UNDERSTANDING MEDIA VIOLENCE FROM

A ROLE-PLAY PERSPECTIVE

Effects of Various Types of Violent Video Games on Players’ Cognitive Aggression

Younbo Jung

Nanyang Technological University, 31 Nanyang Link, Singapore, Republic of Singapore

Keywords: Video game, Violence, Aggression, Role enactment, Enjoyment, Identification.

Abstract: In this study I investigated the effects of depicted character roles and wishful identification with the main

character on aggression and game enjoyment among players in violent video games. The results (n = 36)

showed that character roles (e.g., the police, gangster, and athlete) did not have any significant effect on

post-game aggression. However, there was a significant association among depicted character roles, wishful

identification, and game enjoyment. Implications in terms of the level of aggression, identification with the

game character, and game enjoyment are discussed.

1 INTRODUCTION

Violence pervades our daily lives, regardless of our

awareness. According to the hypothesis of innate

nature of human violence from evolutionary

psychology, it is not so surprising to see human

beings’ enjoyment of violent media because the

sensitivity to violence has evolved from the Stone

Age. That is, the innate modules of violence reside

deeply inside all human minds. Virtual experience of

violence may wake up the innate instinct of violence

(Pinker, 2002). Because the human mind still

regards violence as a rewarding activity for

reproduction and survival, human responses to this

wake-up call for violence (i.e., exposure to violent

media) could be perceived as something enjoyable,

exciting, and satisfactory. As such, the relationship

between exposure to media violence and its various

negative effects has been one of the most popular

research topics in media-effect studies (Sparks and

Sparks, 2002). Research on violent video games has

also demonstrated that exposure to violent video

games tended to increase aggressive affect,

aggressive cognitions, and aggressive behaviors

among users (Bushman and Anderson, 2002;

Carnagey and Anserson, 2005).

However, it is often ignored that there are

different types of violence in video games,

depending on the depicted role of main characters of

the games. For example, violent actions by the

police in video games may influence players

differently, compared to similar actions by the

criminal. It is possible that violent actions by a nice

character or our “guy” are perceived to be more

acceptable to players when players enact the role of

a positive character during their play. According to

Pinker (2002), violence has been historically

justified within the stereotypical dichotomous view

of a good guy versus a bad guy. Skillful violence

against animals to get food and against nemeses to

protect one’s own tribe was regarded as a sign of

being a “real man” (p. 309), and thus was adored by

people. Fellow humans who are outside the circle of

“we” are often treated like animals that do not

deserve our sympathy. Violence by the protagonist

is accepted well without any disturbance because

people tend to identify with the protagonist—the

“our guy, thus a good guy” thinking (Oatley, 1994).

To summarize, playing violent video games could

influence players’ level of aggression and judgment

of violence differently, depending on how the role of

the main character is depicted in video games and

how players identify with the character.

Therefore, the purposes of the current paper are

(1) to investigate how different types of violence in

video games could influence players’ level

aggression; and (2) to discuss a way to reduce

possible negative effects of playing violent video

games by altering the context features of violent

games. A total 36 subjects participated in the

experiment with different types of violence as a

between-subject factor (police violence vs. gangster

261

Jung Y..

UNDERSTANDING MEDIA VIOLENCE FROM A ROLE-PLAY PERSPECTIVE - Effects of Various Types of Violent Video Games on Players’ Cognitive

Aggression.

DOI: 10.5220/0003497802610266

In Proceedings of the 13th International Conference on Enterprise Information Systems (ICEIS-2011), pages 261-266

ISBN: 978-989-8425-56-0

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

violence vs. sports violence). Findings and

implications in terms of the level of aggression,

identification with the game character, and game

enjoyment are discussed.

2 LITERATURE REVIEW

Despite a relatively small size of research

conducted, compared to traditional media violence

(Bushman and Huesmann, 2001, Paik and

Comstock, 1994), studies on violent video games

demonstrated that exposure to violent video games

could cause increase in aggression. The recent meta-

analysis (Anderson, 2004) confirmed a significant

effect of violent video game on aggressive thoughts,

affect, and behaviours, the association of which is

greater than the level of association in the

relationship between condom use and decreased

HIV risk, exposure to passive smoke at work and

lung cancer, and calcium intake and bone mass

(Bushman and Huesmann, 2001).

In spite that previous research showed a

significant relationship between violent video games

and aggression among players, there have been

relatively a few studies that examined specific

factors within such violent games that may influence

the degree of post-game aggression. In the following

sections, I will review the literature on factors within

game play such as character identification and role

enactment and their relationships with game-play

experience.

