 

1339::
l IJBRARY
20061 Michigan State

I Ink 'nmml
VI "van on,

 

 

 

 

This is to certify that the
thesis entitled

AVATAR IDENTIFICATION: HOW SHVIILARIT Y IN
APPEARANCE NFLUENCES AGGRESSIVE RESPONSES

presented by

ASHLEY ORGAN

has been accepted towards fulfillment
of the requirements for the

MA. degree in Communication

 

 

Q2n§50wgéﬂwa

Major Professor’s Signature
4'33'09

Date

MSU is an Afﬁrmative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 K:IProj/Acc&Pres/ClRC/DateDuecindd

AVATAR IDENTIFICATION: HOW SIMILARIT Y IN APPEARANCE INFLUENCES
AGGRESSIVE RESPONSES

By

Ashley Organ

A THESIS

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

MASTER OF ARTS
Communication

2009

ABSTRACT

AVATAR IDENTIFICATION: HOW SIMILARIT Y IN APPEARANCE INFLUENCES
AGGRESSIVE RESPONSES

By

Ashley Anne Organ
Researchers argue that identiﬁcation with violent video game characters heightens
exposure's inﬂuence on aggression. Recently, scholars have argued that attributes of
gaming technology make character identiﬁcation in video games stronger than in other
media. One notable attribute said to heighten character identiﬁcation is the technology’s
ability to increase perceived similarity, or player perceptions that they are similar to a
game character. Social Cognitive Theory suggests that individuals are more likely to
identify with and imitate behavior from others they see as similar to the self. This study
tested the proposition that avatar similarity can increase identiﬁcation with characters in a
ﬁrst person shooter video game, and that the resulting identiﬁcation will increase
subsequent imitative behavior in the form of aggressive response. In an experiment,
avatar similarity in a FPS game was manipulated by customizing characters’ facial
appearance togemulate/deviate from the players’ appearance using face mapping and

morphing software. Identiﬁcation and state aggression were measured after game play.

TABLE OF CONTENTS

LIST OF TABLES ................................................................................... iv
LIST OF FIGURES ................................................................................. v
Introduction .......................................................................................... 1
Identiﬁcation and Aggression ..................................................................... 4
Avatar Similarity and Identiﬁcation .............................................................. 7
Method ................................................................................................ 10
Overview .................................................................................... 10
Participants .................................................................................. l 1
Procedure .................................................................................... ll
Stimulus Material .......................................................................... 12
Measures .................................................................................... 14
Results ................................................................................................ 16
Measurement Model ....................................................................... 17
Descriptive Statistics ...................................................................... 17
Hypothesis Testing ........................................................................ 19
Discussion ........................................................................................... 21
Limitations and Future Research ......................................................... 22
Conclusion .................................................................................. 23
Appendix A......... ................................................................................. 25
State Aggression Measure ................................................................ 25
Empathy Measure .......................................................................... 26
Identiﬁcation Measure ..................................................................... 28
Psychoticism Measure ..................................................................... 30
Appendix B .......................................................................................... 31
State Aggression ........................................................................... 31
Identiﬁcation ............................................................................... 32
Empathy ..................................................................................... 33
Bibliography ......................................................................................... 34

LIST OF TABLES

TABLE 1
Descriptive Statistics for Key Variables in Study .............................................. 18

TABLE 2
Zero-order Correlations among Key Variables in Study ...................... - ................. 19

TABLE 3
State Aggression Component Matrix ............................................................ 31

TABLE 4
Identiﬁcation Component Matrix. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32

TABLE 5
Empathy Component Matrix. . 33

LIST OF FIGURES

FIGURE 1
Morphed Image in Moderate Similarity Condition ............................................ 14

Avatar identiﬁcation: How similarity in appearance inﬂuences aggressive responses

On April 16, 2007, a massacre occurred on the Virginia Tech campus in
Blacksburg, Virginia. The massacre took the lives of 32 people and injured many more,
making it the deadliest school shooting in US. history. Violent acts such as this reﬂect
the violent culture in which we live. For the past several years, the literature on violence
has developed out of growing concern about the potential causes of such catastrophes.
Many scholars, non—violent activists, and media watchdogs (e. g., National Coalition on
Television Violence) claim that the media, speciﬁcally the violent video game industry, is
much to blame for the rise in violence and aggressive behavior. Though not unchallenged
(Aguilera & Mendiz, 2003; Sherry, 2001), the concerns over the medium’s harmful
effects may be warranted. Several studies have empirically conﬁrmed that there exists a
relationship between violent video game play and aggression (Bushman & Anderson,
2002; Gentile, Lynch, Linder & Walsh, 2004; Tamborini, Eastin, Skalski, Lachlan,
Fediuk & Brady, 2004). Several meta-analysis, such as the 2004 piece by Anderson, note
increases in aggressive behavior, cognition, affect, and physiological arousal as a result
of violent video game play.

With the growing concern about the medium’s aggressive effects, many scholars
have been interested in how identiﬁcation with violent video game characters relates to
the inﬂuence play has on aggression. Previous research efforts have speculated on
attributes of video games thought to heighten identiﬁcation with violent video game
characters, and the increased aggression expected to result (Anderson & Bushman, 2001;
Sherry, 2001; Anderson, 2004; Eastin, Appiah, & Cicchirillo, 2007; Schneider, Lang,

Shin & Bradley, 2004; Barlett, Harris & Baldassaro, 2007; Tamborini, 2004; Eastin &

Grifﬁths, 2006). Much of this research has focused on study of the ﬁrst-person shooter
video games.

Most scholars who study ﬁrst-person shooter (FPS) video games suggest that the
user perspective technology found in these games heightens identiﬁcation with players’
violent game characters by allowing the players to move through the violent video game
world through the eyes of violent game characters (Schneider et al., 2004; Barlett et al.,
2007; Tamborini, 2004; Eastin & Grifﬁths, 2006). Yet, user perspective technology may
not be the only attribute of game technology that can heighten identiﬁcation. One notable
attribute that might heighten character identiﬁcation is the game’s ability to increase
perceived similarity, or player perceptions that they are similar to a game character. This
. study tested the proposition that game technology capable of heightening avatar
similarity can increase character identiﬁcation. In the realm of violent FPS video games,
character identiﬁcation is ultimately expected to increase negative imitative behavior. To
date, these effects have not been widely studied.

The logic underlying this study is based on beliefs that video game technology
has recently added features to game play with the ability to heighten identiﬁcation
between video gamers and their avatars, ultimately increasing imitative behavior
expected to result from playing the game. Game designers have acquired the capacities to
create avatars which adeptly resemble the gamers that control them in both appearance
and behavioral representation. More importantly, avatars can be made to resemble game
players’ appearance with nearly photographic realism (Bailenson & Blasccovich, 2004).

New gaming technologies allow players to create their own avatars by matching various

physical features. Face mapping technologies grant even more realism by providing
gamers with the ability to create avatars from digital pictures of their heads and faces.

