Myth 4. Violent media affect everyone in the same way

Many people assume that, to be considered valid, media violence effects must be unidimensional—that is, that everyone must be affected by becoming more aggressive and violent. While that is one of the documented effects, it is not the only one. Meta-analyses (studies that analyze data presented across large numbers of studies) have shown that there are at least four main effects of watching a lot of violent media. These effects have been called the aggressor effect, the victim effect, the bystander effect, and the appetite effect (Donnerstein, Slaby, & Eron, 1994). The aggressor effect describes how children and adults who watch a lot of

violent entertainment tend to become meaner, more aggressive, and more violent.

The victim effect describes how children and adults who watch a lot of violent entertainment tend to see the world as a scarier place, become more scared, and initiate more self-protective behaviors (including going so far as to carry guns to school, which, ironically, increases one’s odds of being shot). The bystander effect describes how children and adults who watch a lot of violent entertainment tend to habituate to gradually increasing amounts of violence, thereby becoming desensitized, more callous, and less sympathetic to victims of violence (both in the media and in real life). The appetite effect describes how children and adults who watch a lot of violent entertainment tend to want to see more violent entertainment. Simply put, the more one watches, the more one wants to watch. These effects are well-documented in hundreds of studies. What is less well known is which people are more prone to which effects (although these effects are not mutually exclusive). In general, females tend to be more affected by the victim effect, whereas males tend to be more affected by the aggressor, bystander, and appetite effects. But it is still unclear how to predict exactly how any given individual will be affected by any given media violence presentation. However, the fact that we cannot yet make this prediction reliably should not be taken as evidence that there is no effect. Furthermore, that everyone is not affected in the same way does not mean that everyone is not affected. To understand where children learn their attitudes, values, and patterns of behavior, we can consider the effects of various proximal and distal sources of influence. The family is closest to children, and children clearly have their attitudes, values, and behavior patterns shaped and modified by their families. The behaviors defined as “normal” within each family affect the behaviors of the individuals within that family. Beyond the level of the family, the norms of the community affect the norms of families and individuals within it. Beyond the level of the community, the norms of society affect the norms of communities, families, and individuals within it. The media operate at this societal level, and media effects can be seen at all levels. Thus, the media can affect us not only one-on-one, when we are watching TV, for example, but they also affect us by affecting the norms, expectations, and patterns of behavior of our families and communities. This is another aspect of the media’s subtlety—they can affect us through multiple directions at once. Although this makes it likely that everyone will be affected by violent media in some way, it also makes it likely that the effects may not be identical for all people.

Myth 5. Causality means “necessary and sufficient.”

Determining if and when something “causes” something else is a problem that has plagued philosophers and scientists for centuries. In the social sciences, it is a surprisingly complex problem to solve. For many people, however, it has become oversimplified—something is a cause if it can be shown to be necessary and/or sufficient as a precursor. This position has been used to argue against the effects of media violence. Ferguson (2002), in a response to Bushman and Anderson’s (2001) meta-analyses of media violence and aggression, stated that: (a) because humans have always been violent, “violent media, then, are not a necessary precursor to violent behavior” (p. 446), and (b) because many people who are exposed to media violence never commit violent behavior, “violent media, then, are not sufficient to cause violent behavior” (p. 446). This argument seems, on its surface, to be reasonable. Yet this argument actually betrays a grossly oversimplified idea of causation. Consider, for example, a rock on the side of a hill. Assume that you give the rock a push and it begins rolling down the hill. Did you cause the rock to roll down the hill?

By the argument laid out above, you did not. Rocks have rolled down hills for centuries without someone coming along and pushing them. Therefore pushing it is not necessary. Furthermore, many rocks that are pushed do not roll down hills. Therefore, pushing it is not sufficient. Although pushing the rock was neither necessary nor sufficient to make it roll down the hill, that does not mean that it was not a cause of the rock’s beginning to roll. Most complex issues of interest (such as aggressive behavior) are multicausal. In the present example, many other issues interact to determine whether the push you gave to the rock caused it to roll down the hill: the force of gravity, the mass of the rock, the shape of the rock (round rocks require less of a push than square ones), the friction of the hill surface, the slope of the hill surface, the direction of the push, the force of the push, how deeply the rock is embedded into the ground, and so on—all interact to determine whether your push makes the rock begin to roll or not. Recognition that the issue is multicausal does not mean that your push is not one of the causes; in fact, it may have been a significant determiner or catalyst for the ultimate outcome, without which the other causes would not have been activated or sufficient. Aggressive behavior, too, is multicausal. Media violence is likely to be one of the pushes that interacts with other forces at work. In most situations, it is neither necessary nor sufficient. However, that does not mean that it is not a cause—it just means that it is one of the causes.

This conception of causality is similar to the idea of “proximate cause” in law, where the goal is to assign legal responsibility for an action. The proximate cause is the last action to set off a sequence of events that produces an injury. Yet, the goal of social science is not the same as that of law. Social science is concerned with all of the causes for some behavior, not only the necessary, sufficient, most recent, or largest causes. Because media violence has been shown to increase the likelihood of aggressive behavior, it can be a cause of aggressive behavior, even if it alone is not a necessary or sufficient cause.