Syndicate content

Lab Psychiatry and Field Economics: does media violence cause crime?

Berk Ozler's picture

Today's post comes from guest blogger Arianna Legovini (World Bank).

Over the kitchen counter and with the sound of video games coming from the family room, my friend Marco was telling me how mirror neurons act out the violence and killings we see on movies. Mirror neurons allow humans to read and share emotions. They also mimic what we see, good and bad, and prepare us to act.

Marco Iacoboni pioneered the research on mirror neurons. He is Professor of Psychiatry and Biobehavioral Sciences and Director of the Transcranial Magnetic Stimulation Lab at the School of Medicine at UCLA and the author of Mirroring People: The New Science of How We Connect with Others. In chapter 8, Marco talks about the possibility that imitative violence is induced by media violence. The results from lab experiments with children are unequivocal: “exposure to media violence has a strong effect on imitative violence”, at least during the span of the experiment.

So, when I fell upon Gordon Dahl and Stefano Della Vigna’s Does Movie Violence Increase Violent Crime? I thought the results quite interesting. Exploiting variation in movie attendance, they estimate the causal link between violent movies and crime and find that “violent crime decreases on days with larger theater audiences for violent movies”. They explain the effect through substitution and self-selection: the people who are violent are more likely to choose violent movies. By doing so, they essentially remove themselves from the streets, and from too much drinking and trouble. What is their conclusion? Contrary to expectations, violent movies may deter as many as 1,000 assaults on an average weekend.

How to reconcile Marco and Stefano’s findings? Easily: violent movies create a taste for violence which is contained by attending more violent movies. But from the few hours of a lab experiment to the length of a weekend, the question still remains whether in the longer term we would be better off without violent movies.

A more recent study by Scott Cunningham, Benjamin Engelstätter and Michael R. Ward investigates whether regulation that reduces access to violent video games may benefit society. They argue that even though the lab experiments have established the causal link between violent games and violent behavior, the effect on society should examine “the time use effects of video games, which incapacitate violent activity by drawing individual gamers into extended gameplay”. In line with Dahl and Della Vigna, they use weekly variation in the volume of sales of violent video games to measure video games’ effect on violent crime incidents. They find that an increased volume of violent video game sales are associated with small overall decreases in crime rates over the next 5 weeks.

While the combined results reduce the case for regulation on access to violent movies and games, my own take is that we still don’t know the sign of the overall effect. The studies only capture the effects of movies and games given the current level of violent disposition in the population. If virtual violence caused some of that violent disposition in the population (and violent entertainment today keeps those people at bay), it does not tell us whether we would be better off protecting our children from it.

 

Comments

Submitted by Dave on
Stefano and Gordon offer their own reconciliation with the lab findings in the paper that you cite: Lab findings demonstrate that violent movies incite violence relative to non-violent movies. But the Stefano-Gordon paper finds that violent movies decrease violence relative to the outside option. Since violent people are selecting into violent movies, their alternative is not to attend Bambi, but rather to go get drunk and then beat someone up (apparently). I find that more convincing than the argument presented here - that violent movies are like a drug: they induce more desire which is then satisfied by higher drug consumption - because I would expect to see spill-over of the increased taste for violence outside of the movies (since that's what the laboratory results find, no?).

Submitted by Matt on
Berk/Arianna - your link to the Cunningham et, al study takes readers to Orbitz instead.

Submitted by Berk Ozler on
Matt, thanks for the heads up -- not sure how that happened...