Social Cognitive Theory (SCT) explains the process of identiﬁcation and
imitation of aggressive thought and behavior from game characters with which players’
identify (Anderson & Bushman, 2002). Applied to video game play, SCT would suggest
that through observational learning processes, players can directly learn aggressive
responses from the violent characters with which they identify. Moreover, SCT reasons
that individuals are more likely to identify with and imitate behaviors from others they
see as similar to the self. If true, we might expect that avatar similarity should increase
perceived character similarity and identiﬁcation with the gamers’ avatars.

Previous research speculates on attributes of avatars which increase perceived
similarity with video game players. Some studies suggest that garners more strongly
identify with physical characteristics of video game characters such as sex and ethnicity
(Eastin, 2006; Eastin et al., 2007). Other research shows that individuals identify more
strongly with facially similar others. In other words, facial similarity is suggested to
heighten individuals’ identiﬁcation with others (Bailenson, Iyengar, Yee & Collins,
2007).

The present study attempted to extend the literature on technological game
features thought to heighten identiﬁcation by varying the similarity of player and
character facial appearance through the customization of players’ avatars. The proposed
study will offer a new approach to testing the prediction that identiﬁcation with the

characters in violent video games increases aggressive reactions resulting from violent

game play. More speciﬁcally, a more advanced understanding of how character similarity
inﬂuences the identiﬁcation process will be granted.
Identiﬁcation and Aggression

Identiﬁcation during game play is a process that can inﬂuence the way players
experience and interpret the video game medium as well as the characters in the game
(Cohen, 2001). The literature on identiﬁcation is fraught with challenges stemming from
conceptual confusion surrounding use of the construct in social science. However, in
simple terms for use in the present study, identiﬁcation with a ﬁctional video game
character can be thought of as the extent to which a person embodies the character by
adopting some dissimilar aspects of the characters’ identity (Hefner, Klimmt, &
Vorderer, 2007). In this sense, to embody the character can be thought of as the extent to
which the player feels almost as if s/he were actually the character. For example, during a
battle or ﬁght, players who normally do not think of themselves as powerful may feel as
though they are more physically adept or courageous than they would feel outside of
game play. In this sense, video games that enable avatars to resemble their game player’s
identity should allow for a more complete embodiment of game characters. An avatar that
adeptly resembles its game player in physical appearance, for example, may allow the
player to feel as though he or she is the avatar in physical form and is maneuvering the
avatar’s behaviors in the three—dimensional game world. When players look like their
game characters, they can then make active decisions to perform aggressive behavior as
an embodiment of their violent character.

Many scholars suggest that empathy is an important dimension of identiﬁcation

with media characters (Zillmann, 1994; Tamborini, 1996). Zillmann (1994) deﬁnes

empathy as “a response (a) to information about circumstances presumed to cause acute
emotions in another individual and/or (b) to the bodily, facial, paralinguistic, and
linguistic expression of emotional experiences by another individual and/or (c) to another
individual's actions that are presumed to be precipitated by acute emotional experiences,
this response being (d) associated with an appreciable increase in excitation and (e)
construed by respondents as feeling with or feeling for another individual.” In line with
Zillmann’s deﬁnition, empathic reactions to media characters are concordant reﬂexive or
reﬂective emotional reactions in response to the witnessing of characters’ emotional
reactions. Tamborini(1996) suggests that empathy consists of many dimensions such as
perspective-taking, emotional contagion, ﬁctional involvement, and others.
Consequently, the empathetic reactions a player may have with their game character may
depend largely on their capabilities of these dimensions. In other words, a player may
have stronger empathetic reactions with their game character, and ultimately stronger
identiﬁcation, if the player is able to adapt the perspective of another or become highly
involved with a narrative or ﬁctional piece. Players should be expected to empathize with
their game characters when witnessing the emotional reactions of those characters (e. g.,
celebrating after completing a mission, distressing after being injured or killed), and may
display concordant physical and emotional expressions to that of their avatar. For
example, upon witnessing their avatar celebrating after completion of a difﬁcult level in
the game, players may similarly jump up and down in happiness.

Though most discussion is consistent with the notion that identiﬁcation with
violent video game characters can heighten aggressive effects (Anderson, 2004;

Anderson & Bushman, 2001; Anderson & Bushman, 2002; Lindsay & Anderson, 2000),

there remains a considerable need for empirical research examining the inﬂuence of
identiﬁcation with violent video game characters on heightened aggression. Aggressive
responses resulting from play are thought to be a direct consequence of the aggression
displayed in violent video games. In research on violent media, aggression has been
deﬁned as behavior that has the intention of harming and is directed toward a target
known to not want the behavior (Anderson & Bushman, 2002). In violent video games,
players move characters with which they identify through the game world committing
aggressive acts of virtual violence (e. g., striking another character, killing another
character). In most cases players purposefully harm targets in the gaming world to
advance their rank in the game.

SCT explains how video game players can learn aggressive responses from the
violent characters with which they identify. SCT suggests that people identify more
strongly with and are more likely to imitate behavior from characters that are seen as
similar, and if these characters act aggressively, aggressive outcomes are likely to occur
(Anderson & Bushman, 2002). Though research on identiﬁcation with violent video
game characters is sparse, this logic is consistent with research on other media use
demonstrating that identiﬁcation with violent characters can increase aggressive
responses. The sizable body of research on ﬁlm and televised violence shows a
convincing relationship between exposure and increased levels of aggression (Anderson,
2001). Particularly relevant in this regard is research by Berkowitz and others showing
the relevance of similarity cues to aggressive response. Though most of this research
demonstrated that aggression aimed at a real life target was increased by the extent to

which the real life target was similar to the target of violence in media portrayals

(Berkowitz, 1965; Berkowitz & Geen, 1966; 1967), a study by Rosekrans (1967) showed
that the depicted similarity between a viewer and an aggressive model increased the
likelihood of aggressive response, and that this effect remained regardless of the
consequences associated with the observed violent behavior. Scholars of violent media
effects use SCT to reason that this type of aggression can be attributed to identiﬁcation
between the television viewer and the violent television characters (Huesman, Moise-
Titus, Podolski, & Eron, 2003; Bushman & Huesmann, 2006; Anderson & Bushman,
2002).

Much like the research on television violence suggests that identiﬁcation with
violent media characters can lead to aggression in viewers, it can be reasoned that video
game players exhibit similar effects after play as a result of identiﬁcation with violent
video game characters. During game play, players are thought to make similarity
judgments between themselves and the game characters. Upon an initial assumption
about a similarity with a game character, the player is thought to make other inferences
about similarity with the avatar, leading to greater identiﬁcation with that character
(Eastin et al., 2007). Notably, attributes of technology afford new media, such as the
video game medium, the ability to simulate, and therefore create, greater identiﬁcation
with characters.

