Discussion of the harmful effects of media violence is as old as the media themselves. There is no medium that has not been suspected of stimulating real-world aggression. Spectacular violent acts such as those in Littleton, Colorado (where in 1999 two teenagers murdered 13 people before committing suicide), or Erfurt, Germany (where in 2002 a 19 year old killed 16 people and himself), regularly instigate public interest on this topic. Although there is a huge body of research on the effects of media violence, public discussion rarely reflects scientific knowledge. This may be due to the fact that the results are complex, sometimes contradictory, and do not allow for simple conclusions. However, on the basis of this research some widely accepted assertions about the effects of violence in the media are possible.
Research has mainly focused on the effects of fictional television violence. Recently, however, there has been much interest in violent content in music and music videos (e.g., Smith & Boyson 2002; Anderson et al. 2003), the Internet (e.g., Slater 2003), and especially in computer games (e.g., Carnagey & Anderson 2004; Sherry 2007). Typically, research examines the media depictions of personal violence (i.e., intended physical and/ or psychic damaging of a person, living beings, and inanimate objects by another person).
Theories Of Pro- And Anti-Social Effects
Research on the effects of mediated violence has been conducted within several theoretical frameworks. The roots of catharsis theory, which was very popular for a long time, can be tracked back to Aristotle, for whom catharsis meant “the homeopathic cleansing of affects” through compassion and fear aroused by tragic theatre. According to modern catharsis theory (Feshbach 1961), the viewing of media violence would lead to an engagement in fantasy aggression that permits the purging of one’s feelings and the discharge of aggressive tendencies such that the viewer is less inclined to behave aggressively after viewing violent media content. Catharsis theory could not be confirmed by research and has been abandoned by the scientific community, although it is still very popular in non-scientific discussions.
An alternative explanation for sporadic findings of aggression reduction is inhibition theory, which assumes a deterrent effect of media violence that causes fear and inhibits aggressive behavior. Grimm (1999) assumes a process of “negative learning” by which the overt depiction of horrible consequences for the victim leads to a critical reflection of the violent perpetrator’s behavior – unless empathy with the victim is so intense that it triggers an aggressive impulse of revenge (the “Robespierre effect”).
The simple assumption that media violence is imitated directly through a suggestion process has been clearly refuted. However, there may be special conditions that allow imitation of violent acts. Contagious effects of media violence have been studied concerning homicide, running amok, and suicide. The imitation of suicides is also labeled the “Werther effect” because Goethe’s The Sorrows of Young Werther (published in 1774) allegedly led so many people to commit suicide that in some countries the book was banned. Phillips (1974) showed that after intense coverage of suicides (e.g., Marilyn Monroe’s) the suicide rate rose in the USA and in Great Britain. In Germany, statistical proof for imitations of suicide was provided by a study on a television film in the early 1980s in which a 19-yearold high school student committed suicide by throwing himself in front of a train (Schmidtke & Häfner 1988). Imitations of this type of suicide could most clearly be proved for people most similar to the model, male youths aged 15 to 19 in this case. However, media content seems to be only one of many more important causes or the final trigger for an action planned long before. In any case, the details of the media content and the attributes of the recipient must be considered.
Although the simple suggestion mechanism has been refuted by research, public discussion on the contagious effects of violent media depiction is ongoing, and this discussion on the risks of media violence holds a risk in itself: Aggressors often tend to justify their behavior by referring to role models on television. This rationalization of one’s own behavior can happen both before an offense (thus facilitating violent behavior) and after it. Denying responsibility by blaming it on the media is so popular because the delinquents can present and/or see themselves as led by external forces. The habituation theory emphasizes cumulative, long-term effects. It assumes that media violence causes emotional blunting and desensitization. As some followers of the habituation theory maintain, this not only reduces emotional and physiological reactions to violent media content and increases the need for stronger stimuli, but also affects attitudes toward violence in real life. It reduces empathy with the victims of violence, increases tolerance for violent behavior in others, and reduces the inhibition threshold for one’s own aggressive behavior. Although some studies seem to back it, habituation is not yet scientifically proven.
Another approach that deals with long-term effects of media violence is cultivation theory. It assumes that heavy television viewers suffer from a distorted view of social reality. Viewing violence thus may cultivate fear of crime and the belief that the world is a mean and scary place. Research is currently concentrating on intervening variables in that process such as the experience of victimization, exploring the effects of special media content (e.g., crime movies), and seeking to explain information processing leading to cultivation in more detail. Additionally it has to be clarified to what extent fear is not the result but the cause of heavy television viewing (escapism).
According to excitation transfer theory different types of media content (violence, but also eroticism, humor, sports, etc.) cause a state of unspecific arousal that intensifies subsequent actions. The behavior being intensified depends on the particular situation and does not relate to the content of the media stimulus. For example, erotic media stimuli may intensify aggression while media violence may foster caring behavior.
