In contemporary society, a wide variety of violent content is reaching children through a variety of media including television, movies, and video games. Often, exposure to violence occurs with little adult or parental supervision. Several content analyses have examined the amount and content of violence on television. These analyses have shown that as of the late 1990s nearly two-thirds of the programs on television contained some amount of violence. There are no comparable figures for movies or video games because there is no central source or collection of programming; however, a variety of content analyses investigations examining the most popular video games, for example, appear to point to the same findings—violence is prevalent in these formats.
Research on the effects of exposure to media violence has included examinations of the effects of violence in films, television, video games, and music videos on aggressive behavior, thoughts, attitudes, and emotions following exposure. The research has consistently revealed a substantial, statistically significant association between exposure to violence in the media and violent behavior measured in the laboratory, in the field, and across substantial time spans. A set of well-articulated theories explain why aggression generally increases after exposure to violence in the media. Additional research on viewer characteristics has refined notions of who is most likely to be influenced by media violence and under what circumstances they are likely to be affected. More recent research has focused on the effects of interactive media such as video games. This research demonstrates effects that are equivalent for this form of media to older media use.
Effects of Television and Movie Violence on Aggressive Behavior
Most research on media violence and viewer aggression has focused on viewers who are passively exposed to movies and television portrayals. A relatively large number of laboratory experiments in which subjects are randomly assigned to view film or television violence and compared with control groups have been conducted over the past 50 years. Several types of aggression toward others have been assessed in these studies, including verbal and physical aggression. These experiments have consistently found that young people who watched violent scenes subsequently displayed more aggressive thoughts, emotions, and behaviors than those who did not. Usually, these laboratory studies measure the immediate impact of violence exposure on aggression. Results from these studies have shown that, for example, children who watched television violence were more likely to be rated as high on physical assault (hurting other children, wrestling, as well as other types of aggression) by observers who did not know which type of film the children had seen. Field experiments in which boys at a summer camp had been assigned to view violent or nonviolent films and then observed revealed that boys who had been assigned to the violent film conditions engaged in significantly more physical assaults on fellow campers. This effect was particularly pronounced for boys who were individually higher on trait aggression. Other research has demonstrated that combining violent stimuli with other arousing activities or portrayals can enhance the aggression effect following exposure. For example, college students who have been provoked by others or who have seen sexually arousing films that portray sex and violence exhibit pronounced increases in retaliatory behavior as indexed by their willingness to deliver what they believe are electric shocks to other subjects.
Randomized laboratory experiments have also demonstrated desensitization effects, whereby children subjects exposed for prolonged periods to media violence were slower to call an adult to intervene when they saw two younger children fighting. Adults exhibit similar desensitization effects including acceptance of physical aggression toward females and hostile behavior after prolonged exposure to violent movies compared with adults in control conditions.
Meta-analyses that have computed the overall effect sizes for randomized experiments have generally concluded that the size of the effect for the media violence aggressive behavior effect is moderate to large ranging from r = .3 to r = .4 for aggression; effect sizes for criminal violence are smaller.
Survey research in which cross sections of elementary school children and adolescents have been surveyed regarding their exposure to violence in the media and measured on various indexes of general aggression have yielded results similar to those in laboratory and field experiments. These surveys show that children and adolescents who report violence viewing also exhibit higher levels of aggressive behavior. Longitudinal studies designed to study the effects of television violence on behavior over time and thus are able to measure exposure to violence in television before aggressive behavior is assessed provide evidence of a media violence aggression causal link over time. Studies that have measured assault or physical fights resulting in injury have found that exposure to violence in television at age 14 significantly predicted assault and fighting at later ages including at 22 and 30 years.
Less well-known are the effects of exposure to news violence, such as news of executions or assassinations, on violent behavior. Likewise, a few experimental studies of music videos and behavior have been conducted, but the research is rather sparse. Several studies of music videos have shown that adolescents assigned to view violent rap music videos increased endorsement of violent behavior.
Studies that have examined the introduction of television into communities that have not had it have also been undertaken. This work has tended to reinforce findings from laboratory experiments and surveys. One study, for example, found an increase in children’s level of aggression in a Canadian community after the introduction of television. However, caution must be taken in the interpretation of these studies, which often measures total television viewing and not the amount of violent programming to which children have been exposed.
Violent Video Games and Aggression
Randomized experiments involving violent video games have also been conducted. Children spend much time with these games and the process of playing them involves repetition and deep involvement. These characteristics should theoretically increase the influence of violent video games on aggressive behavior. Laboratory studies have demonstrated that college students who played violent video games were more likely to deliver high-intensity punishments than those who played a nonviolent video game. There are fewer cross-sectional surveys of video game use and aggressive behavior and little information that would allow for strong longitudinal conclusions to be drawn. Meta-analyses of violent video game effects have revealed that for studies that have the soundest methodological designs, the effect size for exposure to violent video games on aggressive behavior, aggressive attitudes, and decreases in prosocial or helping behavior is comparable with that for televised and movie violence.
