University of Southern California Dissertations and Theses

Children's triggered displaced aggression

(USC Thesis Other)

Children's triggered displaced aggression

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content CHILDREN’S TRIGGERED DISPLACED AGGRESSION
by
Robin L. Toblin
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)
August 2007
Copyright 2007 Robin L. Toblin
ii
Acknowledgements
I am grateful for the dedication and time put forth by the following research
assistants: Christopher Bravo, Victor Cota, Stephanie Green, Michael Kane, Emily
Spears, Michael Pasternak, Hedieh Bolour, and Katie Jackson. I would also like to
thank Dr. Tom Halvorsen (Superintendent) and Dr. Vicki Hernandez (Principal) of
the Savanna School District for allowing the study to be conducted at Hansen
Elementary School. I appreciate the insightful and useful comments made by Drs.
Gayla Margolin, Stanley J. Huey, Jr., and Michele D. Kipke, who served on my
dissertation committee. Finally, I would like to thank Drs. David Schwartz and
Norman Miller for their financial support of the project, and most importantly, for
their mentorship throughout the study.
iii
Table of Contents
Acknowledgements ii
List of Tables v
Abstract vi
Introduction 1
Childhood Aggression 1
Triggered Displaced Aggression 3
Displaced Aggression 3
Triggered Displaced Aggression 5
The General Aggression Model (GAM) 6
and Other Theories of Aggression
GAM Framework 7
Inputs 8
Present Internal States 11
Outcomes 13
Children’s Triggered Displaced Aggression 14
Within the GAM Framework
The Present Study 17
Rationale 17
Hypotheses 17
Potential Confounding Variables 18
Exploratory Hypotheses 20
Method 20
Overview 20
Participants 22
Recruitment of Sample 22
Informed Consent 25
Sample Size Analysis 26
Procedure 27
Procedural Outline 27
Experimental Manipulation 29
Other Measures 33
iv
Results 36
Manipulation Checks 36
Provocation 36
Trigger 37
Triggered Displaced Aggression 38
Group Difference for Potential Confounders 41
Exploratory Analyses 42
Discussion 47
Future Directions 57
References 61
Appendix A. Recruitment Flyer 67
Appendix B. Protocol for CTDA: Children 68
Appendix C. Questionnaire Measures 76
v
List of Tables
Table 1. Demographic and questionnaire variables by experimental group 44
Table 2. Bivariate correlations among questionnaires 46
vi
Abstract
The present study aimed to validate the construct of triggered displaced
aggression (TDA) in a middle childhood sample. The sample was comprised of 9-12
year old boys from the Los Angeles area (n=28) who were randomly placed into four
groups: provocation plus trigger, provocation only, trigger only, or no provocation
and no trigger. Participants and their caregivers completed a series of questionnaires.
Participants then played in a series of two computer games which provided the
vehicle for the provocation and trigger manipulations, respectively. There were no
significant differences on aggression between any of the four groups. Exploratory
analyses found that children who were more likely to believe they could harm their
opponent were more likely to make an aggressive move in the game. Potential
explanations for the lack of significant results are discussed. Limitations and future
directions for research are also addressed. The current study can serve as pilot data
for future researchers who aim to validate the TDA construct in a childhood sample.
1
Introduction
Research on childhood aggression is largely based upon individual
differences. While such characteristics are also examined in adult populations, a
large body of literature in social psychology has focused on situational factors which
may encourage aggressive behavior exceeding the typical level expected for a
person’s individual traits. In particular, a great deal of research has been conducted
on the arousing nature of frustration resulting in aggression as its endpoint. One
noteworthy subtype of situational aggression in adulthood that has received attention
in recent years is triggered displaced aggression (TDA). Triggered displaced
aggression occurs when a person is provoked, but is unable to retaliate against the
provocateur and, subsequently, acts aggressively toward a different person who
provides a second triggering provocation. I hope to extend the childhood aggression
and TDA literatures by demonstrating empirical support for this construct in a
middle childhood/early adolescent population and placing it in the greater context of
theories of the development of aggression in children.
Childhood Aggression
Human aggression is defined as “any behavior directed toward another
individual that is carried out with the proximate intent to cause harm” (Anderson and
Bushman, 2002, p. 28). The operationalization of aggression changes as children
develop, however. By the time children reach middle childhood, often thought of as
ages 6-11, children’s aggressive behavior has already gone through a number of
2
changes (Loeber & Hay, 1997). As infants, children begin showing anger and rage
and certain behavioral signs of aggression (Loeber & Hay, 1997; Tremblay et al.,
2004). As toddlers, children display temper tantrums and aggressive behaviors
including biting, kicking, and hitting. They also have similar risk factors for high
levels of aggression to those of older children with high levels of aggression
including low family income, low levels of maternal education, and hostile and
inconsistent parenting (Tremblay et al., 2004; Cote, Vaillancourt, LeBlanc, Nagin, &
Tremblay, 2006). Toddlers are also more likely to display instrumental types of
aggression than retaliatory types of aggression (Loeber & Hay, 1997). There are few,
if any, gender differences prior to school entry (Loeber & Hay, 1997).
Upon interacting with a larger social group as children enter the context of
school, distinct gender differences begin to emerge, typically by age 4 (Lahey,
Waldman, & McBurnett, 1999). Males begin to display more aggression as a whole.
For both genders, retaliatory aggression surfaces (Loeber & Hay, 1997). During this
time (i.e., the preschool years), however, most children display a reduction in overall
physical aggression from early childhood and, instead, use interpersonal skills to
resolve conflicts. It is for these reasons that Tremblay and colleagues (Tremblay et
al., 2004; Cote et al., 2006) suggest that the preschool years are the best time to
intervene to prevent aggression.
Only a subsection of children still engage in high rates of aggressive
behaviors during the school-age years (i.e., middle childhood; Loeber & Hay, 1997).
A recent nationally- representative sample of Canadian children found that 16.6% of
3
all children and 20.0% of boys are on a trajectory of demonstrating consistently high
levels of physical aggression that begin as young as young as age two and continue
all the way through middle childhood (Cote et al., 2006). A significant body of
literature has been amassed which suggests that the age of onset of problematic
levels of aggression is a critical factor in deciding the trajectory of later behaviors
(Lahey et al., 1999). Children who display unusually high amounts of aggressive
behavior at younger ages and continue into adolescence are at greater risk for
negative outcomes including academic failure, criminal charges, violence,
relationship problems, and financial difficulties than those who do not engage in
aggression until they are adolescents (Lahey et al., 1999) and may cost society
upwards of two million dollars over the course of their lifetime (Dodge, 2006).
Elucidating mechanisms by which a child may aggress helps to understand the full
picture of human aggression and assists in the creation of interventions to reduce
aggression (Anderson & Bushman, 2002).
Triggered Displaced Aggression
Displaced Aggression
Triggered displaced aggression is related to the phenomenon of displaced
aggression. Displaced aggression is an attack against a person/object that is not the
original source of frustration. This may occur because the original provocateur is
unavailable (e.g., no longer in the person’s presence), intangible (e.g., traffic jam), or
could implement punishment if provoked (e.g., a child’s teacher). This construct saw
4
an explosion in the literature following Dollard et and colleagues 1939 monograph
(Dollard, Doob, Miller, Mowrer, and Sears, 1939), but received little attention from
social psychology following that burst. There were no empirical studies of displaced
aggression in the 1990s and textbooks rarely mentioned the phenomenon (Marcus-
Newhall, Pedersen, Carlson, & Miller, 2000). A meta-analytic review of displaced
aggression demonstrated, however, that displaced aggression is a robust and
consistent finding across experimental studies (Marcus-Newhall et al., 2000).
Further, similarity between the target and original provocateur increases displaced
aggression, but similarity between the participant and the target does not. Finally, the
more negative the setting, the greater the displaced aggression. The various
moderators provided numerous ways to experimentally manipulate the various
factors contributing to the field’s understanding of displaced aggression.
Although there is a body of literature on displaced aggression during
adulthood (Miller, Pedersen, Earleywine, & Pollock, 2003), studies investigating
displaced aggression in children are few and far between. A few elements of the
phenomenon have been elucidated through these studies. Children’s understanding
of displaced aggression increases with age (Miller & DeMarie-Dreblow, 1990) with
children as young as 5 having some understanding of it (Weiss & Miller, 1983). By
age 9, children are aware that a person may have discrepant mental and behavioral
states (i.e., the overt-covert discrepancy) (Miller & DeMarie-Dreblow, 1990.) Moore
(1964) found that children did, in fact, engage in displaced aggression. Specifically,
when given a choice of cards with varying images to shoot, boys who lost a card
5
game targeted their shooting at a card most similar to the image on the original card
while boys that did not lose chose targets that were dissimilar to the original image
(Moore, 1964.) This is consistent with meta-analytic findings that increased
similarity between provocateur and target is positively related to the level of
displaced aggression (Miller et al., 2003.) As a whole, very little work has been
conducted regarding displaced aggression and children.
Triggered Displaced Aggression
Another relevant and similarly understudied topic is that of triggered
displaced aggression (TDA). Of greater interest than displaced aggression for its
potential level of ecological validity (Miller & Marcus-Newhall, 1997; Pedersen,
Gonzales, & Miller, 2000) and intervention implications, TDA describes the
phenomenon in which someone aggresses against a minor triggering provocation or
frustration following a strong initial provocation that prohibits retaliation. Dollard et
al. (1939) wrote about the potential influence of minor triggers. They also wrote
about the synergistic effect that a provocation and a subsequent minor trigger could
have eliciting an aggressive response of greater magnitude than either event would
educe on its own. Few empirical studies have tested this proposal, however. Prior to
Miller and colleagues’ studies (Miller & Marcus-Newhall, 1997; Pedersen et al.,
2000; Miller et al., 2003; Pedersen, Aviles, Ito, Miller, & Pollock, 2002; Vasquez,
Denson, Pedersen, Stenstrom, & Miller, 2005; Bushman, Bonacci, Pedersen,
Vasquez, & Miller, 2005), there were only 4 studies that examined TDA and only 2
6
contained both a provocation and trigger within the same design (Baron & Bell,
1975; Worchel, 1966). Neither of these two studies demonstrated the interaction
between provocation and trigger.
Miller’s group developed a paradigm which has reliably demonstrated this
synergistic effect in which receiving both a provocation and trigger led to a greater
level of aggression than the resultant additive aggression from the trigger and
provocation. It is essential to this paradigm that the provocation is of greater
intensity than the trigger (Miller et al., 2003). If they are of equal magnitudes, the
two events might warrant independent aggressive responses and triggered displaced
aggression would not occur because each event would be its own initial provocation
(Vasquez et al., 2005).
These studies also demonstrated a contrast effect for those participants who
received a provocation, but no trigger. This group displayed less displaced
aggression than any other group. This appears to be based upon the comparison of
social interactions for the participant. Following an unpleasant interaction during the
provocation, a neutral social interaction in the second phase of the experiment would
seem more positive than if a neutral interaction occurred at both times.
The General Aggression Model (GAM) and Other Theories of Aggression
A number of theories contribute to the understanding of both childhood
aggression and triggered displaced aggression including Cognitive Neoassociation
Theory (Berkowitz, 1989, 1990, 1993), Excitation Transfer Theory (Zillman, 1988),
7
and social cognitive theories including social-information-processing deficits such as
hostile attribution bias (Crick & Dodge, 1994; Dodge, 2006) and aggressive scripts
(Huesmann, 1988). Anderson and Bushman have posited a unifying framework
under which all of the above theories can be incorporated called the General
Aggression Model (GAM). Triggered displaced aggression in children will be
discussed through the GAM framework utilizing research from the aforementioned
aggression theories.
GAM Framework
The GAM is based on research on the development and use of knowledge
structures such as perceptual schemata which identify phenomena as complex as
social events like a personal insult, person schemata which include beliefs about a
certain person or groups of people and behavioral scripts which contain information
about people’s behavior under various situations. Knowledge structures: develop out
of experience; influence perception at multiple levels, can become automatized with
use; can contain affective states, behavioral programs, and beliefs; and can guide
people’s interpretations and behavioral responses to their social and physical
environment (Anderson & Bushman, 2002).
GAM focuses on the “person in the situation” in an “episode”, or one cycle of
an ongoing social interaction. The cycle begins with a social encounter which
activates inputs based on person and situation factors which influence the internal
present states of cognition, affective states, and arousal in deciding outcomes by
8
making appraisals and decisions. This process guides either a thoughtful or
impulsive action and, ultimately, leads to a second social encounter.
Inputs
Person factors. Person factors include traits such as genetic predispositions,
narcissism and a propensity toward hostile attributions, male gender, beliefs,
attitudes, values, long-term goals, aggressive scripts, and environmental factors.
Factors that would be considered inputs in GAM that have been shown to contribute
to aggression in middle childhood include genetic factors, temperament, social-
cognitive factors, and predisposing environmental factors (i.e., family, peers,
neighborhood, and media) (Loeber & Hay, 1997), and the interactions between
person and environmental factors (Kupersmidt, Griesler, DeRosier, Patterson, &
Davis, 1995). The social-cognitive theories mentioned earlier fit into this part of the
GAM.
Social-cognitive theories of aggression suggest that some children display
social-information-processing deficits which prevent them from generating non-
aggressive solutions to interpersonal conflict or lead to a misinterpretation of other
children’s intentions (Crick & Dodge, 1994). These include attributing hostile intent
to ambiguous, or even prosocial, overtures (Dodge, 1980) as well as being more
attentive to aggressive cues.
Recently, Dodge (2006) has proposed that “individual differences in
aggressive behavior occur as a function of characteristic styles of attributing hostile
9
intent (or not) to other’s provocative behavior” (p. 794). In his theory, he proposes
that while all children are born with hostile attributions and are likely to match intent
with outcome (i.e., “If he bumped into me and it hurt, he must have meant it to
hurt.”), most children also develop benign attributional styles with the onset of
theory of mind at age three. However, some children fail to learn to make benign
attributions and develop a chronic hostile attributional bias. Early life experiences
which may leave children at particular risk for this maladaptive attributional style
include physical abuse, modeling of hostile attributions by adults and peers, failure
in important life tasks, and rearing in a culture that values self-defense, personal
honor, and retaliation. These characteristics often lead to the development of
aggressive and hostile schemas.
In addition to their attributional biases, children’s attitudes and beliefs toward
aggression have been found to predict aggressive behavior with some children
developing normative beliefs about aggression often learned in the family
(Huesmann & Guerra, 1997). Bandura (1977) has described efficacy beliefs (i.e., if
the aggressor has the ability to successfully execute the aggressive behavior) and
outcome beliefs (i.e., if the aggressive behavior will have the intended affect) as
integral to whether children choose to aggress in any given social encounter.
Based upon their attitudes and beliefs, children develop scripts of how they
expect social interactions to occur based on their early learning experiences. In
middle childhood, children begin to develop worldviews, schemata, and mental
scripts based on their experiences. For instance, boys learn that fighting is masculine.
10
Simultaneously, girls’ aggression decreases more dramatically compared to boys’
aggression (Loeber & Hay, 1997). Hostile attributional bias (Dodge, 2006) is a
hostility-related script. Scripts and attitudes also shape available responses and
choice of response when placed in an ambiguous or confrontational situation.
Finally, although Lewin first proposed person-environment interactions in his
Field Theory in 1939, it has only been recently that empirical literature has begun to
examine the interactions between individual biological and psychological
characteristics and environmental factors through the biopsychosocial model. In fact,
this interaction was not focused upon in the initial theoretical paper on the GAM
(Anderson & Bushman, 2002). More recent papers on the GAM have, however,
focused on the importance of such interactions (Anderson & Huesmann, 2003;
Anderson & Carnagey, 2004) including the influence of micro factors (i.e., genetics
and biological predispositions) and macro factors (i.e., community, neighborhood,
and media) on a person’s propensity to aggress.
Situation factors. Also considered inputs are the situational factors of
frustration and provocation. This brings to mind frustration-aggression theory which
was first explicated by researchers in 1939. Dollard et al. (1939) described
frustration as an unexpected prevention of the ascertainment of an anticipated goal
and offered that it was this blockage of anticipated goals that accounted for most
human aggression. Frustration and provocation are also critical elements of the
Cognitive Neoassociation Theory (Berkowitz, 1989, 1990, 1993) which will be
written about further with the discussion of the role of affect as a present internal
11
state. Other situational factors considered inputs by Anderson and Bushman (2002)