2.1 Character Identification

Sherry (2001) argued that despite the existence of an

association between violent video games and

aggression, the effect size is small and can vary with

factors such as types of violence contained in games.

For instance, Schneider and colleagues (2004) found

the effects of narrative on players’ emotional,

motivational and psychological responses to first-

person shooter games. Their argument is that the

presence of a storyline provides a context for

engaging in violent acts, which makes people

relatively easy to justify such violent acts.

Studies in traditional media suggest stronger

effects of justified violence on aggressive

behaviours (Berkowitz, Park, Leyens, and West,

1974; Tan, 1981). For instance, youths are likely to

be attracted to and identify with the “real heroes,”

portrayed by the media as aggressive and powerful

figures (Huesmann and Eron, 1986). When this

happens, they have the tendency to justify the

violent acts that the “real heroes” perform, which, in

turn, may result in their increased level of aggression

both in the short run (Perry & Perry, 1976) and in

the long run (Huesmann & Eron, 1986).

In contrast to several studies conducted in

traditional violent media, there are a few relevant

studies conducted in the context of violent video

games that investigated the effects of identification

with character on aggression. In an effort to bridge

the gap, Konijin, Bijvank and Bushman (2007)

conducted a video game study to examine factors

that can affect individuals’ identifications with the

attractive role model in a violent game. They made

distinction between similarity identification and

wishful identification and suggested that one’s

wishful identification, a concept close to the

vicarious learning (Bandura, 1986), was linked to

one’s increased aggression. Taken together, the

more players wish that they were main characters in

violent video games, the more likely they show post-

game aggression and enjoyment as shown in

research on traditional media effects and video

games

2.2 Proteus Effect and Role Enactment

The underlying framework that can further illustrate

the relationship between one’s identification with

the aggressive “heroes” in violent media and the

correspondingly increased aggression of players is

the Proteus effect (Yee & Balienson, 2007). The

Proteus effect is based on self-perception theory

(Bem, 1972, Goldstein & Cialdini, 2007) and

applied to a new context of virtual environments

(VEs).

Self-perception theory posits that individuals

define themselves by observing their own actions

from a third-person perspective and behave in line

with the inferred “self-concept” (Bem, 1972). In

light of self-perception theory, Goldstein and

Cialdini (2007) further contended that people would

infer their own attributes by observing the

behaviours of the “psychologically merged others”

(i.e., whom they share merged identity with) as if

they were observing themselves performing the

specific behaviour. Therefore, a sense of merged

identity with “others” is a key precondition for one

to shape the self-concept from observing others and

to behave in a way that conforms to the salient self-

concept.

In line with this, Yee and Balienson (2007)

coined the term, “the Proteus Effect” to elaborate

how an avatar that one plays in VEs can in turn

impact one’s behaviour. As individuals are

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

262

deindividuated in virtual environment (Mckenna &

Bargh, 2000), it is very likely that they adhere to the

identities inferred from their avatars. As such,

according to the self-perception theory (Goldstein &

Cialdini, 2007), they tend to perform behaviours that

conform to the perceived identities and stereotypes

of their avatars. For instance, Yee and Balienson

(2007) found that individuals assigned to a more

attractive avatar in VEs appeared to be more

intimate with the confederate than those assigned to

a less attractive avatar; and individuals assigned to a

taller avatars showed more confidence than those

assigned to a shorter avatar. Similarly, a study of

gender-role enactment in VEs (Jung and

McLaughlin, 2008) further proved supporting

evidence to the Proteus effect. The researchers found

that by enacting gender roles in VEs, people showed

corresponding gender self-concepts and gender-

stereotyping behaviors. More specifically,

participants assigned to a male role kept a greater

physical distance between themselves and the

computer agent than those assigned to a female role,

regardless of their biological sex.

In summary, people may show various degrees

of aggression and enjoyment, depending on the role

and visual representation of the main character in

violent video games in ways that conform to role

stereotypes. For example, the American TV drama

“Prison Break” gives a good demonstration of the

potential impact of character roles by showing how

the criminal “Michael” and his brother can become

“heroes” in the eyes of audiences. In the context of

video games, previous studies suggested games

containing human and fantasy violence (Sherry,

2001) and presence of a storyline (Schneider, Lang,

Shin & Bradley, 2004) are more likely to justify the

violent acts and thus make the players more

aggressive and enjoyable. As such, the depicted role

of a main character in violent video games, together

with wishful identification, may influence players’

post-game experience such as aggression and

enjoyment. Based on this, I propose the following

research question and hypothesis:

 RQ 1: What is the effect of a character role on

the level of aggression and game enjoyment among

players in a violent video game?

