Scientists gave kids real guns for an experiment. Now ethicists are weighing in.

The children watched snippets of a Nicolas Cage film before getting 20 minutes to play with toys. In the lab’s playroom, pairs of kids found Legos, Nerf guns and board games. Many also found a 0.38-caliber handgun with a sensor installed on its trigger, tucked away in a cabinet drawer.

Using a hidden camera, their parents watched them. The adults knew the gun was there — modified so it couldn’t go off. Their children did not.

“It wasn’t light like a plastic gun,” Kelly Dillon, one of the study’s authors and assistant professor of communications at Wittenberg University, said by phone. “It [was] very clear that this wasn’t a toy when they went to play with it.”

A research assistant sat in a separate greeting room, where kids could come by any time if they had questions. Almost 83% of the children found the gun; roughly 27% of them informed the assistant or handed it over. A small number of children aimed the weapon at their playmate or out the window and at people in the street. In one case, a child pulled the trigger 26 times — including while pointing the gun at another child’s temple.

About 1.7 million American children live in homes with unlocked firearms, and around 25% of U.S. TV programs depict gun violence, according to a 2007 study. Considering these circumstances, it’s important to understand the connection between media and real-life gun use — but how scientists should research it is up for debate.

Having kids watch violent movies after hiding a gun in their toys is one way to go about it. Some ethicists, however, have doubts.

Giving guns to children — for science

The study took place in fall and winter 2015. Researchers asked kids to watch 20 minutes of PG-rated movies with or without gun violence in them. They recruited 104 children ages 8 to 12, and then split them into randomly assigned pairs. They watched excerpts of Disney’s The Rocketeer or National Treasure. Some saw the clip as is — with guns — while others watching an edited version without guns. Then they got to play.

“Violence in the media increases aggressive behavior. It causes it. We’ve done many experimental studies, and there’s really no debate about that,” Brad Bushman, one of the study’s authors and a professor of communication and psychology at the Ohio State University, said by phone. “But we can’t give people knives and see what they do with them, so we have to look at correlational evidence between exposure to violence in the media and aggressive behavior.”

The study’s results suggest there is a connection: Roughly 42% of the play partners handled the gun, but those who had watched the violent movie spent a median of about 53 seconds holding it; kids who watched the nonviolent clip spent just 11 seconds doing so. The group of children who watched the film with guns also tended to pull the trigger more.

Why ethicists have some questions

First of all, it’s important to note that the scientists gave their experiment ample thought. Not only did an institutional review board approve their methods, but they took a number of safety measures: The gun was inspected by a police officer, and parents were told that they could stop the experiment at any time. Scientists also debriefed the kids after the fact, explaining why they planted the gun and facilitating discussions about gun safety. They also handed parents resources and followed up with a week later.

Even so, an ethicist said the potential harm might not have been worth it. “In any kinds of ethics evaluation, we have to balance the risk against the benefit,” said Celia Fisher, director of the Center for Ethics Education at Fordham University, in a phone call. “I think the study’s results are not of great scientific importance because we already know what the result is going to be. We have decades of scientific research showing that kids will imitate aggressive behavior they view on the TV screen.”

Allowing kids to play with a real gun could send a dangerous message.

“I don’t understand it. It doesn’t seem necessary to have had real guns in that scenario,” Fisher added. “For any child that lives in a home where there is a gun, this could be a very dangerous message. They could interpret that [playing with guns] was implicitly condoned by the adults running the study as well as their parents who gave permission.”

Some kids might have thought it was okay to use the gun because they were in a laboratory — “a place of power,” said John Oates, leader of the British Psychological Society’s Code of Human Research Ethics.

“I think it would have a small added effect of normalizing guns just being left around,” Oates said. “However, that could possibly be neutralized through the debriefing to some extent.”

If the study had been screened by a review board in the U.K., “we would be very unlikely to approve its protocol,” Oates said. “But that would be based partly on big cultural differences.” By that, Oates meant that guns appear to be more common in American households.

The ethicists’ concern is that kids are too young to really understand what’s happening. If they come across a gun in the future, for example, they might think they can safely handle it because they’ve already had the experience before. The median age of study participants was 9.9 years old; at that age, “it’s really difficult to know over the child’s next year the extent to which they would be influenced by this experiment,” Fisher explained. “As good as the debriefing was, it doesn’t necessarily outweigh the experience of that child having an adult implicitly condone them picking up a gun.”

Kids might start thinking they’re violent when they’re not

There’s a phenomenon called “inflicted insight,” which is when an experiment leaves a participant thinking they’ve found out something really bad about themselves.

The most classic example is the 1963 Milgram experiment, when participants were told to administer increasingly high-voltage electric shocks to another human being every time they incorrectly recalled a word from a list.

There’s a phenomenon called “inflicted insight,” which is when an experiment leaves a participant thinking they’ve found out something really bad about themselves.

“[The test subjects] left the study with insight into themselves that they otherwise might not have had — insight which might not even be true, but rather enforced by the conditions of the study,” Fisher said. “Telling a child they did something retrospectively that could’ve been harmful is inflicted insight that could make the child feel ashamed. It could affect their self-esteem.”

In the safety debriefing after the test, scientists did come across some regretful feelings from kids — but nothing they couldn’t smooth out, Dillon said.

“We were very careful. … Some of the kids were upset and started to cry and say that they didn’t realize what they did could have potentially hurt someone. But we talked through it,” Dillon said. “We told [parents] that this is not a commentary on them as a person, nor on you as a parent.”

The researchers considered the risks

It’s not that the researchers behind this paper didn’t consider these risks — they did. But exposing children to real weapons has the potential to be controversial in any circumstance, especially when kids don’t know whether the gun that’s being pointed in their direction is loaded or not.

“I would say that the researchers did the right thing in having their protocol reviewed by an appropriate institutional review board,” Oates said. “So I have to respect [that].”

At the end of the day, the information that tips the scale one way or another is missing: Were there any damaging effects on these kids? So far, there’s no evidence of that.

“That’s the tradeoff in any research: What are the potential costs and what are the potential gains?” Bushman said. “In this study for sure, the potential gains outweighed the costs, and parents agreed.”