If you’re a teenager, how do you know whether it’s cool to smoke cigarettes, curse or get a cartilage piercing? Look around: To find out what’s socially acceptable, impressionable adolescents generally turn to their peers. Now, new research finds that this social dynamic also plays out when it comes to more violent behaviors.

A new study, published yesterday in the American Journal of Public Health, draws on surveys of thousands of teens to reveal how the people around you influence your tendency to engage in violence. The authors report that adolescents are far more likely to commit a violent act if a friend has already done so—adding evidence to a mounting theory that violence in communities can spread like a disease.

The study was born of an unusual collaboration between Ohio State University social psychologist Brad Bushman and OSU political scientist Robert Bond. Bushman, who has written and lectured extensively on humans and violence, was interested in exploring the model of violence spreading like a contagious disease that had beens popularized by University of Chicago at Illinois epidemiologist Gary Slutkin. Bond had expertise in analyzing social networks. "We just really hit off and decided that we should to try find a way to merge our research interests," Bond says.

For the study, the two tracked the behavior of more than 90,000 American teenagers at 142 schools, who were surveyed in class starting in the mid-1990s as part of the National Longitudinal Study on Adolescent to Adult Health. By accessing follow-up interviews that were done with nearly 6,000 of the teenagers years later, the researchers were able to see whether they had practiced violent behavior in the past year—namely, getting into a serious fight, pulling a weapon on someone or hurting someone badly enough that they needed medical attention.

The teenagers were then asked to identify five male and five female friends, who were subsequently interviewed by the surveyors about their violent behavior. With this web of data, Bond and Bushman were able to piece together nodes of violence and their effect on the people connected to them.

What they found was a contagious model. Teenagers were 48 percent more likely to have been in a serious fight, 140 percent more likely to have pulled a weapon and 183 percent more likely to have hurt someone badly enough to require medical attention if they knew someone who had done the same. Moreover, the influence of one violent person can spread through up to 4 degrees of separation. In other words, if your friend's friend's friend's friend practices violent behavior, it's more likely you will too.

"People who exhibit these kinds of behaviors tend to be friends with one another," Bond says, adding: "They're teenagers. They're still sort of learning how to navigate their social environment."

For years, social scientists have theorized that violent behavior can spread from person to person like an illness, infecting whole neighborhoods and communities. This contagious theory was pioneered by Slutkin, who spent his early career working to prevent the spread of communicable diseases such as tuberculosis in San Francisco and Somalia, and AIDS in Uganda.

After returning to the U.S., Slutkin was troubled by the amount of violent crime he saw present in American culture. “I saw that these kids were killing each other,” he says. Soon, he started to see parallels between how violence was being viewed and treated by officials and how the AIDS epidemic was mismanaged and underfunded. “[Violence] is the only contagious epidemic that is not being managed by the health sector,” Slutkin says. “It's been fundamentally misdiagnosed.”

In 2000, Slutkin founded the movement Cure Violence to gain support for viewing violence as a contagious disease as opposed to solely a criminal justice issue. Cure Violence uses epidemiological techniques to target the people most at risk of spreading violence, and by working to stop its spread by “interrupting” violence before it starts. Slutkin has given a TED Talk on his approach, which was featured in the 2011 documentary The Interrupters. Cure Violence’s model, however, has faced resistance from law enforcement suspicious of treating violent criminals as victims, and Chicago dropped the program from its policing efforts in 2013.

Slutkin says that Bushman and Bond’s study adds to the now “thousands of studies that show the contagion of violence.” It also shows evidence that different forms of violence can be similarly contagious, from physical fights to violence using weapons, he says. This supports what he’s seen in his work. “We all unconsciously copy each other, especially with violence,” Slutkin says.

When it comes to other communicable diseases—say, a virus—the best way to avoid falling ill is to avoid the bug in the first place. Bushman thinks that this avoiding exposure is also the best for prevent violent behavior in teenagers. He also believes that the same contagious model could be used to spread non-violent behavior: By training teenagers to practice more empathy, schools and social workers could unleash positive behavior into social networks that would spread to people who don't receive treatment directly, he says.

Bond pointed to school-based violence prevention programs already in place across America to train students to practice peaceful conflict resolution, and said that their research could lead to better targeting of teenagers who would have the most social influence on their networks. "Those types of programs might be a lot more effective," Bond says, "because they're affecting not only who is directly affected by it, but the other people who see the changes in those people's behavior."

For future research, Bond is considering collecting his own data on how teenagers process and react to violence in some kind of a laboratory setting, while Bushman is interested in studying how violence could spread through other kinds of social networks, such as networks of terrorists on social media or in neighborhoods worldwide.

Slutkin, meanwhile, still hopes that people and governments will someday adopt his model of ending preventable violence. He draws parallels between his model and the new theory of our solar system proposed by astronomer Galileo Galilei, who faced opposition when his observations of the planets and moons didn’t fit with the prevailing theory of an Earth-centered solar system. “The theory was wrong,” Slutkin says. “It required a new theory.”