 H1: Wishful identification with the main

character will be positively associated with the level

of aggression and enjoyment among players.

3 METHODS

3.1 Participants

A total of 36 undergraduate students in a university

in Singapore participated in this study (12

participants per each condition). They were given

course credits and five-dollar incentive for their

participation. Each participant was scheduled to

come to a lab to participate in a 30 minute

experiment individually. Upon their arrival, the

participants were greeted and escorted to a game

room with a comfortable chair and a 40 inch flat

screen TV to play a certain title of violent video

game assigned to them. Then, they were asked to

complete the reaction-time task and online survey

about their game experience.

3.2 Manipulations of Different Roles

Three different types of violent video games were

selected in this study in terms of the distinct role

enacted in each condition: the police officer,

gangster and athlete. In order to reduce bias caused

by different titles within each condition, six titles of

violent video games (i.e., two titles per each

condition) were used in this study. All the game

titles are rated M for mature audience in terms of the

level of violence.

In the condition of violence from a positive role,

participants played game titles such as

“Stranglehold” and “Resident Evil” that portray a

role of the main character as a police officer who has

to fight violently with criminals or monsters to

accomplish given missions. Two game titles for the

condition of violence from a negative role are “Kane

& Lynch” and “Grand Theft Auto TV.” In “Kane &

Lynch,” participants played a criminal who breaks

out the prison and fights against the police. In

“Grand Theft Auto TV,” one of the most popular

violent video games, participants played a criminal

who performs violent behaviors such as stealing cars

and killing people in the city, without being caught

by the police. In the condition of violence from an

athlete role, two sports game titles were selected

such as “UFC” and “SMACKDOWN VS RAW.”

Although they are categorized as sports games, they

are also rated M for mature audience due to their

violence. Participants in this condition fought with

other fighters to play in the context of wrestling and

ultimate fighting.

UNDERSTANDING MEDIA VIOLENCE FROM A ROLE-PLAY PERSPECTIVE

- Effects of Various Types of Violent Video Games on Players' Cognitive Aggression

263

3.3 Procedures

3.3.1 Pre-experiment

Each participant was assigned to one of the six game

titles and played the assigned video game upon their

arrival at a laboratory. Firstly, participants were

given a brief instruction about the experimental

process. Secondly, the participants were given a

brief narrative about the game and the character to

help them understand the background of the game

and the character. A photocopy of the storyline of

the video game was also presented to the

participants. The participants were asked to read it

carefully to ensure that they have enough time and

exposure to be familiar with the role of the main

character. Finally, the participants were given an

enlarged photocopy of the control panel to help them

understand how to play the video game.

3.3.2 Post-experiment

After playing a assigned video game for about 30

minutes, participants conducted a reaction-time task.

They were asked to read aloud a series of words

displayed on the screen, immediately after playing

the violent game. The lab was kept completely quite

during the reaction-time task in order to prevent

unintended errors from noises. At last, participants

completed an online survey about their video-game

experience.

3.3.3 Measures

For the reaction time task, a simpler version of the

Genov Modified Stroop Task (Anderson et al., 1996,

Kirsh et al., 2005) was used to measure participants’

reaction times (RTs) in milliseconds to 20 negative

words and 20 neutral words. When each word was

displayed on a computer monitor, participants were

asked to read the word aloud as quickly as possible.

For the training purpose, participants first tried the

task with eight words: ROSE, CURVE, CALL,

WILLOW, FAIL, GLOOM, FIRE, and SORROW

(Kirsh et al., 2005). After the training session,

participants completed the RT task with 40 words.

The presentational order of the words was

randomized. The RT for each word was logged in

the database automatically in the unit of millisecond.

The final index for RTs was the difference in the

average RT between the neutral words and the

negative words. We took the difference in order to

control for individual differences in RTs. We used

the 40 stimulus words matched for length and

frequency in previous research (e.g., Kirsh et al.,

2005, Kucera & Francis, 1967, Sharma & McKenna,

2001). The 20 negative words were; FEAR,

CRASH, GRIEF, DEATH, PAIN, ANGRY,

MURDER, HATE, SHOCK, CANCER, ENEMY,

AFRAID, MISERY, EVIL, KILL, TRAGIC,

THREAT, RAGE, PANIC, and BEATEN. The 20

neutral words were; GATE, NOTE, CLOCK,

THUMB, FIELD, LEVEL, LEAGUE, WIRE,

BREAD, AUTUMN, ANCHOR, SHOT, NAVAL,

SENIOR, EXCEED, LINK, PLATE, DIVIDE,

CALL, and FOOT. Higher scores indicated faster

RTs to negative words, which meant more salient

cognitive-associative networks for aggression

(Bushman, 1998).