Avatar Similarity and Identiﬁcation

In the gaming world, an avatar is a digitally represented video game character that
performs a player’s executed behaviors in real time (Bailenson & Blasccovich, 2004).
Today, game designers can create avatars which adeptly resemble the players that control

the avatar in both appearance and behavioral representation. In fact, advanced gaming

technologies allow players to create avatars that match various physical features,
allowing the avatars to resemble the game players’ appearance (Bailenson &
Blasccovich, 2004). The Nintendo Wii allows players to create a “Mii ” avatar by
choosing from similar body sizes, facial structures, hair colors, and other additional
accessories such as sunglasses and hats. With even more realism, face mapping
technologies give gamers the ability to construct avatars from digital pictures of their
heads and faces. In this way, avatars can be used to establish perceived character
similarity with gamers.

Based on the logic of SCT, gamers’ perceived similarity with their characters
should increase the players’ identiﬁcation with those avatars. Under this understanding, a
player will more strongly identify with a video game character that is believed to look
and behave in a manner similar to the player than with a character that is dissimilar in
appearance and behavior (Bandura, 1992). Therefore, video game players who use the
aforementioned technologies to enhance the perceived similarity of their game avatars
should be expected to more strongly identify with and be more strongly inﬂuenced by
those avatars during play.

Character identiﬁcation in games has been shown to be strongly inﬂuenced by
physical characteristics of video game characters such as sex and ethnicity (Eastin, 2006;
Eastin et al., 2007). Eastin (2006) found that females elicited more post game play
aggression when their character was female and not male. These ﬁndings are consistent
with previous SCT research showing that identiﬁcation may be inﬂuenced by shared
demographic characteristics such as sex and ethnicity (Appiah, 2001; Jose & Brewer,

1984; Bandura, 1986; Bandura, Ross & Ross, 1963). Eastin, Appiah & Cicchirillo (2007)

found an interaction between character and participant race on aggressive thoughts. The
authors predicted that black participants would more strongly identify with black versus
white game characters and white participants would not show a signiﬁcant effect on
avatar identiﬁcation. It was also predicted that white participants would have more
aggression after playing with black versus white game characters due to primed racial
stereotypes; the authors predicted that black participants would not show this effect.
However, their data did not demonstrate the predicted patterns of identiﬁcation with the
game characters expected to result from different combinations of participant and avatar
race. Black players did not differ in post game play aggression between black and white
characters. However, white game players displayed greater post game play aggression
after playing with a black compared to a white character.

If physical characteristics such as sex and ethnicity can inﬂuence identiﬁcation
with video game characters, we might expect characteristics of facial similarity to have
similar or even greater inﬂuence on identiﬁcation. Facial similarity should not only
provide cues of sex and ethnic similarity, but should go beyond this to provide other
similarity cues. Though little is known about facial similarity in video games, research by
Bailenson, Iyengar, Yee, and Collins (2007) shows the inﬂuence of facial similarity on
perception and its potential to heighten identiﬁcation with others. In research on political
perception, Bailenson et a1. morphed photographs of political candidates with
photographs of potential voters to produce photographs of various faces. When the
photographs were presented to the potential voters, the results suggested a strong
preference for candidates high in facial similarity. In other words, voters showed a

greater preference for the photographs of candidates who were morphed with their own

faces. The same results were found for both familiar and unfamiliar candidates.
Additionally, after asking the participants to name one person whom the candidate
resembled, only two out of the 36 participants detected their own face from the morphed
photographs.

SCT logic explaining the inﬂuence of perceived similarity and empirical evidence
from research by Bailenson et al. (2007) can be combined to reason that identiﬁcation
with video game characters should be enhanced by face mapping technology that allows
players to construct avatars from digital pictures of their heads and faces. When used in
violent FPS video games, identiﬁcation with violent game characters produced by avatar
similarity should result in heightened aggression. Based on the above logic, the following
hypotheses were tested:

H1: Video game players assigned to conditions in which player and avatar facial

similarity is higher will score higher on the state aggression scale than players in

other conditions where player and avatar facial similarity is lower.

H2: Video game players assigned to conditions in which player and avatar facial

similarity is higher will score higher on scales measuring identiﬁcation with their

avatar than players in other conditions where player and avatar facial similarity is
lower.

H3: The effect of assignment to player/avatar facial similarity conditions on state

aggression scale scores will be mediated by player scores on identiﬁcation with

their avatar.

Method

Overview

10

Participants were randomly assigned to play Tom Clancy ’s Rainbow Six Vegas 2
video game in one of three conditions designed to vary player/avatar facial similarity
(low, medium, and high) with a video game character. After playing individually for 20
minutes, participants completed questionnaires measuring identiﬁcation with their avatar
and state aggression, along with several psychographic and demographic characteristics.
Participants

A sample of thirty undergraduate students ( 13 male and 17 female) in
introductory communication classes from a large Midwestern university was recruited to
participate in this study. Of these students, 27 were Caucasian and 3 were African
American. The average age of the students was 19.13 (SD = 1.43). The students who
chose to participate in this study received class credit for their involvement.

Procedure

Photographs of all participants were gathered prior to the study. In total, two
photographs (one frontal picture of the face and one proﬁle picture) were obtained for
each of four people varying by sex (male/female) and race (black/white). These
photographs were used to create versions of the game avatar (differing by sex and race)
that vary player/avatar similarity according to the experimental conditions. In all
conditions, only pictures showing neutral facial expressions were used to create versions
of the game avatar; there were no pictures showing clear emotions (i.e., smiling to
suggest happiness).

Upon arrival to the laboratory, participants trained with Tom Clancy’s Rainbow
Six Vegas 2 until they were comfortable manipulating the game’s controls; participants

were not given longer than ten minutes to train. The training allowed participants to

11

develop the necessary skills for game play, permitting them to focus on the content and
responsibilities of the game (as suggested by Tamborini et al., 2004). After the training
was complete, participants were randomly assigned to play the video game individually
for 20 minutes under one of the three experimental conditions. After the time had
expired, participants completed questionnaires measuring identiﬁcation with their avatar
(Hefner et al.,2007; Cohen, 2001), state aggression (Farrar & Krcmar, 2006),
psychoticism (Zillmann & Weaver, 1997), empathy (Tamborini, unpublished) and
demographics. Participants were then debriefed and dismissed.
Stimulus Materials

The Tom Clancy ’5 Rainbow Six Vegas 2 game is the ninth video game in the
Rainbow Six video game series. The game is the ﬁrst-person shooter sequel to Tom
Clancy ’3 Rainbow Six: Vegas involving a counter-terrorist unit called Rainbow. Rainbow
is run by its own intelligence service, with other intelligence contacts around the globe.
All players performed the role of the lead soldier, Logan Keller, carrying out violent
missions in an attempt to stop the brutality and terrorism in the city of Las Vegas. For
each mission set in the Las Vegas Strip, only the lead soldier made up the unit, Team
Rainbow. The main Team Rainbow mission involves battling terrorists who are holding
the innocent people of Las Vegas hostage. The game is set in the present tense and the
task force battle various terrorists, depending on the speciﬁc missions set forth for Team
Rainbow.