The state of arousal and the current situation also are important factors in stimulation theory, which is based on frustration aggression theory. According to Berkowitz (1970), certain types of media violence (e.g., violence that is presented as justified) enhance violent behavior under circumstances in which the viewer is emotionally aroused (e.g., due to frustration) and meets situational cues that are associated with the actual feeling of anger or with past experiences, or that generally instigate aggression (e.g., weapons). A state of emotional arousal brought about by frustration leads to an aggressive disposition and can result in violent behavior if watching media violence bears a resemblance to the real situation.
Later, Berkowitz (1984) focused on the possibility that media violence could prime thoughts of aggression, making actual violent behavior more likely. The concept of priming is based on “cognitive neo-assocation theory” and means that in brain cognition, emotions and behavioral tendencies are connected via associative pathways or neuronal networks. If a stimulus (e.g., media violence) activates a node inside this network (priming), activation spreads to related nodes, activating associatively connected thoughts, emotions, and behavior. This process happens automatically and unconsciously, and it influences the interpretation of new stimuli and makes aggressive behavior more probable in the near future. However, priming theory also assumes that through repeated stimulation, activation may become chronic, resulting in long-term effects. Scientific results support priming theory but priming processes still have to be described more precisely.
Priming is closely connected to script theory. Scripts are mental routines that are stored in memory and applied to guide behavior and to solve problems. Scripts contain information about the typical course of an event (e.g., consultation of a doctor, dealing with provocation), people’s behavior, and the outcome of actions. It is assumed that children who watch much media violence will develop scripts concentrating on violent solutions to a problem. Situational cues will activate the matching scripts. Scripts that are frequently repeated can be accessed more easily. Media content can help to develop new scripts and to activate existing ones.
Short- And Long-Term Effects
Bandura’s theory of social learning seems to be the most appropriate approach to explain the quite heterogeneous results of medium-to-long-term studies on media violence. It postulates that people adopt patterns of behavior by observing other people’s actions (in reality or in the media). However, these patterns do not necessarily have to be acted out. Normally, they remain latent. Violent actions usually underlie inhibiting conditions (e.g., social norms, fear of punishment or revenge, sense of guilt). Violent patterns of behavior only transfer into manifest action under adequate conditions. This may happen if the person encounters a situation that resembles one observed before and if he or she has the means necessary for imitation (e.g., special weapons).
The most important factor, however, is the consequence of a certain behavior (success or failure, reward or punishment) experienced by the role model and/or experienced or expected by the observers themselves. Factors considered by the theory of social learning comprise the attributes of media contents (e.g., importance, explicitness, comprehensibility, efficiency, justification, rewards for violence) and of the observer (e.g., character, cognitive abilities, state of arousal, interests, former experiences with a certain behavior) and social conditions (e.g., socialization, norms and values in family and peer group). Social learning theory takes into account that different observers perceive the same contents differently and derive different consequences for their own actions. Against this background, the finding that children who possess no preference for violent media contents will not show any tendency toward violent behavior, even after long contact with it, does not contradict the theory.
From the perspective of social learning theory, the general patterns of television violence facilitate learning and implementation of violent patterns of behavior. As a rule, the plot is suspenseful and exciting, assuring the necessary attention. Violent acts normally are simple (i.e., easy to perceive and just as easy to imitate). Since the “good” heroes also massively use violence to achieve their aims (e.g., justice), they are suited for identification. Furthermore, violence almost always pays off. Hence, observer can conclude that he or she only needs to be careful enough to avoid the punishment most of the “baddies” suffer at the end of a plot, although up to then they have successfully used violence in many sequences. Overall, violence is shown as a normal, everyday behavior. Although the acts of violence as such are highly unrealistic (hardly any wounds, injuries, etc.), the models themselves are depicted realistically.
New results indicate that in judging the harmful effects of media violence, it is important not only to consider the attributes of violent role models. The results rather suggest that the recipient often identifies with the victim of media violence, resulting in a deterent effect on violent actions (Grimm 1999).
An Integrative Model
The latest theoretical development in media violence research is the general aggression model (Carnagey & Anderson 2004), which tries to integrate different concepts such as learning theory, priming, script theory, and excitation transfer theory. The model suggests that behavior is the result of two input variables: the person (including character, attitudes, values, scripts, current state) and the situation (environmental factors like aggressive cues, rewards, provocation, etc.).
These input variables influence the appraisal of a situation and the subsequent impulsive or thoughtful choice of behavior by influencing the current internal state of a person. This happens through three main routes of impact that may interact: cognition (aggressive constructs become more accessible), affect (evoking of hostility), and increase of arousal. The person’s behavior results in reactions in the environment that may influence the input variables and lead to reinforcement or inhibition of the chosen behavior in the future.