For game enjoyment, I used six adjectives on a

five-point scale ranging from “describes very

poorly” (1) to “describes very well” (5); for

example: “enjoyable,” “entertaining,” and “fun.:

Higher scores indicated more positive game

experience (α = .93).

For wishful identification, I used

four Liket-type

items on a five-point scale adapted from a study by

Konijin and colleagues (2007). The four items

include “I wish I were like the (main) character in

the game,” and “I wish I could do the same things

the character does.” Higher scores indicated stronger

feelings of wishful identification with the game

character (α = .87).

4 RESULTS

I used Analysis of Variance (ANOVA) to test mean

differences among three difference conditions. The

results showed no significant difference in reaction-

time differences among three conditions, F(1, 32) =

0.54, n.s. However, the trend showed that the

participants who played gangster games reacted to

negative words faster than the participants who

played police games. The participants who played

sports games showed the slowest reaction times to

negative words, which implies becoming less

aggressive after playing violent games. I believe that

statistically non-significant difference is due to the

small sample size in this study (see Table 1).

Table 1: Means and standard deviations of all measured

variables.

Police Gangster Sports

M SD M SD M SD

RT 5.75 60 18 49 -2.7 51

Enjoyment 3.91 0.64 2.97 1.11 3.29 1.02

Identification 2.25 1.0 1.29 0.54 1.35 0.47

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

264

In addition, the results showed a significant

difference in wishful identification between the

police violence condition and the other two

conditions, F(1, 33) = 6.83, p < .01. Specifically, the

participants who played police games wished to be

like the violent character in the games more than the

participants who played gangster games and sports

games. It makes sense that people wish to be like a

good character more even in the context of violence.

In the following regression analysis, the results

showed that wishful identification was a significant

predictor for game enjoyment (standardized

regression coefficient = .4, p < .05).

5 CONCLUSIONS

In the current study, I investigated the effects of

various depicted roles in violent video games on

players’ post-game experience measured by

cognitive aggression and enjoyment of the games.

The results showed no significant difference among

three different role conditions: police vs. gangster

vs. sports. Nevertheless, the trend suggests that

future research is needed to examine the potential

effects of contextual factors on game experiences.

The participants in the police condition showed

the strongest feeling of wishful identification and

enjoyed playing games the most. As such, I found a

significant association between wishful

identification and game enjoyment. This finding

implies that the more people wish to be like the main

character in the game, the more they enjoy the game.

For the game designers, games can attract more

players if the main characters in the games have

desirable characteristics that make players wish to be

like the characters. In other words, the narrative of

games is important for players to enjoy games.

There are a few limitations in this study. First of

all, the sample size is small, which could have

resulted in non-significant results on aggression.

Secondly, the time to play games was short. Future

research needs to consider a longitudinal study for

more robust outcomes. Lastly, I only measured

cognitive aggression. Future research can measure

affective and behavioural aggression, in addition to

the cognitive dimension for a more comprehensive

understanding of post-game aggression.

As a final remark, given the results, it may not be

a bad idea to have violent games with good-guy

characters such as the police although it is rather

counter-intuitive. Such contexts may potentially

reduce the level of postgame aggression and yet can

make players wish to be like the main characters

more, which, in turn, result in more positive game

experience.

ACKNOWLEDGEMENTS

This research was supported by the Wee Kim Wee

School of Communication & Information, Nanyang

Technological University (Grant No.

RCC3/2009/SCI). The author thanks Chen Ruili and

Shi Cong Kristy for their help in data collection.

REFERENCES

Anderson, C. A., 2004. An update on the effects of

playing violent video games. Journal of Adolescence

27 (2004) 113–122.

Anderson, C. A., Anderson, K. B., Deuser, W. E., 1996.

Examining an affective aggression framework:

Weapon and temperature effects on aggressive

thoughts, affect, and attitudes. Personality and Social

Psychology Bulletin, 22(4), 366-376.

