What if Horror Films Actually Stop Crime, Not Cause It?

The annual ritual of Hollywood horror films each fall prompts another tradition: a debate over whether violent scenes onscreen inspire real-life violence offscreen. The debate is usually between those who consider such films dangerous – and indeed, survey studies by psychologists suggest copycat violence is possible — and those who hold they’re “harmless.”

But what if, when it comes to preventing real-life violence, horror films are actually helpful?

That’s what economists Gordon Dahl and Stefano DellaVigna found when they analyzed the impact of different blockbuster movies released in the United States over a decade. According to their analysis, for every million people who view a violent film on a given day, violent crime decreases across the nation by 1.2 percent.

To put this in perspective, this year’s Halloween blockbuster “Happy Death Day” pulled almost 4 million people during its opening weekend. Put in other terms, the researchers estimate that, on a weekend when an average number of viewers go see violent movies, the films deter nearly 1,000 assaults.

How is this possible?

According to Dahl and Della Vigna, people who might otherwise commit crimes are drawn into movie theaters when a violent film is released and so aren’t available to commit assaults. In addition, the economists found violent film attendance led to particularly large decreases in assaults involving alcohol and drugs and had a larger deterrent effect for potential offenders just above the legal drinking age. This suggests that violent films prevent crime in part by reducing potential criminals’ alcohol consumption (since few movie theaters allow alcohol on their premises). Importantly, though studying the long-term effects of violent films on crime wasn’t possible in this context, the researchers found “no evidence of medium-run effects up to three weeks after initial exposure” to violent films.

These findings should ease fears about the impact of horror film releases on criminal behavior on Halloween (and every other day of the year), but they also offer a key insight about fighting crime. They suggest that creating attractive diversions for prospective criminals (like opportunities to watch violent films in theaters) can reduce violence.

A recent small study conducted in Cape Town, South Africa, by the nonprofit behavioral science consultancy ideas42 relied on this very same insight — that attractive diversions can be used to fight crime — to tremendous effect. In the study, 156 low-income, at-risk youth in Cape Town were randomly assigned to either a control group that went about their lives as usual or an intervention group that interacted with a computer program designed to help them find and plan fun, safe weekend activities. The program presented users with a series of different suggested weekend pastimes — like starting a pickup soccer game — until a suggestion was accepted. Then it helped a user plan where, when and with whom they would enjoy the suggested activity.

Survey data showed that youths randomly assigned to the intervention group were half as likely to participate in unsafe activities over the following weekend and half as likely to experience violence relative to the control group. The success of this small pilot program suggests huge opportunities for fighting crime by finding and suggesting appealing activities to prospective criminals that can “substitute” for getting into trouble.

A more general lesson from these studies is that an understanding of the psychology of violence — for instance, that it often results from unoccupied time on Friday and Saturday nights — can suggest useful tactics for reducing it. This can add to the arsenal of traditional approaches used for fighting crime like better or more policing and stiffer penalties.

A fascinating paper published earlier this year by a team of economists and psychologists offers yet another example of how understanding the psychology of crime can be useful to crime prevention efforts. This paper examines a program called Becoming a Man (BAM) — designed by the nonprofit group Youth Guidance — that was rolled out to a random sample of 4,804 students in very disadvantaged neighborhoods on the south and west sides of Chicago in 2009-2010 and 2013-2015.

BAM reduced total arrests among those randomly assigned to participate in the program by 28 percent to 35 percent and violent crime arrests by 45 percent to 50 percent. What’s amazing about BAM is not just its effectiveness but the secret sauce that led to its success.

Like many programs designed for at-risk youth, BAM involved regular interactions with an adult mentor as well as in-school and after-school programming. However, an unusual aspect of BAM was the inclusion of curriculum content that taught participants about how to avoid mistakes that people commonly make when they think too quickly and act too impulsively, or as the researchers studying BAM put it, content “designed to reduce common judgment and decision-making problems related to automatic behavior and biased beliefs.”

An evaluation of why BAM worked so well suggests it didn’t change students’ access to adult supporters, understanding of the returns to education, conscientiousness, social skills or emotional intelligence — instead, a key ingredient was that it helped teenagers (an age-group that tends to be particularly impulsive) learn to “slow down in high-stakes settings and re-examine their automatic assumptions.” The researchers evaluating BAM developed a simulated provocation and looked at how BAM participants responded relative to students who hadn’t gone through the program.

All students were given $10 and told another student “partner” had also been given $10. Their partner could take away some of the student’s allocated $10. Via walkie talkie communication, each student learned their partner had decided to take $6 from them. (In reality there was no partner — only a researcher posing as one.) Students then had the opportunity to retaliate by taking money from their partner, but they knew this would be a multi-round interaction so retaliation could lead to a vicious cycle. The BAM students took 80 percent longer than others to respond to the provocation of having $6 taken from them, suggesting they had learned to slow down and think rather than merely reacting on first impulse. And reacting on impulse is a well-known cause of suboptimal decisions (like decisions to commit crimes). The team evaluating BAM found that in areas where BAM was most effective at crime prevention, this slowing down effect was largest, and they estimated that reduced automaticity accounted for “about a third of the total effect of BAM” on arrests.

Together, these findings from behavioral science are cause for celebration. They suggest new, highly effective ways to reduce crime and improve the lives of potential victims and criminals alike. During Halloween season, as you shudder in fear at the sight of spooky costumes or scenes from scary movies, remember that real horrors can be reduced not only with improved approaches to law enforcement — but also with a little insight into the psychology of crime.

Katherine Milkman is a behavioral economist and associate professor at The Wharton School, where she studies decision-making and how to improve it. She has a secondary appointment at the Perelman School of Medicine, and is co-director of Wharton People Analytics. Follow her on Twitter @katy_milkman.

This article originally appeared in The Washington Post.