See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/285776156

Violent video game exposure and aggression: A literature review

Article · March 2004

CITATIONS READS
91 7,650

2 authors:

Nicholas L. Carnagey Craig A Anderson

n/a Iowa State University
14 PUBLICATIONS   2,185 CITATIONS    239 PUBLICATIONS   25,127 CITATIONS   

SEE PROFILE SEE PROFILE

All content following this page was uploaded by Craig A Anderson on 17 October 2016.

The user has requested enhancement of the downloaded file.

 REVIEWS
MINERVA PSICHIATR 2004;45:1-18

Violent video game exposure and aggression

A literature review

N. L. CARNAGEY, C. A. ANDERSON

Over the last 30 years, the video game industry Department of Psychology
has grown into a multi-billion dollar business. Iowa State University, Ames, IA, USA
More children and adults are spending time
playing computer games, consoles games, and
online games than ever before. Violence is a
dominant theme in most of the popular video
games. This article reviews the current litera-
ture on effects of violent video game exposure tronic paddles. In less than 30 years, this
on aggression-related variables. Exposure to humble beginning of the video game indus-
violent video games causes increases in aggres- try has grown into countless companies with
sive behavior, cognitions, and affect. Violent annual sales totaling $ 20 billion worldwide.1
video game exposure also causes increases in The Playstation video game console, which
physiological desensitization to real-life vio- began as a side project at Sony, now repre-
lence and decreases in helping behavior. The
current video game literature is interpreted in sents $ 6 billion of the company’s $ 20 billion
terms of the general aggression model (GAM). in annual sales.1 In recent years, video game
Differences between violent video game expo- annual sales have steadily been higher than
sure and violent television are also discussed. movie ticket sales.2-4
Key words: Child - Video games - Violence - The history of video games can be divid-
Aggression. ed into 3 eras.5 The first era (1977-85) was the
“Atari era”, due to the fact that Atari consoles
dominated the video game market. These
Development and advances first video games contained little violence.
of video game industry The little violence in these early games was
quite abstract. Nolan Bushnell, the founder of
Atari, said, “We had an internal rule that we
I n 1972, a product was released to the
American market that would open the door
for the development of a multi-billion dollar
wouldn’t allow violence against people. You
could blow up a tank or you could blow up
industry. Pong, the first commercially avail- a flying saucer, but you couldn’t blow up
able video game, had entered the homes of people”.6 Another reason for the relatively
America. In Pong, 2 players tried to “hit” an low amount of violence in the early video
electronic “ball” back and forth with elec- games was that the graphical ability of the
Atari games was so low that only simple
graphics could be displayed.
Address reprint requests to: Dr. N. L. Carnagey, Department Over the next decade computers became
of Psychology, W112 Lagomarcino Hall, Iowa State University,
Ames, IA 50011-3180. E-mail: [email protected] more sophisticated, enabling more complex

Vol. 45, N. 1 MINERVA PSICHIATRICA 1

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

video game graphics. As graphics developed, games. It is a “third-person” game because the
so did the potential for profits. Violence also player can see the character that he or she is
began to appear more, even in children’s controlling. It is a “fighting” game because
games. The second era (1985-95), also known virtually all of the game action consists of
as the “Nintendo era”, was dominated large- fighting other game characters. A variety of
ly by Nintendo console games. The Nintendo third-person fighting games were very pop-
console introduced a more powerful plat- ular in this era. Street Fighter is one such
form than its Atari predecessor and began game. As in Mortal Kombat, the main theme
introducing violent themes in numerous is that the player engages in a series of fights
games. Even the seemingly innocuous Super with various opponents. Another interesting
Mario Brothers games included the capacity feature of many third-person fighting games
to destroy harmful creatures by jumping on is that the player can choose who he or she
top of them or by throwing fireballs at them. wants to “be” from a variety of male and fe-
The increased computing power of the sec- male characters. In part, this was an attempt
ond era consoles enabled more complicated to attract more female consumers.
graphics, including more realistic portrayals of First-person shooter games were another
violence. It was also during this era that vid- type of violent video game that developed
eo games became common on desktop com- during the “Nintendo era.” In these games,
puters and in hand-held mini-game systems the player sees the scenario through the eyes
such as Game Boy. As it became apparent to of the main character. The player can see his
manufacturers that violent games sold well, or her own hands and arms, as well as the
the level of violence in the games also in- weapons being used, but does not see his
creased. Truly violent video games came of or her whole character. The games are re-
age in this era with the killing games Mortal ferred to as “shooters” because most of the ac-
Kombat, Street Fighter, and Wolfenstein 3D.7 tion involves shooting enemies with one kind
Mortal Kombat led the way in 1993 by be- of weapon or another. Wolfenstein 3D was
coming the most popular video game of the one of the first very popular 3-dimensional
year.2 In Mortal Kombat, the player controls “first-person shooters.” In one version of
a character enrolled in a fighting tournament Wolfenstein 3D, the player assumes the role
where the only way to advance is to kill your of B. J. Blascowitz, an American soldier
opponent. Players also receive extra rewards caught and taken prisoner as a prisoner of
for using extreme violence (e.g. ripping war by the Nazis during World War II. The
opponent’s spine out or decapitating oppo- player’s job is to escape the prison and shoot
nent). Both Sega and Nintendo released his or her way through Castle Wolfenstein,
home console versions of the popular arcade killing everything that moves (both prison
game Mortal Kombat at about the same time. guards and guard dogs), with the ultimate
However, Nintendo sold a sanitized version goal of assassinating Adolf Hitler. The graph-
of the game, removing the most graphically ics of this game were very violent for this
violent features, depictions of blood, and the era. A successful player would see multiple
worst of the fatal moves. Sega released the full bloody murders and hear victims scream and
version and outsold the Nintendo version by groan. In Wolfenstein 3D the human hero
about 3 to 1. When Nintendo released Mortal can choose from an array of weaponry in-
Kombat 2, it included all of the blood, gore cluding a revolver, automatic weapons, a
and fatal moves of the Sega version. This flamethrower, and a knife.
time, the Nintendo version outsold the Sega We currently are in the third video game
version, probably because Nintendo consoles era (1995-present). The console game market
were already dominant in the marketplace. is largely dominated by the Sony Playstation
Some of the basic characteristics and la- and the most current platform, Playstation 2.
bels of video games also emerged in this era. Their graphic capabilities have been greatly
Mortal Kombat represents a type of game enhanced not only by improvements in com-
now known as “Mortal Kombat fighting” puter technology but also by Sony’s decision

2 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

(emulated by others) to switch from cartridge- games can now be downloaded from the
based systems to CD-ROM, and even more re- Internet. This includes “demo” versions of
cently, DVD-ROM based systems. Currently, extremely violent games that include most or
in addition to Playstation 2, video game buy- all of the graphic features of the full game.
ers also have options of purchasing Ninten- Such demos can be downloaded at no
do’s latest console (Nintendo Gamecube) or charge by virtually anyone with a computer
Microsoft’s X-Box (Microsoft’s first endeavor and a modem. Walsh 11 found that 32% of all
into the console market). With these chang- boys surveyed who play video games have
es in computing power and graphic quality, downloaded such “demo” games from the
the growth of video gaming has been phe- Internet.
nomenal in recent years. In 2001, despite an
economic recession, the video game industry
experienced 43% increase in sales, boosting Video games, violent content,
American sales to $9.4 billion.8 Of course, and preference for violence
video gaming on computers has also evolved
into more violent gaming with more realistic The content of video games has drastical-
graphics. ly changed from the decade dominated by
An emerging current trend is the growth of Atari. Recent content analyses of video games
online gaming. There are numerous games show that as many as 89% of games contain
that one can play over local area networks some violent content,12 and about 50% of the
and over the Internet. The most recent ad- games include serious violent actions toward
vancement is Microsoft’s X-Box Online net- other game characters.12-14
work. X-Box owners can use a broadband In addition, many children prefer to play vi-
connection to access the X-Box network, olent games. Even older surveys of school
where they can play a variety of games with children (4th through 8th grade) showed that
or against other online players. Some of these more than 50% preferred games dominated
games are simply more complex versions of with themes of human violence or fantasy
first person shooters, in which groups of gam- violence.15, 16 In surveys of paired children
ers can play with or against other gamers in and parents, about 2/3 of children named vi-
real time. Perhaps the most interesting trend, olent games as their favorites. Most parents,
though, is the emergence of subscription- however, are not likely to know what their
based online role-playing games, known as child is playing. Only 1/3 of parents were
massively multiplayer online role playing able to correctly identify their child’s favorite
games (MMORPGs). To play these games, video game. In 70% of the incorrect parental
one must subscribe (currently, about $25-$40 responses, children described their favorite
per month) to the company hosting the game. game as violent.17
Players create their own character for the Not only is violence a dominant theme in
game, and can increase the skills and power current video games, but video game compa-
of that character by playing the game. nies are marketing those violent games to-
Characters can kill and can be killed by oth- wards youth. A Federal Trade Commission
er players as well as characters built into the report 18 revealed that 70% of the M-rated
game. Everquest is currently the largest games (games suitable for people 17 or old-
MMORPG, with around 400 000 players er according to the Entertainment Software
worldwide.9 Players can be heavily invested Rating Board) were marketed to children
in their online quests, by buying and selling under 17. Fifty-one percent of the M-rated
their created characters via online auctions for game titles researched had at least one adver-
hundreds of dollars.10 Indeed, this could be tising plan that blatantly included targeting
seen as the beginning of a new video gam- children under 17. Additionally, 10 of the 11
ing era. companies surveyed had documents that in-
The best selling console video games are cluded males under 17 as part of the target
also usually available on computers. Many audience for their M-rated games.

