2. INDIVIDUAL AGGRESSION

In his testimony to Congress, Toronto's Jonathan Freedman stated that a thorough review-in-progress of the new studies about the relationship between media and real-life violence has reinforced his conclusions of a decade earlier: "The research demonstrates either that media violence has no effect on aggression, or that if there is an effect, it is vanishingly small."

Laboratory experiments measure responses to contrived stimuli in controlled environments. From them, social scientists have gathered the strongest evidence that after witnessing an intentionally harmful act in a movie or on TV, a person is more likely to act harmfully. After watching a film of a teacher kicking a blow-up Bobo doll, children battered Bobo, too. Students who watched boxing films were more willing than those who didn't to administer shocks to an errant research assistant. In other studies, people who watched media with violent content responded to questions about hypothetical provocative situations, and, more than those in the control group, imagined themselves striking or punishing others. But the further you move from the hermetic atmosphere of the laboratory, the weaker the links between media and aggression become. In field experiments-where the stimulus is controlled, but the reaction is recorded in such natural settings as a school or hospital-the results have been less clear than in the lab. Lab and short-term field studies suffer from many of the same problems.

Experiments on the effects of adrenaline have found any activity that stimulates this fight or-flight hormone, whether watching an exciting TV show or riding a stationary bicycle, will increase just about any feeling or behavior the researcher tests for, whether it is generosity, punitiveness or anger." Criminologists Zimring and Hawkins suggest that when the child punches the Bobo doll, he could simply be exhibiting excitation, or "physical tension and the need to discharge it," with "no important link to the propensity to commit a serious assault on another human being." The catharsis of hurting the doll could even lessen the likelihood of taking out any frustration against another person."

"It is seldom acknowledged," wrote behavioral scientists Kenneth Gadow and Joyce Sprafkin in one review of the major field studies, "that television programs specifically produced to encourage pro-social behavior can also disinhibit aggressive behavior." They cite one study conducted in the late 1970s, in which the aggressiveness of a group of normally pacific preschoolers tripled after watching Sesame Street and Mr Rogers.

The much-quoted Grossman, leader of his own invented academic discipline of "killology," expresses opinions; he does not report social scientific findings. Perhaps the largest investigation ever of video-game play, and particularly of aggressive content in games, was a recently completed four-year study by the Australian government. Its conclusions contradict Grossman's claims. Watching children and teens in arcades and at computer screens, researchers witnessed "high levels of enjoyment," excitement, challenge, friendly competition, and much laughter and talking. "Verbal or physical aggression toward others was negligible," the report said, and what there was came softened by joking. "The main type of aggression was robust treatment of the equipment."" Australians play the same video games as Americans. Even if you looked to commercial video games for killing lessons, they wouldn't help you. "I don't see how anyone would learn to fire a weapon accurately from these games without some form of mentoring," said Colonel Ron Krisak, who conducted firearms training at Fort Dix .

In order to test one factor at a time in the lab, investigators screen only one class of shows or games-say, very violent or not at all violent-to each group of subjects within a short period of time. This makes sense from the point of view of experimental efficiency and purity, or

"elegance." But this is rarely the way media are used. In real life, a video gamer may desire the kill-or-be-killed thrill of Quake 11 for 20 minutes, then feel like rebuilding civilization with Civilization. He's also probably playing with other kids, joking, competing, commenting and resting. Similarly, a violent TV show is interrupted by commercials, channel surfing, chats with family members and trips to the kitchen. All these activities alter the messages, mood and effects of the media experience.

For obvious ethical reasons, these studies can measure nothing more than behavior toward inanimate objects or an unseen or hypothetical person. As a result, the subject can behave sadistically with no real-life inhibitions. Even preschoolers know that the Bobo doll, unlike little Jennifer or Jamal, feels no pain when they punch it. Equally important: Bobo doesn't punch back. Such studies may even subtly elicit meanness in their subjects. The child in the study not only knows she will escape punishment, she might even conclude that the adult kicking the doll or showing a violent film approves of bad behavior. Thus, she may imitate the behavior to please the experimenter. Psychologists call this a sponsor effect. Despite the fact that experiments measure aggression toward objects and imaginary people, not real people, researchers commonly infer that aggressive play with toys shows a tendency to be aggressive toward people. In 1995, for instance, Irwin and Gross had boys play video games and then play both with toys and with

other kids. They found that the boys who played violent video games moved about rowdily and treated the toys roughly, more than those who played nonviolent video games. But neither group bashed other children. Still, the researchers concluded that violent video games caused 11 aggression."" University of Utrecht communications scholar Jeffrey Goldstein pointed critically to this conclusion as typical of much work in the field. "What the researchers actually found," he said, "was an increase only in harmless aggression against objects, most likely the result of increased excitement generated by the aggressive video game."