The game allows players to customize their avatars as seen from a ﬁrst-person
shooter perspective. Players can alter their clothing, weapons, and facial structures. The

game’s Xbox Live face mapping technology also allows players to take digital pictures of

12

their faces and transpose those images on their avatars to create game character
lookalikes. Visualizations of the avatar’s face were able to be seen when ducking, hiding,
and spying on enemies during play, allowing the shooting and violent acts to be viewed
from a ﬁrst-person perspective. Due to the difﬁculty of the game, all participants played
the videogame in God mode and with unlimited ammo, making the players invulnerable
to damage and death.

In the low player/avatar facial similarity condition, participants played the game
with an avatar depicting an unknown person’s head and face of a different sex and race.
Therefore, when a white male participant who is randomly assigned to this condition
arrived at the laboratory, he was assigned to use a previously created black female avatar.
Avatars created in this condition used characters already provided with the game’s
software.

In the medium player/avatar facial similarity condition, participants played the
game with an avatar depicting proﬁle pictures (i.e. the head of the avatar) of the
participant and a morphed face of the participant and an unknown person of the same sex
and race. Morphed pictures were created using Magic Morph software. Therefore, when a
white male participant who was randomly assigned to this condition arrived at the
laboratory, he was assigned to use a previously created white male avatar depicting a

morphed head shot that combined his face with that of an unknown white male.

13

FIGURE 1
MORPH IMAGE IN MODERATE SIMILARITYCONDITION

      

 

*The pictures on the left and right seen above are of two participants in this study.
The picture between them represents the morphed head-shot of the two.

In the high player/avatar facial similarity condition, participants played the game
with an avatar depicting their own face. Thus, upon arrival to the laboratory, participants
assigned to this condition were given a previously created avatar featuring a head-shot
and proﬁle pictures of their face and head on the character.

Measures

Identiﬁcation. Identiﬁcation with the game character was assessed using an eight-
item scale adapted from Hefner, Klimmt, and Vorderer (2007). Responses were measured
on a seven-point Likert-type scale ranging from “Strongly disagree” to “Strongly agree,”
with items such as “I forgot myself during the game,” “I felt like I was the game
character,” and “The goals of the character became my own goals.” Cohen’s ten item
scale (2001) was revised to apply to video games and also used to assess participant
identiﬁcation. Responses were measured on a seven-item Likert scale ranging from
“Strongly disagree” to “Strongly agree,” with items such as “While playing the game, I
forgot myself and was fully absorbed,” and “While playing the game, I could feel the
emotions of Logan Keller.”

State Aggression. Participants’ state aggression was measured using a modiﬁed

version of the Buss and Perry (1992) trait scale developed by Farrar and Krcmar (2006).

14

The original scale by Buss and Perry measured four dimensions of state aggression:
verbal aggression, physical aggression, anger, and hostility. Farrar and Krcmar’s
modiﬁed version measured the ﬁrst two dimensions: state verbal aggression, and state
physical aggression. In development of this scale, the authors ﬁrst created two scenarios
thought to prime high and low aggression. For the present study, only the high aggression
scenario was used, as it produced the strongest effect in previous research. The measure
asked participants to imagine that they have completed the survey and are leaving the
study. The scenario continued with . .someone bumps into you, spilling your drink and
the contents of your bag. They then begin to curse and step toward you aggressively.”
After given this scenario, participants completed the modiﬁed state aggression scale. The
scale contains 11 items such as “If this person hit me, I’d hit back,” “Given the
provocation, I may hit this person,” and “When this person annoyed me, I would tell
them what I thought of them.”

Control Measures. Several additional scales were included to provide control
measures. These scales included measures of psychoticism, empathy, demographics (sex,
ethnicity, age), and face recognition. Psychotocism was measured using Zillmann and
Weaver’s (1997) modiﬁed version of Eysenck psychoticism scale (Eysenck et al., 1985).
Participants answered “Yes” or “No” to 12 items such as “Would you like other people to
be afraid of you,” and reversed items such as “Is it better to follow society’s rules than go
your own way?” Empathy was measured using Tamborini’s (unpublished) scale that
includes four items measuring emotional contagion (e. g., “I cannot continue to feel okay
if others around me are feeling depressed”), ﬁve items measuring perspective taking (e. g.,

“Before criticizing someone, I try to imagine how I would feel in their place”), seven

15

items measuring empathetic concern (e. g., “I am the type of person who is concerned
when other people are unhappy”), ﬁve items measuring communicative responsiveness
(e. g., “I am the type of person who can say the right thing at the right time”), and ﬁve
items measuring ﬁctional involvement (e.g., “I really get involved with the feelings and
characters in a novel or ﬁlm”). Sex, ethnicity and age were standard single items
measures used to assess these demographic characteristics. Face recognition was a single
item used to determine if the respondent recognized their face on the avatar in the game
(coded not recognize = 2 and recognize = 1). All scales appear in Appendix A.
Results

Analysis began with an inspection of the measurement model to test the quality of
the composite scales used in the study, and a report of related descriptive statistics.
Following this, ANCOVA was used to test the ﬁrst two hypotheses predicting that (H1)
Higher avatar facial similarity conditions will result in higher scores on state aggression,
and (H2) Higher avatar facial similarity conditions will result in higher scores on
identiﬁcation with the avatar. The failure to support H1 and H2 made moot testing of the
third hypothesis predicting that (H3) the effect of avatar facial similarity conditions on
state aggression will be mediated by identiﬁcation.
Inspection of the measurement model began by conducting principle component factor
analyses on all scale items for measures of identiﬁcation, aggression and empathy. The
results of factor analyses were used to determine the items included in composite scales
for these measures by applying the following decision rule. In order to be included, an
item must load above .6 on one of the factors and below .4 on all other factors. The factor

loadings for all scales can be found in Appendix B.

16

Measurement Model

Factor analysis on items used to measure empathy showed two factors could be
extracted. Items three, nine and 22 met the decision rule for inclusion in the ﬁrst factor
which was labeled empathic caring. The second factor included items ﬁve, 10 and 15 and
was labeled ﬁctional involvement. Because of the nature of the dichotomous
psychoticism scale, no items needed to be excluded from analysis.

Factor analysis on the state aggression measure revealed that two factors could be
extracted. Based on the four components of the state aggression scale, measures of state
verbal and physical aggression were computed. Items one and two met the criteria for
inclusion in the ﬁrst factor and were averaged to comprise a variable labeled state verbal
aggression. Items three, ﬁve, six, seven, nine and 10 met the criteria for inclusion‘in the
second factor. However, item three was excluded based on poor face validity. The other
ﬁve items were averaged to comprise a variable labeled state physical aggression.

Factor analysis on the combined items from both identiﬁcation measures determined that
two factors could be extracted. The ﬁrst factor included items two through ﬁve, and
seven and eight of the Hefner measure along with items one through three, and six and
seven of the Cohen measure. These items were averaged in a composite scale labeled
identiﬁcation. Only one item met the criteria for inclusion in the second factor. Because
of concern regarding the reliability of single item measures, it was excluded from
analysis.