Vol. 45, N. 1 MINERVA PSICHIATRICA 3

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

There is a large discrepancy between what Effects of viewing media violence

the video game industry considers violent as
compared to the public. The video game in- The vast amount of research conducted
dustry and its ratings board (Entertainment on the effects of violent television and movies
Software Rating Board) claim to see much on aggressive behavior spans several decades.
less violence in their games than do parents 19 Eighty studies had been published on the ef-
and other research groups.20 For example, fects of media violence on aggressive behav-
many games contain cartoon-like violence, ior by 1975. A meta-analysis conducted on
(known as mild animated violence) which these early studies revealed that exposure to
the industry claims is appropriate for all ag- media violence (both in the laboratory and in
es, but parents and children disagree.17 real-life settings) causes increases in aggres-
sive behavior.25 However, although the scien-
tific research clearly demonstrated that ex-
Time spent playing video games posure to media violence led to increases in
and parental control aggressive behavior, the news media’s cover-
age of this issue painted quite a different sto-
As the financial aspect of the video game ry. Since 1975, research on media violence
industry has grown, so has time spent play- has yielded even stronger evidence of causal
ing video games. In the mid 1980s children effects on aggression, but the news cover-
were spending an average of 4 h a week age in following years portrayed the media vi-
playing video games, both at home and in olence effects as weaker than did earlier news
arcades.21 By the mid 1990s, video game us- reports.25
age had increased to 4.5 h per week for 4th Despite how the news media continues to
grade girls and 7.1 h per week for 4th grade portray the effects of media violence, the re-
boys.15 search is clear: youth who view violent tele-
Recent estimates have shown that video vision tend to become more aggressive
game usage has grown for both young and adults.25-31 Viewing violent television causes
older children alike. Children ages 2 to 7 have increases in aggressive cognitions, affect, and
been shown to play video games an average behavior. In a recent meta-analysis, Bushman
of 3 to 5 hours a week.22 School-age children and Anderson 25 found that the correlation of
(both boys and girls) spend an average of viewing violent television and aggression is
about 7 h per week playing video games.22 greater than correlations of being exposed
These numbers are even higher for slightly to asbestos and contracting laryngeal cancer,
older youth, with 8th and 9th grade students consuming calcium and increased bone mass,
reporting an average of 9 h (13 h for boys, 5 or wearing a condom and not contracting
h for girls) a week spent playing video HIV.
games.23 In 1999, 2.5% of entering college Another media violence effect besides in-
men reported playing video games over 20 h creasing aggressive behavior is desensitization
per week.24 to violence. There is an empirical basis as
In addition, parental supervision of well as a theoretical basis for the desensitiz-
children’s video game use is almost entirely ing effect of violent media. Research has
absent. Walsh 11 reported that 89% of teens shown that participants exposed violent me-
surveyed said their parents never put a limit dia (slasher films, police television programs,
on the amount of time spent playing video violent boxing matches) are less physiolog-
games. Also, 90% of the youth surveyed in ically aroused by real world violence than
grades 8-12 reported that their parents nev- are those exposed to nonviolent media.32-36
er check the ratings of video games before al- Exposure to violent media also changes
lowing the youth to purchase them. Only 1% peoples’ perceptions of violence. Research
of the youth surveyed reported their parents has shown that after viewing several sexual-
had ever prevented them from purchasing a ly violent movies, participants rated the last
video game because of its rating. movies in the set as less violent 32, 37 and

4 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

showed less sympathy and assigned more pg/s. Microsoft’s Xbox, released in 2001, in-
responsibility to a rape victim compared to creased graphic capability to 125 million pg/s.
those who viewed nonviolent movies.37, 38 The goal for PlayStation 3 is 1 billion pg/s.
Exposure to media violence has also been The dramatic increase in speed and graphic
shown to cause decreases in helping a vio- capability has allowed for more realistic vio-
lence victim.39-41 lence than ever before. At this rate, the differ-
ence of graphics between video games and
television may not be an issue in the very
Differences between television near future.
and video games
Why violent video game exposure may be
The effects of violent television and movie more detrimental
exposure have received considerably more
There are several reasons that violent video
research attention than the effects of violent
games could have larger effects than television
video games. The main reason for this is be-
programs. Some of these differences are that
cause video games are still a relatively “new” playing violent video games involve almost
medium of entertainment compared to tele- complete attention and involvement, more iden-
vision and movies, which have received tification with violent characters, more rein-
decades of examination. Due to their similar- forcement of violent acts, and higher frequen-
ity to television programs and movies, we cy of violent scenes.7, 42, 43
can hypothesize that violent video game ex-
posure will have similar effects on aggres-
sion and aggression related variables. LEVEL OF ATTENTION
However, besides their obvious similarities, Television or movies do not necessarily
there are also a variety of differences that require a large amount of attention from
need to be acknowledged. There are sever- viewers. Viewers are not forced to focus their
al theoretical reasons why exposure to violent attention to the television. Programs may be
video games may have greater or weaker ef- running in the background while individuals
fects than exposure to violent television on are completing other tasks (reading newspa-
aggressive behavior, cognitions, affect, and per, talking with others, leaving the room,
desensitization. These differences are ad- etc...). Attention from television or movies
dressed below. can shift without having any effect on the
program itself. Video games, however, re-
Why violent television exposure may be more quire a higher level of attention from the
detrimental player. A player must typically focus his or her
attention towards the video game or a failure
The most obvious difference between vi- of goals will occur. This means that the player
olent video games and violent television pro- is constantly watching the screen and is fo-
grams is the level of graphical quality. Current cused on any potential violence that will be
and past video game consoles do not have shown.
the graphic capability compared to televi-
sion and movies. This difference, however,
could become a moot point in the near future, LEVEL OF ACTIVE INVOLVEMENT
as the graphical capabilities of video games Viewing television or movies can be a rel-
are increasing a high rate. The original Sony atively passive process. Violence will occur
Playstation processed 350 000 polygons per during the television program regardless of
second (pg/s). Sega Dreamcast increased the what the viewer does. In video games, this is
graphical capability by over 9 times in 1999, not the case. What happens next on the
when it processed over 3 million pg/s. screen depends on how the player controls
Playstation 2 blew Dreamcast out of the wa- the characters in the game. Video game
ter when it’s new system processed 66 million players are responsible for the violence they