One of the main theories undergirding the research in this field, as well as most common sense thinking, is "social learning" the idea that a child who sees a Halloween movie or plays Quake will adopt the attitudes and imitate the behaviors portrayed on the screen. At least since Albert Bandura's famous Bobo doll experiments in the 1960s, an oversimplified interpretation of social learning theory has trickled down through the ranks of research psychologists to the news-hour talking heads and to frightened parents: Monkey see, monkey do. A big piece of social learning theory is left out of this interpretation: the larger world of relationships and meanings in which the child views a show, associates the images in it with things he knows and feels, and behaves when the picture is turned off. This is the context. The first part of the context of media violence is what happens inside the story and how the story is told. Most studies of the "incidence" of violence in the media are nothing more than a tally of scenes of force wielded with the intent to hurt. Such "neutral" bullet-counting implies that the effect of seeing any scene of force-from a Roadrunner cartoon to a Terminator film-is to inspire enthusiastic approval or blase dismissal of violence. The point is not that such portrayals have no emotional or intellectual impact. Rather, the meaning of violence depends largely on the context-whether the violence is rewarded or punished, banal or calamitous, humorous or serious. And while reactions to a given scene vary from person to person, the context affects every viewer and determines whether she comes away seared, angry, amused, excited or altogether unaffected.

The other part of context is the human environment in which a child consumes media. In a letter to the British journal The Psychologist, psychologist Anne Sheppard suggested that aggressive behavior regularly elicited in the lab might be hard to create in everyday family life. "Unlike the experimenter, some parents have strategies for coping with their children's behavior after viewing violent TV, such as Power Rangers," wrote Sheppard, who carried out five years of research on the effects of TV on children. "They alter the antecedent and/or reinforcement conditions, so that unacceptable behavior is either not displayed or is not encouraged." In other words, parents talk to their children about whether attacking your little sister with an AK-47 (or even kicking her in the shins) is the way to resolve a dispute about who gets to ride the new bike. And if the child does kick his sister, the parents chastise him. "Social learning theory also emphasizes the importance of cognition on behavior," Sheppard continued. "It is the meaning that children construct from what they see on TV that will determine how they react once away from the screen."

Every critical report of violence in the media trots out terrifying numbers about how many thousands of simulated acts of murders and mayhem a child witnesses during his formative years. These statistics imply a scary chain of events: each bloody scene etches a lesson in the child's brain. Impression is laid upon impression, so that eventually any values of peace and compromise are crowded out by the maxim that might makes right. The next time the child witnesses a bully pushing around a smaller kid, he won't intervene. If someone challenges him, he'll put up his dukes. Over time, the fear is, the media will desensitize him to belligerence by others and disinhibit him from resorting to it himself.

Some kids spend great amounts of time in front of various screens. It is understandable that parents worry that over time, this experience could turn their children into surly or hurtful people. However, science does not support this fear.

One of the most ambitious and frequently cited longitudinal studies assessed groups of boys and girls ages 9 to 11 from the U.S., Finland, Poland, Australia, Israel and the Netherlands over six years in the 1970s and early '80s. This work, steered by research psychologists Leonard Eron and L. Rowell Huesmann, yielded much useful information about the relationships among such factors as parental punishment, socioeconomic status, intelligence, television viewing and aggression. But contrary to the American authors' claims, the study did not provide convincing evidence that watching more violent TV contributed to children's antisocial behavior over time and in different countries. In the U.S., the correlations showed a small increase. In Finland, the correlations for boys increased, then decreased, then increased again; for girls, they decreased, increased, then declined again. In Poland, the graph was similarly bumpy. Commented University of Toronto psychologist Jonathan Freedman, "There is no discernible pattern in the changes.

The Dutch researchers in the study above strongly dissented from the claims of Eron and Huesmann. When they took away the effects of low intelligence and the propensity to aggression that some kids displayed at the debut of the study, they found, "the relationship [between TV and aggression] disappeared almost completely."" Another study considered state-of-the-art in design and method" came to conclusions that were almost exactly the opposite of Huesmann's and Eron's. Psychologists J. Ronald Milavsky and Horst Stipp assessed more than 3,000 Midwestern students over three years. "Measures of violence exposure were conceptualized in eight different ways" (realistic shows versus cartoons, high levels of violence versus low levels, etc.). "Effects were sought among every different theoretically plausible subset of the sample, such as children who had a history of prior aggressive behavior, children without fathers, poor children, children who lived in families and with peer groups in witch aggressive behavior was normative and children whose parents disciplined them with physical punishment." But all that manipulation yielded "only tiny, statistically insignificant" numbers indicating any relationship between exposure to TV violence and antisocial behavior. "Television viewing was not a factor in the development of aggressive behavior among the children in the sample," the authors concluded.'