Her latest study, "Violence in E-rated Video Games," was published in the August 1 issue of the Journal of the American Medical Association.

In the study, Thompson and co-author Kevin Haninger, doctoral student in the University-wide Program in Health Policy, reported that video games rated as "E," meaning suitable for everyone, are not necessarily violence free. In fact, Thompson and Haninger found that 64 percent of a sample of 55 E-rated games involved intentional violence. They reported that the average amount of violent play for these games was more than 30 percent, with the worst offender offering more than 90 percent violent game play.

"We didn’t expect to see as much violence as we did in some of these games," said Thompson, "particularly given their E rating."

She said the health effects of exposure to video-game violence are uncertain and need further study. Her paper is not a report on whether children become aggressive after playing violence-laden games. Instead, she said, the study is the first quantitative look at the content of an entire classification of video games.

"This study is focused on what’s in the games themselves so that parents, physicians, and policymakers can talk about the nature of the games," said Thompson.

In choosing their sample of games, Thompson and Haninger cut across genre lines and included games in categories such as action, sports, role-playing, racing, puzzle, casino, and shooting.

Thompson said she found the genres to be informative, identifying differences in violent content between categories. For example, she said, all of the "action" games had violence in them, whereas "sports" games had very little. That difference may be explained by how the authors defined violence, which they described as "acts in which the aggressor causes or attempts to cause physical injury or death to another character." Physical force normal in sports, such as tackling, was not considered violent.

Thompson and Haninger also investigated whether content descriptors found on the games’ boxes accurately represented the nature of the play. Of the 55 games they sampled, 23 games displayed a content descriptor for violence, and these games were more violent as a group than the 32 games that did not receive a content descriptor for violence. However, the absence of a content descriptor does not mean the games are "violence free." The researchers found that more than 40 percent of the games that did not display a violence descriptor did, in fact, involve violent game play.

Public concern over violence in video games first erupted in 1976 when Exidy games unveiled a game that awarded points to players who ran over stick figures, said Thompson. The company ended up withdrawing the game from the market.

Since then, video games and their accessories have become much more sophisticated, with sound effects, striking imagery, and, in some cases, controls that recoil to suggest the feel of a gun blast.

In 1994, the Entertainment Software Rating Board (ESRB) was established as a branch of the Interactive Digital Software Association to categorize video games and provide content descriptors, similar to the system used to rate movies.

Thompson said the information the ESRB provides is useful. She added she would like to see the ESRB standardize content descriptors and inform consumers about the duration of violence in a game, so that a game with little violence is not lumped into the same category as another with lots of aggression.

She also suggested that members of the ESRB actually play the games before they issue a rating. She explained that the ESRB currently gives ratings based on information and excerpts submitted by game manufacturers.

The video game study is part of a long-term effort by Thompson to analyze risks to children. She recently has focused on media directed towards kids. In June, she and doctoral student Fumie Yokota co-authored a study that found nearly half of all G-rated animated feature films showed alcohol and tobacco use as normal behavior and offered no warnings about the health consequences of using the substances.

Last year, she and Yokota found that violence in G-rated movies is common and has been growing over the years.

Her goal is to create a comprehensive catalog of kid’s risks for children, parents, physicians, and policy makers.

"Children are one of our greatest resources," said Thompson. "The risks they face are definitely an area that needs more research."

For more information about the Kids Risk Project and Thompson’s research, visit the web site at http://www.kidsrisk.harvard.edu

Related Story:

Study Finds Significant Amounts of Violence in Video Games Rated as Suitable for All Ages -- press release of July 31, 2001.