Vol. 45, N. 1 MINERVA PSICHIATRICA 5

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

see on the screen. Players are the ones who aggression is likely to increase when it is re-
are pulling the trigger and throwing the warded.
punches. Research has shown that learning is
enhanced when people are actively rather FREQUENCY OF VIOLENT SCENES
than passively involved.44-46 This active com-
ponent could enhance the negative effects Even in the most violent movies, the vio-
of violent video games, relative to more pas- lence is not completely constant. There are
sive forms of viewed violence. typically less intense scenes with a romanti-
cally or comical theme that enhance the plot
of the TV program or movie. This is not nec-
IDENTIFICATION WITH VIOLENT CHARACTERS
essarily the case for video games. Despite
People who view violence in television the more recent addition of “cut scenes” (di-
programs or movies might identify with vio- gitized movie clips that enhance the plot of
lent characters, but not necessarily. Violent the game), the violence in video games is al-
video game players are essentially forced to most continuous. Player must constantly be
identify with the character they are control- ready to shoot the next enemy and witness
ling. Players are required to take on the iden- the bloody repercussions. Players are almost
tity of a video game persona and “become” continuously exposed to scenes of gore,
the violent character. In first person video blood, and screams of pain. Together these
games, the player sees the virtual world negative sights and sounds are accompanied
through the eyes of the main character. The with positive events (reward of points, new
perspective on the screen is everything that weapons, advancement to higher levels) that
the main character would see. In some third activate positive emotions. Because of this
person games, players are allowed to alter factor, desensitization is likely to occur at a
the appearance, gender, and name of the higher rate from exposure to violent video
character they are controlling. This can al- games compared to violent television if the
low the player to create a visual replica of graphical nature of both forms of media is
oneself in the video game. In some more so- fairly equivalent.
phisticated games, the player can import
scanned images of faces directly onto charac- Empirical issues in the video game vs televi-
ters in the game (called “skins”). Research sion debate
on violent television has demonstrated that
identifying with a violent character increases Currently, it is unclear whether exposure to
the media violence effect.29, 47-49 violent video games or exposure to violent
television has a larger impact on aggressive
behavior. The main reason for this haziness
REINFORCEMENT OF VIOLENT ACTS
is because it is unclear exactly how to ex-
When viewing television or movies, a view- amine this issue in a laboratory setting. One is-
er may only receive indirect rewards for vio- sue is how to equate video games and TV
lent actions of the characters (e.g., witnessing programs on a variety of components (amount
when a violent character is rewarded for his of violence, graphical nature of violence, ac-
or her actions). When individuals play violent tive participation, etc. ...) or whether it is legit-
video games, there is direct (and typically imate or possible to equate these factors. At
instant) reinforcement for their choice of ac- the current time, it seems that each of these
tion. This reinforcement can come in numer- differences must be addressed one issue at a
ous forms: visual effects, sound effects (e.g., time. It still seems unclear how to control
groans of pain from an injured target), verbal certain differences between these 2 types of
praise (e.g., when a target is hit the comput- media (e.g., if trying to control frequency of
er says “well done” or “impressive”), points violent scenes, how do you account for par-
for various violent actions, and advancing to ticipants who are better players and view
the next game level after obtaining certain more violence because they accomplish more
goals. Bandura 50, 51 has demonstrated that in the given amount of time?).

6 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

Effects of violent video games more violent games also had higher hostile at-
tribution biases.56 People with hostile attribu-
Because violent video games are a rather tion biases have been shown to act aggres-
new type of violent media, the literature ex- sively and are likely to be socially maladjust-
amining its negative effects on players is rath- ed.59 These same biases have been found in la-
er small, but a rather clear consensus has al- boratory settings. Bushman and Anderson 60
ready been reached. This consensus is virtu- and Kirsch 61 showed that young adults who
ally identical to the conclusions reached in the had just played a violent video game generat-
violent television literature: playing violent ed more aggressive endings to story stems than
video games increases aggression. Several those who had played nonviolent video games.
studies, both correlational and experimen- Besides hostile attribution biases, aggres-
tal, have demonstrated that playing violent sive cognitions can be measured several dif-
video games can have a wide variety of neg- ferent ways. For example, Anderson and Dill 7
ative effects on players.52 Recent meta-anal- showed that playing a violent video game
yses 52, 53 have demonstrated that exposure to increased the relative speed with which the
violent video games increases aggressive be- person could read aggression-related words
havior, cognition, affect, and physiological (aggressive thoughts). In addition, studies
arousal, and decreases helping behavior. measuring cognitive responses to playing vi-
olent video games have shown that aggres-
Violent video games increase aggressive be- sive thoughts are increased compared to play-
havior ing nonviolent video games.7, 56, 61-63
Recent meta-analyses have shown the aver-
Correlational studies have revealed a pos- age effect size across studies between violent
itive relationship between playing violent vid- game play and aggressive cognitions is 0.27.52
eo games and aggressive behavior. Anderson These effects have also been found in chil-
and Dill 7 showed a positive relationship dren and adults, in males and females, and in
between violent video game exposure and experimental and non-experimental studies.
self-reported aggression on the National Youth
Survey, which includes items assessing as-
Violent video games increase aggressive affect
sault and robbery. Gentile et al.23 found that
young adolescents who played more violent Empirical studies have also demonstrated
video games reported more frequent aggres- that playing violent video games can lead to
sive behaviors, such as arguing with teachers an increase in aggressive affect. Anderson
and getting involved in physical fights. and Bushman’s 52 meta-analyses found that
Experimental studies have revealed the the effect size of playing violent videogame
same results: participants exposed to violent on aggressive affect is 0.26. In one study,
video games behave more aggressively than adults’ state hostility and anxiety levels were
participants not exposed to violent video increased after playing a violent game com-
games.7, 54-58 The average effect size across pared to control conditions.64 In a study of 3rd
studies between violent game play and ag- through 5th grade children, playing a violent
gressive behaviors was 0.19.52 These effects game increased frustration levels more than
have been found in children and adults, in playing a non-violent game.17 Results involv-
males and females, and in experimental and ing aggressive affect are less than clear be-
non-experimental studies. cause nonviolent video games can also cause
increases in negative affect by being to diffi-
Violent video games increase aggressive cog- cult and frustrating the player.
nition
Violent video games decreases prosocial be-
There is both correlational and experimen-
havior
tal evidence that playing violent video games
increases aggressive cognitions. In a correla- Prosocial behavior has been shown to de-
tional study, young adolescents who played crease with exposure to violent video games.

Vol. 45, N. 1 MINERVA PSICHIATRICA 7

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

Studies measuring emotional responses to violence than participants who had played a
playing violent video games have shown that nonviolent video game. This study is the first
prosocial behaviors are decreased compared to demonstrate that violent video games can
to playing nonviolent video games.58, 65-67 have a physiological desensitizing effect to
Carnagey, Bushman and Anderson 68 showed real-world violence.
that participants who played a violent video
game were slower at providing help to a vi-
olence victim than participants who played a Criticisms of the video game literature
nonviolent video game. Anderson and
Bushman 52 meta-analyzed 8 independent There are many criticisms of the current
samples and found the average effect of these violent video game literature. Some of these
studies was -0.19. criticisms are justified, while some of them are
completely unjustified.
Violent video games increase physiological
arousal Justified criticisms
Studies have found that playing violent SMALL SAMPLE SIZES
video games tends to increase heart rate, sys-
tolic and diastolic blood pressure compared Many current studies contain less than ideal
to playing non-violent video games.69 The sample sizes. Small samples sizes can cloud
a research area, by artificially generating what
average effect size across studies between
appear to be inconsistent effects. For exam-
violent game play and physiological arousal
ple, if the average effect size for violent vid-
was 0.22.52 For example, Ballard and Wiest 70
eo game exposure on aggressive behavior is
showed that a violent game (Mortal Kombat
about r = 0.20,52 then the number of study
with the blood “turned on”) resulted in high-
participants should be at least 200 for 0.80
er systolic blood pressure responses than ei-
power (likelihood of being able to find a le-
ther a nonviolent game or a less graphically
gitimate difference between groups). Studies
violent game (Mortal Kombat with the blood
with smaller samples may yield results that
“turned off”). appear inconsistent, with some apparently
Other physiological reactions have also “working” and others failing to find a statis-
been found. Adult males’ brains have been tically significant effect, or even finding op-
shown to release dopamine in response to posite effects. Two ways to overcome this
playing violent video games.71 In addition, obstacle are to increase sample sizes in future
Lynch 72, 73 has found that the physiological ef- research and use meta-analytic strategies to
fects of playing violent video games may be summarize previous findings rather than us-
even greater for already aggressive children. ing the more traditional narrative review.
Children high in trait hostility showed great- When this is done, a more consistent effect of
er increases in heart rate, blood pressure, video game violence is revealed.52
epinephrine, testosterone, norepinephrine
and cortisol in the blood. These findings sug-
gest that video game exposure effects may be NO DIFFERENCES IN VIOLENT CONTENT BETWEEN
CONDITIONS
greater for children who are already at a high
likelihood for aggressive behavior. Some studies have been criticized for the
Even though violent video game exposure lack of difference in violent content between
has been shown to increase certain physio- violent and nonviolent conditions. Typically,
logical measures, it has also been shown to either the violent condition resembles a non-
cause physiological desensitization to real- violent game or the nonviolent game resem-
life violence. Carnagey et al.68 demonstrated bles a violent game. When the “violent” and
that participants who previously had played “nonviolent” comparison conditions have on-
a violent video game for 20 min had lower ly a small difference in violent content, the ef-
heart rates while watching scenes of real-life fect sizes will also be small. When the differ-