Descriptive Statistics

Table 1 shows descriptive statistics for key variables in this study. Measures of

state physical aggression (or = .89) and state verbal aggression (a = .70) were fairly

17

reliable. From this table we notice the sample was generally low on state physical
aggression (M=3.37, SD=1.41) and relatively high on state verbal aggression (M=5.50,
SD=1.22). The sample tended to have a wide range from a minimum of 2 to7 on state
verbal aggression and 1 to 6 on state physical aggression. The identiﬁcation measure was
very reliable (a = .96). Participant character identiﬁcation was relatively moderate
(M=4.31, SD=1.45) with a range of 1.55 to 6.64. Measures of empathetic caring (a = .69)
and ﬁctional involvement ((1 = .82) were moderate. The sample tended to be relatively
high on measures of empathetic caring (M=4.06, SD=0.68) and ﬁctional involvement
(M=5. 19, SD=1.19) with a minimum of 3.00 on empathetic caring and 2.67 on ﬁctional
involvement. The sample showed little psychotic tendency (M=0. 12, SD=0.09) with a
maximum of 0.33. Lastly, few people recognized their face in the game. Only 10% of the
sample or 3 of the 10 participants in the high participant/avatar similarity condition

recognized their face on their avatar in the game.

 

 

TABLE 1
Descriptive Statistics for Key Variables in Study
Mean SD Minimum Maximum
State verbal aggression 5.50 1.22 2.00 7.00
State physical aggression 3.37 1.41 1.00 6.00
Character identiﬁcation 4.31 1.45 1.55 6.64
Empathic caring 4.06 0.68 3.00 5.67
Fictional involvement 5.19 1.19 2.67 7.00
Psychoticism 0.12 0.09 0.00 0.33
Face recognition* 0.10 0.31 0.00 1.00

 

*Coded as a dichotomous variable with 0 = did not recognize and l = did recognize.

18

Table 2 shows the associations among key variables in this study. From this table we
notice a positive correlation of character identiﬁcation with state verbal but not physical
aggression. We also note that ﬁctional involvement is not correlated with the similarity
conditions. Notably, ﬁctional involvement is not correlated with empathetic caring.
Finally, we note that, though not signiﬁcant, ﬁctional involvement is negatively related to

measures of aggression while empathetic caring is not negatively related to measures of

 

 

aggression.
Table 2
Zero-order Correlations among Key Variables in Study
1 2 3 4 S 6
l. Similarity condition [.00
2. State verbal aggression .05 1.00
3. State physical aggression .21 .38* 1.00
4. Character identiﬁcation -.08 .47** .32 1.00
5. Empathic caring .16 .27 .15 .27 1.00
6. Fictional involvement .00 -.23 -.35 .l 1 .09 1.00

 

Note. Similarity condition was coded such that 1 = low, 2 = moderate, and 3 = high.

*indicates p < .05, two-tailed.

Hypothesis Testing

The ﬁrst hypothesis predicted that participants assigned to conditions in which
similarity was high would score higher on measures of state aggression. ANCOVA was
conducted on both state aggression outcome variables controlling for identiﬁcation,
empathetic concern, ﬁctional involvement, sex, ethnicity, age, psychoticism and face
recognition. The results of ANCOVA did not reveal a signiﬁcant main effect for

participant/character similarity on measures of state physical aggression, F(2, 27) = .73,

19

p: .50, n2: .07. Participants in the low similarity condition scored lower on state
physical aggression (M: 2.84, SD: 1.72) in comparison to participants in the moderate
condition (M: 3.70, SD: 1.32) and the high condition (M: 3.56, SD: 1.41). However,
sex was found to have a signiﬁcant effect on state physical aggression, F(1, 28) = 4.90,
p<.05 , n2: .21. Females scored signiﬁcantly lower on measures of state physical
aggression (M: 2.75, SD: .30) in comparison males (M: 4.17, SD: .34). The results of
ANCOVA also failed to show a signiﬁcant effect on the state verbal aggression measure,
F(2, 27) = .33, p: .72, n2: .03. Participants in the low similarity condition (M: 5.45, SD:
1.07) and moderate similarity condition (M: 5.45, SD: 1.67) scored lower on state
aggression in comparison to participants in the high condition (M: 5.60, SD: .99). Thus,
it can be concluded that the data was not consistent with the hypothesis. Notably,
identiﬁcation was found to have a signiﬁcant effect on verbal aggression, F(2, 27) = 5.30,
p<.05, n2: .22. This is consistent with the results of descriptive analyses showing that
identiﬁcation had a signiﬁcant positive correlation with verbal aggression.

The second hypothesis predicted that participants assigned to conditions in which
similarity was high would score higher on measures of identiﬁcation. An ANCOVA was
conducted on the outcome variable controlling for empathetic concern, ﬁctional
involvement, sex, ethnicity, age, psychoticism and face recognition. Analysis did not
reveal a signiﬁcant main effect for participant/character similarity on identiﬁcation F(2,
27) = .30, p: .74, n2 = .03. Moreover, in contrast to predictions, participants in the low
similarity condition scored higher (M: 4.55, SD: 1.53) on the measure of identiﬁcation
than participants in the moderate condition (M: 4.11, SD: 1.72) and the high condition

(M: 4.27, SD: 1.17). Thus, it can be concluded that the data was not consistent with the

20

hypothesis. Based on the ﬁndings mentioned above, an analysis of mediation was not
conducted on the third hypothesis which suggested a mediation of aggression scores with
player/avatar similarity by identiﬁcation scores.

Discussion

The hypothesis suggested that identiﬁcation with game characters would mediate
the effect predicted for similarity in physical appearance between game players and their
game characters on heightened state aggressive responses. The results failed to support
the hypothesized model. The ﬁrst two hypotheses predicted that similarity between game
players and game characters would heighten identiﬁcation and that those with stronger
identiﬁcation would show stronger state aggression. These two hypotheses failed to be
supported, and COnsequently, no hypothesized mediation was shown.

Although the proposed model was not supported in this study, there are some
results worth mentioning. Speciﬁcally, identiﬁcation had a notable effect on the outcome
measure in this study. In this study, identiﬁcation had a signiﬁcant positive correlation
with state verbal aggression, and approached signiﬁcance with state physical aggression.
Thus, although avatar similarity did not increase identiﬁcation as expected, identiﬁcation
did predict increased aggression. In view of that, we are left to determine what aspects of
game experience predict identiﬁcation. In part, the answer to this question may depend
on how identiﬁcation is deﬁned.