were aggressive cues, pain and discomfort, drugs, and incentives.
Present Internal States
These aforementioned input variables influence the outcome behavior
through the mediating present internal states that they create. The three present
internal states most important to GAM are cognition, affect, and arousal which are
highly connected and intertwined.
Cognition. The internal state of cognition includes hostile thoughts and
scripts. Hostile thoughts can be elicited by priming, based on a situational activation
which makes the concept accessible for a short time such as media violence, or by an
aggressive concept that is frequently activated and, therefore, chronically accessible.
Some of the input variables make the relative accessibility of aggressive cognitions
more prominent in memory. Cognition also includes the use of aggressive scripts
including those accompanying a hostile attributional bias.
Affect. The internal state of affect incorporates mood and emotion as well as
expressive motor responses which serve as a cue as to how people should feel. Input
variables influence the kind of emotions and moods a certain social situation will
elicit (i.e., ease of anger activation, pain, uncomfortable temperature). It is the idea of
frustrations and provocations and their influence on negative affect that is at the heart
of Cognitive Neoassociation Theory.
12
In Berkowitz’s reformulation of the frustration-aggression theory (1989), he
limited the relationship of an unexpected prevention of the ascertainment of an
anticipated goal and aggression by arguing that aggression occurs only when
negative affect accompanies a person’s frustration from an aversive event. In his
theory, it is the negative affect which generates aggressive tendencies. Thus, both
expected and unexpected obstructions of the intended goal are prone to aggressive
reactions as either can produce negative affect. Negative affect may be further
intensified by the strength of the desire to reach the goal or the magnitude by which
the goal is blocked. The resulting aggression is frequently retaliatory and non-
instrumental.
This sequence of events can be described by the cognitive neo-
associationistic (CNA) model of aggressive behavior (Berkowitz, 1990, 1993). The
CNA model suggests that negative events produce negative affect which, in turn,
triggers “fight” and “flight” responses that are associated with both anger and fear.
These fight-and-flight responses are moderated by input variables of situational
influences, genetic predispositions, and prior learning. These variations lead to
individual differences in the way the two systems are activated and which is likely to
be the stronger of the two tendencies for each situation. Thus, people with
personalities that tend toward aggression would be more likely to respond
aggressively to activating events. The CNA model also posits that negative thoughts
tend to activate negative memories which may lead to anger and aggressive
inclinations.
13
Arousal. Input variables can also influence arousal which can, in turn,
influence aggression in three ways. First, people who are already aroused that
receive a provocation may engage in a heightened aggressive response. Second, is
based on the mechanism suggested by Excitation Transfer Theory. Zillman (1988)
posited that physiological arousal dissipates slowly and that if two arousing events
occur within a short amount of time, arousal from the first event may be
misattributed to the second event. Thus, if the first event elicits anger, then a second
arousing event will lead to unjustifiable levels of anger based upon the content of the
event. Third, is a new and untested hypothesis of Anderson and Bushman (2002)
which posits that unusually high or low levels of arousal are aversive and may
stimulate aggression in the same way as other aversive stimuli.
Outcomes
Following the influences of inputs and routes through which inputs influence
outcomes, people go through the appraisal and decision processes that lead to the
outcome of either a thoughtful or impulsive action. People who are able to examine
the outcomes and find that their behaviors may elicit negative outcomes engage in
reappraisal until an appropriate solution is chosen. By middle childhood, most
children have been socialized to manage their reactions to frustrations including
masking their reactions to adults and have developed overriding inhibitions based on
a moral standard that they have learned (Anderson & Bushman, 2002). Generally,
children’s problem-solving abilities continue to develop during this time period as
14
well, particularly for girls (Loeber & Hay, 1997). However, some children still have
difficulty controlling their impulses and emotions due to both early child experiences
as well as individual differences in neurologically-mediated tendencies to respond
impulsively (Dodge, 2006). Cognitive deficits that may prevent some children from
developing the skills to avoid conflict through appropriate channels of
communication during this developmental period include low intelligence and
hyperactivity, especially for boys (Lahey et al., 1999; Eron, 1987). Children with
limited cognitive or emotional resources may immediately jump to impulsive actions
without weighing whether the outcome of the action will be important and will lead
to an unpleasant outcome (Anderson & Bushman, 2002). The interaction of
individual characteristics and present internal states can influence the likelihood of a
thoughtful or impulsive action.
Children’s Triggered Displaced Aggression Within the GAM Framework
Triggered displaced aggression fits nicely into the GAM framework. Because
TDA is based upon the situation inputs of provocation and frustration, it is expected
that, to a certain extent, all children display TDA. Situational forms of childhood
aggression might help explain uncharacteristic aggression in otherwise non-
aggressive children. In addition, situational factors contributing to aggression interact
with the inputs of aggression-oriented person variables (i.e., individual risk factors)
to exacerbate aggressive tendencies in at-risk children (Valois, MacDonald, Bretous,
Fischer, & Wanzer Drane, 2002; Lahey et al., 1999).
15
For instance, upon receiving a provocation by an adult (i.e., situational input),
children would likely become unpleasantly aroused (i.e., present internal state). They
may also become angry or sad (i.e., internal present state) and may potentially
generate aggressive scripts (i.e., internal present state). If a child with hostile
attribution bias was to be provoked (i.e., input variable), the situation might bring up
scripts of being treated unfairly (i.e., internal present state) and beliefs that the
person was intending to harm the child (i.e., input variable) and might make an
aggressive response more likely. However, a child’s efficacy beliefs and outcome
beliefs (i.e., input variables) might be called into question given that the target of his
aggression would be an adult. Following these present internal states, the child
would quickly realize the negative outcomes resulting from aggressing against an
adult and would choose to inhibit aggression (i.e., outcome). This outcome would
serve as the basis for the next social encounter. Shortly thereafter, if that same child
was to be teased by a peer, the same situational and person factors would be invoked,
but all the present internal states would be altered. In this situation, the earlier
provocation would be a priming event in which aggressive scripts would be highly
accessible (i.e., present internal state). Further, scripts involving peers would be
elicited (i.e., present internal state) and would, therefore, include positive efficacy
and outcome beliefs (i.e., input variables). Further, the child would be highly aroused
from the initial provocation (i.e., present internal state) and would simultaneously
experience negative affect (i.e., present internal state) from the prior provocation and
might even attribute arousal from the previous event to the current event. These
16
various states might lead to aggression through two paths (i.e., outcome). It might
reduce the ability to make thoughtful decisions and lead directly to an immediate
impulsive action, or the various states might lead to a thought process in which the
child decides that the impulsive action would elicit desired results. This theory
explains how TDA might be activated in children and how it might prove
particularly problematic for at-risk children. The congruence of children’s TDA with
GAM indicates that it is quite likely that children exhibit TDA in much the same way
as adults. These interactions may further place children displaying high levels of
aggression on a continued aggressive trajectory and, subsequently, may lead to
alignment with other aggressive peers, rejection by non-aggressive peers, and
difficulty in interactions with teachers and parents (Anderson & Carnagey, 2004).
With the recent surge of research and funding allocated to the investigation
and prevention of bullying and aggression, learning more about TDA as a potential
contributing factor seems warranted given the number of aggressive and violent
interactions that seem to greatly exceed a reaction that would be warranted based on
the activating event. By carrying out this study, I attempted to elucidate one
mechanism by which children aggress in general and by which the propensity toward
aggression for at-risk children may be aggravated. In this present study, the goal was
simply to validate TDA as a construct applicable to children as well as to develop a
successful paradigm that could lead to a program of research on TDA in children.
Future studies may examine differences in the expression of TDA between
aggressive and non-aggressive children.
17
The Present Study
Rationale
In the present study, I investigated the construct of triggered displaced
aggression in children. As far as I am aware, no other study has examined TDA in a
childhood population, though children have been shown to experience displaced
aggression (Moore, 1964). Because the constructs of displaced aggression and
triggered displaced aggression are similar (Miller et al., 2003; Pedersen et al., 2000),
it seemed likely that children would exhibit TDA. Furthermore, children probably
encounter situations with adults or older children on a daily basis during which they
cannot retaliate such as a playground aide telling a child not to run around or an
older child demanding the younger child’s seat on the bus. At the same time, they
interact with numerous age-mate peers that may include situations containing the
necessary trigger for TDA. It seems possible that children’s circumstances place
them in situations that might elicit TDA more frequently than adults.
Hypotheses
To conduct this research, I used a modified version of the TDA paradigm
established by Pedersen et al. (2000) and Miller et al. (2003). In order to establish the
synergistic effect of the initial provocation and subsequent trigger, four conditions
were tested in which the presence or absence of the provocation was paired with the
presence or absence of the trigger. I predicted that those children in the condition that
received both a provocation (Time 1) and a trigger (Time 2) would display more
18
displaced aggression than children in the other groups in which children experienced
either the provocation or trigger only or received neither provocation nor trigger. I
expected a statistical interaction between provocation and trigger and that this
interaction would lead to a synergistic effect and would exceed mere additive effects
of isolated provocation and trigger events. Furthermore, due to a contrast effect, I
hypothesized that children in the condition with a provocation (Time 1), but no
trigger (Time 2) would display less displaced aggression than children in the other
three groups.
Potential Confounding Variables
Research in childhood aggression, child development, and social psychology
has elicited a host of potential covariates (inputs and internal present states in terms
of the GAM) that may influence the association between variables. Thus, data were
collected on a number of variables to be able to conduct analyses that would ensure
that results were not influenced by these potential confounders. The demographic
variables of age and ethnicity were collected. The number of moves the participant
made during the car racing computer game was also collected as an indicator of both
the comprehension of the instructions and the ability to engage in typical childhood
game playing behavior (i.e., not particularly impulsive or particularly apprehensive).
Finally, four questionnaires administered to participants and their caregivers were
collected.
19
Consistent with the GAM, Miller et al. (2003) suggest that a participant’s
current mood and trait aggressiveness may affect a person’s reaction to situational
frustrations (Miller et al., 2003). Questionnaires that examined each trait were
administered to measure these aspects of a child’s personality. Further, children’s
hostile attributions may be activated while experiencing negative affect (Orobio de
Castro, Slot, Bocsh, Koops, & Veerman, 2003) and may influence their ability to
make reappraisals beyond their immediate appraisals difficult and more likely to lead
to impulsive behaviors (Anderson & Bushman, 2002). By measuring children’s
beliefs about aggression and retaliation, it was possible to assess the impact of this
variable. Parents also filled out a demographic questionnaire to obtain exact age and
ethnicity/race.
Data was collected in an ethnically-diverse part of Los Angeles ensuring an
ethnically diverse sample. Participants for the study were between the ages of 9-12.
Three studies demonstrate that 9 years old is an appropriate age for successful
establishment of TDA in children. First, Eron (1987) established that individual
differences for this behavior tend to stabilize around 8 or 9 years old. Second,
children understand the basic tenets of a behavior-thought split by age 9 (Miller &
DeMarie-Dreblow, 1990) and would, therefore, be able to understand the goal of the
study during the debriefing. Third, at age 9, children are able to differentiate between
intentional and unintentional provocation suggesting that aggressive responses to
provocations in the experiment would be based on the intent of the provocateur
(Shantz & Voydanoff, 1973) rather than the outcome of the game. The upper end of
20
the range was chosen to allow for a large enough range to sufficiently recruit
participants without being so wide that results were not generalizable due to
developmental differences.
Exploratory Hypotheses
The GAM framework suggests that person variables may serve as inputs that
influence the course of events in any given social encounter (Anderson & Bushman,
2002). Thus, while my main hypotheses focus on the interactive effects of two social
encounters on children’s behavior, it is possible that their own person variables may
influence their behaviors during the computer game. Three such person variables for
which we have data are children’s aggressive schemas, their tendency to behave
aggressively as rated by both the children and their parents, and their mood state over
the last few months prior to participation in the study. Though we have no a priori
hypotheses, exploratory analyses were conducted to examine the associations
between these variables and outcome measures of aggressive responding during a
competitive game.
Method
Overview
The study design was a 2 (provocation/no provocation) x 2 (trigger/no
trigger) between-subjects factorial (7 boys per cell for a total of 28 participants).
Every four participants were randomly assigned so that there would be one
21
participant in each group. Then, the next four participants would be randomly
assigned again to each of the cells so that the number of participants in each cell
increased evenly to avoid threats to internal validity such as a change in
experimenter or an external event that could alter children’s reactions to stimuli (i.e.,
history effects). The four groups included a Provocation/Trigger group, a
Provocation/No Trigger group, a No Provocation/Trigger group, and a No
Provocation/No Trigger group (i.e., control group).
The experiment was posited to participants as a study in which experimenters
wanted to observe children’s game-playing behavior in competitive and non-
competitive games. The provocation was a computer malfunction in which the
experimenter blames the child for the malfunction and takes away all prizes. The
trigger was a competitor who taunts the participant prior to a competitive car racing
game. The dependent variable was the participant’s likelihood of making an
aggressive first move in the competitive game. There is precedent for experimental
studies of childhood aggression in which the dependent variable is dichotomous
(Dodge, 1980; Swaim, Henry, & Kelly, 2006; Cohen & Prinstein, 2006), suggesting
it is a valid form of assessing childhood aggression. To reduce threats to internal
validity, I relied on a design that incorporated double blind procedures. Insofar as
possible, the child experimenter (i.e., research assistants) and participant were
unaware of the experimental condition. Practically, however, it was necessary for the
experimenter to become aware of the condition at a late point in the procedure.
22
Participants
Recruitment of Sample
A total of 82 participants were recruited. However, because the triggered
displaced aggression effect has not been demonstrated by past researchers in a
childhood sample, the study procedure was extensively piloted prior to the final
experiment. Indeed, a primary objective of this dissertation was to develop and
validate an efficient experimental paradigm. As a result, my analyses were restricted
to a subset of the children recruited. There were also four participants who were
dropped from analyses. One participant exceeded our age restrictions, a second
appeared to have a reading level that was lower than necessary to complete the self-
report measures, and the remain two participants demonstrated a lack of
understanding as determined through observations and outlier data.
Pilot procedures. 50 participants (34 boys, 16 girls) were involved in pilot
iterations of the study design. There were a number of design elements that were
removed after data review as well as theoretical discussion with the two co-principal
investigators.
First, the final sample included boys only. While girls were initially recruited
into the sample, due to resource limitations (i.e., time and money), it was determined
that there would not be sufficient power to conduct separate analyses by gender. This
was deemed necessary by the substantial theoretical literature that suggests that boys
are considerably more likely to be aggressive than females. Indeed, my data
supported the literature. Data for female pilot participants showed females displaying
23
little aggression toward their peers resulting in data with constrained variability.
Across four measures of aggression (first move, first two moves, first five moves,
and total number of aggressive moves), the girls had lower mean scores and more
restricted ranges than boys. These mean differences were statistically significant for
the first move and the total number of aggressive moves.
Second, earlier versions of the study used a frustration rather than a
provocation to induce TDA. This had originally been proposed as no empirical
studies had examined frustration as opposed to provocation. In the earlier protocols,
rather than blaming the child for the computer malfunction, the RA was genuinely