Bandura, A., 1986. The social learning perspective:

Mechanisms of aggression. In H. Toch (Ed.),

Psychology of crime and criminal justice (pp. 198–

236). Prospect Heights, IL: Waveland Press

Bem, D., 1972. Self perception theory. In L. Berkowitz

(Ed.), Advances in experimental social psychology

(Vol. 6, pp. 2-57). New York: Academic Press

Berkowitz, L., Parke, R. D., Leyens, J. PH., West, S.,1974,

“Reaction of young delinquents to justified and less

justified movie violence”. Journal of Research in

Crime and Delinquency , 11, 16-24

Bushman, B. J., 1998. Priming effects of media violence

on the accessibility of aggressive constructs in

memory. Personality and Social Psychology Bulletin,

24(5), 537-545.

Bushman, B. J., Anderson, C. A., 2002. Violent Video

Games and Hostile Expectations: A Test of the

General Aggression Model. Personality and Social

Psychological Bulletin, Vol. 28 No. 12, December

2002 1679-1686.

Bushman, B. J., Huesmann, L. R., 2006. Short-term and

long-term effects of violent media on aggression in

children and adults. Archives of Pediatrics &

Adolescent Medicine, 160. 348-352.

Carnagey, N. L., Anderson, C. A., 2003. Theory in the

study of media violence: The general aggression

model, in Gentile, D. A. (2008). Media Violence and

children: A complete guide for parents and

professionals. Connecticut: Praeger

Goldstein, N. J., Cialdini, R. B., 2007. The Spyglass Self:

A model of vicarious self-perception. Journal of

Personality and Social Psychology, 92, 402-417.

Huesmann, L. R., Eron, L. D., 1986. Television and the

Aggressive Child: A Cross-national Comparison.

Lawrence Erlbaum.

UNDERSTANDING MEDIA VIOLENCE FROM A ROLE-PLAY PERSPECTIVE

- Effects of Various Types of Violent Video Games on Players' Cognitive Aggression

265

Jung, Y., McLaughlin, M. L., 2008. Role enactment based

on gender stereotypes in interactive media: Effects of

role play on the management of self-concept and

physical distance in virtual reality environments.

Annual Conference of International Communication

Association, Montreal, Canada

Kirsh, S. J., Olczak, P. V., Mounts, J. R. W., 2005. Violent

video games induce an affect processing bias. Media

Psychology, 7, 239-250.

Konijin, E. A., Bijvank, M. N., Bushman, B. J., 2007. I

wish I were a warrior: The role of wishful

identification in the effects of violent video games on

aggression in adolescent boys. Developmental

Psychology, 43(4), 1038-1044.

Kucera, H., Francis, W., 1967. Computational analysis of

present-day American English. Providence, RI: Brown

University Press.

McKenna, K. Y. A., Bargh, J. A. 2000. Plan 9 from

cyberspace: the implications of the Internet for

personality and social psychology.Personality and Social

Psychology Bulletin, 4, 57-75.

Oatley, K., 1994. A taxonomy of the emotions of literary

response and a theory of identification in fictional

narrative. Poetics, 23, 53-74.

Paik, H., Comstock, G., 1994. The effects of television

violence on antisocial behavior: A meta-analysis.

Communication Research, 21(4), 516-546.

Perry, D. G., Perry, L. C., 1976. Identification with film

characters, covert aggressive verbalization, and

reactions to film violence. Journal of Experimental.

Research in Personality, 10, 399–409.

Pinker, S., 2002. The blank slate: The modern denial of

human nature. New York: The Penguin Group.

Sharma, D., McKenna, F. P., 2001. The role of time

pressure on the emotional Stroop task. British Journal

of Psychology, 92, 471-481.

Schneider, E. F., Lang, A., Shin, M., Bradley, S. D., 2004.

Death with a story: How story impacts emotional,

motivational, and physiological responses to first-

person shooter video games. Human Communication

Research, 30(3), 361-375

Sherry, J. L., 2001. The effects of violent video games on

aggression: A meta-analysis. Human Communication

Research, 27, 409-431.

Sparks, G. G., Sparks, C. W., 2002. Effects of media

violence. In J. Bryant & D. Zillmann, Media effects:

Advances in theory and research (2nd ed., pp. 269-

285). Mahwah, NJ: Lawrence Erlbaum Associates

Tan, 1981. Mass communication theories and research.

Columbus, OH: Grid Publishing.

Yee, N., Bailenson, J., 2007. The Proteus Effect: The

Effect of Transformed Self-Representation on

Behavior, Human Communication Research 33 (2007)

271–290.

ICEIS 2011 - 13th International Conference on Enterprise Information Systems

266