Also of interest in these ﬁndings is the fact that state verbal and physical
aggression were unrelated to measures of empathetic caring and ﬁctional involvement.
Though not quite signiﬁcant, empathetic caring was positively related to measures of

aggression and ﬁctional involvement was negatively related to measures of aggression. If

21

identiﬁcation enhances aggression, this may suggest that identiﬁcation has an association
with state aggression that is distinct from some aspects of empathy.
Limitations and Future Research

The limitations to this study begin with the inherent qualities of the video game
used during participants’ play, Tom Clancy ’5 Rainbow Six Vegas: 2. While the capability
to use morphing technology to vary game character similarity allowed for an interesting
test of the predicted model of similarity’s inﬂuence on identiﬁcation with game
characters and subsequent aggressive outcomes, there were other procedural problems
that occurred as a result of the use of the game that may have affected the results in this
study. First, Tom Clancy ’s Rainbow Six Vegas: 2 is a very difﬁcult game for
inexperienced users to play. As such, in an attempt to overcome game difﬁculty barriers
for study participants, God mode was enabled, allowing for less difﬁcult play. An
unanticipated consequence of using God mode was that participants modified their game
play to adjust to the low level of difﬁculty (e. g., playing less defensively). When playing
the game on God mode, players had little need to engage in defensive positions such as
hiding, ducking and taking cover. These positions would have allowed the players to see
their own face on the screen. Without taking these defensive positions, players rarely saw
their own images. Ultimately, this limited the study’s ability to test the overall
hypothesized model.

While the gaming environment in which the participants played was the biggest
limitation to this study, there are other factors that might have contributed to the
unreliability of this study’s results. First, the number of participants used in this study

was particularly low. Furthermore, although attempts were made to balance gender,

22

failure of participants to show up for scheduled appointments resulted in a situation
where gender was not evenly distributed across similarity conditions. The low similarity
condition included 7 females and 3 males, the moderate similarity condition included 6
females and 4 males and the high similarity condition included 4 females and 6 males.
Although an attempt was made to control statistically for this imbalance, the likelihood
that gender inﬂuenced the relationships of similarity condition with identiﬁcation and
aggression cannot be overlooked.

Conclusion

If the results of this study are a valid test of avatar and game player similarity and
its inﬂuence on the manner in which violent game play inﬂuences aggressive responses
in game players, the ﬁndingsshould be interpreted to suggest that avatar similarity does
not increase identiﬁcation and subsequent aggressive outcomes. In this study,
identiﬁcation did not predict aggressive outcomes, and similarity did not predict
identiﬁcation or aggression. At the same time, procedural issues leave open the
possibility that this study did not provide a valid test of avatar similarity’s inﬂuence on
aggressive responses.

Future research should more closely consider features of the game environment
which may hinder the ability to see the inﬂuence of avatar similarity on player
identiﬁcation with their game character. The game software and technology used to
create character similarity should provide players with greater opportunity to see their
similarity to the game character in order to better examine the inﬂuence of similarity on
identiﬁcation with their game characters. Future research should also consider additional

game software and technology attributes which may lead to or contribute to identiﬁcation

23

with video game characters.

24

Appendix A
State Aggression Measure

Image that after you are done completing the survey, you leave the building and someone
bumps into you, spilling your drink and the contents of your bag. They then begin to
curse and step toward you aggressively. Please respond to the following questions by
writing the number next to each question that most closely co I neither agree nor disagree
responds with your answer.

Disagree Disagree Disagree Undecided Agree Agree Agree
Strongly Moderately Slightly Slightly Moderately Strongly
1 2 3 4 5 6 7

1. _ I would tell this person openly that I disagreed with him.

2. _ I would ﬁnd myself disagreeing with this person.

3. __ When this person annoyed me, I would tell them what I thought of them.

4. _ This person would say I’m somewhat argumentative.

5. __ I could not control my urge to strike this person.

6. _ Given the provocation, I may hit this person.

7. __ If this person hit me, I’d hit back.

8. _ I’d get into a ﬁght with this person a little more than the average person.

9. __ If I had to resort to violence to protect my rights, I would.

10. _ This person would have pushed me so far that we came to blows.

11. _ I can think of no good reason for ever hitting this person.

25

Empathy Measure

Please respond to the following questions by writing the number next to each question
that most closely corresponds with your answer.

Disagree Disagree Disagree Undecided Agree Agree Agree
Strongly Moderately Slightly Slightly Moderately Strongly
1 2 3 4 5 6 7
1. I cannot continue to feel okay if others around me are feeling depressed.
2 Before criticizing someone, I try to imagine how I would feel in their place.
3. I am the type of person who is concerned when other people are unhappy.
4. I am the type of person who can say the right thing at the right time.
5. I really get involved with the feelings and characters in a novel or ﬁlm.
6. I don’t become upset just because a friend is acting upset.
7. I sometimes try to understand my friends better by imaging things from their
perspective.
8. When I see someone being taken advantage of, I feel kind of protective toward
them.
9. Even though I often try to console someone who is feeling bad, I never seem to

be able to say the right thing.

10. __ When I am reading an interesting story or novel, I imagine how I would feel if
the events were to happen to me.

11. _ I become nervous if others around me seem nervous.

12. _ I sometimes ﬁnd it difﬁcult to see things from another’s perspective.

13. _ I often have tender, concerned feelings for people less fortunate than myself.
14. _ I usually respond appropriately to the feelings of others.

15. _ After acting in a play or seeing a play or a movie, I have felt partly as though I

were one of the characters.
16. The people around me have a great inﬂuence on my moods.

17. _ I try to look at everyone’s side of a disagreement before I make a decision.

26

18. _ I would describe myself as a pretty soft-hearted person.

19. __ Others think of me as an empathetic person.

20. _ When I watch a good movie, I can easily put myself I the place of the lead
character.

21. __ When I am upset, I usually try to put myself in his or her shoes for awhile.

22. __ I sometimes don’t feel very sorry for people when they are having problems.

23. _ My friends come to me with their problems because I am a good listener.

24. __ I become very involved when I watch a movie.

25. __ Other people’s misfortunes do not usually disturb me a great deal.

26. _ I am often touched by the things that I see happen.

 

Scale Item Scoring (‘R’ indicates reverse scored-items): Emotional Contagion: 1, 6R, ll,
16; Perspective Taking: 2, 7, 12R, 17, 21; Empathetic Concern: 3, 8, 13, 18, 22R, 25R,
26; Communicative Response: 4, 9R, 14, 19, 23; Fictional Involvement: 5, 10, 15, 20, 24.

27

Identiﬁcation Measure (1)

Please respond to the following questions by writing the number next to each question
that most closely corresponds with your answer.

Disagree Disagree Disagree Undecided Agree Agree Agree

Strongly Moderately Slightly Slightly Moderately Strongly
1 2 3 4 5 6 7

1. While I was playing the game, I forgot everything around me.

2. I had the feeling I was literally in the character’s skin.

3. I forgot myself because I was focusing so much on the game character’s actions.

4. I had the feeling I was the game character more so than myself.

5. The game character’s goals became my goals.

6. While I was playing the game, the game world was more real to me than my

“actual reality.” ”

7. I felt as if I was really participating in the shown/depicted happenings.

8. I almost had the feeling of actually being the character.

28

Identiﬁcation Measure (2)

Please respond to the following questions by writing the number next to each question
that most closely corresponds with your answer.