Researchers in psychology, communications, and sociology have developed theoretical models that account well for the relationship between media violence and aggression. These theories are best described as social cognitive in nature and focus on how people learn, think, and behave in their social world—a world that contains interactions with humans such as parents and peers and a virtual world created by the media. Psychologists have generally distinguished between theoretical mechanisms that create short-term effects and those responsible for longer-term outcomes. Short-term effects are thought to be due to cognitive priming, temporary imitation, arousal, and excitation.
Priming explanations rely on the concept of an associative neural network in which ideas are activated (primed) by stimuli in the social environment. Exposure to violent scenes may activate related thoughts, feelings, and scripts involving aggression. These aggressive thoughts, once activated, become an interpretational filter so that ambiguous events are more likely to be interpreted as aggressive and thus stimulate aggressive behavioral tendencies.
Arousal explanations focus on the fact that violent media are arousing and exciting for children and adolescents. The residual excitement left from media violence viewing may serve to fuel dominant response tendencies after exposure to violence. Placed in a situation whereby an aggressive response is possible, the aroused individual may be more likely to be aggressive. Video games may be especially likely to provoke this form of arousal and the nature of video game playing, involving repeated and long-term use, may facilitate aggressive responding.
Both short-term and long-term effects are also thought to be the result of observational learning. Learning aggression from media portrayals of violence is facilitated by several factors. Violent models performing behavior that is similar to or attractive to the viewer are likely to increase aggression in viewers. Aggressive behavior following exposure to media violence is also more likely when there is high viewer identification with the model, the context in which the violence is presented is realistic, and the violent behavior portrayed in the media is followed by rewarding rather than punishing consequences. For the effect to become a long-term outcome, the social environment must reinforce the behaviors learned in the media. Furthermore, learning need not be limited to the specifics of the violent media portrayals. General scripts for behavior and social interaction that later guide perceptions and attitude formation may be learned from violent media.
Priming effects are usually thought of as short-term effects, but social cognitive research has shown that such effects can have lasting influences. Frequently primed aggressive thoughts and emotions may become chronically accessible in a media-violence-saturated environment. The impact of this chronic accessibility of violent thoughts and scripts for action may be that neutral social interactions are interpreted in an aggression-biased way.
Long-term exposure to violence in the media may also result in an emotional desensitization effect that appears to operate much like the habituation that occurs through therapeutic processes such as systematic desensitization—a procedure successful in treating phobias. Exposure to violence in the media appears to reduce the anxiety or fear associated with violence and causes viewers to be less physiologically aroused by violence later presented in real-life situations. The relationship between desensitization to violence and aggressive behavior is unknown.
Not all media presentations of violence have the same effect. Several stable individual differences have been identified by researchers who moderate the impact of violent media on aggressive behavior. Viewer age appears to make a difference, at least under some circumstances, but the relationship between age and effects of media violence is not resolved at this time. Gender of viewer also appears to interact with exposure to violence in the media in complicated ways. Initial research found greater effects for boys than for girls; however, more recent research did not confirm this difference. It is more likely that males and females display differing aggression patterns in general, with boys exhibiting greater tendencies toward direct physical aggression and girls displaying tendencies toward indirect forms of aggression. Early exposure to violence in the media may increase indirect aggression tendencies in females (telling lies, taking other people’s things out of anger) but not in males. Aggressiveness of the viewer also appears to interact with media exposure. Children who are especially aggressive may be both likely to seek out violent media and later, when exposed to it, more likely to learn aggressive scripts or be cognitively stimulated by violent depictions. Viewer intelligence appears not to be related to media violence effects. However, children with certain perceptual tendencies such as greater identification with violent actors and the perception that the media violence is realistic appear to display aggressive tendencies well after media exposure.
Some media portrayals carry more risk than others for increasing aggressive behavior in viewers. There is evidence that viewers are influenced by aggressive characters who appear to be similar to themselves. Other research has demonstrated that violent perpetrators who are charismatic or generally attractive are more likely to be imitated. The consequences portrayed for the violent behavior in the media may also be important for predicting imitation effects. Findings from experiments that manipulated whether violence was justified increased the likelihood that angered subjects would later exhibit increased aggression. Media violence perpetrators who are rewarded are also more likely to be imitated. On the other hand, showing negative consequences for violent behavior portrayed in the media appears to reduce later aggression. Media portrayals that are especially bloody or gory while increasing desensitization to violence in viewers do not necessarily appear to reduce violence as a result of the portrayal of negative consequences to the victim.