sympathetic to the child’s plight and apologetically explained that without knowing
the number of points accrued, no prizes could be awarded. Concerns were raised,
however, that no prior studies had used frustration and that the manipulation might
be too gentle to elicit any significant aggression. Thus, it was recommended that a
provocation be used for the Time 1 manipulation in which the RA blamed the child
for the computer malfunction.
Third, the instructions given prior to the car racing game in which the
dependent variable was measured were potentially tapping into a child’s competitive
tendencies rather than their aggressive tendencies. Participants had been told the
person who won the race would win a better prize than the person who lost the race,
but that they could prevent their opponent from winning a prize at all if they could
keep them from reaching a certain marker in the race. Thus, participants may have
also thought that they might have lost for the chance for a prize if they did not reach
24
that same marker and may have, therefore, focused on moving their own car forward
rather than thinking about the preceding events. Thus, the instructions were changed
so that there was no opportunity for the participant or the opponent to lose a prize.
Fourth, a trigger based on a hearsay insult rather than a direct insult from a
peer was used originally. Initially, the experimenter would enter the room in which
the RA and participant were located and would tell the participant that the other child
was ready. In the trigger manipulations, the experimenter would add, “Oh and the
other kid says he knows he can beat you.” Two changes were made to the trigger. A
number of participants expressed questions regarding the existence of the other
participant. Thus, it was determined that a different method should be used for the
trigger in which the opponent seemed more realistic. Thus, a tape recording of an
appropriate age-mate played over two-way radios was used to convey the trigger.
Further, it was determined that the content of the trigger may have been viewed as
competitive rather than hostile so the words of the trigger were changed to reflect a
more personal attack (i.e., I know I can beat you because I’m cooler than you.)
Finally, recruitment methods were changed. The monetary incentive was
changed from $15, to $10 with a $200 lottery to the final amount of $25 with a $200
lottery. It is unclear if these changes altered the flow of participants calling into the
study. In addition, efforts were focused on methods in which large numbers of
parents could be reached at one time. The most cost-effective (i.e., money and time)
recruitment strategies were having fliers distributed at schools, Boys Scouts, and bi-
weekly postings on the volunteers section of www.craigslist.org.
25
Current sample. Following review of the experimental protocols, the study
was then revised to increase internal validity and address the aforementioned
concerns. 28 male participants took part in the final iteration of the experiment.
Research assistants, hereafter called RAs, (i.e., trained advanced undergraduates) and
I recruited 28 boys between the ages of nine and 12 (mean age = 10.86) and their
parents/ caregivers (hereafter called parents) from the Los Angeles and Anaheim
communities. Recruitment efforts combined partnering with local schools and Boy
Scout troops to send home fliers in English and Spanish to all of the parents of
children in the 3
rd
through 7
th
grades and advertising on listserves, websites, and
other child-focused local publications. The diverse surroundings of the USC campus
naturally elicited a multi-ethnic and balanced sample. Participants were: Mixed Race
(25.0%), White (25.0%), Hispanic (21.4%), Black (14.3%), Asian/Pacific Islander
(10.8%), and Armenian (3.6%).
Informed Consent
Recruitment fliers framed the study as a computer game study for children
(Appendix A). Upon seeing the flier, parents called or emailed the study to find out
more information at which point they were entered into a drawing for a $200 prize.
The flier also offered that parents would receive $25 for their time if they were to
participate and that children would have the chance to win small USC prizes. When
RAs spoke to the parents about the study, they informed parents that the study was
really investigating aggression and explained the TDA concept and study procedures.
26
They also reinforced that it was important that the children did not know that the
study focused on aggression and that children would be debriefed following the
procedure and awarded all prizes offered throughout the experiment.
Upon arrival to the study, parents signed an Informed Consent form
following an opportunity to either read the form or have an oral explanation of the
form by the parent experimenter (usually me) while the child experimenter (a
research assistant) explained the Informed Assent form to participants who then
signed the form. All questions regarding informed consent were answered to their
fullest extent.
Sample Size Analysis
A simple sample size analysis equation was used to determine the number of
participants per cell necessary to detect statistically significant group differences. Per
Valente (2002, p. 115), the following calculation was used in which the sample size
necessary to detect differences between proportions is obtained: n = 2z
2
pq/d
2
in
which z is the z-score associated with the chosen confidence interval (in this case,
95%, making the appropriate z-score, 1.96), p is the proportion of people who
experience triggered displaced aggression while q is 1-p, or the proportion of the
population who do not experience triggered displaced aggression, and d is the effect
size for the phenomenon. The results of the calculation are the necessary sample size
for each cell in the design. In this case, the most conservative estimates were used for
proportions which are .5 for p and .5 for q assuming that the phenomenon is evenly
27
split in the population. For the effect size, .55 was used. This choice of effect size is
based upon the effect size elicited in the meta-analysis conducted on the literature
regarding displaced aggression (Marcus-Newhall et al., 2000). Several researchers
have suggested that triggered displaced aggression is more ecologically valid and is
likely to elicit even stronger effects than displaced aggression; thus, using the
displaced aggression effect size is a conservative estimate. Using the aforementioned
formula, the required sample size is 6.35 per cell, or in practicality, 7 per cell. Based
on conservative estimates for both the proportions and effect size, these results may
be an overestimate of the actual sample size necessary to achieve sufficient power to
make inferences.
Procedure
Procedural Outline
The exact protocol for the study can be found in Appendix B. Parents and
children arrived to the study site and were told that there was another research
assistant (RA) working with another family in a room down the hall and that the
other participant would be playing the same games; however, there was not really
another child or RA. The participant was told that the parent experimenter’s role was
to go back and forth between the rooms to make sure everything was going
smoothly. The child was then told that they would be playing a game against the
other child later in the study. The RA and I both had two-way radios used for
communication. Then the parent and child were told that they would both fill out
28
questionnaires (Appendix C). Parents filled out the Demographic Questionnaire and
the Reactive-Proactive Questionnaire- Parent Version (RPQ-P; Raine et al., 2006)
and were provided popular magazines to read. Children were taken into a room that
had the questionnaires, a laptop computer, and hidden prizes. Subsequently, they
independently filled out three self-report measures: How I Feel (Walden, Harris, &
Catron, 2003), Reactive-Proactive Questionnaire (RPQ; Raine et al., 2006), and
Normative Beliefs About Aggression Scale (NOBAGS; Huesmann & Guerra, 1997)
questionnaires which had been deemed appropriate for fourth graders (i.e., the
youngest age of participants). The instruments were administered in counter-
balanced order. RAs explained how to fill out each of the surveys and read the first
question on each to explain the surveys further. Participants were told to fill out their
surveys and to ask the RA if they had questions. The RAs left the room to allow the
child privacy so that they would be able to answer honestly. Children had been
screened for fourth grade reading level according to their parents.
Following the completion of the questionnaires, participants played the two
computer games after instructions from the RA that comprised the main part of the
experiment described below in more detail. After they completed the procedure, the
child was debriefed and received all five prizes that were offered throughout the
game while parents received their $25 compensation. Parents were given the option
of providing their address on an envelope so results of the study could be sent home.
29
Experimental Manipulation
Provocation procedure. The participants were then asked to play Space
Cadet, a 3-D pinball game found on Windows XP operating systems. This game was
chosen because there are no aggressive elements to it, thereby, ensuring that later
measures of aggression were not primed by aggressive content of the game (Irwin &
Gross, 1995; Anderson & Bushman, 2002). The child experimenters (i.e., the RAs)
told the participants that I wanted to assess how children their age can use their
minds to play and win different games as well as how they behave, independently
and against another child. They were given instructions to ensure some skill with the
game with the goal of collecting a certain amount of points to ostensibly have a
chance at wining prizes. They were then shown prizes of increasing value, which
correlated to the number of points they could have earned during the game, and told
that they had three minutes to score as many points as possible. Finally, the RA
ostensibly called over to the “other RA” to make sure the “other participant” was
ready to play although they were actually contacting me with the two-way radio. I,
then, played a tape recording of the other RA (i.e., a woman’s voice recorded on a
tape player) who said that the participant had gone to the bathroom. This was done to
increase the face validity of the existence of the “other participant.” Fifteen seconds
later, I would contact the RA and play the tape recording of the other RA saying they
were ready to begin playing the pinball game. The RA left the room so participants
could play without oversight.
30
In the provocation condition, participants played the pinball game for one
minute which was enough time for participants to reach at least the first prize point
and maybe the second. At that point, the computer was pre-programmed to
malfunction and produce an error message. The participant then notified the RA who
appeared to attempt to understand the error. The experimenter then chastised the
child stating, “What did you do? This hasn’t happened with any of the other kids.
Well, you must have done something to mess it up. This has never happened before.
I don’t have time to try again since I have other people coming in and this might be
your fault anyway. I guess I’ll just have to keep going with the experiment, but I
can’t give you any of the prizes since I don’t know your score.” In the no-
provocation condition, the participants were allowed to play for three minutes and
were given the prizes that they earned.
Trigger procedure. Following the pinball machine game, the RA instructed
participants that the next game was a car racing computer game that they would play
against the “other kid.” Instead, the child played against the computer, but the game
was designed to appear as though participants were playing against another child.
The participant received the instructions and rules for the task. On each turn,
participants rolled dice on the computer to determine which player moved next. This,
too, was programmed so that it seemed that the dice roll was arbitrary. In the end,
however, each player was scheduled to move an equal number of times so that the
winner depended on the speed with which the child made their moves making the
outcome arbitrary. For each move, the participant could choose to advance their car
31
or move their opponent’s car backward and the player who was furthest ahead after 3
minutes won a prize slightly better than that of the consolation prize, but of lesser
value than the minimal prize for the pinball game. Then, the RA would tell the
participants that they were going to have a chance to meet their opponent over the
two-way radios and that they were to state their name and school when the RA told
them it was their turn.
The RA used the two-way radio to communicate with me that they were
ready for the car racing game. At that point, I played a tape recording on which the
“other RA” said that she and the “other kid” were also ready and that introductions
should begin. The “other kid” (i.e., a tape recording of a child’s voice who was the
brother of a research assistant) stated his name and school into the two-way radio.
For participants in the trigger condition, the participant then heard their
opponent state the same introduction, but with the added sentence, “Oh, and I know I
can beat you because I’m cooler than you.” This taunt served as the trigger.
Participants in the non-trigger condition only heard the recording state the child’s
name and school. At this point, the game was begun and the experimenter left the
room so that the child could make his or her moves without pressure from an adult
and returned when the game was over to award the child the winner or consolation
prize.
Dependent variable. With each move participants made, they were able to
advance their car forward one space or push his opponent back one space. However,
it was emphasized to the participants that these two moves were equivalent in terms
32
of the outcome of the game. In this way, motives of competition were removed and
participants’ choice to move their opponent backwards would be made out of
aggressive motives instead. The dependent variable was the dichotomous choice of
an aggressive (i.e., move opponent’s car backwards) or non-aggressive move (i.e.,
advance their own car) at their first opportunity to make a move. The first move was
taken as the best indicator of aggressive behavior because after the first move,
participants respond in a tit-for-tat fashion, consistent with norms of reciprocity
(Axelrod, 1984; Gouldner, 1960; Lindsay & Anderson, 2000).
This portion of the game shares some characteristics with a well-validated
laboratory measurement of aggression in adults, the Point Subtraction Aggression
Paradigm (PSAP) (Cherek, Schnapp, Moeller, & Dougherty, 1996). In the PSAP,
during several 25-minute sessions, participants are able to enhance their chances of
winning money by pressing a button one hundred times, or they can subtract points
from their opponents’ point total by pressing a second button ten times, though they
win no money for doing so. When they subtract a point from their opponent, there is
a grace period during which none of their own points can be subtracted. However, by
subtracting points from their opponent, they are losing opportunities to earn money.
Except for the grace periods, their “opponent” (i.e., the computer program), is able to
subtract points from the participant and win additional money by doing so. The
dependent variable in this paradigm is the amount of aggressive responses (i.e.,
number of times the second button is pressed) during the session. By using elements
33
of a well-validated procedure and mapping it onto the TDA paradigm, it was hoped
that the procedures used in this study would have high internal validity.
Other Measures
Demographic questionnaire. Parents filled out a short, one-page
questionnaire requesting information about their child’s gender, age, birthday,
ethnicity, and language(s) spoken at home. Children’s precise age was calculated by
subtracting their birth date from the day they participated in the study.
Baseline emotional state. Children completed the How I Feel measure
(Walden et al., 2003), a 30-item self-report rating scale that assesses children’s
emotional arousal and control. This device has three factors: positive emotions (e.g.,
“I was excited almost all of the time”; ! = .87), negative emotions (e.g., “I was mad
very often”; ! = .89), and positive and negative emotional control (e.g., “When I felt
scared, my scared feelings were very strong”; ! = .84). Items are rated on a 5-point
rating scale, with points ranging from "0- not at all true of me" to "4- very true of
me." Because children’s emotional arousal and control may interfere with their
ability to adaptively cope with aversive situations (Walden et al., 2003), I measured
this as a potential covariate. Summary scores for each of the three scales were
created by taking the mean of participant’s answers on the related questions for each
scale leading to scale scores between zero and four.
Baseline aggressive behavior. Children completed the Reactive-Proactive
Aggression Questionnaire, a 22-item checklist assessing behavioral indicators of
34
aggression (RPQ; Raine et al., 2006). The participant’s parent filled out the parent
version of the RPQ (RPQ-P) which alters the wording so items refer to their children.
The RPQ has one item that was dropped (“I bring weapons to school”) as it is illegal
to bring weapons to school in the Los Angeles Unified School District and there
were concerns about confidentiality and legal issues related to the study. This scale
has two factors: proactive and reactive aggression (Raine et al., 2006). Items include
proactive aggression items (“I yell at others so they will do things for me”; ! = .86)
and reactive aggression items (“I get angry when frustrated"; ! = .84). Items are
rated on a 3-point scale: never, sometimes, and often. These scales were used to
assess children’s baseline aggression levels as a covariate so that results are not
influenced by a child’s own typical level of aggression as trait aggressiveness may
affect a person’s reaction to situational frustrations (Miller et al., 2003; Anderson &
Bushman, 2002). Summary scores for each of the three scales were created by taking
the mean of participant’s answers on the related questions for each scale leading to
scale scores between zero and two.
Bivariate correlations were performed between the two scales of the RPQ and
the RPQ-P (i.e., reactive and proactive scales) to assess if the data between the
informants could be collapsed into a singular aggressive behavior variable. There
was not a statistically significant correlation between parent and child responses to
items on the proactive scale (r = .18, p = .14). Data from the two informants was
positively and significantly correlated for items comprising the reactive scale of the
RPQ (r = .32, p = .008); however, the correlation was modest. These results are
35
consistent with prior data from this measure that suggests parent-child correlations
are modest (r=.23; Raine et al., 2003). Thus, the measures of aggressive behavior
were examined separately in group difference analyses.
Baseline beliefs about aggression. Children also took the Normative Beliefs
About Aggression Scale (NOBAGS; Huesmann & Guerra, 1997), a 20-item scale
which measures children’s schemata regarding aggression as a normative behavior
(! = .84). There are two scales: Retaliation Beliefs and General Beliefs. In the
