Study: Play of ‘Risk-Glorifying’ Video Games Can Exacerbate Deviant Behavior

Earlier this week, a new study began making the press rounds (we caught it at that linked the play of particular video games to teens' propensity for risky behavior.

And no, this one is not from Craig Anderson, Brad Bushman or Douglas Gentile!

No, this study comes from Dartmouth College who surveyed about 5000 US teenagers over the phone, four separate times over a period of four years.  During each call, interviewers queried participants on how often they played video games and if they played any of three games the study identified as "violent and risk-glorifying in content" – Spiderman II, Grand Theft Auto III, and Manhunt.

Participants were then asked how often they consumed alcohol and smoked cigarettes.  Later calls, asked 18-and-older participants about sexual activity and deviant behavior (such as getting into fights).

This sounds exactly like a study Dartmouth did a few years ago concerning video game play and dangerous driving habits.  Reading the old study, we found that the authors, procedure and makeup of the study participants was exactly the same.

We contacted lead author Professor Jay Hull for clarification:

"The previous [study] dealt with reckless driving," says Hull.  "This deals with alcohol (drinking frequency, quantity, binge drinking), cigarette smoking, aggression, risky sex, and delinquent behavior.  They are from the same multi-wave data set.  Although the driving paper came out separately, we wanted to package all the rest together to make the point that the deviant behavioral effects are very broad."

Both studies found that teens who played mature-rated, risk-glorifying video games were more likely to engage in risky behavior.

"Up to now, studies of video games have focused primarily on their effects on aggression and violent behaviors," said Dartmouth professor of pediatrics and co-author James Sargent.  "This study is important because it is the first to suggest that possible effects of violent video games go well beyond violence to apply to substance use, risky driving and risk-taking sexual behavior."

Hull added, "With respect to playing deviant video game characters, we feel it best to follow the admonition of Kurt Vonnegut in Mother Night: 'We are what we pretend to be, so we must be careful about what we pretend to be.'"

The study was published Monday in the Journal of Personality and Social Psychology.  If you'd like to read it, you'll find a PDF linked below (thanks to Jay Hull for hooking us up!).

Reporting from San Diego, GamePolitics Contributing Editor Andrew Eisen
Tweet about this on TwitterShare on FacebookShare on Google+Share on RedditEmail this to someone


  1. 0
    Sleaker says:

    And again Correlation != causation.  Good job guys you've simply shown a correlation between people that have deviant behaviour and the listed games.

    The graph looks deceptively laid out it doesn't use percentages. it presents a baseline as 0 and then uses -.6 to +.6. How can you have negative amounts/percentages of people? So many flaws in this just to spin a specific point instead of just laying out the facts of how it was done.

    Edit also a few annoyances with the site: The bottom pdf link is broken. and if I try to click the Image to get a better view of it it just opens this article over and over.

  2. 0
    Thomas Riordan says:

    Smoking cigarettes, drinking, having sex and getting into fights. You mean things that hormonally driven teenagers have been doing forever. Yep, must be those violent video games.

    We are what we pretend to be? Really? So Arnold Swartzenager is a robot from the future that kills people?

  3. 0
    ChrowX says:

    With measurements like "Sometimes" and "all the time" it's really hard to take this seriously. especially because the effect seems to be so minor that it may as well be part of the margin of error.

  4. 0
    prh99 says:

    Just being a teen is linked to risky behavior.

    Half a percent on self reported data, ok. Also it has the same problem as other studies that claim to link gaming to various negative personality traits, no causal link or accounting for other influences. As for aggression and video games:

  5. 0
    Avalongod says:

    I think it's a better study than a lot of the stuff out there…the trouble is it's oversold.  The effect sizes are really small, suggesting that video games predict, for most variables from less than half a percent increase in the behavior in question to maybe three and a half percent…that's small stuff and easily explained by the limitations of the study or other variables they didn't control.  I wish scholars didn't overadvertise their studies like this. 

Leave a Reply