8 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

ences are large, the effects tend to be large.74 ing aggressive behavior such as hitting an
Future studies need to do a better job of as- inanimate object. Future studies need to use
sessing the violent content of the video games better measures of aggression.
being compared and ensure there is a sig-
nificant difference between games. NO LONGITUDINAL STUDIES
Currently, there has not been a longitudi-
ADDITIONAL DIFFERENCES BETWEEN VIOLENT AND
nal study conducted on the negative effects
NONVIOLENT GAMES
of violent video games. The main reason for
Some studies utilized video games that this is because major funding is needed to
have other differences between them besides conduct a large-scale longitudinal study and
violent content. The nonviolent game could so far, no such funding has been issued. Until
be more boring or frustrating than the violent a large-scale study is funded, one must look
game. These additional differences can yield to the longitudinal studies conducted in the
inaccurate results. The obvious solution for TV/movie violence domain to get a reason-
future studies is to do more pilot testing or able guess as to the likely long-term effects.
manipulation checks on such aggression-rel-
evant dimensions. Unjustified criticisms
LACK OF RESULTS REPORTING TOO FEW STUDIES TO WARRANT ANY CONCLUSIONS
Some previously published studies have If a research area has too few studies, it
neglected to include all the necessary infor- may be impossible to detect a small effect
mation needed to include them in meta-anal- size. However, this is not the case for the
yses. For example, this has occurred in stud- video game literature. As noted earlier, meta-
ies in which 50% of the participants played a analyses have been conducted and clear
video game while the other 50% merely ob- findings have already demonstrated that ex-
served. Reported means were then collapsed posure to violent video games cause increas-
across this play vs observe dimension. Future es in aggressive behavior, cognitions, affect,
reports should include the individual means and decreases in prosocial behavior.52 If any-
necessary for including in future meta-anal- thing, it is remarkable that such reliable ef-
yses. fects have emerged from such a relatively
small number of studies (compared to TV
QUESTIONABLE MEASURES OF AGGRESSIVE BEHAV-
and movie violence studies) that vary so
IOR
much in method, sample population, and
video game stimuli.
Modern definitions of aggression restrict
its application to behaviors that are intended LACK OF EXTERNAL VALIDITY
to harm another person.43, 53, 75 A surprising
number of past studies have used inappropri- An old standby criticism used on a variety
ate dependent measures. For example, self- of research areas has been that experimental
reports on “Trait Aggression” scales have studies lack external validity, due to demand
been used as measures of aggressive behav- characteristics, participant suspicion, trivial
ior in short-term experiments. But there is measures, artificial settings, and unrepresen-
no way for a short-term manipulation of ex- tative participants. These arguments have
posure to violent vs nonviolent video game been successfully debunked many times by
(e.g., 20 min) to influence one’s past frequen- empirical analyses and have found little cause
cy of aggression. In this short-term context, for concern.76-79 In addition, examinations of
such a trait measure might possibly be con- these issues in the aggression domain have
ceived as a measure of cognitive priming, consistently found evidence of high external
but it clearly is not a measure of aggressive validity, and have done so in several very
behavior. Another example is operationaliz- different ways.80-83

Vol. 45, N. 1 MINERVA PSICHIATRICA 9

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

CORRELATION IS NOT CAUSATION college students’ reports of violent video game

play in prior years were positively related to
Psychology instructors teach this mantra
aggressive actions that would be considered
to introductory psychology students, and
criminal (e.g., assault, robbery) if known to
hope that they will gain a much more so-
police. Similarly, Gentile et al.23found signifi-
phisticated view of methods and scientific cant links between violent video game expo-
inference by the time they are seniors. Whole sure and fighting in school.
fields of science are based on correlational
data (e.g., astronomy). Correlational studies
are used to test causal theories by providing VIOLENT MEDIA AFFECT ONLY A SMALL PORTION OF
THE POPULATION
opportunities for falsification. A well-con-
ducted correlational design, one that attempts There are certain reasons to believe that
to control for likely confounding variables, some sub-populations may be more suscep-
can provide a large amount of useful informa- tible than others to the negative effects of vi-
tion. The main key for a particular research olent video game exposure. However, there
area is that it provides a consistent pattern of has been no evidence that particular popula-
results across studies that differ in design, tions are completely “immune”. Even some
procedure, and measures. The current re- populations thought to be at low risk have
search on violent video games and aggression nonetheless yielded significant violent vid-
related variables does yield consistent re- eo game exposure effects.7, 23
sults.52
EFFECTS OF MEDIA VIOLENCE ARE TRIVIALLY SMALL
AROUSAL ACCOUNTS FOR ALL VIDEO GAME EFFECTS
One of the most well known criticisms is
Another unjustified criticism of the video that even though there may be a violent vid-
game literature is that the arousal caused in eo game exposure effect, the effect is so small
violent video games accounts for the differ- that it is trivial. This is simply not true. Violent
ences in aggressive related variables. This video game effects have been shown to be
criticism has several flaws. First, arousal dis- larger than the effects of passive tobacco
sipates fairly quickly and the criticism cannot smoke on lung cancer, exposure to lead on
apply to studies that measure aggressive be- I.Q. scores in children, and calcium intake
havior more than 30 min after game play has on bone mass.25, 52
occurred. Also, this criticism generally does Critics have used all of the unjustified crit-
not apply to correlational studies, but they icisms discussed above to dismiss and trivial-
show a significant link between violent video ize the current video game literature. By us-
game exposure and aggression.52 Further- ing unjustified criticism “correlation does not
more, there are some experimental studies in equal causation”, one can ignore all of the
which the violent and nonviolent game con- correlational research. In addition, with use
ditions were equated on arousal, still reveal- of unjustified criticism “lab studies are trivial”,
ing significant violent content effects.7, 68 all of the experimental studies can be similar-
ly ignored. The only way to satisfy these un-
NO STUDIES LINKING VIOLENT VIDEO GAME PLAY
justified criticisms is to conduct a longitudi-
TO SERIOUS AGGRESSION
nal field experiment in which young chil-
dren are assigned to high or no exposure to
Another unjustified criticism of the video violent video games over the course of their
game literature is that there are no studies life while measuring their violent, criminal
that examine the relationship between violent activities. Obviously, such a study is com-
video game exposure to serious aggressive pletely unethical and could never (nor should
actions. This criticism is simply not true. A ever) take place. It is not an accident that all
number of correlational studies have linked ethically feasible types of studies are dis-
repeated violent video game play to serious missed by the industry and its supporters and
aggression. Anderson and Dill7 showed that the only way to prove (in their eyes) the dan-

10 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

gers of their products is to conduct a study

that would never be allowed.
Inputs Person Situation
Social
The general aggression model: encounter
a theoretical explanation
Present internal state:
Routes Affect
A theory developed in recent years can be
used in understanding the media violence Cognition Arousal
research is the general aggression model
(GAM).53, 75 GAM is an integration that com- Thoughtful
bines key ideas from earlier models: social action
Apraisal
learning theory and related social cognitive Outcomes & decision
theory concepts,50, 84-88 Berkowitz’s cognitive processes
neoassociationist model,89-91 Dodge’s social Impulsive
action
information-processing model,59, 92 Geen’s af-
fective aggression model,93 Huesmann’s script
Figure 1.—The general aggression model: episodic pro-
theory,94 and Zillmann’s excitation transfer cesses.
model.95 GAM describes a cyclical pattern of
interaction between the person and the en-
vironment. Three main points compose the present internal state: cognition, affect and
cycle: input variables of person and situa- arousal.
tion, present internal state of the individual,
and outcomes resulting from various apprai- COGNITION
sal and decision processes. Input variables can influence internal states
by making aggressive constructs more read-
Input variables ily accessible in memory. Constructs can be
According to GAM, a person’s behavior is either temporarily or chronically access-
based on 2 main kinds of input variables: the ible.96, 97 As a construct is repeatedly accessed,
person and the situation (Figure 1). The per- its activation threshold decreases. This means
that the construct requires less energy neces-
son variables are all the factors a person
sary for activation, making it chronically ac-
brings into the current situation, including
cessible. A situational input (e.g., a violent
traits, current states, beliefs, attitudes, values,
film) results in a temporary lowered thresh-
sex, scripts, and aggressive personality. old of activation, making the construct ac-
Situation variables are simply the environ- cessible for a short time. This temporary in-
mental factors surrounding the individual, crease in the accessibility of a construct is
including factors in the environment that often called “associative priming”. Situational
could affect the person’s actions, like aggres- variables may also activate aggressive
sive cues, provocation, pain, rewards, and scripts.94 Aggressive scripts can bias the inter-
frustration. pretation of a situation and the possible re-
sponses to that situation. Repeated access of
Routes aggressive scripts can also make them more
Input variables, sometimes interactively, readily accessible and more likely to be acti-
affect an individual’s appraisal of a situation vated in future situations, guiding future be-
and ultimately affect the behavior performed havior.
in response to that appraisal, primarily by in-
fluencing the present internal state of the in- AFFECT
dividual. According to GAM, these influences Input variables can also influence the
can occur through 3 main aspects of the present internal state through affect, which in