Retaliation scale, children are given scenarios followed by possible responses (e.g.,
“Suppose a girl says something bad to a boy. Do you think it’s wrong for the boy to
hit her?”; ! = .82). In the General Beliefs section, questions ask about a child’s
global beliefs (e.g., “It is wrong to insult other people.”; ! = .77). Items are
measured on a 4-point scale ranging from “It’s really wrong” to “It’s perfectly ok.”
Aggressive beliefs tend to be associated with aggressive behavior and hostile
attributions. Because the experiment deals with concerns of retaliation, I am
administering this measure as another potential covariate. If children were to endorse
aggressive beliefs, they may be more likely to act aggressively during the
competitive game even if they did not endorse aggressive behavior. Summary scores
for each of the two scales were created by taking the mean of participant’s answers
on the related questions for each scale leading to scale scores between zero and three.
36
Results
Manipulation Checks
Provocation
I wanted to ensure that children were sufficiently provoked and frustrated
following the initial provocation. Without sufficient provocation, triggered displaced
aggression will not occur (Miller et al., 2003). A multi-informant approach was used
to determine if the provocation manipulation was effective. Participants took a nine-
item survey about the pinball machine game in which six items were related to the
provocation and three were dummy variables about the game itself to enhance face
validity. The six items analyzed included: a) Was the game frustrating to you? b) Do
you feel happy right now? (reverse coded) c) Do you feel excited right now? (reverse
coded), d) Do you feel angry right now?, e) Do you feel sad right now?, and f) Do
you feel disappointed right now? Each item was assessed on a 5-point scale ranging
from 0 (Not at all) to 4 (Very much). This scale had good internal consistency
(Cronbach’s ! = .80). Participants in the two groups that received the provocation
were more likely to have higher mean scores on these items (M = 1.20, SD = 0.98)
than participants in the two groups that did not receive the provocation (M = 0.38,
SD = 0.39), t(26) = 2.90, p = .007.
While participants were filling out the survey regarding the pinball machine
game, experimenters filled out a 9-item checklist based upon their memory of
observations of the participants’ reaction to the results of the results of the pinball
machine game (i.e., provocation or no provocation). Items asked if the participant: a)
37
got frustrated, b) got angry, c) got sad, d) grunted, e) whined, f) complained, g)
fidgeted, h) noticeably changed behavior, i) had an overall negative response. Each
item was assessed by a dichotomous response (1-Yes or 0-No). This scale had
adequate internal consistency (Cronbach’s ! = .75). Participants in the trigger
conditions were rated as having a more negative reaction to the results of the pinball
machine game (M = 0.29, SD = 0.21) than participants in the non-trigger conditions
(M = 0.07, SD = 0.15), t(26) = 3.23, p = .003. Thus, the provocation manipulation
appears to have been effective.
Trigger
The measure of the dependent variable (i.e., triggered displaced aggression)
was based upon the participants moving their opponent’s car backwards in the car
racing game even though this behavior would not harm the opponent’s chance to win
a prize; thus, any such move was conceptualized as being made solely to be mean to
the other participant. This caveat was explained to participants (i.e., “Your chances
of winning aren’t improved by moving forward or moving your opponent
backwards; they are both equal.”) However, due to the complex nature of the concept
and its importance in verifying that children played the game under equivalent
understanding of the rules, it seemed important to test participants’ comprehension.
In a three-item questionnaire in which two variables were questions used for face
validity, a third question asked, “Did you feel that you were hurting your opponent’s
chances of winning a prize when you moved them backwards?” The item was
38
assessed using a 3-point scale (e.g., No, Maybe, or Yes). For all participants, 65% of
participants answered “Yes” or “Maybe” indicating that they did not understand this
concept. There were no significant differences between trigger conditions and no-
trigger conditions for this item, t (26) = 0.22, p = .83. However, the association
between the child’s belief that they could harm their opponent with a move
backwards and their likelihood of moving their opponent backwards on the first
move was significant (r =0.41, p = .03). The belief in the ability to harm the
opponent was not significantly associated with either the first five moves (r =0.25, p
= .20) or with all of the participants’ moves (r = 0.11, p = .57) suggesting a linear
trend that supports the notion of the norms of reciprocity. No items were used to
assess the effectiveness of the trigger itself and, therefore, no comparisons can be
made between the intensity of the provocation and trigger nor can the trigger be
shown to be effective.
Triggered Displaced Aggression
To assess aggression towards the target (i.e., the participant’s opponent in the
game), a participant’s first move in the computer game was examined. The children
had the opportunity to move their own car forward or his opponent’s car backwards
in each move in the game. This measure was scored in a dichotomous manner;
moving the opponent backwards was considered aggressive while moving the
participant’s own car forward was considered non-aggressive. Of the 28 participants,
28.6% (n=8) moved the opponent backwards on the first move.
39
Logistic regression was conducted and revealed no main effects of
provocation (Odds Ratio (OR) = 4.50, 95% CI, 0.34, 60.15) or trigger (OR = 2.40,
95% CI, 0.16, 34.93), and no significant provocation X trigger interaction (OR =
0.22, 95% CI, 0.01, 7.20). As predicted, the trigger did not seem to affect children’s
likelihood of moving their opponent backwards without provocation (M = .296, SD =
.488 for the no-provocation/yes-trigger cell compared with M = .143, SD = .378 for
the no-provocation/no-trigger cell). However, the provocation alone did not elicit
lower levels of aggression than the other groups as predicted (M = .429, SD = .535)
nor did the provocation/trigger group elicit higher levels of aggression as predicted
(M = .296, SD = .488). There were no group differences between the four conditions
indicating that triggered displaced aggression was not elicited in this study design.
To ensure that the results from just the first move were not due to participants
becoming acquainted with the game, the first five moves in the game were
aggregated into an aggression score. Five moves allowed children to become
acquainted with the game, but not settled enough to educe the aforementioned tit-for-
tat game playing behavior. Each move was scored in a dichotomous manner (0 or 1)
and the scores of the five moves were summed. Thus, this variable had a range of 0
to 5 with higher scores indicating greater aggression. The mean number of times out
of five that a participant moved his opponent backwards was 2.07 with higher scores
indicating greater aggression. The mean number of times out of five that a
participant moved his opponent backwards was 2.07. Just one participant moved the
opponent back all five times (3.6%), three moved the opponent back four times
40
(10.7%), 8 moved the opponent back three times (28.6%), 6 moved the opponent
back twice (21.4%) and 5 participants each moved the opponent back one or zero of
the five opportunities (17.9%).
A 2 X 2 ANOVA was conducted using the first five moves as the dependent
variable and revealed no main effects of provocation, F (1, 27) = 1.71, p = .20, or
trigger, F (1, 27) = 0.07, p = .80, and no significant provocation X trigger interaction,
F (1, 27) = 0.07, p = .80. Again, the means did not differ between conditions. The
means were 1.57 (SD = 1.51) for the provocation/trigger group, 2.43 (SD = 1.27) for
the no-provocation/no-trigger group, 2.43 (SD = 1.72) for the no-provocation/trigger
group, and 1.85 (SD = 1.21) for the provocation/no-trigger group. Thus, it appears
that there were no effects of triggered displaced aggression once the children were
acquainted with the game.
Finally, the total number of aggressive moves was examined as a dependent
variable consistent with the Point Subtraction Aggression Paradigm (Cherek et al.,
1996). The range of this variable was 0-36 aggressive moves with nine moves being
the median number of moves. Three-fourths of the sample (n=21) moved their
opponent back 13 times or less, and just two participants moved the opponent back
more than 19 times. A 2 X 2 ANOVA was conducted on this dependent variable and
also revealed no main effects of provocation, F (1, 27) = 0.06, p = .81, or trigger, F
(1, 27) = 0.40, p = .53, and no significant provocation X trigger interaction, F (1, 27)
= 0.01, p = .91. Again, the means did not differ between conditions. The mean
number of aggressive moves was 8.71 (SD = 8.58) for the provocation/trigger group,
41
10.00 (SD = 6.61) for the no-provocation/no-trigger group, 10.42 (SD = 11.84) for
the no-provocation/trigger group, and 7.57 (SD = 5.65) for the provocation/no-trigger
group. Thus, using all of the children’s aggressive moves did not change the results;
there were no effects of triggered displaced aggression.
Group Differences for Potential Confounders
With cell sizes of seven, these data do not have sufficient power to conduct
analysis of covariance (ANCOVA) which would control for possible covariates
while conducting analysis of variance between the independent (i.e., experimental
group membership) and dependent variable (i.e., level of aggression). Instead,
analyses were conducted to elicit any statistically significant group differences for
the potential covariates that may have influenced the results. Due to the use of a
randomization process in assigning participants to a group, it was expected that there
would be no differences. Questionnaire data was separated into their empirically
validated scales that were discussed in the method section.
A 1 x 4 one-way analysis of variance (ANOVA) was used to compare group
differences for potential covariates that could confound the results of the
hypothesized associations between independent and dependent variables. These
comparisons are displayed in Table 1. Table 1 displays F-values, which report if
there were statistical differences between groups, and group means for each of the
experimental groups for the aforementioned potential confounder variables. There
were no significant differences between the experimental groups on any of the
42
potential confounder variables. This indicates that it is unlikely that results of the
main experimental hypotheses were influenced by these non-experimental, external
variables.
Exploratory Analyses
The questionnaires used in the study were given to serve as potential
covariates for the main association at hand (i.e., experimental manipulations and
resulting aggression). However, given the individual differences that shape a
person’s propensity to aggress (i.e., person input variables), I believed that there may
be direct correlations between aggressive schemata, aggressive behavior, and
negative mood with a child’s aggressive responding in a video game. The
correlations between these variables are displayed in Table 2. Specifically, Table 2
shows the bivariate correlations between the empirically-validated scales of the four
questionnaires described in the method section (i.e., mood state, aggressive beliefs,
parent-rated aggressive behavior, and child-rated aggressive behavior) and the three
variants of the dependent variable as previously discussed. Neither the reactive nor
proactive aggression scales rated by parents had significant associations with any of
the seven child-rated scales though the proactive and reactive scales rated by parents
were significantly related (r = .66, p < .001). Parents rating of their child’s reactive
aggression did have a negative association with the overall number of aggressive
responses a child made. However, given the parents’ seemingly inconsistent
knowledge of their child’s behavior and the large number of variables examined, it
43
seems that this relation may be due to chance. In contrast, a number of relations
appeared between the child-rated measures. There was a significant correlation
between the two scales of the aggressive beliefs questionnaire (r = .69, p < .001).
Emotional control and positive emotional state were positively associated (r = .43, p
= .02) though neither were significantly associated with negative emotional state.
There was a moderate negative association between retaliatory beliefs and emotional
control (r = -.52, p = .005). Finally, there were significant relations between both
retaliatory beliefs and negative emotional state with child-rated reactive aggression
(r = .40, p = .03 and r = .51, p = .005, respectively) and significant relations between
both retaliatory beliefs and general aggression beliefs with child-rated proactive
behavior (r = .52, p = .004 and r = .75, p < .001, respectively). However, none of the
child measures, including endorsement of retaliatory beliefs, had any significant
associations with the dependent measures of a child’s first move, their first five
moves, or all of a child’s aggressive moves.
44
Table 1. Demographic and questionnaire variables by experimental group
Mean (SD) by experimental group
Main effect of Provocation/ Provocation Trigger
experimental group Trigger Only Only Control
Variable (F level) (n = 7) (n = 7) (n = 7) (n = 7)
Age (9.00-12.99 years of age) 2.42 11.46 (1.01) 11.07 (0.81) 10.42 (0.74) 10.48 (0.66)
Ethnicity (% White) 1.80 14.29 (37.80) 57.14 (53.45) 14.29 (37.80) 14.29 (37.80)
Ethnicity (% Mixed Race) 0.65 14.29 (37.80) 14.29 (37.80) 43.86 (53.45) 28.57 (48.80)
Number of moves in computer game 0.03 29.57 (5.50) 28.43 (13.94) 27.71 (16.74) 29.14 (10.09)
Normative beliefs about aggression
Retaliatory aggression beliefs 0.40 0.67 (0.32) 0.65 (0.41) 0.48 (0.37) 0.65 (0.41)
(0-to-4 rating)
General aggression beliefs 0.35 0.30 (0.40) 0.27 (0.50) 0.11 (0.23) 0.23 (0.35)
(0-to-4 rating)
Aggressive behavior- child report
Reactive aggression 0.07 0.68 (0.15) 0.64 (0.16) 0.62 (0.35) 0.66 (0.33)
(0-2 rating)
Proactive aggression 0.22 0.12 (0.14) 0.13 (0.16) 0.07 (0.09) 0.12 (0.20)
(0-to-2 rating)
45
Table 1, continued
Mean (SD) by experimental group
Main effect of Provocation/ Provocation Trigger
experimental group Trigger Only Only Control
Variable (F level) (n = 7) (n = 7) (n = 7) (n = 7)
Aggressive behavior- parent report
Reactive aggression 0.13 0.81 (0.17) 0.83 (0.28) 0.74 (0.24) 0.78 (0.39)
(0-to-2 rating)
Proactive aggression 0.24 0.09 (0.11) 0.10 (0.17) 0.14 (0.12) 0.17 (0.29)
(0-to-2 rating)
Mood state
Positive emotion 0.60 3.07 (0.36) 2.57 (1.02) 3.04 (0.99) 2.77 (0.71)
(0-to-4 rating)
Negative emotion 0.06 0.98 (0.59) 1.05 (0.66) 0.94 (1.14) 0.87 (0.82)
(0-to-4 rating)
Emotional control 0.69 3.03 (0.27) 2.91 (0.58) 2.63 (1.27) 2.46 (0.85)
(0-to-4 rating)
Note.
*
p< .05.
46
Table 2. Bivariate correlations among questionnaires
Variable 2 3 4 5 6 7 8 9 10 11 12
1. Retaliation Beliefs .69
**
-.34
!
.02 -.52
**
-.05 -.14 .40
*
.52
**
-.11 -.10 -.17
2. General Aggression Beliefs – -.01 .10 -.12 .11 .09 .33 .75
**
.01 -.03 -.04
3. Positive Emotion – – -.16 .43
*
.30 .24 -.28 .10 -.14 -.05 -.15
4. Negative Emotion – – – -.09 .11 -.06 .51
**
.12 .17 .34 -.04
5. Emotional Control – – – – .20 .27 -.36
!
-.17 .02 -.02 .21
6. Reactive Aggression- Parent – – – – – .66
**
-.09 .05 -.18 -.24 -.50
**
7. Proactive Aggression- Parent – – – – – – -.25 -.04 -.04 -.10 -.27
8. Reactive Aggression- Child – – – – – – – .29 .32 .32 .18
9. Proactive Aggression- Child – – – – – – – – .03 .05 -.08
10. First Move in Game – – – – – – – – – .60
**
-.50
**
11. First 5 Moves in Game – – – – – – – – – – .60
**
12. All Moves in Game – – – – – – – – – – –
Note.
!
p<.075 (marginally significant).
*
p<.05.
**
p<.01.
***
p<.001.
47
Discussion
The results of the data analysis do not support the main hypothesis or the
secondary hypothesis. I had predicted that those participants who received both an
initial provocation and a subsequent trigger would display greater levels of
aggression as measured through moves on a competitive car racing game than
participants who had received a provocation only, a trigger only, or no trigger and no
provocation. The results showed that there were no group differences; there were no
significant differences between any of the four experimental groups.
It is important to explore potential explanations for the lack of significant
findings. One possibility is that, in fact, children do not exhibit triggered displaced
aggression. An independent study would be required to test this using the
aforementioned theory as the main hypothesis. Perhaps there is no interactive effect
between social encounters for children and each event is independent. However, in
that case, one would expect that the groups receiving a trigger would have a higher
level of aggression in the car racing game than children receiving no trigger and the
results do not support that theory. Further, the notion of independent reactions to
sequential events would suggest that rumination from a first encounter does not
affect the interpretation of a second social event which would be a departure from
the adult literature (Bushman et al., 2005). There is evidence that children engage in
ruminative thinking in a similar manner to adults and that such thinking impacts
children’s emotions and cognitions at a later point in time (Abela, Brozina, & Haigh,
2002), however, making the theory of independent social events having no impact on
48
one another unlikely. Another explanation that would be consonant with the notion
that children do not exhibit TDA is that situational triggers do not affect children in
the same way as adults. However, research has demonstrated that situational triggers
do exacerbate aggressive tendencies (Valois et al., 2002) making this theory
improbable.
An alternative explanation might focus on potential threats to the internal
validity of the study. These include concerns of insufficient power, potential
confounders that were not examined ineffective manipulations of the experimental
variables, and poor construct validity for the dependent variable. First, the study may
have been insufficiently powered to detect significant effects. While the sample size
calculations I conducted suggested that a cell size of seven would be sufficient, the
calculations were based upon the effect size found for displaced aggression in adults.
The effect size might differ for TDA and the effect size for either displaced
aggression or TDA may differ for children. In the published TDA study most similar
to the present study, a cell size of 12 demonstrated significant effects with adults that
supported hypotheses similar to mine (Pedersen et al., 2000). However, one
published TDA study with significant results published in the top social psychology
journal had a cell size of seven (Bushman et al., 2005). Thus, while more data may
have elicited significant results, previous studies have detected TDA using
reasonably similar procedures with as many participants as were analyzed in the
present study. It is possible that the translation of the TDA paradigm to a child-
friendly experiment may have altered the strength of the phenomenon. After
49
collecting 50 participants used to pilot the procedures, resource limitations only
permitted the collection of seven participants per cell based upon the sample size