Disagree Disagree Disagree Undecided Agree Agree Agree
Strongly Moderately Slightly Slightly Moderately Strongly
1 2 3 4 5 6 7

1. While I was playing the game, I felt as if I was part of the action.

2. While playing the game, I forgot myself and was fully absorbed.

3. I was able to understand the events in the game in a manner similar to that in

which Logan Keller understood them.

4. I think I have a good understanding of Logan Keller.

5. I tend to understand the reasons why Logan Keller does what s/he does.

6. During playing the game, I felt I could really get inside Logan Keller’s head.

7. At key moments in the game, I felt I knew exactly what Logan Keller was going
through.

8 While playing the game, I wanted Logan Keller to succeed in achieving his/her

goals.

9. While playing the game, I could feel the emotions of Logan Keller.

10. When Logan Keller succeeded I felt joy, but when Logan Keller failed, I was
sad.

29

Psychoticism Measure

Please respond to the following questions by circling the word next to each question that
most closely corresponds with your answer.

1) Do you take much notice of what people think? .......................................... Yes No
2) Would being in debt worry you? ................................................................. Yes No
3) Would you take drugs which may have strange or dangerous effects? ....... Yes No
4) Do you prefer to go your own way rather than act by the rules? ................. Yes No
5) Do good manners and cleanliness matter much to you? .............................. Yes No
6) Do you think marriage is old-fashioned and should be done away with? ...Yes No
7) Do you enjoy co-operating with others? ...................................................... Yes No
8) Does it worry you if you know there are mistakes in your work? ............... Yes No
9) Do you think people spend too much time safeguarding their future with

savings and insurances? ............................................................................... Yes No
10) Do you try not to be rude to people? ........................................................... Yes No
11) Would you like other people to be afraid of you? ....................................... Yes No
12) Is it better to follow society’s rules than go your own way? ....................... Yes No

 

Scale Item Scoring: YES: 3, 4, 6, 9, 11; NO: 1, 2, 5, 7, 8, 10, 12

30

Appendix B

 

 

 

 

 

TABLE 3
STATE AGGRESSION COMPONENT MATRIX
Component
2 3
State Aggression .096 .754 .356 .247
State Aggression .348 .635 .271 —.245
State Aggression .666 .273 -.392 .392
State Aggression .563 .420 -.491 -.10
State Aggression .836 -.126 .195 -.14
State Aggression .859 .042 .228 .07
State Aggression .791 -.231 —.025 .341
State Aggression .634 .057 -.109 -.68
State Aggression .725 -.236 .265 .332
State Aggression .735 -.358 .297 -.21
State Aggression -.542 .114 .500 .00

 

 

 

 

 

31

TABLE 4
IDENTICATION COMPONENT MATRIX

 

 

 

 

 

 

 

 

 

2

ldentitication(1) 1 .333 —.268 -.208 .797
ldentitication(l) 2 .782 -.166 -.234 -.253
ldentitication(1) 3 .883 -.131 -.011 .13
Identification( 1) 4 .821 -.076 -.322 -.136
ldentitication(1) 5 .880 .141 .060 .192
ldentitication(1) 6 .824 .184 -.425 -.09
ldentitication(1) 7 .881 -.031 -.373 -.08
ldentitication(1) 8 .814 -.O48 -.362 -.26
ldentitication(2) 1 .735 -.094 -.229 .25
ldentitication(2) 2 .880 -.022 -.008 .25
ldentitication(2) 3 .734 -.317 .234 -.067
ldentitication(2) 4 .698 -.375 .486 .02
ldentitication(2) 5 .703 -.278 .579 -.01
ldentitication(2) 6 .820 -.036 .391 -.201
ldentitication(2) 7 .813 -.201 .128 -.13
ldentitication(2) 8 .452 .743 .200 .249 ‘
ldentitication(2) 9 .751 .542 .062 -.12
ldentitication(2) 10 .586 .709 .245 .02

 

EMPATHY COMPONENT MATRIX

TABLE 5

 

 

 

 

 

 

 

 

 

 

 

 

 

Component
3 4
Empathy .059 .684 .470 -.034 .168 .044
Empathy .507 .235 -.413 -.337 .081 .196
Empathy .155 .742 -.140 -.476 .114 .046
Empathy .018 .458 .285 -.256 -.595 .053
Empathy -.050 .031 .740 -.419 .304 .087
Empathy -.174 -.338 -.292 -.247 .545 -.302
Empathy .927 -.128 -.093 -.101 -.004 -.038
Empathy .884 -.095 -.020 .142 -.105 .01 1
Empathy .492 -.746 .033 .278 .103 .075
Empathy .850 -.174 .282 -.099 -.1 1 1 -.102
Empathy .801 .015 -.035 .012 .026 -.051
Empathy .620 -.366 -.065 .327 .253
Empathy .91 1 -.003 -.008 -.226 -.126
Empathy .923 -.015 -.081 .002 -.030
Empathy .724 -.119 .445 -.269 -.125
Empathy .878 -.190 .050 -.201 .056
Empathy .846 .058 -.098 .130 -.1 13
Empathy .632 .231 i .153 .346 .390
Empathy .216 .610 .102 .613 -.058
Empathy -.039 -.009 .784 .325 .019
Empathy .163 .570 -.382 .180 .452
Empathy -.227 -.703 .305 -.328 -.055
Empathy .155 .575 -.101 .530 -.257
Empathy -.087 .204 .651 .194 .296
Empathy -.183 -.519 .422 .395 -.033
Empathy .216 .530 .408 -.358 .169

 

 

 

33

BIBLIOGRAPHY

Aguilera, M. D., & Mendiz, A. (2003). Video games and education: Education in the face
of a “parallel school.” ACM Computers in Entertainment, 1, 1-14.

Anderson, C. A. (2004). An update on the effects of violent video games. Journal of
Adolescence, 2 7, 113-122.

Anderson, C. A. (2001). TV violence and viewer aggression: A cumulation of study
results 1956-1976. Public Opinion Quarterly, 4, 314-331.

Anderson, C.A., Benjamin, A.J., Bartholow, BB. (1998). Does the gun pull the trigger?
Automatic priming effects of weapon pictures and weapon names. Psychological
Science, 9, 308-314.

Anderson, C. A., & Bushman, B. J. (2001). Effects of violent video games on aggressive
behavior, aggressive cognition, aggressive affect, physiological arousal, and
prosocial behavior: A meta-analytic review of scientiﬁc literature. Psychological
Science, 12, 353-359.

Anderson, C. A., & Bushman, B. J. (2002). Human Aggression. Annual Review of
Psychology, 53, 27-51.

Anderson, C. A., & Bushman, B. J. (2002). The effects of media violence on society.
Science, 295, 2377-2379.

Anderson, C. A., & Dill, K. E. (2000). Video games and aggressive thoughts, feelings,
and behaviors in the laboratory and in life. Journal of Personality and Social
Psychology, 78, 772-790.