Vol. 45, N. 1 MINERVA PSICHIATRICA 11

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

turn can impact later behavior. For example, Outcomes

pain and uncomfortable temperatures can
Typically, the individual will appraise the
produce increases in state hostility (anger)
current situation and then select a behavior
and general negative affect.98 Exposure to vi-
appropriate for the situation before that behav-
olent movies, TV, or video games also in-
ior is emitted. Depending on the situational
creases state hostility.52, 99-103 Besides situa-
variables and resources present, appraisals
tional variables, personality variables are al-
may be made hastily and automatically, with-
so related to hostility-related affect. Self-re-
ported trait hostility has been shown to be out much (or any) thought or awareness, re-
positively related to state hostility.98, 99 sulting in an impulsive behavior. However,
frequently the individual will have the time
and resources to reappraise the situation and
AROUSAL perform a more thoughtful action. Of course,
The final aspect of the present internal both impulsive and thoughtful actions may
state that can be influenced is arousal. This be aggressive or nonaggressive.
can be accomplished a number of ways. An This action will then be followed by a re-
increase in arousal can strengthen an already action from the environment, which is typi-
present action tendency, which could be an cally other people’s response to the action.
aggressive tendency. For example, if the per- This social encounter can alter the input var-
son has been provoked at the time of in- iables, depending on the environmental re-
creased arousal, aggression is more likely to sponse. This encounter could then modify
be an outcome than if the arousal increase situation variables, person variables, or both,
didn’t occur. Geen and O’Neal 104 demon- resulting in a reinforcement or inhibition of
strated this phenomenon by showing that similar behavior in the future.75
loud noise increased arousal and aggression.
A second way in which arousal could in- Applying GAM to media violence
crease aggression is commonly associated
with excitation transfer theory.95 Arousal elic- GAM can be used to interpret the effects of
ited by other sources (e.g., exercise) may be exposure to violent media. Theoretically, vi-
misattributed as anger in situations involv- olent media can affect all 3 components of
ing provocation, thus more likely producing internal state. As mentioned earlier, the liter-
anger-motivated aggressive behavior. A third, ature on violent video games has shown that
and untested, way is that unusually high and playing them can temporarily increase ag-
low levels of arousal may be aversive and gressive thoughts, affect and arousal.52 Also
may stimulate aggression in a similar manner noted earlier, exposure to violent media can
as other aversive stimuli.53 reduce arousal to subsequent depictions of vi-
olence. Playing a violent video game can al-
so influence the person’s internal state
INTERACTION BETWEEN ROUTES through the affective route by increasing feel-
Not only input variables can influence cog- ings of anger, and through the arousal route
nition, affect, and arousal, but these 3 routes by increasing heart rate.52
may also influence one another. The idea Despite’s the GAM’s primary focus on the
that cognitions and arousal influence affect episode, it is not restricted to short-term ef-
dates back all the way to William James 105 fects. The cyclical process of GAM lends it-
and was first popularized among social psy- self to addressing long-term effects of expo-
chologists by Schachter and Singer.106 Affect sure to media violence. With repeated ex-
has also been shown to influence cognition posure to certain stimuli (e.g., media vio-
and arousal.107 Research has shown that peo- lence), particular knowledge structures (e.g.,
ple often use their affective state to guide in- aggressive scripts) become more readily ac-
ference and judgment processes.108, 109 Thus, cessible. Figure 2 displays this process and
hostility-related affect may make hostile cog- several common types of long-term changes
nitions more accessible, and vice versa. that may occur. Over time, the individual will

12 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

Repeated violent media exposure:

Learning, rehearsal, and reinforcemety6nt of aggression-related knowledge

Aggressive beliefs Aggressive perce- Aggressive expecta- Aggressive behavior Aggression

and attitudes ptual schemata tion schemata scripts desensitization

Increase in aggressive personality

Personological variables Situational variables

e.g., Aggressive personality e.g., Social situations
new peer group

General aggression model, as in Figure 1

Figure 2.—The general aggression model: personality processes.

employ these knowledge structures and pos- search on violent video games. First, one
sibly receive environmental reinforcement for must consider the research that has been
their usage. Over time, these knowledge struc- conducted on violent video games. Even
tures become strengthened and more likely to though this research area is still relatively
be used in later situations. Research supports small compared to research on the effects of
this notion by demonstrating that repeatedly ex- violence in TV and movies, there is sufficient
posing children to media violence produces evidence to state that playing violent video
aggressive adults.110 Such long-term effects re- games can cause significant increases in ag-
sult from the development, automatization, gression in the short term, demonstrated both
and reinforcement of aggression-related knowl- in laboratory-based situations and in real
edge structures. In essence, the creation and au- world contexts.
tomatization of these aggression-related knowl- Second, one must consider the sheer size
edge structures and the desensitization effects of the research literature on media violence
change the individual’s personality. Long-term in general. This literature has demonstrated
consumers of violent media, for example, can that media violence has significant causal ef-
become more aggressive in outlook, perceptu- fects on aggression and on interpersonal vi-
al biases, attitudes, beliefs, and behavior than olence in both the short and long term. The
they were before the repeated exposure, or processes underlying the effects of TV and
would have become without such exposure. movie violence are very similar to those pre-
sumed to underlie the effects of violent vid-
eo games, making findings from the TV do-
Public policy implications main very relevant to implications for video
and parental supervision game research.
Third, one must consider the various pro-
There are 3 main factors to consider when cesses underlying human aggression in gen-
discussing the potential implications of re- eral. Numerous advances in understanding

Vol. 45, N. 1 MINERVA PSICHIATRICA 13

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

human aggression have been achieved in re- binding system of access restrictions for chil-
cent years; advances that allowed the formu- dren. The video game industry has been fol-
lation of the general aggression model. As lowing in their footsteps in this regard. Other
Kurt Lewin 111 once said, “There is nothing so countries, however, do have governmental
practical as a good theory.” Though no the- agencies that evaluate a variety of types of en-
ory is ever complete in all respects, a good tertainment media for the primary purpose of
one allows accurate predictions about the restricting access to inappropriate material
likely effects of certain actions. Social by children without parental permission (e.g.,
scientists’ understanding of human aggres- Australia’s Office of Film and Literature
sion in general and of media violence effects Classification).
in particular is now good enough to allow
fairly accurate predictions of the likely ef- Parental supervision
fects of actions that policy makers and parents
might take. Based on survey research of parents and
children, we know that parental supervision
for obtaining and playing violent video games
Potential public policy action
is relatively low.11 Parents can correct this in
Public policy makers can aid parents in a number of ways. First, parents need to dis-
their attempts to provide a healthy media cover what kinds of media, including video
diet to their children by giving them back games, their children are ingesting, and then
some of the parental control that has been take steps to ensure that their media diet is a
lost to the media industry and their market- healthy one. Some parents lament that it is
ing departments. Legislation that restricts ac- impossible to totally control what their chil-
cess of minors to violent media by requiring dren see and do and therefore throw up their
parental permission for minor children to hands in dismay. It is true that they cannot to-
buy, rent, or play violent video games in ar- tally control their children’s media diets, but
cades, is likely to be helpful. Creating a sin- this is no reason to give up all attempts at
gle, unified rating system for various types of media diet control. For example, it is difficult
entertainment media would also help par- to control what games your child plays at a
ents regain some control by simplifying the friend’s house. However, it is also difficult to
current confusion of systems. The new rating control what your child eats when at a friend’s
system should be applied by a group that is house, but the solution is not to abandon your
truly independent of the entertainment media parental responsibility to provide nutritious
industries. It should also be based on the meals at home. As the general aggression mo-
best available research. We now know from del illustrates, the long-term negative effects
research, for example, that strictly age-based of exposure to violent video games is related
systems encourage children to seek out me- to frequency of exposure. Steps that reduce
dia that are “too old” for them.112 A rating such exposure in any environment, especial-
system that more clearly labels the content of ly the home, can be beneficial. Such steps
the video game might well be more infor- might include removing the TV, the video
mative to parents and might provide less en- game console, and the computer from the
couragement to youth to violate age targets. child’s room to an area that is more easily
Research also shows that even “cartoon” vi- monitored by the parent. One can also mon-
olence has a negative impact on children, itor and control what kinds of computer
especially on the youngest ones for whom the games are on whatever computers the child
cartoon violence is supposedly created. Of uses and can restrict the web sites that the
course, feasible types of public policies vary computer can access while in use by the child.
widely across countries around the world. Along these same lines, parents should en-
In the US, for example, the TV and movie courage the parents of their children’s friends
industries have been quite successful in their to provide a healthy media diet rather than a
efforts to prevent the creation of a legally violent one. None of these monitoring and