calculations; future studies should use larger cell sizes to increase power to ensure
the ability to detect significant effects.
Further, there are a number of individual variables that were not assessed
which may have accounted for differences, or lack thereof, and should be collected
in future studies. It is possible that a child’s comfort level with video games as well
as their expectations of winning such games may have altered their reaction to an
obstruction of their goal of winning. Future studies might ask children about their
frequency of computer game play and their competency in winning computer games
as well as the importance of winning the car racing game and how the decision was
made to move their car forward or their opponents’ car backward on the first move.
Then these variables could be compared across groups to examine group differences
and could be examined as a possible correlate of aggression to ascertain whether
these efficacy and outcome beliefs regarding computer games might affect
aggressive behavior.
Another possibility for the lack of significant results is that either the
provocation or trigger manipulation was not effective in eliciting TDA. Regarding
the manipulation check for the provocation, the RAs filled out the checklist of
children’s behaviors following the pinball machine game. Thus, they were aware of
the child’s provocation manipulation status (i.e., provocation or no provocation) and
this knowledge may have contaminated their ratings leading to experimenter bias
50
(Epley & Dunning, 2006). However, the similarity in the RA observations to the
participants’ self-report measure of their emotional state following the pinball
machine game indicates that the ratings were likely a valid representation of the
participants’ reactions. The participants in the provocation group endorsed
significantly more self-reported negative affect and displayed significantly more
behavioral signs of negative affect as rated by the research assistants than
participants receiving no provocation making it very likely that the provocation
manipulation was effective.
While the manipulation check for the provocation appeared to have been
effective, future studies may want to videotape the children so independent
observational ratings of the children’s reactions can be made to ensure internally
valid measurement of the provocation manipulation check. Further, it may be helpful
to inquire whether participants felt justified in their anger or if they felt shame, guilt,
or inhibition by using a scale administered by the examiner such as the State-Trait
Anxiety Inventory for Children (STAIC), which is designed for this exact age group
(i.e., 9 - 12 years old) (Spielberger, 1973).
Unfortunately, no manipulation checks were made on the trigger
manipulation to be able to ascertain the validity of the trigger. Vasquez et al. (2005)
demonstrated that a trigger that was too strong would elicit its own aggressive
responding and would be seen as an entirely separate provocation while a trigger that
was too minor would not elicit any aggressive responding. Thus, it is possible that
the participants found the trigger manipulation of the taunt by his peer to be a strong
51
provocation. While no formalized measure was obtained of children’s responses to
the trigger, personal communications from the research assistants suggested that
many children in the provocation group smiled or did not seem fazed by the
opponents’ taunt while others took it as a motivation to play competitively.
According to these communications, few participants displayed the same intensity of
negative affect as were seen accompanying the provocation manipulation. Thus, it is
more likely that the trigger was not strong enough.
Perhaps the audio recording of another child’s voice was not proximal
enough and a visual cue would have enhanced the effect of the trigger. However,
other studies suggest that it is the intensity of the content of the trigger rather than
the type of cue used that will influence the effectiveness of a hostile interaction
(Dodge, 1980). Dodge’s experiment with aggression in children in 1980 used a
similar procedure with hearing another child’s voice broadcast from another room. In
this study, however, in the manipulations in which the opponent (i.e., confederate)
was overtly hostile, the opponent’s comments and actions were much more clear in
the intent to harm the participant, whereas in the current study, the opponent could
have been viewed as competitive.
Similarly, TDA studies with adults have directly insulted the participants
based on performance in an earlier part of the study whereas in the current study, the
opponent had no basis for their insult so the participant could easily ignore the
comment or view it as competitive; thus, the participants may not have interpreted
the comment as an interpersonal attack. Future iterations of this study should use
52
manipulation checks for the trigger and should consider making the trigger stronger
either by directly indicating the opponent’s desire to harm the participant in some
way (Miller et al., 2003) or by making a direct insult of the participant’s character
based on other performance (i.e., “I can’t believe you only got 180,000 points. You
must be really bad at video games. It’ll be so easy to beat you.”)
Finally, a fourth threat to internal validity may have been the lack of
construct validity of the dependent measure. The measure used was based on
literature which has validated dichotomous dependent variables for aggressive
behavior (Dodge, 1980; Swaim et al., 2006; Cohen & Prinstein, 2006) and literature
that has used aggressive moves made in a computer game as a representation of
aggressive behavior (Cherek et al., 1996). In this study, however, there were no
significant correlations between the participants’ questionnaire data and the
dependent variable. If the dependent variable was a good measure of aggression,
there ought to have been significant correlations between the aggressive beliefs,
aggressive behavior, and the mood questionnaires with the dependent variable.
Support for this assertion can be found in a recent empirical investigation of TDA in
which no group differences were found, but in which the questionnaire measures of
aggression were significantly correlated with the dependent variable of aggression
(Borders, 2006). The use of data from three different stages of the computer game as
potential dependent variables also indicates that it was not the amount of time the
participants were able to play the game that led to a lack of significant findings.
53
If, in fact, there were problems with the construct validity of the measure of
aggression, one reason could be that the instructions regarding the dependent
variable may have been flawed. There may have been a lack of motivation to be
aggressive due to the non-competitive nature of the game in which the dependent
variable was measured. The presumption was that if a participant was told that there
was no advantage to moving their car forward or the opponent’s car backward, any
move backward would be clearly aggressive. This statement was given to avoid
competitive game playing behavior which could be mistaken for aggressive
behavior, but also led to the game appearing as though it was based purely on the
luck of rolling the dice. Without any kind of reinforcement for moving the
opponent’s car backwards (i.e., no rewards for the participant and no harm to the
opponent), there may have been no motivation to do so. This may have undermined
the ability to assess aggressive behavior. Without any negative effects on the
opponent, any moves were inherently non-aggressive as they lacked the “intent to
harm” element of aggressive behavior (Anderson & Bushman, 2002).
There is evidence to support this hypothesis. Two-thirds of the sample said
yes or maybe to the question, “Did you think you were hurting your opponent’s
chance of winning when you moved them backwards?” likely indicating unclear
instructions prior to the game. Data suggest that the more a child believed that
moving the opponent backward would hurt his opponent, the more likely he was to
move the child’s car backwards on the first move. Participants who believed they
could harm the opponent’s chance to win may have indeed been acting aggressively
54
when they moved the participant back. Future research might include a design in
which there are varied outcomes based on a participant’s move which would elicit
different benefits and penalties to allow for the measurement of aggression that is
distinct from competitiveness similar to the design of the PSAP (Cherek et al., 1996).
In addition to the potential threats to internal validity, a number of threats to
external validity must also be examined including the representativeness of the
sample by race/ethnicity, other demographic variables, and developmental stage as
well as the ecological validity of the laboratory paradigm. The ethnicity/race of the
sample may have affected both internal and external validity. I recruited a sample
from the Los Angeles area which includes great diversity in both ethnic groups and
socioeconomic status. While the sample was quite ethnically diverse, the use of a
convenience sample (as opposed to a population-based sample) may have led to a
sample that was not ethnically representative of the children in the LA area and may
not be representative of other contexts. However, there were no group differences in
the proportions of the two largest ethnic groups making it unlikely that ethnic/racial
differences would have contributed to the outcome. It is unclear whether there would
be significant differences between ethnicity/race in rates of aggression were sample
sizes to be sufficiently powered to make such comparisons. Such an investigation is
warranted given the literature suggesting different rates of aggression by
ethnicity/race (Lochman & Wayland, 1994).
It is also unclear if, by choosing to participate in the study, the participants in
the sample are unique from those who choose not to contact or participate in the
55
study. It is also unclear whether the rationale for participating and the rewards
promised to the child after the experiment might alter a child’s likelihood of
aggression (e.g., monetary incentive, exposure of child to science and higher
education) (Tomporowski, Simpson, & Hager, 1993). Family income, parental
involvement, family education level, parenting styles, and children’s academic
engagement were not assessed; thus, the sample cannot be compared to the greater
Los Angeles area. Future studies may want to systematically ask questions regarding
the means by which families were recruited into the study as well as questions on
child and parent engagement, parenting styles, and family demographics.
In addition, children’s stage of social development may alter their reactions
to provocation. Indeed, researchers have long shown that children’s ability to
differentiate intent varies with age which may directly contribute to their aggressive
behavior (Shantz & Voydanoff, 1973). Thus, the protocol used might apply to a
limited age range and results are not generalizable to all stages of development.
Finally, the ecological validity of the study design is a potential threat to
external validity; that is, how analogous was the design to everyday situations a child
may face. In a meta-analytic investigation of laboratory studies of aggression,
experiments were shown to have strong external validity (Anderson & Bushman,
1997). However, in the current study, procedures and analyses were modeled after
empirical studies of TDA collected with college students. Ethnicity/race was not
reported in those studies (Pedersen et al. 2000, Bushman et al., 2005) making direct
comparisons impossible. It is possible that the samples were different enough in
56
variables such as ethnicity/race or family education as well as developmental stage
that the procedures may have not had the same level of success amongst a child
population.
In the present study, however, I tried to establish a paradigm which used the
essential elements of the TDA paradigm used with adults while maintaining a child-
friendly design that would engage children for recruitment purposes and for effortful
participation during the experiment (i.e., computer games as the vehicle for the
paradigm). It is unclear if children would be too inhibited in the laboratory to express
aggressive behavior due to the presence of an adult figure. While the RA exits the
room during the car racing game in order for the participant to feel less social
desirability, it is still possible that social desirability may have impacted a child’s
decisions. On the other hand, one study has found that participants are more likely to
be aggressive if they are playing an opponent on a computer than in a face-to-face
setting (Williams & Clippinger, 2002). Though many other experimental studies
have used computer and video games with children, due to the aforementioned
ambiguity, it is uncertain if the design has ecological validity at this point. Future
studies may incorporate a way to model aggressive reactions to adult blame and
generate social support for aggression through peer interactions to avoid inhibited
responses.
In addition to the child-friendly framework of a computer game study, the
design includes some elements of a highly ecologically-valid paradigm, the Point
Subtraction Aggression Paradigm (PSAP; Cherek et al., 1996), such as the use of a
57
computer game and a fake opponent as well as the option to move themselves
forward or their opponent backward. However, we were not able to use the exact
paradigm. The PSAP’s use of multiple 25-minute sessions would make recruitment
of children and their families very difficult unless a significant increase in
compensation were offered. This was not possible with the resources available for
the current project. In addition, the software for the PSAP is licensed and immutable.
This might lead to a problem in that the participant does not win money for
subtracting points from his opponent while the opponent does win money for
subtracting points from the participant. This could serve as a trigger, in and of itself,
and might make it difficult to have a non-trigger group. Future researchers should
use a design that is close enough to validated paradigms to ensure ecological validity
while still being child-friendly enough to ensure child participation.
Future Directions
The first step for researchers is to design a paradigm which reliably elicits
TDA in children. First, researchers should aim for a larger cell size, perhaps the more
typical 12 participants per cell seen in TDA studies (Pedersen et al., 2000), to
increase power. Second, they should alter the trigger so it more directly relates to the
participant’s performance (Miller et al., 2003) and a manipulation check should be
added to validate the effectiveness of the trigger. Such a measure might include
questions regarding the intent to harm and the meaning of a move forward or
58
backwards as well as the interpretation of the trigger (i.e., did the participant
interpret the trigger interpersonally?).
In addition, the second game should be designed in a way that elicits
aggression that is separate from competition. The findings demonstrated that children
were more likely to make an aggressive move if they believed they could harm their
opponent. Thus, it seems critical to make certain that the participant is cognizant of
their ability to harm the opponent in some way.
The PSAP should serve as a good base for a design that elicits aggression
which is distinct from competition. In this paradigm, it is far easier to subtract points
from the opponent (i.e., taking monetary earnings away from them) than to earn
one’s own points. In addition, once the participant subtracts the opponent’s points,
they receive a 250 second grace period in which their own points cannot be removed.
However, the time they spend harming their opponent is time they cannot earn
money. In this case, subtracting points from the opponent is clearly aggressive as
there is an obvious and immediate negative result when they act aggressively (i.e.,
they see the point total of the opponent decrease). However, it is not competitive in
that the goal is not to win a prize by being the leader at the end of the game, but
rather, each person wins the amount of money they have earned. Researchers will
have to take care to ensure that the game playing rules are the same for both players
with another form of trigger utilized rather than the unequal version of one player
being able to earn the opponent’s subtracted points. Alternatively, this imbalance in
point earning abilities could be utilized as the trigger while the non-trigger game
59
would not allow either player to earn points for subtracting points from their
opponent.
If TDA is demonstrated in a child sample, the next step would be to see if the
phenomenon was influenced by person variables such as hostile attributional biases,
aggressive beliefs, and values which support aggression as posited by the GAM
(Anderson & Bushman, 2002). Such an investigation might include a sample of
children who had been identified as aggressive by their peers and a sample of
children who had been identified as non-passive and non-aggressive by their peers to
serve as controls. While TDA would be expected for all children, it might be
expected that the magnitude of the aggression would be greater for aggressive
children.
It might also be interesting to examine which person variables account for the
greatest proportion of the variance in TDA. In my master’s thesis which was later
published, we found that aggressive children who are not victimized by their peers
are more likely to have strong self-efficacy, expectancy, and outcome beliefs
regarding aggression, but normative levels of emotion regulation and hyperactivity
as rated by their teachers. On the other hand, aggressive children who are victimized
by their peers have equivalent levels of positive beliefs about aggression compared
to the control group, but were rated as emotionally dysregulated, hyperactive, and
not well-liked by their peers (Toblin, Schwartz, Gorman, & Abou-Ezzeddine, 2005).
Another study could compare aggressive, non-victimized children (i.e., bullies) and
aggressive, victimized children (i.e., aggressive victims) on the magnitude of TDA
60
exhibited by each group. If person variables are found to be risk factors for
exhibiting TDA, explaining situations that elicit TDA could be incorporated into
already existing intervention programs which work to reduce social-cognitive
deficits (Hudley & Graham, 1993; Sukhodolsky, Golub, Stone, & Orban, 2005).
In summary, this study aimed to validate the construct of TDA in a child
sample. There were no group differences found between groups receiving a
provocation, trigger, both, or neither. Some explanation for why the study may have
not worked include insufficient power, a trigger that was too mild, and a poor
measure of aggression due to instructions which led the dependent variable used to
assess aggression to rely on luck more than on an intent to harm. The data analyzed
in this study may serve as pilot data for future researchers who can further refine the
paradigm to reliably elicit TDA in children and engage in a line of research
regarding TDA as a situational contributor to aggression in children.
61
References
Abela, J. R. Z., Brozina, K., & Haigh, E. P. (2002). An examination of the response
styles theory of depression in third- and seventh-grade children: A short-term
longitudinal study. Journal of Abnormal Child Psychology, 30, 515-527.