Appiah, O. (2001). Ethnic identiﬁcation on adolescents’ evaluations of advertisements.
Journal of Advertising Research, 41, 7-22.

Bailenson, J .N ., & Blasccovich, J. (2004). Avatars. In W.S. Bainbridge (Ed.),
Encyclopedia of human-computer interaction (pp. 64-68). Great Barrington, MA:
Berkshire Publishing Group.

Bailenson, J .N., Iyengar, S., Yee, N., & Collins, N. (2007, in press). Facial similarity
between voters and candidates causes inﬂuence. Public Opinion Quarterly.

Bandura, A. (1986). Social foundation of thought and action: A social cognitive theory.
Englewood Cliffs, NJ: Prentice-Hall.

Bandura, A. (1992). Exercise of personal agency through the self-efﬁcacy mechanism. In

R. Schwarzer (Ed.), Self-eﬁicacy: Thought control of action (pp. 3-38).
Washington, DC: Hemisphere.

34

Bandura, A., Ross, D., Ross, SA (1963). Vicarious reinforcement and imitative learning.
Journal of Abnormal Psychology, 67, 601-607.

Barlett, C.P., Harris, R.J., & Baldassaro, R. (2007). Longer you play, the more hostile
you feel: Examination of ﬁrst person shooter video games and aggression during
video game play. Aggressive Behavior, 33, 486-497.

Baron, R. M., & Kenny, DA. (1986). The moderator-mediator variable distinction in
social psychology research: Conceptual, strategic and statistical considerations.
Journal of Personality and Social Psychology, 51,1173-1 182.

Berkowitz, L. ( 1965). Some aspects of observed aggression. Journal of Personality and
Social Psychology, 2, 359-369.

Berkowitz, L., & Geen, R. G. (1966). Film violence and the cue properties of available
targets. Journal of Personality and Social Psychology, 3, 525-530.

Berkowitz, L., & Geen, R. G. (1967). Stimulus qualities of the target of aggression: A
further study. Journal of Personality and Social Psychology, 5, 364-68.

Berkowitz, L., & Powers, PC. (1979). Effects of timing and justiﬁcation of witnessed
aggression on the observer’s punitiveness. Journal of Research in Personality, 13,
71-80.

Buss, A.H., & Perry, M. (1992). The aggression questionnaire. Journal of Personality
and Social Psychology, 63, 452-459.

Bushman, B. J ., & Huesmann, L. R., (2006). Short-term and long-tem effects of violent
media on aggression in children and adults. Archives of Pediatrics & Adolescent
Medicine,160, 348-352.

Cohen, J. (2001). Deﬁning identiﬁcation: A theoretical look at the identiﬁcation of
audiences with media characters. Mass Communication & Society, 4, 245-264.

Eastin, M., Appiah, O., & Cicchirillo, V. (2007). I am the man in the mirror:
Identiﬁcation and the inﬂuence of cultural stereotyping on post game play
hostility. Paper presented at the annual meeting of the International
Communication Association.

Eastin, M. S. (2006). Video game violence and the female game player: Self- and
opponent gender effects on presence and aggressive thoughts. Human
Communication Research, 32, 351-372.

Eastin, M. S., & Grifﬁths, R. P. (2006). Beyond shooter games: Examining presence and
hostile outcomes among male game players. Communication Research, 6.

35

Eysenck, S.B.G., Eysenck, H.J., & Barrett, P. (1985). A revised version of the
psychoticism scale. Personality and Individual Diﬁ‘erences, 6, 21-29.

Farrar, K., & Krcmar, M. (2006). Measuring state and trait aggression: A short,
cautionary tale. Media Psychology, 8, 127-138.

Geen, R.G., & Stonner, D. (1973). Context effects in observed violence. Journal of
Personality and Social Psychology, 25, 145-150.

Geen, R.G., & Stonner, D. (1974). The meaning of observed violence: Effects on arousal
And aggressive behavior. Journal of Research in Personality, 8, 55-63.

Gentile, D. A., Lynch, P. J ., Linder, J. R., & Walsh, DA. (2004). The effects of violent
Video game habits on adolescent hostility, aggressive behaviors and school
performance. Journal of Adolescence, 27, 5-22.

Hefner, D., Klimmt, C., & Vorderer, P. (2007). Identiﬁcation with the player character as
determinant of video game enjoyment. In L. Ma, R. Nakatsu, and M. Rauterberg
(Eds), Lecture Notes in Computer Science. (39-48). IFIP International Federation
for Information Processing.

Huesmann, L.R., Moise-Titus, J ., Podolski, C.L., & Eron, L. D. (2003). Longitudinal
Relations between children’s exposure to TV violence and their aggressive and
violent behavior in young adulthood: 1977-1992. Journal of Developmental

Psychology, 39, 201-221.

Jose, P.E., & Brewer, W.F. (1984). Develpoment of story liking: Character identiﬁcation,
suspense, and outcome resolution. Developmental Psychology, 20, 911-924.

Lachlan, K.A., Smith, S.L., Tamborini, R. (2005). Models for aggressive behavior: The
attributes of violent characters in popular video games. Communication Studies,
56, 313-329.

Lindsay, J .J ., & Anderson, CA. (2000). From antecedent conditions to violent actions: A
general affective aggression model. Personality & Social Psychology Bulletin, 26,
533-547.

Nowak, K.L., & Rauh, C. (2006). The inﬂuence of the avatar on online perceptions of
anthropomorphism, androgyny, credibility, homophily, and attraction. Journal of
Computer-Mediated Communication, 11, 153-178.

Rosekrans, M. A. ( 1967). Imitation in children as a function of perceived similarity to a
social model and vicarious reinforcement. Journal of Personality and Social

Psychology, 7, 307-315.

Schneider, E.F., Lang, A., Shin, M., & Bradley, SD. (2004). Death with a story: How

36

Story impacts emotional, motivational, and physiological responses to ﬁrst-person
shooter video games. Human Communication Research, 30, 361-375.

Sherry, J. (2001). The effects of violent video games on aggression: A meta analysis.
Human Communication Research, 2 7, 409-431.

Tamborini, R. (1996). A model of empathy and emotion reactions to horror. In J .B.
Weaver & R. Tamborini (Eds), Horror ﬁlms: Current research on audience
preferences and reactions (pp. 103-123). Hillsdale, NJ: Lawrence Erlbaum.

Tamborini, R., Eastin, M. S., Skalski, P., Lachlan, K., Fediuk, T. A., & Brady, R. (2004).
Violent virtual video games and hostile thoughts. Journal of Broadcasting &
Electronic Media, 48, 335-357.

Zillmann, D. (1994). Mechanisms of emotional involvement with drama. Poetics, 23, 33-
51.

Zillmann, D., & Weaver, 1.3. (1997). Psychoticism in the effect of prolonged exposure to
gratuitous media violence on the acceptance of violence as a preferred means of
conﬂict resolution. Journal of Personality and Individual Differences, 22, 613-
627.

37