14 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

control tasks is easy, but a lot can be accom- If 2 or more answers are “yes”, think very
plished by a committed parent. carefully about the lessons being taught be-
Second, parents can actively teach their fore allowing your child access to the game.
children the reasons behind the restrictions
they place on certain types of media, why
such media can be bad for them. In a sense, Future research
this is teaching them to become more media
savvy. Similarly, discussing alternative nonvi- The evidence is now clear that playing vi-
olent solutions to interpersonal conflicts with olent video games increases aggressive be-
one’s children can help teach more positive havior and decreases prosocial behavior in
values as well as practical guides to life. This children and in young adults. Despite these
can be done in the context of violent media conclusions, there are still many unanswered
themselves, as well as in numerous every- questions and much more work is needed.
day situations as conflicts arise. The following is a list of research needs in this
Third, reducing the amount of time chil- domain.
dren spend on electronic media and substi- 1) Does explicitly gory violence desensitize
tuting increased time in social contexts and video game players more so than less gory vi-
activities is likely to improve social skills and olence? If so, does this desensitization in-
functioning over time. crease subsequent aggression? Does it de-
In summary, parents need to be on the crease helping behavior?
alert for any video game that encourages or 2) What features increase the game player’s
allows the player to harm another creature, identification with an aggressive character in
human or nonhuman. Such games are very video games? Prior research and theory in
likely teaching the game player subtle but the media violence domain suggest that the
harmful aggression lessons, regardless of how impact of exposure to violent video games is
cute the game characters are or how unrealis- likely to be greater when the game player
tic the violence appears. closely identifies with an aggressive game
character.29
Six questions parents should ask themselves re- 3) What features, if any, could be added to
garding violent video games violent video games to decrease the impact
on subsequent aggression by the game
When examining video games your child player? For instance, does the addition of
is being (or could potentially be) exposed pain responses by the game victims make
to, it is best to get a first-hand demonstra- players less reluctant to reenact the aggression
tion of the game and ask yourself 6 ques- in later real-world situations, or do such pain
tions. If you, like many parents, are not vid- responses in the game further desensitize the
eo game savvy enough to operate the game player to others’ pain?
yourself, have someone else demonstrate it 4) Can exciting video games be created
for you. Then ask yourself the following ques- that teach and reinforce nonviolent solutions
tions: to social conflicts?
— Does the game involve some charac- 5) What are the long-term effects of expo-
ters trying to harm others? sure to violent video games?
— Does this happen frequently, more than 6) What types of people are most suscep-
once or twice in 30 minutes? tible to violent video game effects, and who
— Is the harm rewarded in any way? is relatively immune?
— Is the harm portrayed as humorous? 7) As mentioned earlier, does exposure to
— Are nonviolent solutions absent or less violent television or violent video games have
“fun” than the violent ones? larger impacts on aggression related variables
— Are realistic consequences of violence and desensitization?
absent from the game? Answers to these questions are vital in

Vol. 45, N. 1 MINERVA PSICHIATRICA 15

CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

understanding the effects of video game vi- References

olence, but will require considerable effort 1. Cohen A. Time 2000, October 30;156:58-60.
by the research community and considerable 2. Elmer-Dewitt P. The amazing video game boom. Time
funding by federal and other research agen- 1993, September;66-73.
3. Hettrick S. Video games on target for $22 billion.
cies and foundations. Violence in contempo- Hollywood Reporter 1995, May 11.
rary American culture is a major social con- 4. Walsh DA. Sixth annual video and computer game re-
cern, and media violence plays an important port card [On-line]. National Institute on Media and
the Family 2001. Retrieved September 5, 2002 from:
role. To date, researchers in the United State http//www.mediafamily.org/research/vgrc/2001-2.-
have studied violence and video game ques- shtml.
5. Gentile DA, Anderson CA. Violent video games: the
tions with virtually no governmental sup- newest media violence hazard. In: Gentile DA, editor.
port. As video games continue to evolve in A Falling Anvil: The Real Effects of Media Violence on
Children. Westport, CT: Ablex [in press].
the digital age, becoming ever more realis- 6. Kent SL. The First Quarter: A 25-year History of Video
tic and violent, a commitment by federal Games. Bothell, Washington: BWD Press; 2000.
governments is needed to fund basic re- 7. Anderson CA, Dill KE. Video games and aggressive
thoughts, feelings, and behavior in the laboratory and
search that will more adequately inform pol- in life. J Pers Soc Psychol 2000;78:772-90.
icy makers about the role that these games 8. Markoff J. Recession? Don’t tell the video game indus-
try. New York Times 2002, May 24. Retrieved June 20,
play in the development of childhood ag- 2002 from http://www.nytimes.com/2002/05/24/tech-
gression. What we have learned so far has nology/24GAME.html?ex=1023256208&ei=1&en=
taken almost 20 years, in part because of b044bbd10bda69d2.
9. Sony Online. 2002. Official webpage http://ever-
the lack of research funding; let’s hope that quest.station.sony.com/
it will not take another 20 years to complete 10. Ebay Online Auctions. Retrieved March 20, 2003, from
http://www.ebay.com
the next round of studies on this vital topic. 11. Walsh DA. Interactive violence and children: Testimony
submitted to the Committee on Commerce, Science,
and Technology, United States Senate. Minneapolis,
MN: National Institute on Media and the Family, 2000,
Riassunto March 21 Available at: http://www.mediafamily.org/-
press/senateviolence-full.shtml.
12. Children Now (2001). Children and the Media.
Esposizione a videogiochi violenti e aggressività: una Retrieved July 1, 2001, from http://www.children-
review della letteratura now.org
Nel corso degli ultimi 30 anni, l’industria dei vi- 13. Dietz TL. An examination of violence and gender role
portrayals in video games: implications for gender so-
deogiochi si è trasformata in un business plurimiliar- cialization and aggressive behavior. Sex Roles
dario. Un numero sempre crescente di bambini e di 1998;38:425-42.
adulti dedica del tempo ai videogiochi per computer, 14. Dill KE, Gentile DA, Richter WA, Dill JC. Violence,
ai videogiochi per console e ai videogiochi on-line. race, sex and age in video games: A content analysis.
La violenza figura come uno dei motivi dominanti Paper presented at the annual meeting of the American
Psychological Association, San Francisco, CA 2001,
nella maggior parte dei videogiochi più popolari. Il August 25.
presente articolo offre una review aggiornata della let- 15. Buchman DD, Funk JB. Video and computer games in
teratura circa gli effetti dell’esposizione a videogiochi the ’90s: Childrens’ time commitment and game pref-
violenti sulle variabili comportamentali legate all’ag- erence. Child Today 1996;24:12-6.
gressività. L’esposizione a videogiochi violenti de- 16. Funk JB. Reevaluation of the impact of violent video
games. Clin Pediatr 1993;32:86-90.
termina un aumento dell’aggressività nella sfera com- 17. Funk JB, Flores G, Buchman DD, Germann JN. (1999).
portamentale, in quella cognitiva e in quella affettiva. Rating electronic games: violence is in the eye of the
L’esposizione a videogiochi violenti determina an- beholder. Youth Soc 1999;30:283-312.
che un incremento nella fisiologica desensibilizza- 18. Federal Trade Commission. Marketing violent enter-
tainment to children: A review of self-regulation and in-
zione agli episodi di natura violenta che accadono nel- dustry practices in the motion picture, music recording,
la vita reale e una diminuzione dei comportamenti di & electronic game industries. Report of the Federal
tipo solidale con chi è in difficoltà. La recente lette- Trace Commission, 2000. Available online at: http://
ratura scientifica riguardante i videogiochi viene in- www.ftc.gov/bcp/congress/index.html
terpretata nei termini del Modello Generale di 19. Walsh DA, Gentile DA. A validity test of movie, television,
and video-game ratings. Pediatrics 2001;107:1302-8.
Aggressione. Vengono inoltre discusse le differenze 20. Thompson KM, Haninger K. Violence in E-rated video
tra l’esposizione a videogiochi e violenti e a pro- games. JAMA 2001;286:591-8.
grammi televisivi violenti. 21. Harris MB, Williams R. Video games and school perfor-
mance. Education 1985;105:306-9.
Parole chiave: Età pediatrica - Videogioco - Violen- 22. Gentile DA, Walsh DA. A normative study of family
za - Aggressività. media habits. J Appl Develop Psychol 2002;23:157-78.