Anderson, C. A., & Bushman, B. J. (1997). External validity of “trivial”
experiments: The case of laboratory aggression. Review of General Psychology,
1, 19-41.
Anderson, C. A., & Bushman, B. J. (2002). Human aggression. Annual Review of
Psychology, 53, 27-51.
Anderson, C. A., & Carnagey, N. L. (2004). Violent evil and the general aggression
model. In A. Miller (Ed.), The Social Psychology of Good and Evil (pp. 168-
192). New York: Guilford Publications.
Anderson, C. A., & Huesmann, L. R. (2003). Human aggression: A social-cognitive
view. In M. A. Hogg & J. Cooper (Eds.), The Sage Handbook of Social
Psychology (pp. 296-323). Thousand Oaks, CA; Sage Publications.
Axelrod, R. (1984). The evolution of cooperation. New York: Basic Books.
Bandura, A. (1977). Social learning theory. Englewood Cliffs, N.J.: Prentice-Hall,
Inc.
Baron, R. A., & Bell, P. A. (1975). Aggression and heat: Mediating effects of prior
provocation and exposure to an aggressive model. Journal of Personality and
Social Psychology, 31, 825-832.
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.
Borders, A. (2006). The effects of mood-specific rumination on aggression.
Unpublished doctoral dissertation.
62
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.
Cherek, D. R., Schnapp, W., Moeller, F. G., & Dougherty, D. M. (1996). Laboratory
measures of aggressive responding in male parolees with violent and nonviolent
histories. Aggressive Behavior, 22, 27-36.
Cohen, G. L., & Prinstein, M. J. (2006). Peer contagion of aggression and health risk
behavior among adolescent males: An experimental investigation of effects on
public conduct and private attitudes. Child Development, 77, 967-983.
Côté, S. M., Vaillancourt, T., LeBlanc, J. C., Nagin, D. S., & Tremblay, R. E.
(2006). The development of physical aggression from toddlerhood to pre-
adolescence: A nation wide longitudinal study of Canadian children. Journal of
Abnormal Child Psychology, 34, 71-85.
Crick, N. R., & Dodge, K. A. (1994). A review and reformulation of social-
information-processing mechanisms in children's social adjustment.
Psychological Bulletin, 115, 74-101.
Dodge, K. A. (1980). Social cognition and children’s aggressive behavior. Child
Development, 51, 162-170.
Dodge, K. A. (2006). Translational science in action: Hostile attributional style and
the development of aggressive behavior problems. Development and
Psychopathology, 18, 791-814.
Dollard, J., Doob, L. W., Miller, N. E., Mowrer, O. H., & Sears, R. R. (1939).
Frustration and aggression. New Haven, CT: Yale University Press.
Epley, N., & Dunning, D. (2006). The mixed blessings of self-knowledge in
behavioral prediction: Enhanced discrimination but exacerbated bias. Personality
and Social Psychology Bulletin, 32, 641-655.
Eron, L. D. (1987). The development of aggressive behavior from the perspective of
a developing behaviorism. American Psychologist, 42, 435-442.
Gouldner, A. W. (1960). The norm of reciprocity: A preliminary statement.
American Sociological Review, 25, 161-178.
63
Hudley, C. A. & Graham, S. (1993). An attributional intervention to reduce peer-
directed aggression among African-American boys. Child Development, 64, 124-
138.
Huesmann, L. R. (1988). An information-processing model for the development of
aggression Aggressive Behavior, 14, 13-24.
Huesmann, L. R., & Guerra, N. G. (1997). Children’s normative beliefs about
aggression and aggressive behavior. Journal of Personality and Social
Psychology, 72, 408-419.
Irwin, A. R. & Gross, A. M. (1995). Cognitive tempo, violent video games, and
aggressive behavior in young boys. Journal of Family Violence, 10, 337-350.
Kupersmidt, J. B., Griesler, P. C., DeRosier, M. E., Patterson, C. J., & Davis, P. W.
(1995). Childhood aggression and peer relations in the context of family and
neighborhood factors. Child Development, 66, 360-375.
Lahey, B. B., Waldman, I. D., & McBurnett, K. (1999). Annotation: The
development of antisocial behavior: An integrative causal model. Journal of
Child Psychology, Psychiatry, and Allied Disciplines, 40, 669-682.
Lewin, K. (1939). Field theory and experiment in social psychology: Concepts and
methods. American Journal of Sociology, 44, 868-896.
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.
Lochman, J. E., & Wayland, K. K. (1994). Aggression, social acceptance, and race
as predictors of negative adolescent outcomes. Journal of American Academy of
Child and Adolescent Psychiatry, 33, 1026-1035.
Loeber, R., & Hay, D. (1997). Key issues in the development of aggression and
violence from childhood to early adulthood. Annual Review of Psychology, 48,
371-410.
Marcus-Newhall, A., Pedersen, W. C., Carlson, M., & Miller, N. (2000). Displaced
aggression is alive and well: A meta-analytic review. Journal of Personality and
Social Psychology, 78, 670-689.
64
Miller, N., & Marcus-Newhall, A. (1997). A conceptual analysis of displaced
aggression. In R. Ben-Ari, & Y. Rich, (Eds.), Enhancing education in
heterogeneous schools: Theory and application (pp. 69-108). Rarnat-Gan, Israel:
Bar-Ilan University Press.
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.
Miller, P. H., & DeMarie-Dreblow, D. (1990). Social-cognitive correlates of
children’s understanding of displaced aggression. Journal of Experimental Child
Psychology, 49, 488-504.
Moore, S. G. (1964). Displaced aggression in young children. Journal of Abnormal
and Social Psychology, 68, 200-204.
Orobio de Castro, B., Slot, N. W., Bosch, J. D., Koops, W., & Veerman, J. W.
(2003). Negative feelings exacerbate hostile attributions of intent in highly
aggressive boys. Journal of Clinical Child and Adolescent Psychology, 32, 56-
65.
Pedersen, W. C., Aviles, F. E., Ito, T. A., Miller N., & Pollock, V. E. (2002).
Psychological experimentation on alcohol-induced human aggression.
Aggression and Violent Behavior, 7, 293-312.
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.
Raine, A., Dodge, K. A., Loeber, R., Gatzke-Kopp, L. M., Lynam, D. R., Reynolds,
C., & Stouthamer-Loeber, M. (2003). Proactive and reactive aggression in
adolescents. Unpublished manuscript.
Raine, A., Dodge, K. A., Loeber, R., Gatzke-Kopp, L. M., Lynam, D. R., Reynolds,
C., Stouthamer-Loeber, M., & Liu, J. (2006). The Reactive-Proactive Aggression
Questionnaire: Differential correlates of reactive and proactive aggression in
adolescent boys. Aggressive Behavior, 32, 159-171.
Shantz, D. W. & Voydanoff, D. A. (1973). Situational effects on retaliatory
aggression at three age levels. Child Development, 44, 149-153.
Spielberger C. D. (1973). Manual for the State-Trait Anxiety Inventory for Children
(STAI-C). Palo Alto, CA: Consulting Psychologists Press.
65
Sukhodolsky, D. G., Golub, A., Stone, E. C., & Orban, L. (2005). Dismantling anger
control training for children: A randomized pilot study of social problem-solving
versus social skills training components. Behavior Therapy, 36, 15-23.
Swaim, R. C., Henry, K. L., & Kelly, K. (2006). Predictors of aggressive behaviors
among rural middle school youth. The Journal of Primary Prevention, 27, 229-
243.
Toblin, R. L., Schwartz, D., Gorman, A. H., & Abou-ezzeddine, T. (2005). Social-
cognitive and behavioral attributes of aggressive victims of bullying. Journal of
Applied Developmental Psychology, 26, 329-346.
Tomporowski, P. D., Simpson, R. G., & Hager, L. (1993). Method of recruiting
subjects and performance on cognitive tests. The American Journal of
Psychology, 106, 499-521.
Tremblay, R. E., Nagin, D. S., Seguin, J. R., Zoccolillo, M., Zelazo, P. D., Boivin,
M., Perusse, D., and Japel, C. (2004). Physical aggression during early
childhood: Trajectories and predictors. Pediatrics, 114, e43-e50.
Valente, T. W. (2002). Evaluating Health Promotion Programs. New York: Oxford
University Press.
Valois, R. F., MacDonald, J. M., Bretous, L., Fischer, M. A., & Wanzer Drane, J.
(2002). Risk factors and behaviors associated with adolescent violence and
aggression. American Journal of Health Behavior, 26, 454-464.
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.
Walden, T. A., Harris, V. S., & Catron, T. F. (2003). How I Feel: A self-report
measure of emotional arousal and regulation for children. Psychological
Assessment, 15, 399-412.
Weiss, M. G., & Miller, P. H. (1983). Young children’s understanding of displaced
aggression. Journal of Experimental Child Psychology, 35, 529-539.
Williams, R. B., & Clippinger, C. A. (2002). Aggression, competition, and computer
games: Computer and human opponents. Computers in Human Behavior, 18,
495-506.
66
Worchel, P. (1966). Displacement and the summation of frustration. Journal of
Experimental Research in Personality, 1, 256-261.
Zillman, D. (1988). Cognition-excitation interdependencies in aggressive behavior.
Aggressive Behavior, 14, 51-64
67
Appendix A. Recruitment Flier
Does your child
like to play
computer games?
Is your child
between
9-12 years old?
If you answered yes to both questions, you and your child
may be able to be in a 1 hour USC psychology research
study that takes place at USC! By calling, you will be
entered into a drawing for $200*. By participating, you
will receive $25 and your child may win USC prizes.
* Participation is not required for an entry in the drawing.
Please contact Robin Toblin, M.A. at
213-740-3638 or
uscpsychstudy@sbcglobal.net
David Schwartz, Ph.D., Faculty
Advisor, Dept. of Psychology
68
Appendix B: Protocol for CTDA: Children
Groups
1=Frustration, Trigger
2=No Frustration, Trigger
3=Frustration, No Trigger
4=No Frustration, No Trigger
Set up for experiment (for RAs dealing with parents).
1. Place “Experiment in Progress” sign on SGM 625 found in the third drawer in
the “Signs” folder.
2. Take out next folder in drawer in numerical order. Folders are in the top drawer
of the gray file cabinet in the back of the storage closet.
3. Order the packets so that the parent forms are together in this order:
a. Parental Consent
b. Demographic Checklist
c. CAQ-P
d. Debriefing Statement for Parents
e. Payment Receipt
f. Envelope to send results home
4. Place a-c on top of one another on top of the folder. Leave d-f in the folder.
5. Bring out “Sample Folder 88” from the third drawer and place on table along
with a pen.
6. Order the packets so that the children’s forms are together.
If the participant ID is an even number (2, 34, 68, etc.), order them like this:
a. Child Assent
b. CAQ-C
c. How I Feel
d. What I Think
e. Experimenter Checklist
f. Space Cadet
g. Car Racing Game
h. Debriefing Statement for Children
If the participant ID is an odd number (9, 43, 75, etc.), order them like this:
a. Child Assent
b. What I think
c. How I Feel
d. CAQ-C
e. Experimenter Checklist
f. Space Cadet
g. Car Racing Game
h. Debriefing Statement for Children
69
7. Place a-d on top of each other face up in the middle of the table in SGM 625A
without the laptop. Place e-h on top of each other on the upper left corner of the
same table face down so the participant doesn’t look at them.
8. Place pencil (with eraser if possible) on the same table.
9. Turn on laptop.
10. Double click on Games folder.
11. Double click on Game 1 folder.
12. Double click on Game 1 icon.
13. Close Game 1 folder.
14. If participant is in Group 1 or Group 3, use drop down menu to answer “yes” for
Interrupt game during experiment?”
g. Type in “60” for “Run game for how many seconds?”
h. Click on Start, but then leave the pinball machine game on the start page
with the graphic that has a guy in a space copter and DON’T click on
Start. The experimenter will click on start when the child is ready.
15. If participant is in Group 2 or Group 4, use drop down menu to answer “no” for
“Interrupt game during experiment?”
i. Click on Start, but then leave the pinball machine game on the start page
with the graphic that has a guy in a space copter and DON’T click on
Start. The experimenter will click on start when the child is ready.
16. Double click on Games folder on desktop (again).
17. Double click on Game 2 folder.
18. Double click on Game 2 icon.
19. Close Game 2 folder.
20. Enter participant’s ID number in space.
21. Click on “Begin Game”.
22. Minimize both game windows so they are on the toolbar.
23. Get $25 from drawer to have ready in folder.
Actual Experiment; C= Child; KE= Kid experimenter; PE= Parent experimenter
24. KE picks up parent and child at main doors to SGM opposite corner from PSA.
Introduces themselves and chats with both (asks questions re: where do they live,
what grade is child in, where do they go to school, weather, etc.)
25. KE brings them up to the room.
26. KE introduces PE to the parent and child by name.
27. PE: Ok, we’ll be splitting up now. I’ll be staying out here with your (relative’s
relation such as mother/father/aunt) and KE will be going into this room over
here with (child’s name).We’ll have the door closed most of the way, so that we
can hear the person we are talking to better. Just like I’m working with you,
(child’s name), a woman named Teresa is working with another kid that is also
doing the study. After I talk to your (child’s relative), I’ll be doing the same thing
70
with the other kid’s relative and then I’ll be going back and forth between the
rooms to make sure everything is ok.
28. KE goes over assent form with child:
29. KE: “You are here today to be a part of a project for Robin Toblin. (If Robin is
there, add on, “who you just met.”). This project will help Robin graduate from
school. We want to learn more about how children think, feel, and act while
playing computer games, by themselves and against another child.
30. If you agree to be in the study, we will ask you to play some computer games and
fill out some surveys. One will ask you about how you’ve been feeling for the
past couple of months, another about what you think about things that could
happen to other kids, and the last one is about the way you act. Your (child’s
relative) will also fill out surveys.
31. We won’t be using your name in this project, but will give you a code number to
put your information into the computer. So you are number (say ID number and
point to it on the corner of the page). We have to keep the papers in a locked
cabinet in a locked room in the lab for seven years, but your answers won’t count
for anything else except this project. You can’t be matched to your answers since
we won’t use your actual name. Your answers won’t be shown to anyone, not
even your parents, and it won’t affect your school grades at all.
32. There are things that we would have to tell your parents if you told us, though. If
you tell us that you are thinking about hurting yourself or other people, we will
have to tell your parent. Or, if you tell us that your parent is hurting you, we will
have to tell the Department of Child and Family Services, and they may send
someone to your home to learn more about it.
33. After you fill out the surveys, we will ask you to play two computer games, and
depending on the amount of points you have, you may win some prizes. Between
the games, you will take another survey. The first game is a pinball game and the
second is a car racing game against that other kid we told you about before. You
won’t meet the other kid, but you will get to introduce yourselves over this
walkie talkie right before you play the game against each other. Until then he will
be doing the same thing as you, but in another room. The winner gets a prize and
the one who doesn’t win will get a small consolation prize. You will also fill out
another survey after the car game. This should all take less than an hour.
34. The project shouldn’t be hard, but you may get frustrated, just like you would if
you were playing a game at home. There are some things about the project that
we can’t tell you now, but we will tell you about it when the project’s over. They
shouldn’t cause you any problems. We have also asked your parents to say it’s
okay for you to participate. But, even if your parents say “yes”, you don’t have to
participate if you don’t want to. Being in this study is your choice and no one
will be upset if you decide you don’t want to be in it, or if you change your mind
later and want to stop. You also don’t have to answer any questions you don’t
want to and if you stop in the middle of it, you will still get some small prizes for
playing the pinball game.
71
35. You can ask any questions about the study and if you have a question later, you
can call me or e-mail me. You can also ask your parent to call. All of our
information will be in a packet we will be giving to you after we are finished.
36. Signing your name at the bottom of the sheet means you agree to be in the study;
remember, you can still change your mind at any time during the study.
37. C: “Okay...”
38. KE: “Do you still want to be in the study?”
39. C: “Yes.”
40. KE: “Okay, so just go ahead and print your name and the date right here (point to
lines). Today’s date is . And then sign right here (point to line).”
a. If they ask which to write their name in (cursive or print), say “Cursive”*
41. Child signs and dates assent form.
42. KE: “Okay now I’m going to sign right here, to show that I was here when you
signed it.” (Sign and date assent form).
43. Explain the three surveys (note: surveys will be in reverse order than the
description below if the child’s id number is odd).
44. E: “Okay so now you’re going to fill out these three surveys. The first survey
(point to survey titled “How I Feel” or “CAQ-C” depending on whether the
participant’s Id is even or odd. If it is odd, skip to 46, and then come back to 44.)
is going to ask you questions about how you’ve felt in the past three months, so
since (count back three months, or say something like since the
beginning of the school year, since the middle of the summer, etc.). Rate how
true you think that statements describe the way you have acted in the past couple
of months by checking one of the lines. Only check one.
45. The second survey (point to survey titled “What I Think”) is going to ask you
some questions about how you think about things that could happen to other kids
and whether or not you think certain behaviors are wrong or are okay. Circle the
answer that best describes what you think, but make sure you only circle one
answer.
46. The last survey (point to survey titled “CAQ-C”) will ask you questions about the
way you act and how often you do certain things. You will read a statement and
then circle which choice is most true for you.
47. Answer the questions as best as you can, but remember if you don’t want to
answer a question, you don’t have to; no one will get mad at you. Do you have
any questions?”
48. Answer any questions they have, or if they don’t move right along:
49. KE: “Okay, I’m going to go into the other room now while you fill these out. If
you have any questions, just call open the door or call my name and I’ll come
back in to answer them. Okay?”
50. C: “Okay.”
51. Go into room with parent. Sit and wait until kid is ready.
52. Go back in room when ready. Turn surveys over on desk and move to other desk
where computer games are set up.
72
53. E: “Okay, so now that you’re done with those, you get to play the first computer
game. This is a pinball machine game; you may have played it before or even
have it at home. You will be playing this game for 3 minutes, and then that will
be your score for this game.
54. Depending on the score you get, you can win three different prizes.
55. (Go over to drawer and take out USC prizes).
56. For 100,000 points, you’ll get these cool USC stickers that you can put on your
books or somewhere else. For 200,000 points, you’ll also get a USC pen. And, if
you get 300,000 points, then you also get a USC megaphone and you can shout
into them during games.”
57. C: nods in agreement.
