous, J., Rennie, C., Kozek,
K., et al. (2000). The neural correlates of orienting: An integration of fMRI
and skin conductance orienting. Neuroreport, 11, 3011-3015.
www.lapdonline.org. Retrieved 01-17-06 from
http://www.lapdonline.org/general_information/
crime_statistics/2005_crime_summary.htm.
Zimmer, C. (2005, November). The neurobiology of the self. Scientific American,
pp. 93-101.
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
The neural correlates of skilled reading: an MRI investigation of phonological processing
PDF
The moderating effect of triggering provocation on direct aggression
PDF
The neural correlates of creativity and perceptual pleasure: from simple shapes to humor
PDF
Behavioral and neural influences of interoception and alexithymia on emotional empathy in autism spectrum disorder
PDF
Heart, brain, and breath: studies on the neuromodulation of interoceptive systems
Asset Metadata
Creator
Denson, Thomas Frederick
(author)
Core Title
The angry brain: neural correlates of interpersonal provocation, directed rumination, trait direct aggression, and trait displaced aggression
School
College of Letters, Arts and Sciences
Degree
Doctor of Philosophy
Degree Program
Psychology
Publication Date
04/12/2007
Defense Date
03/05/2007
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
aggression,aggressive personality,Anger,displaced aggression,fMRI,neuroimaging,OAI-PMH Harvest,rumination
Language
English
Advisor
Miller, Norman (
committee chair
), Lickel, Brian (
committee member
), Lu, Zhong-Lin (
committee member
), Nezami, Elahe (
committee member
), Read, Stephen J. (
committee member
)
Creator Email
denson@usc.edu
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-m380
Unique identifier
UC1325430
Identifier
etd-Denson-20070412 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-403724 (legacy record id),usctheses-m380 (legacy record id)
Legacy Identifier
etd-Denson-20070412.pdf
Dmrecord
403724
Document Type
Dissertation
Rights
Denson, Thomas Frederick
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Repository Name
Libraries, University of Southern California
Repository Location
Los Angeles, California
Repository Email
cisadmin@lib.usc.edu
Tags
aggression
aggressive personality
displaced aggression
fMRI
neuroimaging
rumination