Demographic Variables and Media Violence
Children from families of lower socioeconomic status (SES) watch more television and thus are exposed to more media violence than others. There is not much evidence that low SES itself is causally related to increases in aggressive behavior following media exposure. Parental involvement may be important in moderating the effects of media violence. Children of parents who discuss the appropriateness of aggressive behavior following exposure to violence in the media show fewer aggressive tendencies. However, parental tendencies such as aggressiveness and coldness, other parental personality variables, and parental television viewing habits appear to be unrelated to children’s aggressive tendencies following exposure to violence in the media.
The effects of media violence on aggressive behavior are sufficiently robust for media violence to have been considered a significant public health problem. Because of the large number of children and youths exposed to media violence, the overall effect of the media on behavior may be quite significant. A correlation of .2 between viewing media violence and aggressive behavior may translate into millions of additional aggressive acts, many of them lethal, across the nation. Furthermore, there are few other variables in the violence prediction area that account for substantially more variability. The size of the media violence effect is equal to or even larger than many public health effects we as a society deem large, such as the effects of condom use on HIV transmission or the effects of passive smoking on lung cancer. Public health policy has taken two directions: the development of antiviolence interventions and the creation of media industry policies that are designed either to warn parents about violent media content or permit parents to limit violence viewing.
Few studies have been conducted on effective media violence intervention techniques. Interventions that appear to be the most promising are those that stress children and adolescents engaging in active antiviolence message construction and being able to observe themselves and others in social situations that advocate antiviolence problem solving. Strategies that emphasize parent-child coviewing of television and movies may also be effective in reducing media violence/aggression effects.
An extensive ratings system that includes warnings about both violence and sex has been developed by the television industry in response to political pressure partly generated by the large and consistent body of data developed by researchers. The introduction of the “V-chip,” a device in every new television now sold in the United States, permits parents to bypass programs whose ratings indicate violent content. However, overwhelmingly, most parents in most households report not using the device.
Despite overwhelming scientific evidence of a link between media violence and aggressive behavior, criminal and civil legal actions against the producers of violent content have been extremely limited by the “incitement” standard articulated in Brandenburg v. Ohio (1969). In this case, the U.S. Supreme Court stated that the government has a right to regulate any expression that is “directed to inciting or producing imminent lawless action and is likely to incite or produce such action” (p. 1829). Thus, for violent media content to pass the incitement test, it would need to be shown that viewing the depiction (e.g., via television, movies, or playing violent video games) is likely to lead to imminent lawless action. Furthermore, the Supreme Court said in Brandenburg that the expression falling into this category must be specifically intended to bring about the lawlessness.
The most prominent legal case involving the idea that media violence incites violence involved Florida resident Ronny Zamora, whose lawyer unsuccessfully argued in 1977 that “television intoxication” led him to murder an elderly neighbor at the age of 15. Mr. Zamora’s lawyer tried to portray him as a youngster driven criminally insane by years of watching violent television.
Since the tragedy at Columbine in 1999, there has been a great deal of interest among state legislatures in regulating the access minors have to violent video games. Illinois, Washington, and Michigan have passed limits on what types of games minors can rent or buy. This legislative interest is partly the result of a belief that social scientists have convincingly demonstrated that exposure of minors to such games produces effects such as increased aggressive attitudes and emotions and aggressive behavior. Although still in the early stages of litigation, nearly all federal courts have blocked or struck down these state and local laws that would ban the sale of violent video games to minors, and no court has upheld such statutes. The courts have also questioned whether there is evidence that violent video games cause aggressive behavior and thus need to be regulated by the government.
- Anderson, C. A., Berkowitz, L., Donnerstein, E., Huesmann, L. R., Johnson, J. D., Linz, D., et al. (2003). The influence of media violence on youth. Psychological Science in the Public Interest, 4(3), 81-110.
- Anderson, C. A., & Bushman, B. J. (2001). Effects of violent video games on aggressive behavior, aggressive cognition, aggressive affect, physiological arousal, and prosocial behavior: A meta-analytic review of the scientific literature. Psychological Science, 12, 353-359.
- Bandura, A. (1994). Social cognitive theory of mass communication. In J. Bryant & D. Zillmann (Eds.), Media effects: Advances in theory and research (pp. 61-90). Hillsdale, NJ: Lawrence Erlbaum.
- Berkowitz, L. (1984). Some effects of thoughts on anti- and prosocial influences of media events: A cognitive-neoassociation analysis. Psychological Bulletin, 95, 410-127.
- Brandenburg v. Ohio, 395 U.S. 444 (1969).
- Huesmann, L. R., Lagerspetz, K., & Eron, L. D. (1984). Intervening variables in the TV violence-aggression relation: Evidence from two countries. Developmental Psychology, 20, 746-775.
Return to Criminal Behavior overview.