58. Walk back over to computer to explain the controls (will be color-coded for each
game; pinball= green, car racing=yellow).
59. KE: You hold down the space bar to launch the ball and then you use the “z”
button as your left flipper and the slash button (point to slash button right next to
Shift).
60. “Do you have any questions?”
61. Answer any questions, or move on.
62. KE: Ok, I’m going to call over to the other room to see if they are ready to start.
We want to make sure we start at the same time so you’ll be ready to play the car
racing game at the same time.
63. Call over to Robin on walkie talkie. “We’re ready to play the pinball game.”
64. PE: plays tape of Teresa saying “Oh, darn. He just ran to the bathroom. Let me
call you back when he’s back.” (Wait 15 seconds). “Ok, we’re ready to start.”
65. KE: “Okay, well I’m going to leave the room while you play. If you have any
questions, call my name. If not, I will see you in three minutes.”
66. C: “Okay.”
67. Begin timing with stop watch (should only play for three minutes).
68. If the child doesn’t call you in, then after three minutes, go back into the room
and tell the child to stop. This is the amount of points that they have accumulated
for this game.
69. KE: “Okay, it’s time to stop now. Your score for this game is (child’s score).
That means you win these prizes (give child the prizes he/she has won).”
70. “Cool.”
71. If the child calls you in to tell you there is a problem after one minute (usually
yells name, says “Excuse me”, etc.), go into room, do not act as if you were
expecting this.
72. C: “I don’t know what happened. The game just shut off.”
73. KE: “What did you do? This hasn’t happened with any of the other kids. Let me
see.”
74. Walk over to computer. Read the illegal action box aloud and act surprised.
75. KE: (act a little upset) “Well, you must have done something to mess it up. This
has never happened before. We don’t have time to try again since we have other
73
people coming in and this might be your fault anyway. I guess we’ll just have to
keep going with the experiment, but I can’t give you any of the prizes since I
don’t even know your score.” (Be sure to look at child’s face and listen to what
they say so you can fill out the Experiment Checklist. Take away the prizes and
put them back in the drawer).
76. If child says:
77. a. “Okay.” Keep going with experiment.
78. b. “Can’t I just replay the game?” Say: “No, we don’t have the time. This
hasn’t happened with any of the other kids. We have to just keep going. You
just can’t get any of the prizes.”
79. Bring Space Cadet questionnaire to child to fill out.
80. KE: “This survey is about the game you just played--how much you liked it and
how you feel. I’m going to leave again so you can fill this out. Call my name if
you have any questions. If not, call me when you’re done.”
81. C: “Okay.”
82. Remember to act neutrally towards child. Treat the child the same regardless of
which group they were in.
83. Leave the room.
84. Fill out Behavioral Checklist while waiting.
85. Come back in the room when child is done. Click out of the pinball game (may
already be clicked out of) and bring up the car racing game.
86. KE: “Now, you will play the second computer game. This game is a car racing
game that you will be playing against the other kid. You are going to be the red
car and the red dice where your opponent is going to be the blue car and the blue
dice. The winner of the game will get one of these rubber bouncy balls (open
drawer and show them selection). The person that doesn’t win will get one of
these sports erasers (open drawer and show them selection). We flipped a coin to
determine who would get to roll the dice throughout the entire game and you
won, so you will roll the dice every time. Each time you roll the dice, whoever
has the dice with the highest number after the roll gets to move. Every time it’s
your turn to move, you can move yourself forward or move your opponent
backward. Your chances of winning aren’t improved by moving forward or
moving your opponent backwards, they are both equal. So, you can press “u” to
move yourself forward, “d” to move the other kid back (point to buttons). Do you
have any questions?”
87. C: “Okay.”
88. C: “Cool.”
89. KE: “Okay, let me just call Teresa and tell them we’re ready. Also, like I
mentioned before, we’re going to have you introduce yourselves to each other.
The only things you should say are your name and school.”
90. KE: (in the walkie talkie) Ok, we’re ready to start the game.
91. PE: (play tape with Teresa saying) We’re ready to start the game. Let’s begin
with introductions.
74
92. KE: Say to kid: ok, go ahead and say your name and school into this walkie
talkie.
93. C: My name is _____ and I go to ______.
94. PE: (plays tape with David saying). “My name is David and I go to College Park.
[Oh, and I know I can beat you because I’m cooler than you.] (The bracketed
statement is only said in half of the groups).
95. Ask if they are ready. Click on File, Start. Leave room and wait for three
minutes.
96. After three minutes, go back into other room.
97. KE: “How did it go?”
98. C: I lost/tied/won.
99. E: If lost: “Oh well, at least you win an eraser! Which one do you want?” If tied,
say “That’s cool- that doesn’t happen very much. Then you both win the bouncy
ball.” If they won, “Awesome! Come choose out one of the bouncy balls.”
100. E: “Okay, now you have to fill out this survey (hand child car racing game
survey) about the second computer game--how much you liked it and how
you feel.” (Straighten up papers so they know you aren’t watching them fill
out the survey).
101. C: “Okay.”
102. When child is done, call PE on walkie talkie to tell them you are finished so
they can come in and debrief the parent. Begin debriefing the child:
103. KE: “Do you have any questions about the study?”
104. KE: “What grade do you think David was in?”
105. KE: “Did any of the things we did seem weird or uncomfortable?”
106. If they were in the provocation group (Groups 1 and 3) ask: “In your opinion,
was playing the pinball game frustrating or upsetting?”
107. If they say “yes”, say: “Sometimes an upsetting task, like the one you did,
can affect kids’ feelings. Did you experience any strong feelings during the
task?”
108. If they say “yes”, say: “What were they? Were they strong?”
109. KE: “If you had to guess, out of all the things we just had you do, which do
you think we really want to know most about?”
110. Child answers.
111. KE: “Why?”
112. Child answers.
113. Debriefing statement. Can read pretty much verbatim except for capital
letters which is an add-in.
114. “If you felt that how you did on the pinball game was fair or very unfair, this
was related to the group you were in and not to do with you or your abilities.
Before you got here, each kid was randomly put into different groups. For
some kids, during the pinball game, we had the pinball game shut down on
purpose so we could make sure those kids had a bad experience so they
would be upset. Then we had you play the car racing game to see how you
75
would treat the other child by seeing your moves in the game. We wanted to
see if making you mad would make you meaner during the game. For some
kids, we had the experimenter tell you that the other kid said that he or she
could beat you at the game. Other kids did not have that happen. Again, this
had to do with what group in our study you were in and not how good you are
at games.
115. There was NOT really another kid playing the car race computer game in
another room. We made you believe this so the way you acted and the moves
you made during the game were as real as we could make them. Your moves
in the game will NOT help or hurt another child in any way. We were just
trying to see how the way you were feeling might affect the way you acted
toward other kids. We hope you can understand that we were attempting to
make things as real as possible.
116. KE: “Did you think there was another kid?”
117. RECORD ANSWER on folder below the three lines that include group
number on the folder checklist on the left by just writing Yes or No.
118. The purpose of these computer games is to study how being mad or upset and
getting teased might change a child’s treatment of another child and how this
affects the number of spaces you moved the other kid’s car back. This
experiment was NOT being used to test how you behave while playing
computer games or how fun the games are, but we told you this because
knowing the real reason for the study might make it hard to act normally. We
are sorry if this experiment upsets you, but we hope you can understand that
we thought this would help you try your best.
119. Please don’t talk about this experiment with anybody else because kids that
know about this experiment will act differently than kids that don’t.
120. Remember, you can ask us to not include any of the parts of the experiment
in our study. You can stop participating in this study any time and you won’t
be in trouble at all. If there are any questions on the questionnaires you don’t
want included or you don’t want the results of either or both of the computer
games included, you can let us know that now or later on by contacting us by
email or phone, listed below.
121. If you have any questions about this study, you or your parent can contact us
with the info on this sheet which you can take home.
122. Now one last thing before I let you go. Since you were SUCH a good sport- I
am going to give you all the prizes you could have won (give any prizes they
haven’t won, bringing them out one by one and announcing them- yay, lots of
drama!)
123. Bring child back out to parent.
124. Go back to laptop.
125. Double click on Games
126. Double Click on Game 2
127. Double Click on Game 2 Results.
76
Appendix C: Questionnaire Measures
All About Your Child
Gender: Boy-0 Girl-1
Ethnicity: Please circle the ethnicity that best describes your child.
1-European American 2-Hispanic 3- American Indian/ 4-Black, Non-
(White) Alaska Native Hispanic
5-Filipino 6-Pacific Islander 7- Asian 8-Middle-Eastern
9-Armenian 10-Mixed (specify):_________________
What language(s) does your family speak at home?
1- English 2- Spanish 3-Cantonese 4-Tagalog 5-Armenian
6- Japanese 7-Mandarin 8-Farsi 9-Hebrew 10-Arabic
11-Laotian 12-Korean 13-Thai 14-Russian 15- Cambodian
16-Other______________
Your child’s age is: _________
Your child’s birthdate is
Month Date Year
1- Jan 7- Jul 1992
2- Feb 8- Aug _____ 1993
3- Mar 9- Sep 1994
4- Apr 10-Oct 1995
5- May 11- Nov 1996
6- Jun 12- Dec 1997
77
CAQ-P
How often are the following statements true for your child…
1. My child yells at others when they
annoy him/her.
Never Sometimes Often
2. My child fights others to show who
is on top.
Never Sometimes Often
3. My child gets angry when others
annoy him/her.
Never Sometimes Often
4. My child takes things from other
kids.
Never Sometimes Often
5. My child gets angry when
frustrated.
Never Sometimes Often
6. My child damages or breaks things
for fun.
Never Sometimes Often
7. My child has temper tantrums.
Never Sometimes Often
8. My child damages things when
he/she is mad.
Never Sometimes Often
9. My child gets into fights to be cool.
Never Sometimes Often
10. My child hurts others to win a
game.
Never Sometimes Often
11. My child gets angry or mad when
he/she doesn't get his/her way.
Never Sometimes Often
12. My child uses force to get others
to do what he/she wants.
Never Sometimes Often
13. My child gets angry or mad when
he/she loses a game.
Never Sometimes Often
14. My child gets angry when others
threaten him/her.
Never Sometimes Often
15. My child bullies other kids to get
money or things from them.
Never Sometimes Often
16. My child feels better after hitting
or yelling at someone.
Never Sometimes Often
78
17. My child threatens and bullies
other kids.
Never Sometimes Often
18. My child makes prank phone calls
just for fun.
Never Sometimes Often
19. My child hits others to defend
him/herself.
Never Sometimes Often
20. My child gets others to gang up on
other kids.
Never Sometimes Often
21. My child gets mad or hits others
when they tease him/her.
Never Sometimes Often
22. My child yells at others so they
will do things for him/her.
Never Sometimes Often
79
CAQ-C
How often are the following statements true for you…
1. I yell at others when they annoy me.
Never Sometimes Often
2. I fight others to show who is on top.
Never Sometimes Often
3. I get angry when others annoy me.
Never Sometimes Often
4. I take things from other kids.
Never Sometimes Often
5. I get angry when frustrated.
Never Sometimes Often
6. I damage or break things for fun.
Never Sometimes Often
7. I have temper tantrums.
Never Sometimes Often
8. I damage things when I am mad.
Never Sometimes Often
9. I get into fights to be cool.
Never Sometimes Often
10. I hurt others to win a game.
Never Sometimes Often
11. I get angry or mad when I don't get
my way.
Never Sometimes Often
12. I use force to get others to do what
I want.
Never Sometimes Often
13. I get angry or mad when I lose a
game.
Never Sometimes Often
14. I get angry when others threaten
me.
Never Sometimes Often
15. I bully other kids to get money or
things from them.
Never Sometimes Often
16. I feel better after hitting or yelling
at someone.
Never Sometimes Often
17. I threaten and bully other kids.
Never Sometimes Often
80
18. I make prank phone calls just for
fun.
Never Sometimes Often
19. I hit others to defend myself.
Never Sometimes Often
20. I get others to gang up on other
kids.
Never Sometimes Often
21. I get mad or hit others when they
tease me.
Never Sometimes Often
22. I yell at others so they will do
things for me.
Never Sometimes Often
81
How I Feel
Please rate the sentences below for how true each was of you in the past three
months.
1. I was happy very often.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
2. When I felt sad, my sad feelings
were very strong.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
3. I was in control of how often I felt
mad.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
4. I was excited almost all of the
time.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
5. When I felt scared, my scared
feelings were very powerful.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
6. When I felt happy, I could control
or change how happy I felt.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
7. I was sad very often.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
8. When I felt mad, my mad feelings
were very strong.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
9. I was in control of how often I felt
excited.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
10. I was scared almost all the time.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
82
11. When I felt happy, my happy
feelings were very powerful.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
12. When I felt sad, I could control
or change how sad I felt.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
13. I was mad very often.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
14. When I felt excited, my excited
feelings were very strong.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
15. I was in control of how often I
felt scared.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
16. I was happy almost all the time.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
17. When I felt sad, my sad feelings
were very powerful.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
18. When I felt mad, I could control
or change how mad I felt.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
19. I was excited very often.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
20. When I felt scared, my scared
feelings were very strong.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
83
21. I was in control of how often I
felt happy.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
22. I was sad almost all the time.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
23. When I felt mad, my mad
feelings were very powerful.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
24. When I felt excited, I could
control or change how excited I felt.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
25. I was scared very often.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
26. When I felt happy, my happy
feelings were very strong.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
27. I was in control of how often I
felt sad.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
28. I was mad almost all the time.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
29. When I felt excited, my excited
feelings were very powerful.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
30. When I felt scared, I could
control or change how scared I felt.
___ Not at all true of me
___ A little true of me
___ Somewhat true of me
___ Pretty true of me
___ Very true of me
84
What I Think
Instructions: The following questions ask you about whether you think certain
behaviors are WRONG or are OK. Circle the answer that best describes what you
think. Circle ONE and only one answer.
Suppose a boy says something bad
to another boy, David.
1) Do you think it’s ok for David to
scream at him?
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
2) Do you think it’s OK for David to
hit him?
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
Suppose a boy says something bad
to a girl.
3) Do you think it’s wrong for the girl
to scream at him?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
4) Do you think it’s wrong for the girl
to hit him?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
Suppose a girl says something bad
to another girl, Monica.
5) Do you think it’s ok for Monica to
scream at her?
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
6) Do you think it’s ok for Monica to
hit her?
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
Suppose a girl says something bad
to a boy.
7) Do you think it’s wrong for the boy
to scream at her?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
8) Do you think its wrong for the boy
to hit her?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
85
Suppose a boy hits another boy,
David.
9) Do you think it’s wrong for David
to hit him back?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
Suppose a boy hits a girl.
10) Do you think it’s ok for the girl to
hit him back?
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
Suppose a girl hits another girl,
Monica.
11) Do you think it’s wrong for
Monica to hit her back?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
Suppose a girl hits a boy.
12) Do you think it’s wrong for the
boy to hit her back?
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
13) In general, it is wrong to hit other
people.
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
14) If you’re angry, it is ok to say
mean things to other people.
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
15) In general, it is ok to yell at others
and say bad things.
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
16) It is usually ok to push or shove
other people around if you’re mad.
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.
17) It is wrong to insult other people.
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
18) It is wrong to take it out on others
by saying mean things when you’re
mad.
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
19) It is generally wrong to get into
physical fights with others.
A. It’s really wrong.
B. It’s sort of wrong.
C. It’s sort of ok.
D. It’s perfectly ok.
86
20) In general, it is ok to take your
anger out on others by using physical
force.
A. It’s perfectly ok.
B. It’s sort of ok.
C. It’s sort of wrong.
D. It’s really wrong.