Based on this cyclical process, the general aggression model also posits long-term effects, because repeated exposure to violent stimuli helps to develop easily accessible aggression-related knowledge structures that may be reinforced by successful application and that become increasingly complex, automatized, and resistant to change. Together with desensitization effects, this may lead to the development of an aggressive personality.
Effects Strength And Future Research
Most researchers agree that media violence may cause negative effects. However, these effects have to be considered in a differentiated way. Correlations found in empirical studies on television violence are usually quite small (between r = 0.1 and r = 0.31, which means that at most 9 percent of aggression is explained by media violence; Comstock & Scharrer 2003). Although it is assumed that violence in computer games should be far more dangerous (because of the player’s active role, the current repetition of violent acts that are rewarded immediately, etc.), and research indicates negative cognitive, emotional, and behavioral effects, particularly strong effects of computer game violence have not been found up to now.
These small effects point to the fact that media violence is only one factor within a complex network of causes for real-world aggression. However, a small correlation between media violence and violent behavior that holds true for the average of recipients does not mean that strong effects for particular forms of media contents and for particular recipients cannot be found.
Future research should concentrate on the conditions for negative effects to emerge. In doing so, three main factors of influence must be considered (Kunczik & Zipfel 2006): quality of media content, attributes of the recipient, and his or her social environment. According to the present state of knowledge, negative effects of media violence are most likely to occur to young, male, socially deprived heavy viewers who already possess a violent personality, grow up in violent families with high media (violence) usage, experience much violence from their parents and in school, and belong to aggressive and/ or delinquent peer groups. The last point means that they are exposed to a “double dose” of violent role models.
Violent media contents and their own violent experiences interact and reinforce each other; violent media models are much more interesting and “useful” for children growing up in a violent environment, and additionally, the violent behavior of media models seems to be quite normal to them. These recipients are even more at risk if the media content they perceive presents violence in a realistic and/or humorous context, if violent behavior seems justified and is committed by attractive, successful protagonists with whom the recipient can identify (e.g., also because of the resemblance between them), and especially if violence is not punished and does not harm the victim in a visible way. These factors are not independent of each other, but the interaction patterns are not yet described in detail. This remains an important scientific task to be fulfilled.
- Anderson, C. A., Carnagey, N. L., & Eubanks, J. (2003). Exposure to violent media: The effects of songs with violent lyrics on aggressive thoughts and feelings. Journal of Personality and Social Psychology, 84, 960–971.
- Berkowitz, L. (1970). The contagion of violence. Nebraska Symposium on Motivation, 18, 95– 135.
- Berkowitz, L. (1984). Some effects of thoughts on antisocial and prosocial influences of media effects: A cognitive-neoassociation analysis. Psychological Bulletin, 95, 410–427.
- Carnagey, N. L., & Anderson, C. A. (2004). Violent video game exposure and aggression. A literature review. Minerva Psichiatrica, 45, 1–18.
- Comstock, G., & Scharrer, E. (2003). Meta-analyzing the controversy over television violence and aggression. In D. A. Gentile (ed.), Media violence and children: A complete guide for parents and professionals. Westport, CT, and London: Praeger, pp. 205–226.
- Feshbach, S. (1961). The stimulating versus cathartic effects of a vicarious aggressive activity. Journal of Abnormal and Social Psychology, 63, 381–385.
- Grimm, J. (1999). Fernsehgewalt: Zuwendungsattraktivität, Erregungsverläufe, sozialer Effekt. Zur Begründung und praktischen Anwendung eines kognitiv-physiologischen Ansatzes der Medienrezeptionsforschung am Beispiel von Gewaltdarstellungen [Television violence]. Opladen and Wiesbaden: Westdeutscher.
- Kunczik, M., & Zipfel, A. (2006). Medien und Gewalt [Mass media and violence], 5th edn. Cologne, Vienna, and Weimar: Böhlau.
- Phillips, D. P. (1974). The influence of suggestion on suicide: Substantive and theoretical implications of the Werther effect. American Sociological Review, 39, 340–354.
- Schmidtke, A., & Häfner, H. (1988). The Werther Effect after television films: New evidence for an old hypothesis. Psychological Medicine, 18, 665–676.
- Sherry, J. L. (2007). Violent video games and aggression: Why can’t we find effects. In R. W. Preiss, B. M. Gayle, N. Burrell, M. Allen, & J. Bryant (eds.), Mass media effects research: Advances through meta-analysis. Mahwah, NJ, and London: Lawrence Erlbaum, pp. 245–262.
- Slater, M. D. (2003). Alienation, aggression, and sensation seeking as predictors of adolescent use of violent film, computer, and website content. Journal of Communication, 53, 105–121.
- Smith, S. L., & Boyson, A. R. (2002). Violence in music videos: Examining the prevalence and context of physical aggression. Journal of Communication, 52, 61–83.