16 MINERVA PSICHIATRICA Marzo 2004

VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION CARNAGEY

23. Gentile DA, Lynch PJ, Linder JR, Walsh DA. The ef- 46. Onion CWR, Bartzokas CA. Changing attitudes to infec-
fects of violent video game habits on adolescent aggres- tion management in primary care: a controlled trial of
sive attitudes and behaviors. J Adolesc. [In press]. active versus passive guideline implementation strate-
24. CIRP, Cooperative Institutional Research Program gies. Family Practice 1998;15:99-104.
Survey Results (Ames, Ia.: Office of Institutional 47. Leyens JP, Picus S. Identification with the winner of a
Research, 1998, 1999). fight and name mediation: their differential effects
25. Bushman BJ, Anderson CA. Media violence and the upon subsequent aggressive behavior. Br J Soc Clin
American public: scientific facts versus media misinfor- Psychol 1973;12:374-7.
mation. Am Psychol 2001;56:477-89. 48. Perry DG, Perry LC. Identification with film charac-
26. Anderson CA, Bushman BJ: The effects of media vio- ters, covert aggression verbalization, and reactions to
lence on society. Science 2002;295:2377-8. film violence. J Res Pers 1976;10:399-409.
27. Hearold S. A synthesis of 1043 effects of television on 49. Turner CW, Berkowitz L. Identification with film aggres-
social behavior. In: Comstock G, editor. Public sor (covert role taking) and reactions to film violence.
Communication and Behavior. New York: Academic J Pers Soc Psychol 1972;21:256-64.
Press;1986.p.65-133. 50. Bandura A. Aggression: A social learning theory anal-
28. Huesmann LR. Aggressive behavior: current perspec- ysis. Englewood Cliffs, NJ: Prentice-Hall; 1973.
tives. New York: Plenum; 1994. 51. Bandura A. Psychological mechanism of aggression. In
29. Huesmann LR, Moise-Titus J, Podolski C, Eron LD. Geen RG, Donnerstein EI, editors. Aggression: theoret-
Longitudinal relations between children’s exposure to ical and empirical reviews. New York: Academic Press;
TV violence and their aggressive and violent behavior 1983.p.1-40.
in young adulthood: 1977-1992. Develop Psychol 52. Anderson CA, Bushman BJ. Effects of violent video
2003;39:201-21. games on aggressive behavior, aggressive cognition, ag-
30. Paik H, Comstock G. The effects of television violence gressive affect, physiological arousal, and prosocial
on antisocial behavior: a meta-analysis. Comm Res behavior: A meta-analytic review of the scientific liter-
1994;21:516-46. ature. Psychol Sci 2001;12:353-9.
31. Wood W, Wong FY, Chachere JG. Effects of media vi- 53. Anderson CA, Huesmann LR. Human Aggression: A
olence on viewers’ aggression in unconstrained social Social-Cognitive View. In: Hogg MA, Cooper J, editors.
interaction. Psychol Bull 1991;109:371-83. Handbook of Social Psychology. London: Worchel;2003.
32. Cline VB, Croft RG, Courrier S. Desensitization of children 54. Cooper J, Mackie D. Video games and aggression in
to television violence. J Pers Soc Psychol 1973;27:360-5. children. J Appl Soc Psychol 1986;16:726-44.
33. Lazarus RS, Speisman M, Mordkoff, AM, Davidson LA. 55. Irwin AR, Gross AM. Cognitive tempo, violent video
(1962). A laboratory study of psychological stress pro- games, and aggressive behavior in young boys. J Fam
duced by a motion picture film. Psychol Monogr Violence 1995;10:337-50.
1962;34 (553). 56. Lynch PJ, Gentile DA, Olson AA, Van Brederode TM.
34. Linz D, Donnerstein E, Adams SM. Physiological desen- The effects of violent video game habits on adoles-
sitization and judgments about female victims of vio- cent aggressive attitudes and behaviors. Paper pre-
lence. Hum Comm Res 1989;15:509-22. sented at the Biennial Conference of the Society for
35. Thomas MH. Physiological arousal, exposure to a rel- Research in Child Development (April, 2001),
atively lengthy aggressive film, and aggressive behav- Minneapolis, Minnesota.
ior. J Res Pers 1982;16:72-81. 57. Schutte NS, Malouff JM, Post-Gorden J, Rodasta AL.
36. Thomas MH, Horton RW, Lippincott, EC, Drabman RS. Effects of playing videogames on children’s aggres-
Desensitization to portrays of real life aggression as a sive and other behaviors. Journal of Appl Soc Psychol
function of television violence. J Pers Soc Psychol 1988;18:454-60.
1977;35:450-8. 58. Silvern SB, Williamson PA. The effects of video game
37. Linz DG, Donnerstein E, Penrod S. Effects of long- play on young children’s aggression, fantasy and pro-
term exposure to violent and sexually degrading depic- social behavior. J Appl Develop Psychol 1987;8:
tions of women. J Pers Soc Psychol 1988;55:758-68. 453-62.
38. Dexter HR, Penrod S, Linz D, Saunders D. Attributing 59. Crick NR, Dodge KA. A review and reformulation of so-
responsibility to female victims after exposure to sex- cial information processing mechanisms in children’s
ually violent films. J Appl Soc Psychol 1997;27:2149-71. adjustment. Psychol Bull 1994;115:74-101.
39. Drabman RS, Thomas MH. Does media violence in- 60. Bushman BJ, Anderson CA. Violent video games and
crease children’s toleration of real-Life aggression? hostile expectations: a test of the general aggression
Develop Psychol 1974;10:418-21. model. Pers Soc Psychol Bull. [In press].
40. Drabman RS, Thomas MH. Does watching violence 61. Kirsh SJ. Seeing the world through Mortal Kombat-col-
on television cause apathy? Pediatrics 1976;57:329-31. ored glasses: violent video games and the develop-
41. Thomas MH, Drabman RS. Toleration of real life aggres- ment of a short-term hostile attribution bias. Childhood
sion as a function of exposure to televised violence 1998;5:177-84.
and age of subject. Merrill-Palmer Q 1975;21:227-32. 62. Calvert SL, Tan S. Impact of virtual reality on young
42. Dill KE, Dill JC. Video game violence: a review of the em- adults’ physiological arousal and aggressive thoughts:
pirical literature. Aggress Violent Behav 1998;3:407-28. interaction versus observation. J Appl Develop Psychol
43. Geen RG. Human aggression. Buckingham, England: 1994;15:125-39.
Open University Press; 2001. 63. Graybill D, Kirsch JR, Esselman ED. Effects of playing
44. Atlas R, Cornett L, Lane DM, Napier HA. The use of an- violent versus nonviolent video games on the aggres-
imation in software training: Pitfalls and benefits. In: sive ideation of aggressive and nonaggressive chil-
Quinones MA, Ehrenstein A editors. Training for a rap- dren. Child Study J 1985;15:199-205.
idly changing workplace: applications of psychologi- 64. Anderson CA, Ford CM. Affect of the game player:
cal research. Washington DC: American Psychological short term effects of highlyand mildly aggressive vid-
Association; 1997. eo games. Pers Soc Psychol Bull 1986;12:390-402.
45. Berkowitz L. Some determinants of impulsive aggres- 65. Ballard ME, Lineberger R. Video game violence and
sion: role of mediated associations with reinforcements confederate gender: effects on reward and punish-
for aggression. Psychol Rev 1974;81:165-76. ment given by college males. Sex Roles 1999;41:541-58.