Abstract (if available)

Abstract The present study aimed to validate the construct of triggered displaced aggression (TDA) in a middle childhood sample. The sample was comprised of 9-12 year old boys from the Los Angeles area (n=28) who were randomly placed into four groups: provocation plus trigger, provocation only, trigger only, or no provocation and no trigger. Participants and their caregivers completed a series of questionnaires. Participants then played in a series of two computer games which provided the vehicle for the provocation and trigger manipulations, respectively. There were no significant differences on aggression between any of the four groups. Exploratory analyses found that children who were more likely to believe they could harm their opponent were more likely to make an aggressive move in the game. Potential explanations for the lack of significant results are discussed. Limitations and future directions for research are also addressed. The current study can serve as pilot data for future researchers who aim to validate the TDA construct in a childhood sample.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Frustration within the triggered displaced aggression paradigm

PDF

The moderating effect of triggering provocation on direct aggression

PDF

The impact of naturally-occurring rumination on aggression

PDF

Projection on outgroups: complementary projection in aversive intergroup contexts

PDF

Examining the longitudinal relationships between community violence exposure and aggressive behavior among a sample of maltreated and non-maltreated adolescents

PDF

Emotion regulation as a mechanism linking parents’ marital aggression to adolescent behavioral problems: a longitudinal analysis

PDF

Implicit theory ideology about human attributes and aggression

PDF

Reducing gang involvement through employment: a pilot intervention

PDF

Mediators and moderators in the link between maternal psychological control and peer victimization for Hong Kong Chinese boys

PDF

The successful African American Parenting (SAAP) Program: an evaluation of a culturally sensitive parenting intervention

PDF

An investigation of the factors associated with electronic aggression among college students

PDF

A "second chance" program for vulnerable young adults: a program evaluation and a randomized controlled trial of adjunctive motivational interviewing to improve retention

PDF

Opportunity beliefs and behavioral outcomes in Latinx youth: an exploration of Ogbu’s cultural-ecological model

PDF

Within session therapist behavior in multisystemic therapy

PDF

Evaluating a cultural process model of depression and suicidality in Latino adolescents

PDF

Prior exposure to aggression and young adults' negative expectancies about romantic partner discussions

PDF

Persistence of pollution-induced lung function deficits in early adulthood: evidence from the Children's Health Study

PDF

Social norms intervention to promote help-seeking with depressed Asian and European Americans: A pilot study

PDF

Evaluating social-cognitive measures of motivation in a longitudinal study of people completing New Year's resolutions to exercise

PDF

The economy crunch: a multimedia website devoted to the economy and what we eat

Asset Metadata

Creator Toblin, Robin Laurie (author)

Core Title Children's triggered displaced aggression

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Publication Date 06/13/2007

Defense Date 03/22/2007

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

Tag aggression,Children,computer games,laboratory,OAI-PMH Harvest,triggered displaced aggression

Language English

Advisor Schwartz, David (committee chair), Huey, Stanley J., Jr. (committee member), Kipke, Michele D. (committee member), Margolin, Gayla (committee member)

Creator Email toblin@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m529

Unique identifier UC1153668

Identifier etd-Toblin-20070613 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-499830 (legacy record id),usctheses-m529 (legacy record id)

Legacy Identifier etd-Toblin-20070613.pdf

Dmrecord 499830

Document Type Dissertation

Rights Toblin, Robin Laurie

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