Vol. 45, N. 1 MINERVA PSICHIATRICA 17

 CARNAGEY VIOLENT VIDEO GAME EXPOSURE AND AGGRESSION

66. Chambers JH, Ascione FR. The effects of prosocial and 90. Berkowitz L. On the formation and regulation of an-
aggressive video games on children’s donating and ger and aggression. Am Psychol 1990;45:494-503.
helping. J Genet Psychol 1987;148:499-505. 91. Berkowitz L. Pain and aggression: some findings and
67. Wiegman O, van Schie EGM. Video game playing and implications. Motiv Emot 1993;17:277-93.
its relations with aggressive and prosocial behavior. 92. Dodge KA, Crick NR. Social information-processing
Br J Soc Psychol 1998;37:367-8. bases of aggressive behavior in children. Pers Soc
68. Carnagey NL, Bushman BJ, Anderson CA. Video game Psychol Bull 1990;16:8-22.
violence desensitizes players to real world violence. [In 93. Geen RG. Human Aggression. Pacific Grove, CA:
press]. McGraw Hill; 1990.
69. Murphy JK, Alpert BS, Walker SS. Whether to measure 94. Huesmann LR. Psychological processes promoting
change from baseline or absolute level in studies of the relation between exposure to media violence and
children’s cardiovascular reactivity: A two-year follow- aggressive behavior by the viewer. J Soc Issues
up. J Behav Med 1991;14:409-19. 1986;42:125-39.
70. Ballard ME, Weist JR. Mortal Kombat: the effects of vi- 95. Zillmann D. Arousal and aggression. In: Geen R,
olent video game play on males’ hostility and cardiovas- Donnerstein E, editors. Aggression: Theoretical and
cular responding. J Appl Soc Psychol 1996;26:717-30. empirical reviews. New York: Academic Press;
71. Koepp MJ, Gunn RN, Lawrence AD, Cunningham VJ, 1983.p.75-102.
Dagher A, Jones T et al. Evidence for striatal dopamine 96. Bargh JA, Lombardi WJ, Higgins ET. Automaticity of
release during a video game. Nature 1998;393:266-8. chronically accessible constructs in person X situa-
72. Lynch PJ. Type A behavior, hostility, and cardiovas- tion effects on person perception: it’s just a matter of
cular function at rest and after playing video games in time. J Pers Soc Psychol 1988;55:599-605.
teenagers. Psychosom Med 1994;56:152. 97. Sedikides C, Skowronski JJ. Towards reconciling per-
73. Lynch PJ. Hostility, Type A behavior, and stress hor- sonality and social psychology: a construct access-
mones at rest and after playing violent video games in ibility approach. J Soc Behav Pers 1990;5:531-46.
teenagers. Psychosom Med 1999;61:113. 98. Anderson KB, Anderson CA, Dill KE, Deuser WE. The
74. Anderson CA. Violent video games and aggressive interactive relations between trait hostility, pain, and
thoughts, feelings, and behaviors. In: Calvert SL, Jordan aggressive thoughts. Aggress Behav 1998;24:161-71.
AB, Cocking RR, editors. Children in the Digital Age. 99. Anderson CA. Effects of violent movies and trait irri-
Westport, CT: Praeger Publishers; 2002.p.101-19. tability on hostile feelings and aggressive thoughts.
75. Anderson CA, Bushman BJ. Human aggression. Ann Aggress Behav 1997;23:161-78.
Rev Psychol 2002;53:27-51. 100. Bushman BJ. Moderating role of trait aggressiveness
76. Anderson CA, Lindsay JJ, Bushman, BJ. Research in in the effects of violent media on aggression. J Pers Soc
the psychological laboratory: Truth or triviality? Curr Dir Psychol 1995;69:950-60.
Psychol Sci 1999;8:3-9. 101. Bushman BJ, Geen RG. Role of cognitive-emotional
77. Banaji MR, Crowder RG. The bankruptcy of everyday mediators and individual differences in the effects of
memory. Am Psychol 1989;44:1185-93. media violence on aggression. J Pers Soc Psychol
78. Kruglanski AW. The human subject in the psychology 1990;58:156-63.
experiment: fact and artifact. In Berkowitz, editor. 102. Bushman BJ, Huesmann LR. Effects of televised vio-
Advances in experimental social psychology. New lence on aggression. In: Singer D, Singer J, editors.
York: Academic Press; 1975.p.101-47. Handbook of children and the media. Thousand Oaks,
79. Mook DG. In defense of external invalidity. Am Psychol CA: Sage Publications; 2000.p.223-54.
1983;38:379-87. 103. Hansen CH, Hansen RD. The influence of sex and
80. Anderson CA, Bushman BJ. External validity of “trivi- violence on the appeal of rock music videos. Comm
al” experiments: The case of laboratory aggression. Res 1990;17:212-34.
Rev Gen Psychol 1997;1:19-41. 104. Geen RG, O’Neal EC. Activation of cue-elicited aggres-
81. Berkowitz L, Donnerstein E. External validity is more sion by general arousal. J Pers Soc Psychol 1969;
than skin deep: some answers to criticism of laborato- 11:289-92.
ry experiments. Am Psychol 1982;37:245-57. 105. James W. Principles of psychology. New York: Holt;
82. Carlson M, Marcus-Newhall A, Miller, N. Evidence for 1890.
a general construct of aggression. Pers Soc Psychol 106. Schachter S, Singer J. Cognitive, social, and physiolog-
Bull 1989;15:377-89. ical determinants of emotional state. Psychol Rev
83. Giancola PR, Chermack ST. Construct validity of labor- 1962;69:379-99.
atory aggression paradigms: a response to Tedeschi and 107. Bower G. Mood and memory. Am Psychol 1981;
Quigley (1996). Aggress Violent Behav 1998;3:237-53. 36:129-48.
84. Bandura A. Psychotherapy based upon modeling princi- 108. Forgas JP. Affect in social judgments and decisions: a
ples. In: Bergin, AE Garfield SL, editors. Handbook of psy- multiprocess model. Adv Exp Soc Psychol 1992;25:
chotherapy and behavior change. New York: Wiley; 1971. 227-75.
85. Bandura A, Ross D, Ross SA. Transmission of aggres- 109. Schwarz N, Clore GL. Feelings and phenomenal ex-
sion through imitation of aggressive models. J Abnorm periences. In: Higgins E, Kruglanski editors. Social
Soc Psychol 1961;63:575-82. psychology: Handbook of basic principles. New York:
86. Bandura A, Ross D, Ross SA. Imitation of film-mediated Guilford; 1996.p.433-65.
aggressive models. J Abnorm Soc Psychol 1963;66:3-11. 110. Huesmann LR, Miller LS. Long-term effects of repeat-
87. Mischel W. Toward a cognitive social learning reconcep- ed exposure to media violence in childhood. In:
tualization of personality. Psychol Rev 1973;80:252-83. Huesmann R, editors. Aggressive behavior: current
88. Mischel W, Shoda Y. A cognitive-affective system the- perspectives. New York: Plenum Press; 1994.p.153-86.
ory of personality: Reconceptualizing situations, dispo- 111. Lewin K. Problems of research in social psychology.
sitions, dynamics, and invariance in personality struc- In: Cartwright D, editor. Field theory in social sci-
ture. Psychol Rev 1995;102:246-68. ence. New York: Harper and Row; 1951.p.169.
89. Berkowitz L. Some effects of thoughts on anti- and 112. Bushman BJ, Stack AD. Forbidden fruit versus tainted
prosocial influence of media events: cognitive neoas- fruit: effects of warning labels on attraction to televi-
sociationist analysis. Psychol Bull 1984;95:410-27. sion violence. J Exp Psychol Appl 1996;2:207-26.

18 MINERVA PSICHIATRICA Marzo 2004

View publication stats